Prisoners' Coping Skills and Involvement in Serious Prison Misconduct and Violence

PRISONERS' COPING SKILLS AND INVOLVEMENT IN SERIOUS PRISON MISCONDUCT AND VIOLENCE

A Dissertation Presented by

Ann Marie Kelley Rocheleau

to The College of Criminal Justice

In partial fulfillment of the requirements for the degree of Doctor of Philosophy

in the field of

Criminology and Justice Policy

Northeastern University Boston, Massachusetts April, 2011

PRISONERS' COPING SKILLS AND INVOLVEMENT IN SERIOUS PRISON MISCONDUCT AND VIOLENCE

Ann Marie Kelley Rocheleau

ABSTRACT OF DISSERTATION

Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Criminology and Justice Policy in the College of Criminal Justice of Northeastern University, April, 2011

Abstract

Prison misconduct and specifically prison violence generate serious problems in prisons across the United States. Most of the research on prison misconduct has focused on static rather than dynamic characteristics of prisoners. This study examined a dynamic personal attribute, prisoners‘ ability to cope. The research tested whether prisoners‘ ways of coping affect their involvement in serious prison misconduct and violence. The study also examined the traditional predictors of serious prison misconduct and violence and their relationship to ways of coping.

The research used a mixed methods design and a concurrent data collection strategy. The quantitative data collection, which was the dominant approach, included self-report surveys administered to a random sample of 312 prisoners stratified by involvement in serious misconduct and residing in the medium and maximum security facilities, and extraction from the prison system‘s database. The qualitative research included in-depth interviews with staff and prisoners, and observation of classification and disciplinary hearings.

The study yielded a number of findings about serious prison misconduct and violence.

First, five out of the eight ways of coping in the study were directly related either to violence alone or serious misconduct and violence. Prisoners who learned to elicit both emotional and instrumental support from loved ones, fellow pro-social prisoners, and staff were less likely to be disruptive. Those who coped through venting their emotions, and bravado, and who charged into reacting to stressors were more likely to be disruptive and to misbehave. Second, trait emotions did affect misconduct. That is, when the personal predictors and coping were taken into consideration, prisoners who were angry were less likely to be involved in misconduct and violence, while those who were anxious were more likely to be involved in serious misconduct.

Third, the study yielded at least five categories of prisoners that were more apt to be involved in serious misconduct and violence: prisoners with mental health problems, young prisoners, weak prisoners, gang members, and those prisoners, usually more well-behaved, who fought back when victimized by others. The study also found that policies, practices, and the level of staff skills affected serious prison misconduct and violence. Finally, it was concluded that a reconceptualization of the predictors of serious prison misconduct and violence into static and dynamic predictors would be useful to focus future research and policy recommendations.

ACKNOWLEDGEMENTS

There are many people without whom this dissertation would not have been completed. I would first like to thank Rhode Island Department of Corrections Director Ashbel T. Wall, II and

Jake Gadsden, former Assistant Director of Institutions & Operations, for valuing research despite the risk and allowing me to conduct my study in the Rhode Island Department of

Corrections. Thanks also to Jeff Renzi, Associate Director of Planning & Research, and his staff for all of their help in the application and data collection process. Special thanks to Greg

McCarthy who spent innumerable hours extracting data, explaining details about the INFACTS system and RIDOC policies, and ensuring I connected with the appropriate administrators within the facilities. I am also grateful to Donna Collins, Deputy Warden, who helped me navigate and understand the RIDOC system and to all the facility administrators who provided access and facilitated my data collection. However I could not have completed this study without the participation of over two dozen staff and over 300 prisoners who were willing to bare their souls and share their ideas, their experiences, and their thoughts amid the general peer pressure not to get involved in such endeavors.

I would like to give a heartfelt thanks to Donna Bishop who started out as my advisor at

Northeastern but played multiple roles including teacher, dissertation chair, and mentor. Thanks for your wise advice and direction and your kind words and support during those particularly challenging times. Thanks to Carlos Cuevas and Natasha Frost for serving on my committee, providing statistical advice, and commenting on my rather lengthy drafts. Special thanks to my outside committee member and mentor Robert Johnson from American University. Your work on prisons, especially your concept of mature coping, was the inspiration for the focus of this research. I know my dissertation was much improved as a result of my committee‘s feedback

5 and suggestions. Thanks also to Jack McDevitt for his unending support and guidance from the application stage through the seven years in this doctoral program. I am also grateful to all of the

College of Criminal Justice faculty who contributed to my learning, either formally through class and research projects or informally in discussions.

Of course none of this would have been possible without the constant support and encouragement from my friends and family. The experience would have been quite lonely, less fun, and even more challenging without the backing, friendship, and collaboration among the doctoral students from the program‘s first cohort (oh and yes some of you from the second and third cohorts—you know who you are). Thanks for allowing an ―old lady‖ to join you in both your studies and your fun. Thank you also to Laurie Mastone for your friendship and assistance throughout these years in the program. I would also like to thank Susan Guarino for literally decades of friendship and immeasurable guidance throughout the years at Boston College, in state correctional agencies, and now at Stonehill College. You were literally my cheerleader during this dissertation process. I am extremely grateful for the unending support and love of my family, including my parents Bob and Velma, sister Sandra, brother Bobby, and sister-in-law

Trisha. However, the biggest thanks must go to my husband, Michael Kelley, and my children

Caitlin and Paul. Thanks for all of your understanding and patience, especially whenever I was in my writing mode (which was often); this frequently led to taking over whole rooms with wires, papers, and books, take-out and Trader Joe‘s dinners, and long spans of times away from all three of you. Mike, thanks for all the wonderful fires you made for me to study by; Caitlin and Paul, somehow you both managed to grow up to be wonderful adults of whom I am most proud.

TABLE OF CONTENTS

Abstract of Dissertation ...... 2

Acknowledgements ...... 5

Table of Contents ...... 7

Introduction ...... 9

I. Conceptual Model and Hypotheses ...... 11

II. Literature Review ...... 17

III. Research Setting ...... 50

IV. Methodology, Data, and Analytic Strategy ...... 63

V. Disciplinary Reports, Disciplinary Process, and Sanctions ...... 109

VI. Individual Predictors of Serious Misconduct and Violence ...... 132

VII. Multivariate Analysis of Prison Misconduct ...... 191

VIII. Staff and Prisoner Feedback on Prison Misconduct ...... 230

IX. Discussion ...... 282

X. Future Research and Policy Recommendations ...... 343

Appendix A – RIDOC Disciplinary Severity Scale ...... 365

Appendix B – Prisoner Self-Report Survey ...... 371

Appendix C – Prisoner Interview Schedule ...... 383

Appendix D – Staff Interview Schedule ...... 387

References ...... 389

INTRODUCTION

Prison misconduct and specifically prison violence generate serious problems in county, state, and federal prisons in the United States. While the collective prison unrest of the 1960s to

1980s has subsided (Colvin, 1992), individual-level and gang-related prison violence have increased (Gaes, Wallace, Gilman, Klein-Saffran, & Suppa, 2002; Edgar, O‘Donnell, & Martin,

2003; McCorkle, 1992). Prison violence poses a threat both to staff and prisoners alike, and generates high monetary and political costs for prisons and high personal costs for the people involved (Sykes, 1958; Johnson & Toch, 1988; Johnson, 2002; Hochstetler, Murphy, & Simons,

2004). Prison violence not only affects prisoners and staff members who have been victims of violence, but prisoners and staff who witness it. Some prisoners withdraw and attempt to ―do their own time,‖ whereas others become violent or threaten violence in anticipation of personal victimization (McCorkle, 1992; O‘Donnell, & Edgar, 1998). Finally, when prisoners are caught and found guilty of involvement in prison violence or serious prison misconduct, they are given disciplinary sanctions. These may include loss of visits or other privileges (including good time), being locked in one‘s own cell, reclassification to a higher security facility, removal to a short-term segregation unit, or sentencing to a long-term ‗supermax‘ type of prison or segregation unit. These sanctions further impede the potential for positive change in prison resulting in more time in prison, less contact with family, and less opportunity for rehabilitation

(Poole & Regoli, 1980; Lovell & Jemelka, 1996; Riveland, 1999). They also add significantly to the costs of prison operations—more prisoners in more costly higher security facilities, costs of construction and operation of ‗supermax‘ facilities, and greater numbers of personnel to staff disciplinary hearings and segregation units (Lovell & Jemelka, 1996; Riveland, 1999).

Given the grave consequences and costs of serious prison misconduct and violence, it is essential that research on this problem shed light both on its underlying causes and potential solutions. Most of the research on prison misconduct and violence has focused on identifying correlates of prison misconduct (Zamble & Porporino, 1988). Researchers have found an association between prison misconduct and the individual-level factors of age, race, marital status, level of education, pre-prison employment, mental health status, prior criminal history, prior history of prison misconduct, and gang affiliation. Most studies have focused on static rather than dynamic characteristics of prisoners. Static characteristics are those that cannot be altered such as race, prior criminal history, or prior street gang affiliation. Dynamic characteristics are those that can be changed such as educational level, poor communication skills, and current prison gang affiliation. Several researchers have suggested incorporating dynamic inmate variables into prison misconduct research (Harer & Langan, 2001; Camp, Gaes,

Langan, & Saylor, 2003).

The current research examined a dynamic personal attribute, prisoners‘ ability to cope.

The research tested whether prisoners‘ possession of positive coping skills affects their involvement in serious prison misconduct and violence. For this study, serious prison misconduct and violence included such offenses as escape, possession of a weapon, extortion, riot, threatening to inflict harm on staff or other prisoners, assault on staff or other prisoners, and fighting. In addition, the research explored the types of stresses and hardships with which prisoners must cope. This is useful for two purposes: 1) to provide prison administrators with information which they might use to minimize or address some prison hardships; and 2) to focus future coping skills training programs on those prison stresses and hardships that prisoners must regularly address.

I. CONCEPTUAL MODEL AND HYPOTHESES

The central thesis of this research is that prisoners who are involved in serious prison misconduct and violence are less likely to possess and use positive coping skills compared to other prisoners. This thesis implicitly acknowledges the existence of various prison-related stresses and hardships that most prisoners must face. However it also assumes that prisoners‘ capacity to cope, a dynamic personal attribute, is an important factor.

The current study drew on the social-psychological theory of coping that was developed by Lazarus and his colleagues (Lazarus, 1966, 1981; Lazarus, Averill, & Opton, 1974; Lazarus,

Coyne, & Folkman, 1982; Lazarus & Folkman, 1984; Folkman & Lazarus, 1985). Lazarus and

Folkman define coping as ―constantly changing cognitive and behavioral efforts to manage specific external and/or internal demands that are appraised as taxing or exceeding the resources of the person‖ (Lazarus & Folkman, 1984, p. 141). The coping literature generally recognizes that two separate processes occur when one copes and that each of these processes are regulated by different conceptual systems within one‘s personality (Lazarus & Folkman, 1984; Folkman &

Lazarus, 1985; Epstein, 1990; Carver, Scheier, & Weintraub, 1989; D‘Zurilla & Chang, 1995).

These processes include the primary appraisal of a stressful situation involving an emotion- focused process (problem orientation) and the secondary appraisal involving a problem-focused process (problem-solving proper). While researchers have acknowledged that these two processes are linked, they are considered to be separate processes that affect a person‘s coping, be it positive or negative. Carver, Scheier, & Weintraub (1989) have taken these two separate processes and created subscales that measure specific aspects of both emotion-focused and problem-focused coping. Aspects of emotion-focused coping include positive reinterpretation and growth, denial, acceptance, and use of social support for emotional reasons. Aspects of

11 problem-focused coping skills include planning, suppression of competing activities, restraint coping (holding off on other activities while focusing on the problem at hand), use of instrumental social support, focusing on and venting emotions, behavioral disengagement, and mental disengagement.

Given this social-psychological theory of coping, its emotion-focused and problem- focused processes, and the various aspects of coping, the conceptual model suggests that prisoners who do not possess positive coping skills will use negative ways of coping that are more likely to result in behavior that includes prison misconduct than those other prisoners who do possess positive coping skills. The focal research question was whether the possession of positive coping skills affects a prisoner‘s likelihood of being involved in serious prison misconduct and violence. In addition the study explored which aspects of coping were more likely to affect a prisoner‘s likelihood of being involved in serious prison misconduct and violence.

Coping theory also acknowledges the role of one‘s general emotional state on the outcomes of coping (Carver, Scheier, & Weintraub, 1989; Soderstrom, Castellano, & Figaro,

2001). As mentioned previously, the primary appraisal stage of coping involves an emotion- focused response that may include anger, depression, and anxiety. Thus another underlying assumption of the proposed research is that negative trait emotions, and in particular anger, anxiety, and depression, can affect one‘s ultimate behavior in a stressful situation. It is assumed that prisoners who easily become angry, depressed, and anxious may be more likely to engage in behaviors that are negative such as serious prison misconduct and violence. Researchers who have studied the coping skills of prisoners and other offenders have also studied their trait emotions (Soderstrom, et al., 2001; McDonald, 2006).

State emotions vary in individuals across time and depend on the context of a situation or a specific event. State emotions are those that a person is experiencing at the moment in response to a situation or event. On the other hand, trait emotions are a part of an individual‘s general make-up or personality. Trait emotions denote how often a person experiences an emotion—hardly ever angry or always angry. While state emotions are generally the result of a specific event (e.g. sad and depressed that a friend has died), trait emotions are those that a person might experience on a daily basis (e.g. always depressed). Therefore the study sought to determine whether the general traits of anger, depression, and anxiety were associated with the likelihood of engaging in serious prison misconduct and violence. It explored which of these three emotions, if any, was associated with a greater level of serious prison misconduct and violence. The study also explored the relationship between coping skills and these negative emotions. For example, does the introduction of ways of coping into the equation mediate or moderate the effects of trait emotions on serious prison misconduct? Conversely, do trait emotions influence the relationship between ways of coping and serious prison misconduct and violence?

In addition, the conceptual model for the research acknowledged that there are other factors that affect serious prison misconduct and violence. These factors include those that are measured at the individual level, some of which have been found in previous research to have a significant effect on prison misconduct and others that have not been previously studied. They include both prisoner characteristics (e.g. age, criminal history, prior history of misconduct, race, prior substance abuse history, commitment to convention factors, and gang involvement) as well as other individual prison-related experiences (e.g. length of time already incarcerated, idleness,

13 involvement in programming, and psychiatric treatment in prison).1 The current study controlled these other factors that might affect serious prison misconduct and violence in the study.

However, the relationship was further investigated to determine what individual-level factors still affect a prisoner‘s involvement in serious prison misconduct and violence once one‘s coping skills and associated emotions were taken into account. For example, age has been found to have an inverse relationship to prison misconduct. However, age may be serving as a proxy for the natural maturation process which might result in the development of coping skills. Therefore the research tested whether one‘s coping skills and associated emotions interact with or mediate any of those other individual factors.

Finally the study delineated the types of prison hardships that prisoners experience. It explored what types of stresses and hardships were most likely to affect the likelihood of involvement in serious misconduct and violence.

The conceptual model for the quantitative analysis is below (Diagram 1). Four hypotheses were tested that are directly related to the theory of why prisoners are involved in serious prison misconduct and violence.

Hypothesis 1: The greater the possession and use of positive coping skills by

prisoners, the lesser will be their involvement in serious prison misconduct and violence.

Hypothesis 2: The greater the possession and use of negative ways of coping by

prisoners, the greater will be their involvement in serious prison misconduct and

violence.

1 Institutional-level prison factors such as staffing levels, overcrowding, or racial make- up of staff were not included in this study because it was conducted in only one state and with only four facilities.

Hypothesis 3: Prisoners with stronger levels of negative trait emotions will be

involved in greater levels of serious misconduct.

Hypothesis 4: That the effect of individual level factors (e.g. age, criminal history,

race) on serious prison misconduct and violence will be at least partially mediated by

prisoners‘ possession and use of positive and negative ways of coping.

Diagram 1 - Lack of positive coping skills, presence of negative coping, & the presence of negative emotions result in greater levels of serious prison misconduct and violence

Negative trait emotions (anger, anxiety, depression)

Prison hardships

Lack of positive coping skills; presence of negative coping

Serious prison misconduct and violence

Other individual-level factors including prisoner characteristics (e.g. race, age, education, prior drug use) and prison- related experiences (e.g. # visits, security level, idleness, # enemies)

The conceptual model necessitated the use of a quantitative research design to measure the various concepts listed above and allow one to test the stated hypotheses. The research incorporated a smaller qualitative component in order to gain descriptive and more in-depth knowledge about the actual process of coping to confirm and interpret some of the quantitative findings, and provide contextual information. The qualitative research focused on those

15 prisoners who have been involved in serious prison misconduct and violence and the staff who work with them. It is hoped this attempt at triangulation has resulted in a richer, more in-depth analysis that in turn will inform policy recommendations for decreasing serious prison misconduct and violence in prison. The qualitative research questions included:

 What reasons do prisoners give for being involved in serious prison misconduct and violence?

 What are the prison hardships that they say are most associated with getting involved in serious prison misconduct and violence?

 What do prisoners who are involved in serious prison misconduct and violence think of their abilities to cope?

 What solutions do prisoners who are involved in serious prison misconduct and violence give for reducing misconduct and violence?

 What reasons do staff members who work with this population of prisoners give for inmate involvement in serious prison misconduct and violence?

 What solutions do staff members who work with this population of prisoners give for reducing misconduct and violence?

 Do staff members perceive that prisoners can be assisted to develop more positive coping strategies?

II. LITERATURE REVIEW

This chapter reviews two bodies of literature: prison misconduct and coping. The section on prison misconduct includes a brief history of the research on prison misconduct and violence and a summary of those individual characteristics that over time have been identified as significant predictors of prison misconduct. The second section covers several topics: the general prison adjustment literature, prison stress and the social-psychological concept of coping, the empirical research on coping in prison, research on coping and misconduct, and research on the relationship between coping and negative emotions.

RESEARCH ON PRISON MISCONDUCT

BRIEF HISTORY OF RESEARCH IN PRISON MISCONDUCT Early studies on prison misconduct focused on the characteristics of prisoners who received disciplinary reports (dreports) for misconduct, self-reported misconduct, or who were nominated by staff as presenting discipline problems (Flanagan, 1983). These studies examined a number of variables including age, offense, prior record, length of sentence, time served, education and/or intelligence, marital status, prior employment, race, visits received, and psychosocial indicators (Schnur, 1949; Zink, 1958; Wolfgang, 1961; Johnson, 1966; Ellis, Grasmick, &

Gilman, 1974; Flanagan, 1980 & 1983; Louscher, Hosford, & Moss, 1983). Because a record of disciplinary infractions was one indication of maladjustment in prison (others include withdrawal into one‘s cell, suicide and other forms of self-mutilation, depression, and anxiety) and was used by prison administrators to make many kinds of decisions (movement to lower security, work assignment, parole to name a few), these researchers were interested in differentiating between those prisoners involved in prison misconduct and those who were not.

However, in the late 1970s and early 1980s, researchers began questioning the narrow focus on individual-level factors in prison misconduct research and also questioned the validity of official misconduct as a measure of inmate behavior. Poole and Regoli (1980) questioned the discretionary enforcement of prison rules by correctional staff, and argued that official discipline could reflect staff bias against young offenders, minorities, drug abusers, or others. Around the same time, following an explosive increase in prison populations around the country, other criminologists began studying the effects of overcrowding on prison infraction rates (Nacci,

Teitlebaum, & Prather, 1977; Walkey & Gilmour, 1981; Eckland-Olson, Barrick, & Cohen,

1983; Gaes, 1985; Gaes & McGuire, 1985; Smith, 1988; Anson & Hancock, 1992). The research on overcrowding was mixed. Some studies found that it contributed to prison misconduct and violence; others found its effect on prison misconduct was dependent on such variables as age and type of housing; and others found no association between overcrowding and prison misconduct and violence. Researchers also studied other prison-level characteristics and found several strong correlates of misconduct and violence including composition of inmate population, composition of staff, and prison security level (Sieverdes & Bartollas, 1986;

Wooldredge, Griffin, & Pratt, 2001; Camp, et al., 2003). Most recently, there have been studies of prisons that examined situational factors affecting prison misconduct and violence (Gaes, et al., 2002; Wortley, 2002).

All of this research was conducted amid a larger debate about inmate behavior in general and many researchers adopted either the ―importation‖ or ―deprivation‖ model as a framework within which to understand prison misconduct. Essentially do offenders import their tendencies towards violence and misconduct when they enter prison or do the harsh conditions of confinement result in the use of violence to stay safe in the prison? The deprivation model was

18 not applied to empirical research of prison misconduct until the late 1970s when researchers began studying overcrowding and using the model to begin focusing on prison-level characteristics. Although it was not discussed as the ―importation‖ model until 1970 (Irwin,

1970), early prison research on prison misconduct tested the model by examining the characteristics of prisoners to determine why some got into trouble and others did not.

By the mid-1980s, penologists began integrating the two types of variables in their research of prison misconduct to determine which were more important. In their study of assault, Gaes and McGuire (1985) underscored the importance of examining both individual- and prison-level factors and they believed that over time the most important of these predictors would emerge without the ―trappings‖ of the importation vs. deprivation models. Eventually the integration of individual- and prison-level factors along with situational factors has become the norm, although some researchers still focus on the outdated models of importation vs. deprivation. Although it is clear that factors measured at the prison level and situational factors are important to understanding prison misconduct, the hypotheses of this research focused on individual prisoners‘ level of violent and serious misconduct within one prison system. While all of the factors were able to be measured at the individual level, some were characteristics of the prisoners (e.g. race, age) while others were experiences related to their incarceration (e.g. time incarcerated, prior history of misconduct). Therefore, an examination of the literature of those individual level factors most predictive of prison misconduct will be useful for two purposes.

First it helps determine what factors should be included as control variables in the research.

Second, it might shed light on whether prisoners‘ possession and use of positive and negative coping skills, another set of individual-level factors might mediate or interact with the effects of any of these individual factors.

INDIVIDUAL FACTORS ASSOCIATED WITH PRISON MISCONDUCT The four strongest individual level factors associated with prison misconduct are age, prior criminal history, prior history of prison misconduct, and evidence of mental health problems. Other factors less strongly associated with prison misconduct are race and ethnicity, marriage, prior employment, education level, seriousness of current offense, time served in prison, prior substance abuse, and gang affiliation.

AGE The age of the prisoner has consistently been found to be the strongest and most consistent predictor of prison misconduct from the earlier studies to the present (Schnur, 1949;

Zink, 1958; Wolfgang, 1961; Johnson, 1966; Ellis, et al., 1974; Flanagan, 1980; Ekland-Olson, et al., 1983; Flanagan, 1983; Toch, Adams, & Grant, 1989; Anson & Hancock, 1992; Lovell,

Cloyes, Allen, & Rhodes, 2000; Harer & Langan, 2001; Wooldredge, et al., 2001; Gaes, et al.,

2002; Camp, et al., 2003; Peck, Jr., 2004; Griffin & Hepburn, 2006; Cunningham & Sorensen,

2007; Steiner & Wooldredge, 2008; Kuanliang, Sorensen, & Cunningham, 2008). In their study

Camp et al. (2003) examined all misconduct and subcategories that included violent, property- related, drug-related, security-related, accountability-related (escapes) and other miscellaneous misconduct. They found that younger prisoners were more likely than older ones to be involved in misconduct and found this inverse relationship to be true of the total misconduct category as well as all of the sub-categories except property offenses. Since most criminologists acknowledge that younger persons are more likely to be involved in criminal activities than older persons (Cohen, 1955; Matza, 1964; Hirschi, 1969; Akers, 1985; Gottfredson & Hirschi, 1990;

Anderson, 1999; Warr, 2002; and Laub & Sampson, 2004), it should not be surprising that the problem of inmate misconduct is greater when there are more youthful criminals in prison.

What is it about youthful prisoners, besides a general lack of maturity that might result in their higher involvement in prison misconduct? There is some evidence in prison misconduct studies that young prisoners possess poorer coping skills that result in more negative behavior.

McCorkle (1992) studied ―personal precautions to violence‖ in a southern maximum-security prison for adult males through surveys of 300 prisoners and in-depth interviews with 25 prisoners. In his examination of different types of precautionary behavior, that is behavior aimed to prevent being assaulted, he found that age was the best predictor of aggressive precautionary behavior. Overall, his findings indicate that younger prisoners are more likely than older prisoners to misbehave in general, to be aggressive, and to handle potential violence by initiating violence themselves. Lovell, Cloyes, Allen, and Rhodes (2000) conducted a study of 232 male prisoners residing in Washington State‘s four Intensive Management Units to determine how, if at all, they differed from general population prisoners. Prisoners living in IMU (Washington

State‘s equivalent of a supermax) are those prisoners whom the prison administration has classified as dangerous and disruptive due to their greater involvement in prison misconduct.

Among younger IMU residents, they found two overlapping patterns: some younger prisoners used IMU as an informal strategy to gain protection from other inmates; and some of the younger

IMU residents suffer from serious impulse control problems, often being described by staff as

―explosive,‖ or ―out of control.‖ Both patterns of behavior could possibly be related to a lack of positive coping skills on the part of these juveniles. Finally, age has been found to have an interactive effect with overcrowding, in that the relationship is strongest in facilities housing young prisoners thus indicating that young males are least likely to handle the pressures associated with overcrowding and therefore are more likely to be involved in misconduct as a result (Nacci, et al., 1977; Wooldredge, et al., 2001). Thus it might be a lack of positive coping

21 skills or the presence of negative coping that explains why younger prisoners are more likely to be involved in serious prison misconduct and violence, rather than just age itself.

PRIOR CRIMINAL HISTORY AND PRIOR HISTORY OF PRISON MISCONDUCT Prior criminal history and prior history of prison misconduct are both strong predictors of prison misconduct and violence. Researchers who have operationalized criminal history by examining the number, seriousness, and/or recency of prior arrests or convictions have found a clear link to higher levels of prison misconduct (Johnson, 1966; Toch, et al., 1989; Gendreau,

Goggin, & Law, 1997; Lovell, et al., 2000; Harer & Langan, 2001; Wooldredge, et al., 2001;

Gaes, et al., 2002; Camp, et al., 2003; Peck, Jr., 2004; Griffin & Hepburn, 2006; Cunningham &

Sorensen, 2007; Trulson, 2007; Steiner & Wooldredge, 2008). In the breakdown of prison misconduct into sub-categories (violent, property-related, drug-related, security-related, escapes and other miscellaneous misconduct), Camp and colleagues (2003) found serious criminal history to be a positive and significant predictor for all misconduct and all sub-categories of misconduct, with the exception of drug- and security-related misconduct. A meta-analysis of 39 studies on the predictors of misconduct pointed to criminal history as one of the strongest of the individual-level predictors (Gendreau, et al., 1997). There have been some mixed results regarding the severity of prisoners‘ current offense with some finding it a factor (Harer &

Langan, 2001; Griffin & Hepburn, 2006) and another study finding no such relationship

(Wooldredge et al., 2001). Toch and his colleagues (1989) also found a relationship when they limited prior arrests to those for violent offenses, believing that for some prisoners violent misconduct in prison is a continuation of a violent coping mechanism they had used on the street.

Prisoners with severe prior histories of violence are also more likely to continue their violent offending in prison than are prisoners with no such prior history. Nachshon and

Rotenberg (1977) conducted an experiment in Israel where they measured teens‘ perceptions of violence in movie scenes comparing two groups of teens: 31 violent juvenile delinquents residing in a closed-correctional facility and 54 non-delinquent teens in a residential vocational school. They showed ambiguously violent scenes from animated movies, half of which were defined by a group of judges as violent and half defined as non-violent. Each teen told an interviewer what they saw in each scene. Their responses were reviewed by judges who determined what responses were violent or not and all judges scored the responses of each individual and their scores were averaged into a ―violence score‖ for each youth. They found a statistically significant difference in the perceptions of violence between the two groups, and noted that ―violent tendencies or exposure to violence affect perception by producing a tendency in the subject to perceive violence‖ even when there was none (Nachshon & Rotenberg, 1977, p.457). Therefore, prisoners who either have been exposed to violence or have been involved with violent behavior (e.g. gang violence) might be more likely to interpret situations in a negative and potentially violent manner than are persons who have not been so exposed. This tendency to interpret situations as more violent and to determine that violence is therefore a necessary response in more situations could be linked to poor assessments of a situation and poor problem-solving skills—in other words, a lack of positive coping skills.

More recent research has uncovered the relationship between gang affiliation and prison misconduct. Gaes and his colleagues (2002) discovered gang affiliation to be associated with prison violence in that gang members were more likely than non-gang members to be involved in prison violence. This study in federal prisons determined that core members of a gang were more violent than those who were suspected of being gang members or those loosely associated with gangs. Interestingly, the length of time in the gang was negatively associated with violence.

Possibly new members, going through initiation and needing to demonstrate their loyalty to the gang, were the ones being commanded to assault other prisoners and staff or felt the need to prove themselves, while gang leaders, being in the gang the longest, would no longer do the dirty work.

Subsequent studies (Huebner, 2003; Griffin & Hepburn, 2006; Trulson, 2007) all found evidence that gang membership affected prison misconduct. Huebner (2003) studied both the individual and institutional predictors of violence and found that at the individual level, gang affiliation was associated with prison violence. Similarly, Griffin and Hepburn (2006) found that age and gang membership were the only variables that were significant for all of the individual violent offenses in the bivariate analysis. In the multivariate analysis, they found that gang membership had an effect on all violent misconduct beyond the inclusion of individual-level factors. Trulson (2007) found that prisoners with gang influences were significantly more likely to be involved in prison violence, but not in incidences of institutional disruption.

Finally, DeLisi, Berg, and Hochstetler (2004) explored the relationship between prison violence and various types of gang memberships (street gangs, prison gangs, and both types of gangs) and found mixed results. They found that prisoners with prior street gang involvement were no more likely than other prisoners to engage in prison violence. They also found that those with prison gang involvement were also no more likely than other prisoners to engage in prison violence. However, when the multiplicative of the two was added, prisoners with street gang history and prison gang involvement were significantly more involved in prison violence compared to non-gang prisoners. Interestingly the gang variables were significantly predictive of prison violence only in the full model when various types of gang membership were considered. In addition, the overall effects of gang membership were smaller than some of the

24 risk factors related to chronic offending (i. e. history of violence, escape, adjustment risk, time served, and prior confinement) and other variables (i.e. family ties, education, quadratic age, age, race, family ties, education).

Although less researched than prior criminal history, prior history of prison misconduct has been consistently found to predict future prison misconduct (Harer & Langan, 2001; Camp et al, 2003; Cunningham & Sorensen, 2007). Camp and his colleagues (2003) examined the effect of the prior history of each type of prison misconduct and discovered that the relationship was positive and significant and their strongest individual-level predictor. Finally, Harer and Langan

(2001) found a history of prior escapes or escape attempts to also be a good predictor of prison misconduct.

MENTAL HEALTH Studies that have examined the mental health status of prisoners have found that prisoners with mental health problems are more likely to be involved in prison misconduct (Toch

& Adams, 1986; Toch, et al., 1989; Toch & Adams, 1994; Lovell, et al., 2000; Peck, Jr., 2004;

Toch & Kupers, 2007). Toch and various colleagues have studied the effect of mental health issues and prison misconduct in a number of studies. Toch and Adams (1986) examined the prison violation rates of 10,534 prisoners released from New York state prisons between 7/30/82 and 9/1/83. After the criminal history and social history variables were entered into the model, the effects of whether an inmate had been hospitalized for mental illness or treated on an outpatient basis were significant in predicting prison misconduct. They found violation rates to be highest for those diagnosed with conduct disturbance and antisocial personality, higher than average violation rates for prisoners diagnosed with schizophrenia, but unexpectedly low violation rates for prisoners with substance abuse and anxiety disorders. In their study of

Washington State‘s four Intensive Management Units, Lovell and his colleagues (2000) used five proxy measures of mental health disorders and found that almost 30 percent of IMU prisoners had at least one indicator of a mental health disorder compared to 10 to 15 percent in the general prison population. In addition, those identified with a potential mental health disorder in IMU had higher infraction rates, more violent infractions, and more disturbing or bizarre infractions compared to other IMU residents. They interpreted this pattern as being

―consistent with other findings that mentally ill prisoners have greater difficulty coping with prison settings‖ (2000; p. 36). Many of these researchers have noted the connection between mental health and the inability to cope with the chaos, variable personalities, insecurities, and boredom that characterize prison life thus often leading to disciplinary issues.

OTHER INDIVIDUAL LEVEL FACTORS FOUND TO AFFECT PRISON MISCONDUCT There have been mixed results regarding the effects of race and ethnicity on prison misconduct, and especially violence. Some research has found blacks or minorities had higher rates of prison misconduct (Carroll, 1974; Toch, et al., 1989; Gaes, et al., 2002; Berg & DeLisi,

2006; Trulson, 2007). Others have found that race was not a significant predictor of prison misconduct (Ellis, Grasmick, & Gilman, 1974; Wright, 1989). Still others have found mixed results within their own studies, depending on the location of the prisons, the type of prison misconduct studied, and whether other variables, like prior criminal history were included in the model (Petersilia, 1983; Harer & Steffensmeier, 1996; Camp, Gaes, Langan, & Saylor 2003).

Wright (1989) has reported that much of the prison literature suggests that blacks, because of their experience in the modern urban ghetto, are more resilient to the deprivations of incarceration (e.g. lack of amenities, lack of security) than are whites. He delineates two views of black prisoners: one by Carroll (1988) of an inmate who is tough, domineering, and

26 aggressive and the other by Johnson (1979) of an inmate who is strong, stoic, and unmoved by pressure. Carroll conceptualizes prisons more like ―slum communities‖ than like ―total institutions‖ as described by Goffman (1961). He discusses how over half of the people in prison are minorities who were most likely to have come from communities that would be defined as slums or ghettos. He discusses blacks‘ concern with discrimination by prison staff and other prisoners, the role of gangs, and concludes that blacks may be more resilient and better able to adapt to the prison environment compared to their white peers. As he explains it, ―(i)t is plausible to infer that the subculture of the black urban ghetto is functional for survival in the walled ghetto of the prison. The ghetto inhabitants‘ sense of themselves as victims, reinforced daily by the facts of their existence, provides a rationale that shifts responsibility for their acts from themselves to the system‖ (p. 188). Thus Carroll views prison misconduct simply as an extension of violence and deviance that urban black convicts would have committed on the street

‗out of necessity‘ had they not been incarcerated. He assumes then that blacks are more likely than whites to be involved in prison violence. Johnson also believes that blacks‘ experience of discrimination and harsh living conditions on the street prepares them for the pains of imprisonment. He explains how black prisoners use their ‗facades‘ of being in control and having strength as a means to put people off and avoid victimization in prison. While he acknowledges that injustice and discrimination contribute to their pain of imprisonment, he does not assume that they are more aggressive because their similar experiences on the street help them to address those issues in prison. Instead, he points to whites‘ higher rates of suicide and mental health issues as proof that blacks are better able to adapt to prison life. In his study,

Wright (1989) discovered that with the exception of self-injury (more whites were involved in self-injury than blacks), blacks and whites experienced incarceration similarly—that is, with

27 similar levels of distress, mental health contacts, and aggressive behavior. He reported that both blacks and whites listed similar environmental needs and rated their prison settings similarly. He did find some support in his data that a ―more economically marginal life-style before incarceration‖ was related to successful adjustment, but found it to be independent of race. He concluded, ―(t)hese findings lead to the conclusion that racial distinctions are not universal, and the practice of suggesting that blacks adapt one way and whites another leads to inappropriate conclusions about patterns of prison adjustment‖ (p. 67).

Other individual factors that have yielded mixed results include those ‗commitment to convention‘ variables that researchers have often added together to measure the stability and conventionality of prisoners‘ lives before incarceration and include marriage, employment, and education level (Toch, et al., 1989; Wooldredge, et al., 2001; Harer & Langan, 2001; Kuanliang, et al., 2008; Steiner & Wooldredge, 2008); the severity of current offense (Lovell, et al., 2000;

Harer & Langan, 2001; Wooldredge et al., 2001; Griffin & Hepburn, 2006; Cunningham &

Sorensen, 2007); and time served or sentence length (Zink, 1958; Wolfgang, 1961; Flanagan,

1980; Wooldredge et al., 2001; Griffin & Hepburn, 2006; Cunningham & Sorensen, 2007;

Trulson, 2007; Kuanliang, et al., 2008; Steiner & Wooldredge, 2008). Perhaps, like race, the mixed research results are due to other underlying factors associated with these variables. For example, what does it mean for future prison misconduct if prisoners are married or had successful employment histories prior to being incarcerated? Are these factors indicators of stability or a person‘s ability to cope? Other possible explanations for mixed results could be the measurement of these variables (e.g. most studies examined race using a dichotomous variable) or of prison misconduct in general.

Another individual factor that has yielded somewhat mixed results is pre-prison substance abuse (Gendreau, et al., 1997; Berg & DeLisi, 2006; Dhami, Ayton, & Loewenstein,

2007; Steiner & Wooldredge, 2008). In their meta-analyses of the predictors of prison misconduct, Gendreau and his colleagues (1997) included prior substance abuse in the domain they labeled ―antisocial attitudes and behavior;‖ thus it was not reported how many studies included substance abuse as a predictor. Nevertheless, this data was collected in studies totaling over 16,000 subjects and it was deemed to be a strong predictor of prison misconduct. Dhami and colleagues (2007) included substance abuse as one of their ―quality of life before prison‖ domains, which also included completion of high school, employment and in a relationship prior to prison, and prior incarceration and which was found to be predictive of prison misconduct.

Steiner and Wooldredge (2008) included prior substance abuse as a separate factor in their bi- level model of inmate misconduct and found it predictive of not only misconduct in general but also predictive specifically of assaults, drug/alcohol use in prison, and other nonviolent misconduct. However, one recent study (Berg & DeLisi, 2006) found no relationship between substance abuse and prison violence.

RESEARCH ON COPING

Many researchers have incorporated the concept of coping in their empirical studies of prisons. However, coping has been defined in many ways—sometimes broadly such that it is synonymous with adjusting (Thomas & Foster, 1972; Poole & Regoli, 1980; Parisi, 1982;

Goodstein, MacKenzie, & Shotland, 1984; Wright, 1987; Gover, MacKenzie, & Armstrong,

2000; DeLisi, et al., 2004; Dhami, et al., 2007) and sometimes narrowly such that it refers to specific skills one possesses to deal with stress and strains (Zamble & Porporino, 1988; Zamble,

1992; Sappington, 1996; Biggam & Power, 1999; Harreveld, Van Der Pligt, Classen, & Van

Dijk, 2007). This review will concentrate on both the broad definition and the narrower theory of social-psychological coping and the empirical research that has been carried out in prisons.

ADJUSTMENT AND THE GENERAL CONCEPT OF COPING IN PRISON RESEARCH In the early years of American prison research, Clemmer (1940) and Sykes (1958) discussed the effects of the prison culture on prisoners and the adaptation process that was necessary to survive in prison. Clemmer coined the adaptation process ‗prisonization,‘ which involved prisoners accepting their new inferior role, adapting to rules within a regimented structure, and getting used to dependence on the prison for some needs while foregoing others.

Sykes expanded the concept of prisonization by describing the prisoner subculture and its concomitant prison code that guided inmate behavior with other prisoners and with correctional staff. ―Deprivation theory argues that prisonization is an adaptive process employed by inmates to cope with the social and physical deprivations of imprisonment‖ (Paterline & Petersen, 1999, p. 429). The importation model of prisonization assumes that ―preprison experiences, particularly those involving the adoption of criminal values, and personal characteristics of the inmates affect the degree of assimilation into the inmate subculture‖ (Paterline & Petersen, 1999, p. 429).

The concept of coping has been an integral component of the general study of Clemmer‘s concept of prisonization, and its related debate between importation and deprivation (Wheeler,

1961; Poole & Regoli, 1980; Feld, 1981; Flanagan, 1983; Goodstein, et al., 1984; Wright, 1987;

Gover, et al., 2000; Gillespie, 2003; DeLisi, et al., 2004; Dhami, et al., 2007). In this context it is often used interchangeably with the words ‗prison adjustment‘ and ‗adaptation.‘ Zamble and

Porporino (1988) ably summarized the prisonization studies that emphasized ‗intra-institutional pressures and problems‘ created by the experience of imprisonment.

―Studies based on the deprivation model have examined the relationships between prisonization and a variety of factors, including: (a) the length of time in prison and the time remaining to be served; (b) interpersonal involvements and the social role assumed by the inmate; (c) the type of institution and organizational structure; and (d) the degree of alienation or powerlessness experienced by the inmate‖ (p. 7).

Similarly, they summarize the factors that proponents of the importation model of prisonization include:

―(a) general social history factors such as age, race, educational attainment, and pre-offense socioeconomic and employment status; (b) variables reflecting the individual’s history of criminal involvement, such as the number of prior convictions and the number and length of prior prison terms; (c) identification with criminal values and attitudes toward the legal system; (d) the self-concept of the individual; and (e) identification with broad social, political, racial, and religious ideologies‖ (p. 8).

Interestingly, both types of studies used a variety of indicators to conceptualize prisonization.

These have included: adherence to the inmate code (Wheeler, 1961; Tittle & Tittle, 1964; Feld,

1981; Gillespie, 2003); attitudes toward staff and legitimacy of prison administration (Wellford,

1967; Feld, 1981); inmate violence and misconduct (Ellis, et al., 1974; Feld, 1981; Flanagan,

1983; Poole & Regoli, 1980; DeLisi, et al., 2004); prisoners‘ abilities for autonomous decision making (Goodstein et al., 1984;); post-release success (Thomas & Foster, 1972) and emotions and evidence of psychological distress, including suicide (Wright, 1987; Gover, et al., 2000;

Dhami, et al., 2007). Given the varied operationalizations of the dependent variable, prisonization, it is not surprising that the results of both the deprivation and importation studies on prisonization have been mixed. Possibly as a result, others have attempted to integrate both importation and deprivations factors into a model that can predict prisonization (Thomas &

Foster, 1972; Gillespie, 2003; Dhami, et al., 2007) and have found some evidence to support the integration of the two models.

Porporino and Zamble (1984) have critiqued the direction that prison adjustment research had taken, namely focusing on the concept of prisonization. They did acknowledge the importance of adjustment in prison, highlighting the fact that poor adjustment results in prisoner suicide, self-inflicted injury, stress-related disorders and conditions, and an increase in violence and the threat of violence to both staff and prisoners. However they believe that the sociological analyses of imprisonment have been flawed methodologically and conceptually. Methodological flaws include failure to establish reliability of measures, exclusive use of self-report data and the lack of longitudinal data for a measure that by definition changes over time. However,

Porporino and Zamble point to conceptual problems as being more serious and thus resulting in many of the conflicting results that have been found. They note that studies have yielded mixed results for both importation and deprivation factors causing prisonization as well as mixed results on the effects of prisonization on post-release results. Thus they labeled prisonization as

―possibly of some value as a global theoretical construct, (but) as a measure of how imprisonment affects individuals, prisonization is clearly too general and too crude a concept‖

(1984, p. 406).

Similarly, Porporino and Zamble argue that studies that attempted to measure the psychological effects of imprisonment are similarly flawed. These studies examined the effects of imprisonment on personality functioning, self-esteem, hostility, locus of control, attitudes towards the law, level of authoritarianism, perceptual-motor and cognitive functioning, depression and anxiety and have likewise found mixed results. While they acknowledge that

―the available evidence indicates that gross psychological deterioration is not an inevitable consequence of imprisonment… [they note that] the research may be criticized for using measures that are insensitive to subtle effects of imprisonment… and (that) the effects of other

32 potentially significant factors (e.g., age, prior prison experience) typically have not been adequately controlled for or taken into account‖ (Zamble & Porporino, 1988, p. 11). Porporino and Zamble criticized these studies‘ methodologies for poor sampling, cross-sectional studies of measures that change over time, and failure to include other factors that might affect psychological adjustment such as time remaining in sentence, post-release expectations, and differences in the experiences of prisoners. These studies have often used undifferentiated samples to look for general patterns that omit the experience of particular groups—often omitting from the study special groups of people such as women, young prisoners, and those in protective custody and segregation, indeed often the very subgroups of prisoners that might have the most difficulty with coping (Liebling, 1999).

Goodstein and Wright (1991) have also been very critical of studies that examine prisonization and that focus on the issues of deprivation and importation. They question the

―generalizability of the constructs of prisonization‖ noting that many of them depend on the nature of the prison environment. In addition they question one of its underlying assumptions.

The ―construct of prisonization is based on the assumption that the pains of imprisonment will be accompanied by the harms of imprisonment. It assumes that inmates who adjust according to its prescriptions will be disadvantaged in the long run, that they will leave prison worse off than when they entered, and that releasees are less capable of success than they would have been if they had not employed this adjustment strategy‖ (pp. 241-242). Liebling (1999) was similarly critical of studies of prison suicide that focused on prisonization, noting that they had not ―tapped the subjective, cognitive, or affective contributions prisoners make to their own experiences of prison‖ (p. 287). She was particularly critical of studies that operationalized harm as permanent

33 deterioration in IQ or other types of measurable functioning, and studies that did not investigate psychological or emotional distress or disability before or after incarceration.

Porporino and Zamble (1984) advocated for abandoning the prisonization/deprivation/importation model studies and focusing instead on how prisoners appraise and cope with the many stressful situations and events they must face in being incarcerated. They tout the initial findings of both Hans Toch‘s work on coping in prison and their own work in Canadian penitentiaries as providing preliminary evidence that coping is an important and measurable concept with respect to prison adjustment.

PRISON STRESS AND THE SOCIAL-PSYCHOLOGICAL THEORY OF COPING The focus on coping in prison began with the individual and joint writings of Hans Toch and Robert Johnson (Johnson & Toch, 1988; Toch & Adams, 1986; Toch, 1987; Toch, Adams,

& Grant, 1989; Toch & Adams, 1994; Johnson, 2002). In his study comparing a systematic random sample of prisoners as the control group with prisoners who had attempted suicide or had purposely injured themselves, Toch (1975) discovered that ―almost no inmate‘s institutional career had been free of serious (and potentially disabling) stress‖ but that the main difference between the control group of prisoners and those who had attempted suicide was that the former group used a repertoire of coping strategies to ―weather‖ their problems in prison. Similarly,

Toch and his colleagues studied prisoners with mental health problems, examining what they termed their ―maladaptive behavior‖ that included self-destructive or self-injurious behavior, demonstrations of deficits of perceptiveness or skill, creation of problems for the environment or other persons and repeated failures to accomplish goals (Toch & Adams, 1986; Toch, et al.,

1989; Toch & Adams, 1994). Toch and Adams (1986) found a pattern of disruptiveness

34 including violence among prisoners who were emotionally disturbed and attributed their findings to the limited coping abilities and psychological resources of the mentally disturbed.

―In a predatory environment a victim-prone person can be hounded into preemptive aggression or into explosive panic. In congested settings such as prisons self-insulation or privacy may become difficult to achieve. In settings that oscillate between chaos and routine (as most prisons do), stimulation levels may defy accommodation or exceed thresholds…Recalcitrance in such contexts can express a person’s feelings of impotence or sense of despair‖ (p. 16).

They believed that the deinstitutionalization of mental health patients resulted in an influx of prisoners who were both ―mad and bad‖ increasing the percentage of the prison population that was both disturbed and disruptive.

In their book The Pains of Imprisonment, Johnson and Toch (1988) explore the various sources and consequences of stresses in the prison system. In the chapter that explains the concept of stress and how it pertains to prisoners, Toch draws on a conceptual model by

McGrath (1970) that breaks stress down into four stages. The first stage is the stress transaction, situation, demand, or constraint itself. A demand stressor can either require more of an inmate than he can deliver (overload) or fail to challenge his interests or capacities (underload). The second stage of stress is "reception" because what one prisoner may perceive as a stressful situation another may not—the definition of a stressful event must include its perception. The third stage is the "responses" to the perceived pain—these are psychological feelings (anger, lowered self-esteem, timidity) as well as behavioral responses. The fourth stage is "the consequence of the responses." Toch highlights the importance of the fourth stage noting that it provides a feedback loop and is absolutely pertinent to those prisoners who get into disciplinary problems constantly. So often a prisoner will end up in segregation as a result of a small issue or stressor. But his initial response to that stressor and the consequences of that response escalate

35 the behavior/response pattern. It becomes more and more serious until the prisoner, out of control, ends up in segregation.

Toch (1982) explains Howard and Scott's (1965) typology of adaptation to stress and notes its importance in the prison setting. Adaptations can be assertive—responses that address stress in a head-on fashion in an attempt to solve the problem. Less helpful are those adaptations that are divergent (divert energies away from problem) or inert (no response). Toch does not acknowledge that assertive responses could potentially be negative responses or that in some circumstances doing nothing in response to stress (inert adaptation) may be preferable.

However, the value of the model is that it advocates addressing a problem head on instead of ignoring it. Toch advocates helping prisoners profit from stress so that they can emerge more resilient and capable. He recommends that prison administrators reinforce the use of assertive, yet appropriate responses by giving prisoners problem-solving resources. Toch observes that past successes and failures in dealing with life's situations enhance and reduce one‘s capacity to respond to stress, reflect coping skills, self-confidence, and self-esteem. Successful coping increases self-esteem and future capacity to cope. "Interventions that help people cope with stress help to break the failure/low-self-esteem cycle" (1982, p. 37). Toch quotes Janus (1969) who maintains that successful coping can have a benefit beyond solving a specific problem at hand.

"[T]he central theme is that any instance of bereavement, separation, failure or suffering can be a turning point in the person's life, resulting either in emotional breakdown and sustained personality damage or a marked improvement in personality functioning...Crises are viewed as rare occasions when the person faces serious threats, losses, and demands that are near the limits of his resources for coping. But if the person solves adequately the difficult problems confronting him, his sense of self-confidence is greatly increased, and he may be able to overcome other difficulties that had been interfering with his adjustment in the past" (Janus, 1969, p. 198).

The concept of ―mature coping‖ was developed over time in a number of writings by

Toch and Johnson. Toch first discussed the connection between coping and maladaptation in several writings that mostly focused on the mentally ill in prison (Toch, 1975; Toch & Adams,

1986; Toch, et al., 1989; Toch & Adams, 1994). In their book on prison stress Johnson and Toch

(1988) discussed the connection between prison stresses, a lack of coping skills, and poor behavioral outcomes. Finally Johnson (2002) named the concept of ―mature coping‖ in his book

Hard Times. Johnson explains that prisoners can more easily adapt to prison in a positive manner if they have developed ―mature coping.‖ Johnson writes that,

―Mature coping means, in essence, dealing with life’s problems like a responsive and responsible human being, one who seeks autonomy without violating the rights of others, security without resort to deception or violence, and relatedness to others as the finest and fullest expression of human identity‖ (2002, p. 83).

Johnson delineates three separate elements of mature coping. First, a person is considered to exhibit mature coping when he/she is willing and able to address and handle problems that arise. This relates to the assertive adaptation response to stress discussed previously which is likely affected by one‘s self-efficacy or locus of control. The second element of mature coping is to handle problems without violence or lying. Thus many of the responses common in prison—physical or threatened violence, deceit and manipulation—would not be considered mature coping. Finally, the third element of mature coping is to handle problems by relying on one‘s communication skills and relationships with others to address the problem in a constructive manner. This final element assumes that prisoners empathize with others and if needed, reach out to others for assistance when dealing with their stresses and strains.

Ideally, correctional administrators would prefer to have a prison full of prisoners who demonstrate mature coping—resulting in a population less likely to be deceptive, manipulative, violent, and helpless. Program demonstration and research on the principles of effective correctional treatment of prisoners, including the use of cognitive behavioral therapy, have provided a potential solution on how to increase mature coping skills in prisoners (Gendreau,

1996). In their evaluations of correctional treatment programs, such as boot camps or substance abuse treatment, researchers have measured whether mature coping increased and in turn affected other outcomes (Soderstrom, et al., 2001; McDonald, 2006). Mature coping is the ideal construct to measure in correctional program evaluations or in identifying those prisoners who would be appropriate for correctional treatment because it incorporates the coping skills, the empathy, and, eschewing of violence and deception when approaching problems. However, in the proposed study of serious prison misconduct and violence, the concept of mature coping, as defined by Johnson would prove tautological. This is because one of the main elements of

Johnson‘s construct of mature coping—that a person deals with their problems without violence—is indeed the dependent variable of this study. Therefore, research, such as this, that examines the effects of coping on serious misconduct and violence should focus on the other elements of Johnson‘s (2002) mature coping, such as assertive problem-solving and utilizing support from others—the constructs developed by psychologists in the social-psychological theory of coping.

The social-psychological theory of coping has been developed by Lazarus and his colleagues over many years (Lazarus, 1966, 1981; Lazarus, et al., 1974; Lazarus, et al., 1982;

Lazarus & Folkman, 1984; Folkman & Lazarus, 1985). Lazarus and Folkman define coping as

―constantly changing cognitive and behavioral efforts to manage specific external and/or internal

38 demands that are appraised as taxing or exceeding the resources of the person‖ (Lazarus &

Folkman, 1984, p. 141). The coping literature generally recognizes that two separate processes occur before one copes and that each of these processes are regulated by different conceptual systems within one‘s personality (Lazarus & Folkman, 1984; Folkman & Lazarus, 1985; Epstein,

1990). The primary appraisal of a stressful situation involves an emotion-focused process associated with one‘s experiential system and the secondary appraisal involves a problem- focused process associated with one‘s rational system. Coping is the actual process of carrying out the solutions determined during the secondary appraisal. McDonald (2006) explains that the emotion-focused portion of coping deals with the emotional response to the situation with feelings such as anger, depression, anxiety, or guilt. The second process occurs when efforts are

―directed at defining the problem, generating alternative solutions, weighing the alternatives in terms of their costs and benefits, choosing among them, and acting‖ (Lazarus & Folkman, 1984, p. 152). Empirical research on coping has focused on both processes, but primarily on the problem-solving process and has generated various sub-categories or ways of coping (Lazarus,

1981; Carver, et al., 1989; Amirkhan, 1990; D‘Zurilla & Chang, 1995), including general problem-solving capabilities (whether they are positive or negative), seeking social support, avoidance, and carelessness/impulsivity.

EMPIRICAL RESEARCH ON COPING AND ITS CONSEQUENCES ON PRISONER ADJUSTMENT As mentioned previously, Porporino and Zamble (1984) called for the abandonment of the deprivation/importation model studies and instead suggested that researchers focus on how prisoners appraise and cope with the hardships of prison. They (Zamble & Porporino, 1988 and

1990) conducted their own study to determine how prisoners coped with their environment, including determining what situations prisoners see as problems, how they interpret them, and

39 what they do. Their longitudinal study of prison coping was conducted in male penitentiaries in the Ontario region of the Corrections Service of Canada. It involved face-to-face interviews, self-report surveys, and official data collection from the prison files of 133 male prisoners who had recently entered Ontario prisons. Information about problems faced and situational coping skills used, as well as other potentially dynamic factors, were gathered in three data collection phases over 16 months.

Zamble and Porporino did not suppose that the ‗pains of imprisonment‘ would have inevitable negative effects on prison misconduct and other outcomes, nor did they expect that characteristics of prisoners (such as age, race, or prior criminal history) would solely determine prison outcomes during confinement. Instead, they used an interactionist perspective to examine the appraisals and the general and specific coping behaviors that prisoners used as they

―negotiated‖ their time in prison. They were also interested in the effects of imprisonment and the way that confinement changes (or does not change) established behavior patterns of prisoners. Their perspective has provided much of the framework for this proposed study and has demonstrated the efficacy of measuring coping skills with this population and examining the relation between coping and misconduct.

Zamble and Porporino found that while prisoners experienced high stress and depression within weeks or months of their commitment to prison, these dips in prisoners‘ psychological well-being did not last over the course of the study, but eventually stabilized. Taken together with data from prior psychological studies, they concluded that ―general deterioration does not occur‖ (1988, p. 147). Regarding coping, they found that imprisonment had no positive or negative effect on the ways that prisoners think about or react to stressful situations and circumstances. Prisoners who coped well at the beginning of their sentences generally coped

40 well throughout their time in prison. The same consistent pattern was also found for prisoners with poor coping abilities. Zamble and Porporino concluded that ―individuals who enter prison with inadequate behavior repertoires or maladaptive modes of coping are bound to leave with the same (lack of) capabilities. While men are in prison their outside behavior patterns remain, in effect frozen in time. Indeed, we can characterize imprisonment as the behavioral equivalent of a deep freeze, in which the outside behavior of inmates is stored until their release‖ (1988, p.

152).

While they acknowledge that correctional administrators and prison researchers may breathe an initial sigh of relief that prisons might not be so debilitating, they caution that this

―deep freeze‖ is just as harmful.

―Most of us learn to cope through accumulated experience, but prisoners are deprived of much of that experience. (Thus as well) as having his behavior frozen in its original pattern, the prisoner is frozen developmentally. This might explain why the behavior of habitual offenders resembles that of adolescents in many ways, e.g., in the heavy dependence on peer groups, emphasis on physical dominance, and generally impulsive behavioral style. It also follows that imprisonment at an early age will have more effect on subsequent behavior than at a later time. Finally, from this analysis we can predict…that those who have spent the most time in prison will cope most poorly when they are on the outside. And…this would then lead them to commit new offenses, resulting in their return to prison, and a return to the cycle, ad infinitum‖ (1988, p. 153).

A number of other studies have also successfully measured the coping abilities of male prisoners (Sappinton, 1996; Biggam & Power, 1999; Harreveld, et al., 2007) or female prisoners

(MacKenzie, Robinson, & Campbell, 1989; Nagy, Woods, & Carlson, 1997; Greer, 2002;

McDonald, 2006; Fritch, 2006). These studies either examined dispositional coping or situational coping and its relationship to myriad factors relating to incarceration. Research that asks about dispositional coping will ask participants what they usually do when confronted with

41 stressful situations or events. On the contrary, research that asks about situational coping will ask participants how they coped with a specific stressful event (Carver, Scheier, & Weintraub,

1994).

MacKenzie, Robinson, and Campbell (1989) conducted research on female prisoners and coping and found no support for their hypotheses that those who were imprisoned longer would possess better coping skills than those newly incarcerated or that sentence length affected female prisoners‘ coping abilities. Instead they found (like Zamble and Porporino, 1988) that the ability to cope had more to do with the skills that the prisoners had acquired prior to incarceration.

McDonald (2006) conducted a study of an in-prison substance abuse therapeutic community for female prisoners. Her qualitative research led to the conclusion that participation in a therapeutic community could improve the positive coping skills (improved problem-solving abilities and seeking of social support) and reduce the negative ways of coping (avoidance) of its female participants. Other researchers have also conducted qualitative studies on female prisoners and their coping strategies and skills. Giallombardo (1966), Fox (1970), and Owen (1998) found that female prisoners adapted to prison by constructing relationships with other women, pseudo families that helped them deal with the impersonal nature of imprisonment. Fox also noted that female prisoners often resorted to negative ways of coping such as emotional outbursts or requesting medication to avoid such outbursts.

In summary, Zamble and Porporino (1988) demonstrated, as did those researchers who followed them, that the social-psychological concept of coping is a viable construct whose dimensions can be measured with an incarcerated population. In general it can be viewed as a dynamic trait that prisoners have to varying degrees when they enter prison and that affects their outcomes as they navigate the stresses associated with their incarceration.

RESEARCH ON COPING AND ITS EFFECT ON PRISON MISCONDUCT There are very few studies that have examined the effects of coping on prisoners‘ likelihood of being involved in prison misconduct. Zamble and Porporino (1988) examined prisoners‘ appraisals of their prison problems and their total problems and also general coping behaviors, such as dealing with problems head on (efficacy), suicide attempts, or the use of alcohol. In addition, they examined specific prison behavior and specific types of coping such as socializing, interacting with friends, planning ability, controlling their thoughts, and thinking about the future or past. They found that coping and appraisal factors both individually and grouped were relatively more powerful as predictors of their outcomes (including prison misconduct, medical, depression, anxiety, and anger) as compared to socio-demographic and criminal history factors. With respect to misconduct they found that the individual factors that best predicted the existence of disciplinary issues included (in order of importance): alcohol abuse (a general coping variable), the level of prison problems reported, the number of friends a prisoner had, age, the prisoner‘s planning abilities, family type, custody level, and their efficacy at coping which together accounted for 39 percent of the variance in level of dreports. Zamble and Porporino also divided all of the control variables into four groups: individual-level background variables, general coping variables, specific behavior and coping variables, and appraisals. Background and general coping behavior best predicted involvement in disciplinary issues followed by prisoners‘ appraisal of problems. Together these groups of variables explained 48 percent of the variance in level of dreports.

Sappington (1996) explored the relationship between coping and prisoner misconduct and also studied prisoners‘ beliefs about themselves and their environment. He defined beliefs as

―assertions that are subjectively held to be true.‖ He used Bandura‘s (1977) typology of beliefs:

―(a) response-outcome expectancies, which refer to an individual‘s belief that a particular action will lead to a particular consequence; and (b) self-efficacy expectancies, which refer to an individual‘s belief that he or she is able to perform the action‖ (1996, p. 54). This study of 47 male prisoners in maximum security examined the relationship of several different dispositional cognitive coping styles to prison adjustment variables that included the number of dreports and cognitive tests for anger, depression, and anxiety.

Sappington found evidence that both response-outcome beliefs and self-efficacy beliefs appeared to affect prison adjustment. He found that the number of dreports were higher when prisoners believed that their behavior most likely failed to affect how they were treated by others in prison and when they believed that they did not have control of their actions. With respect to coping, he found that prisoners who used the coping strategy of blaming others were more likely to have a greater level of prison misconduct than those who did not. He also found that those who used ―Pollyanna‖ thinking—that is, a tendency to think more positively about life in general—were less likely to be involved in prison misconduct.

Taken together, these two studies highlight the potential relationship of various coping skills to prison misconduct within an incarcerated population.

RESEARCH ON COPING AND NEGATIVE EMOTIONS As mentioned previously, coping theory also recognizes the role of one‘s emotional status on the outcomes of coping (Carver, et al., 1989; Soderstrom, et al., 2001). For example,

Folkman and Lazarus (1980) found that as depression increases the likelihood of using passive and avoidant coping also increases, but the likelihood of using active coping decreases.

Researchers who have studied the coping skills of prisoners and other offenders have also

44 studied their trait emotions (Soderstrom, et al., 2001; McDonald, 2006; Fritch, 2006). In her study of 101 incarcerated women Fritch (2006) studied the relationship between exposure to trauma prior to incarceration, prisoners‘ coping skills, and their mental and physical health.

While she did not find that coping mediated the relationship between trauma and mental and physical health as hypothesized, she did find that positive coping strategies mediated the associations between prisoners‘ mental health variables (e.g. PTSD, depression, distress) and their physical health and that the use of emotion-focused coping strategies was related to improved mental and physical health. Soderstrom, Castellano, and Figaro (2001) conducted a psychometric assessment of various instruments to measure Johnson‘s concept of ―mature coping‖ among adult and juvenile boot camp samples. Pre- and post-testing of the samples found that participants in both samples exhibited decreases in depression, and increases in self- esteem and problem-solving abilities as a result of their participation in the boot camp. Finally,

McDonald (2006) studied women who participated in a prison therapeutic community and found that the program did increase women‘s positive coping strategies, but that improved coping did not lower women‘s depression or anxiety.

State emotions are generally the result of a specific event, while trait emotions are those that a person might experience on a daily basis. Therefore research that asks about dispositional coping (what one usually does when one confronts a stressful situation or event) usually also asks about dispositional or trait emotions (whether one is generally an anxious, depressed, or angry person) (Carver, Scheier, & Weintraub, 1989). While research that asks about situational coping (how one coped with a specific stressful event) correspondingly asks about how one felt about that particular event (state emotion) (Zamble & Porporino, 1988).

Some of the research on prisoner coping has examined the relationship between the primary appraisal stage of coping (which is an emotion-focused process of assessment) and negative emotions. Starting with the research on appraisals, Zamble and Porporino (1988) found the negative emotion outcomes of anger, anxiety and depression were most strongly predicted by prisoners‘ negative appraisals or assessments of their prison and other types of problems.

Similarly, Sappington (1996) found in his dispositional study of coping that the tendency to blame others during the appraisal of their problems was correlated with general feelings of depression, anxiety, and anger. Interestingly, Harreveld and colleagues (2007) found a positive relationship between psychological well-being and those who engaged in ―silver-lining thinking‖ which they defined as the reevaluation of a negative situation. Thus it appears that prisoners‘ who possess and use positive skills and positive thinking at the appraisal stage of coping might be more likely to benefit and less likely to experience negative emotions.

The research on the secondary appraisal process (which entails the use of actual coping skills) and negative emotions is less clear-cut. When Zamble and Porporino (1988) examined the correlations between actual coping variables on negative emotions, they did not find any relationship between the categories of what they called general coping (which included alcohol abuse, suicide attempts, and efficacy) and the negative emotions of depression or anxiety. They did find that prisoners who were more likely to use alcohol to cope had higher levels of anger and interestingly, those who demonstrated a higher level of efficacy of coping in prison also had higher levels of anger. But prisoners who were more apt to control their thoughts and those who were more likely to think about the future when confronting a stressful situation were also more likely to suffer from anxiety, but not depression or anger. On the other hand, those who dwelled on the past were more likely to feel both angry and anxious. In their regression analysis, both

46 their appraisal category and their social support/planning category were statistically related to depression, anxiety, and anger, but the general coping category was not so related. Nagy,

Woods, and Carlson (1997) found that female prisoners who possessed positive coping skills were less likely to feel depressed or anxious compared to those who used negative coping strategies. Harreveld, et al. (2007) examined the relationship between coping strategies and prisoners‘ psychological and physical well-being. They found that prisoners who accepted social support from others had improved moods, better psychological well-being and better health.

They concluded that prisoners who believed in their own coping strategies were more likely to have a greater level of well-being. ―This suggests the importance of having a sense that one can adequately deal with the negative feelings that can accompany incarceration. It appears that this sense of coping efficacy is quite specific, and not just the result of a more general feeling of optimism and beliefs about one‘s ability to overcome obstacles in life, because there was no significant correlation with optimism‖ (2007, p. 705).

Biggam and Power, (1999) like Toch (Toch & Adams, 1986; Toch, et al., 1989; Toch &

Adams, 1994) focused their study of coping abilities on special groups of prisoners who pose challenges to prison administrators, namely suicidal, bullied, and protected prisoners along with a control group. In their research, a psychologist conducted individual face-to-face interviews and administered cognitive tests of depression, mood, and hopelessness in addition to assessments of prisoners‘ situational problem-solving abilities using hypothetical scenarios.

They found that suicidal prisoners displayed the most elevated levels of distress, followed by bully victims, then protective custody prisoners, and then regular prisoners. In terms of coping, suicidal prisoners had the largest deficit in problem-solving ability; the control group had the best problem-solving skills. Finally with respect to the relationship of negative emotions and

47 coping, Biggam and Power found that a greater level of negative emotions was correlated with deficits in prisoners‘ problem-solving abilities. Overall this study extended the research of

Zamble & Porporino (1988) on the problem-solving abilities of prisoners by comparing the coping abilities and negative emotions of groups of prisoners who are situationally distinct

(being bullied, in protective custody, and recently suicidal) from those prisoners in general population. They concluded that prisons should help prisoners—especially those who are most vulnerable in prison—to increase their repertoire of social problem-solving skills.

In general, the research on the actual coping abilities of prisoners, though somewhat meager, has demonstrated that coping ability is related to prisoners‘ psychological well-being.

Although the relationship of actual coping abilities to negative emotions is mixed and somewhat unclear, it warrants further research.

*****

If indeed, coping skills are a good predictor of prison violence and serious misconduct, it would have important policy implications. Proponents of coping, while acknowledging the existence of myriad strains that prisoners face, focus instead on the coping skills because this is a dynamic factor that prison officials can address through prison initiatives and programming. If indeed the lack of coping skills is the main difference between general population prisoners and those classified as dangerous, prisons could take proactive measures to identify those prisoners with poor coping skills, address them through programming, and reduce violence and other crime within the prison system.

The research also examines those strains that are linked to prisoner misconduct. Some of these strains are inevitable, but others may be alleviated by correctional officials.

The current research will result in several contributions to the literature. Interestingly, the few studies that have examined the effects of coping on prison misconduct have done so within the larger framework of studying prison adjustment. That is, prison misconduct has usually been seen as an indication of whether prisoners have adjusted to prison or not. None of the studies that have focused exclusively on examining the factors related to prison misconduct has included coping as a factor.

Thus the current study will be the first study focused specifically on prison misconduct that included both coping factors and prison strain that also controlled for those individual-level variables that have been found to be related to prison misconduct. Those studies of prison adjustment that did examine coping and its relationship to prison misconduct generally had much smaller sample sizes (the largest sample was 133 and most were under 100) and generally counted the number of total dreports, including those for minor misconduct offenses. Many did not use random samples, but instead purposive ones. The current study used a stratified random sample of 312 prisoners. In addition it focused on those behaviors that are considered serious and/or involves violence—that is misconduct that is most disruptive and threatening to the safety of a prison. These include such offenses as escape, possession of a weapon, extortion, rioting, threatening to inflict harm on staff or other prisoners, assault on staff or other prisoners, and fighting. Finally most studies on prison misconduct have examined prisoners in the general population of the prison often to the exclusion of prisoners in either protective custody or those in segregation. This study included those prisoners in both protective custody and those in segregation or in the High Security. No studies have measured the coping skills of prisoners who are or have been in administrative segregation. Given the growth of supermax prisons/units in this country in the last two decades, ascertaining the general characteristics of these prisoners as well as their coping abilities and their exposure to stress and strains will make an important contribution to the literature.

III. RESEARCH SETTING

THE RHODE ISLAND DEPARTMENT OF CORRECTION

This research was conducted within the Rhode Island Department of Correction

(RIDOC). The RIDOC is a ―full service‖ correctional agency in that it is responsible for every type of correctional supervision and care of adult offenders in the state. In addition to jurisdiction over all state prisons, the RIDOC also oversees probation, parole, people awaiting trial, and some juvenile facilities. The RIDOC‘s adult correctional facilities, collectively referred to as Adult Correctional Institutions (ACI) house over 3700 prisoners and include two facilities for females, and the Intake Services Center, the High Security Center, the maximum security facility, two medium security facilities, and a minimum security facility for males

(RIDOC, 2009).

SENTENCED POPULATION The overall sentenced and awaiting trial population of the RIDOC rose 526 percent between FY 1976 and FY 2009, from about 588 to 3773 prisoners. Rhode Island still has an incarceration rate (235 per 100,000 residents) that is less than half of the total United States average (506 per 100,000 residents) (RIDOC, 2010). Six percent of the total population were women (RIDOC, 2010e). Out of the 3773, 670 were awaiting trial in the Intake Service Center leaving 3103 sentenced prisoners. The average age of the sentenced prisoner was 36 years old.

RIDOC reports that the sentenced population includes 46 percent white, 29 percent black, 22 percent Hispanic, and two percent who are made up of Asian, American Indians, and ‗other‘ races/ethnicities. Forty-two percent had been sentenced for violent offenses, 12 percent for sex offenses, eight percent for drug possession, 13 percent for other drug offenses, nine percent for breaking & entering, two percent for weapons charges, and 14 percent for other non-violent

50 offenses. Sixty percent were serving sentences of six months or less (based on CY 09 statistics)

(RIDOC, 2010f). This is not surprising given that the RIDOC is responsible for all sentenced prisoners in the state and that there are no houses of correction for less serious offenders as there are in other states. At the end of FY 09, 142 offenders were serving life sentences, 27 of whom were serving life without the possibility of parole (RIDOC, 2010e). Of the 3916 new commitments of sentenced offenders to its facilities during calendar year 2009, 65 percent were newly sentenced and 29 percent were committed due to a probation or parole violation. The three-year recidivism rate (those who returned to prison) for prisoners released in 2004 was 54 percent.

DESCRIPTION OF FACILITIES IN STUDY The stratified random sample of prisoners for this study drew from the population residing in the High Security Center (HSC), Maximum Security, and the two medium security facilities. The HSC—a supermax-like facility built in 1981 that is separate but on the grounds of the RIDOC—has an operational capacity for 138 prisoners, but its average population is 90 occupants (RIDOC, 2010b). This self-contained maximum security facility houses those prisoners who for various reasons require close custody, control, and the highest security.

Costing about $157,000 per prisoner per year, this facility houses juveniles convicted in adult courts (who are prohibited by law from living in population until their 18th birthdays), protective custody prisoners, prisoners with mental health problems residing in the Observation Security

Unit (OSU) for observation and stabilization of serious mental health problems, those awaiting trial but deemed to be dangerous, as well as those convicted prisoners deemed dangerous or disruptive and classified to C or B status—the latter group most often having been involved in serious prison misconduct and violence. Because of the nature of the HSC, work opportunities

51 for prisoners are minimal. Educational and treatment programs, especially cognitive behavioral- oriented ones, are available although mandatory separation of the HSC‘s various groups of prisoners makes daily program involvement for any one group impossible resulting in each group being able to access education and programming activities from one to three times per week.

The Maximum Security facility (Maximum) was the first RIDOC prison built in 1867

(RIDOC, 2010f) and was designed on the Auburn style. It was operating above capacity (409) with an average daily population of 442 prisoners during FY 09 (RIDOC, 2010d). This facility operates at an annual cost of over $52,000 per prisoner and houses those prisoners serving long sentences for a variety of offenses as well as those prisoners who have been transferred from medium or minimum security facilities due to serious prison misconduct and violence. The facility contains an industrial area that together with other prison jobs employs almost 60 percent of its population. In addition academic and vocational education programs as well as other rehabilitative programs are available.

The John J. Moran Medium Security Facility (hereafter referred to as the Moran facility) was built in 1991 and is one of the largest facilities with a current capacity of just over 1000 prisoners. During FY 2009, it was operating with an average of 1028 prisoners and costs just under $33,000 a year (FY 08) to house a prisoner in this facility (RIDOC, 2010c). Moran also has the most educational and program opportunities for prisoners. This institution operates a rather large prison industries program, offers a number of vocational and academic educational programs, and offers a substantial variety of treatment and rehabilitative programs geared to various individual problems (substance abuse, suicide prevention, sexual abuse), offender types

(murderers‘ counseling, domestic violence, and sex offender therapy), and other

52 treatment/rehabilitative needs (cognitive behavioral groups, AIDS education and counseling, anger management, meditation, and various religious services).

The Donald Price Medium Security Facility (hereafter referred to as the Price facility) is a dormitory-style prison built in 1932 that went through various changes, including being closed and then reopened again in 1996 as a medium security facility. Price operated a little above its capacity of 324 prisoners with about 345 prisoners residing there in FY 2009 (RIDOC, 2010a).

The annual cost per offender of over $53,000 is slightly higher than either Maximum or Moran, most likely due to its dormitory-style housing, which puts large groups of prisoners together necessitating heavier reliance on uniformed staff to maintain security within the housing units.

Price does not have an industries program but does offer some of the educational and rehabilitative programs offered at the larger Moran facility. Prisoners residing in Price are also classified for medium security, but are usually deemed to need slightly less supervision and to be amenable to dormitory-style living. For many, Price is a stepping stone between Moran and transfer to minimum security.

DISCIPLINARY REPORTS, THE DISCIPLINARY PROCESS, AND SANCTIONING

RIDOC POLICY AND PROCEDURES The issuance of disciplinary reports (called ‖bookings‖ in Rhode Island), the disciplinary process, and the sanctions allowed are governed by RIDOC Policy on Code of Inmate Discipline

(RIDOC Policy and Procedure 11.01-5 DOC, effective date 3/9/09). This policy underwent a review and major revision by administrators in 2008 in order to afford staff more flexibility in their issuance of bookings and to increase the deterrent effect of the disciplinary and sanctioning process. Changes included categorizing bookings into four classes, assigning a range of sanctions to each offense category, increasing the amount of possible time in segregation from

60 days to one year, and making changes to the disciplinary and appeal processes to assure accountability and standardization across facilities. Since that large overhaul of the policy, there have been minor changes in the lengths of a few sanctions, including some that were made during the course of this study.

Officers can choose from four classes of bookings when they are confronted with prisoner behavior that warrants a disciplinary report: Class 1 – Highest; Class 2 – High; Class 3 –

Moderate; and Class 4 – Low Moderate. (See Appendix 1 for RIDOC Disciplinary Severity

Scale taken from the March, 2009 policy, which lists all offenses within each class.) Class 1 –

Highest is further divided into two categories, predatory and non-predatory. In general Class 1 bookings are those offenses which are viewed as extremely serious in that they may or have resulted in the injury of staff or other prisoners, and/or may or have resulted in seriously compromising security at the institution. For example, Class 1 predatory bookings include escape; sexual assault; arson; assault on staff, on other prisoners, or any other persons; extortion; hostage-taking; killing; mayhem; robbery; and rioting. Class 1 non-predatory bookings are categorized (as are all other classes), but the individual offenses under each category are much more serious than those offenses under a similar category in a lower class. For example, Class 1 non-predatory includes contraband (smuggling contraband into a facility); disobedience

(failing/refusing to halt when commanded to do so); disturbance (engaging in or encouraging a group demonstration); escape (possessing escape paraphernalia); mutinous act (recruiting for a gang); sexual (making sexual proposals or threats to another); substance abuse (being intoxicated); substance possession (narcotics trafficking); vandalism (tampering with a security device); and violence (altering any property to make a weapon). Class 2 – High bookings have the same general categories, but these offenses are not considered as serious or as threatening to

54 the security of the facility. For example, Class 2 categories and examples include: contraband

(possessing tattoo equipment); disobedience (failing to comply expeditiously with an instruction from a staff member); mutinous act (partaking in group petitions in any form); sexual (verbal statements that contain amorous expression and/or sexual language to any staff member); vandalism (willfully…damaging…any materials, tools, machinery, and/or any state property); and violence (making verbal threats). Classes 3 and 4 are progressively less serious, with the latter only including about ten offenses such as not making one‘s bunk properly, fishing, littering, and poor job performance.

Staff members who witness or investigate and as a result discover inmate misbehavior write an initial booking report. However, in the case of Class 4 – Low Moderate bookings, staff have the option of offering the prisoner a one- or two-nighter in lieu of receipt of a booking. A one-nighter is ―an administrative loss of institutional privileges for a period of 24 continuous hours and includes the forfeiture of visits, except attorney and RIDOC clergy; telephone, except attorney; furlough, except emergencies approved by Director or designee; recreation; on-grounds classes, school, employment, except at the Warden‘s/designee‘s discretion; out-of-cell time, except meals and/or to seek medical care; television and/or radio, except at the

Warden‘s/designee‘s discretion‖ (RIDOC, 2009, p. 2-3). A two-nighter involves the same loss of privileges except for a period of 48 hours. If a staff member chooses the one- or two-nighter option, the charge and disposition are presented to the prisoner and if the prisoner decides to accept it, he would sign a waiver of his right to appear before a Hearing Officer. He would then be given a copy of the waiver and his one- or two-nighter would commence immediately. If he refuses to sign the waiver, he would then be scheduled to go before the Hearing Officer as he would for any other booking.

With the exception of the one- or two-nighters, there are six mandatory steps associated with any booking: initial booking report, superior officer review, prisoner notification, the disciplinary hearing, administrative review, and the recording of the booking and the results of the hearing in INFACTS (INmate FACility Tracking System) (RIDOC, 2009). After the initial booking report is written, a ―timely review‖ by the superior officer or designee must take place and must ensure that the booking write-up is complete and accurate, and that the charging offense is consistent with the description of the incident (that is, that the prisoner is not charged with a higher class offense than is warranted). The superior or designee also initiates a preliminary review of the matter that may include interviews with the staff person, the prisoner charged and any other witnesses. The superior or designee then makes a recommendation to either dismiss the booking or proceed with the disciplinary hearing. Concomitantly, by policy mental health staff at each facility checks INFACTS on a daily basis for all dreports to determine whether any of them might have been a result of a prisoner‘s mental health status. If so, after consulting with the Clinical Director Psychologist, the mental health worker may approach the

Warden or Warden‘s designee to discuss the matter. The Warden may then decide to dismiss the dreport or let it proceed to a hearing.

If it is determined that a booking should proceed to a hearing, the superior officer or designee must inform the prisoner orally and in writing of the dreport no later than one day after the charge was filed and at least 24 hours before the disciplinary hearing (RIDOC, 2009). At the same time, the prisoner is asked if he would like a representative at the hearing and the prisoner‘s response is recorded. Representatives are usually an assigned correctional counselor who meets with the prisoner prior to the hearing, and helps him prepare a defense and make a presentation at the formal disciplinary hearing. The disciplinary hearing is presided over by the Hearing

Officer, a lieutenant at each facility who has been designated for such duty. Often, another lieutenant is designated as the superior whose job it is to review and investigate bookings.

Hearing Officers cannot preside over bookings that they issued and are responsible for the recording and distribution of the results of the hearing process. Prisoners are usually present at their disciplinary hearing unless they refuse to appear, unless their behavior ―gives cause for exclusion or removal,‖ and except during the time that confidential information is being presented, or the Hearing Officer recesses for deliberation.

At the hearing, the charge is read, the prisoner is asked to enter a plea: guilty, guilty with an explanation, or not guilty (RIDOC, 2009). Witnesses and evidence are allowed in the hearing at the discretion of the Hearing Officer and the entire proceedings are recorded. After presentation of witnesses and evidence (if applicable), questioning of the prisoner regarding the circumstances and motivation of the offense, and rebuttal by the prisoner (if applicable), the

Hearing Officer has several options. One option is to dismiss the charge. Another is to change the booking to a higher or lower class if it is determined that the original charge was inappropriately classified and then to find the prisoner guilty of that booking. A third option for a Class 3 or 4 booking is to issue a finding of guilt and a reprimand. The final and most prevalent option is to find the prisoner guilty of the booking and to sanction the prisoner.

Sanctions for bookings can include a referral to the Classification Board for a review of the prisoner‘s classification status (and possible downgrade to a higher security facility), temporary loss of privileges for up to 15 days, disciplinary confinement that can range from one day to one year, restitution, or any combination of these sanctions (RIDOC, 2009). The numbers of days of disciplinary confinement is dependent on the class level of the booking. Disciplinary confinement for a Class 1 – Highest Predatory booking can last from 31 days to one year; for a

Class 1 – Highest Non-Predatory, 21 to 30 days; for a Class 2 – High, one to 20 days; and for a

Class 3 – Moderate, three to 15 days. Prisoners found guilty of a Class 4 – Low Moderate booking are not eligible for disciplinary confinement, but instead may lose privileges for up to five days. In addition, by policy, prisoners lose the same number of days of good-time as the number of days in confinement or the number of days of loss of privileges. For example, if a prisoner assaults a staff person and is sanctioned with eight months in disciplinary confinement, he would also lose eight months of good-time. Prisoners who choose a one- or two-nighter in lieu of going before a Disciplinary Board for Class 4 – Low Moderate offenses do not lose good- time, thus it is a popular option since if they went to a hearing and received two days loss of privileges they would end up with the same loss of privileges as they would have received with a two-nighter but also would lose good-time.

After deliberation, the Hearing Officer immediately informs the prisoner of the decision of guilty/dismissal and the sanction and asks the prisoner if he would like to appeal the decision

(RIDOC, 2009). Whenever a prisoner is found guilty of an infraction in his absence or when the sanction is more than 30 days of disciplinary confinement, there is an automatic review by the

Warden or designee. If the prisoner is given a sanction of 30 days or less, he must submit his request for an appeal within five working days of receiving the written result of the hearing. The

Warden has 15 calendar days from receipt of an appeal to review and either confirm or revise the decision. If the sanction is more than 90 days and is upheld by the Warden, it is automatically sent to the Assistant Director of Institutional Operations (ADIO) for review within 20 days. The

Warden or ADIO can uphold the guilty decision and the sanction; can order further investigation or another hearing of the charge; can substitute an infraction at either the same or a lower class; can alter or suspend the sanction; or dismiss the booking outright. Either way, prisoners are

58 notified in writing of the final outcome of appeals and all information is made available to relevant staff and is entered into the INFACTS database.

2005 DISCIPLINARY REPORTS AND SANCTIONS To put the number of bookings within Rhode Island into broader perspective, the amount of serious prison misconduct and violence reported by Rhode Island and other jurisdictions was determined using secondary analysis of disciplinary infraction data from the Census of State and

Federal Correctional Facilities, 2005 (Bureau of Justice Statistics, 2005).2 The dataset included infractions recorded by all of the adult medium and maximum federal and state correctional facilities. The facilities were divided into two groups—Rhode Island facilities and all other medium and maximum facilities where data were either reported or estimated. The annual rate of various groups of dreports per 1000 prisoners was calculated by taking the number reported for 2005, dividing it by the number of prisoners in the facility on December 30, 2005 and multiplying by 1000. A comparison of the means of several of the dreport categories is presented in Table 1. Generally, Rhode Island‘s number of dreports issued during 2005 was on par with most other medium and maximum security facilities. There were no statistical differences between Rhode Island and other states in the number of major disciplinary infractions, staff assaults, staff deaths due to assault, prisoner assaults, or major disturbances.

2 This census is the seventh in a series of such studies that were previously conducted for the years 1974, 1979, 1984, 1990, 1995, and 2000. The census was conducted by the U. S. Census Bureau for the Bureau of Justice Statistics, U. S. Department of Justice. The census collects data on a variety of prison-level characteristics. These include the type of facility, security level, type of population, and who operates the facilities. It also includes numerous aggregate-level characteristics of prisoners and staff, testing and treatment of various diseases, misconduct data, inmate programming information, and suicide prevention. The census asked for the count of prisoners on December 30, 2005.

Fortunately during 2005 there were no staff deaths or major disturbances in Rhode Island, unlike other correctional facilities.

Table 1 - Annual Rate of DReports in 2005 - Rhode Island & Other State/Federal Prisons*

*Per 1000 prisoners Rhode Island Other Prisons

Disciplinary Report Groupings Mean SD Mean SD Major disciplinary infractions t (807, N=809)=-1.22, ρ=.223 1532 1086.42 1014 946.06 Staff assaults t (389, N=391)=-.16, ρ=.876 20 33.54 14 76.52 Staff deaths via assault t (836, N=838)=.08, ρ=.932 .00 .00 .02 .56 Prisoner assaults t (811, N=813)=-.74, ρ=.460 47 30.62 27 60.52 Major disturbances t (806, N=808)=.19, ρ=.844 .00 .00 .35 4.35

Office of Justice Programs, Bureau of Justice Statistics: Census of State and Federal Correctional Facilities, 2005 * ρ<.05 ** ρ<.01 *** ρ<.001

Table 2 shows the distribution of disciplinary infractions reported in the census for both

2000 and 2005 for the four facilities included in this study. Clearly HSC prisoners received the largest number of dreports for major infractions and staff assaults in both 2000 and 2005. The greatest number of prisoner-on-prisoner assaults occurred at the Maximum Security facility.

Among the four facilities, the lowest number of dreports was given out at the Moran facility.

With only a couple of exceptions (most notably, major infractions at Price), the number of dreports given out at each facility dropped from 2000 to 2005. It should also be noted that these figures for dreports precede the implementation of the revised disciplinary policy in 2008.

Table 2 - Rate of Disciplinary Reports – Rhode Island Medium and Maximum Prisons* Major Disciplinary Staff Assaults Prisoner Assaults Facilities Infractions

*Per 1000 prisoners 2000 2005 2000 2005 2000 2005

High Security Center 8403 3000 104 70 65 81 Maximum Security 1652 912 10 5 72 57 Medium 1 - Moran 697 510 3 1 9 3 Medium 2 - Price 1827 2367 0 0 23 0

Office of Justice Programs, Bureau of Justice Statistics: Census of State and Federal Correctional Facilities, 2005 Office of Justice Programs, Bureau of Justice Statistics: Census of State and Federal Correctional Facilities, 2000

Table 3 again uses data from the Census of State and Federal Correctional Facilities to examine the number of prisoners confined in restricted housing units on December 30, 2005.

The report also differentiates between prisoners who are housed in disciplinary segregation

(often referred to as punitive segregation) and those housed in administrative segregation (often referred to as special management housing, control units, or supermaxes). The former have been found guilty of a dreport and have been sanctioned to segregation as a punishment. The latter is an administrative or classification determination, possibly based on dreports, but also based on the assessed dangerousness and disruptiveness of the prisoner. Each facility in the RIDOC has a punitive segregation unit. The HSC houses administrative segregation units as well as disciplinary segregation for those administrative segregation and protective custody prisoners who are also serving punitive segregation. The mean number of prisoners per 1000 housed in protective custody in Rhode Island was higher (mean=47) than in other adult prisons (mean=12).

The differences were statistically significant. The mean number reported to be housed in disciplinary segregation units was also higher in Rhode Island (mean=116 per 1000 prisoners) compared to other adult facilities (mean=34 per 1000 prisoners). The differences were again statistically significant. However, there were no statistically significant differences in the

61 number of prisoners housed in administrative segregation in Rhode Island compared to other prison systems.

Table 3 - Rate in Restricted Housing – Rhode Island & Other State/Federal Prisons, 2005

Medium and Maximum Prisons Rhode Island Other Prisons

Restricted Housing Types Mean SD Mean SD Protective custody t (460, N=462)=-2.26, ρ=.024* 47 104.00 12 33.06 Disciplinary segregation t (598, N=600)=-2.54, ρ=.011* 116 74.48 34 71.74 Administrative segregation t (564, N=566)=.39, ρ=.699 28 62.28 45 96.08

* ρ<.05 ** ρ<.01 *** ρ<.001

*****

In summary, the Rhode Island Department of Correction‘s issuance of dreports to prisoners during 2005 in its medium and maximum facilities was similar to facilities in other states and the federal jurisdiction. Within Rhode Island the number of dreports decreased from

2000 to 2005 in most categories and in most facilities. Prisoners in the HSC received the largest number of dreports for major infractions and staff assaults in both 2000 and 2005; prisoners in

Maximum Security received the most dreports for assaults on prisoners. Finally, Rhode Island

DOC prisoners appeared to be more likely to serve time in protective custody and segregation than prisoners at other facilities. It appears that the RIDOC administration was more likely to use punitive segregation to deal with misbehavior, rather than administrative segregation.

IV. METHODOLOGY, DATA, AND ANALYTIC STRATEGY

As mentioned above, the research study used a mixed methods research design combining qualitative with quantitative research methods. This type of research design has many benefits including: that biases inherent in one method could neutralize or cancel the biases in other methods; that triangulation provides a means for finding convergence across the two types of methods; and that data from one method can help explain the other. Specifically the study used a concurrent data collection strategy, meaning that both quantitative and qualitative data were collected simultaneously. While the quantitative data collection—from prisoner records and prisoner self-report surveys—was the dominant approach, field research was ongoing with prisoners and staff in the HSC, maximum security, and the two medium security facilities within the Rhode Island Department of Correction.

The quantitative data collection involved the use of prisoner self-report surveys and the extraction of socio-demographic, criminal history, and current prison variables from the RIDOC prisoner database. These data were collected on a stratified random sample of current RIDOC prisoners (see details of sampling procedures below). These data were used to test the previously-stated hypotheses regarding whether prisoners‘ ways of coping affected their involvement in serious prison misconduct and violence. The qualitative data collection focused on those prisoners who were involved in serious misconduct and violence. Field research included attendance at classification and disciplinary meetings and in-depth interviews with prisoners involved in serious prison misconduct and violence and staff working with those prisoners in medium, maximum, and the HSC.

SAMPLING

SAMPLING IN PRISON MISCONDUCT AND COPING RESEARCH The majority of the research on prison misconduct and violence focuses on the general prison population and differentiates between those prisoners who have received dreports for serious prison misconduct and those who have not. The research studies that examined both prisoner-level and prison-level factors associated with higher disciplinary histories drew large samples and collected data generated from correctional databases (Gaes & McGuire, 1985;

Wooldredge, et al., 2001; Gaes, et al., 2002; Camp, et al., 2003). However studies that collected data from self-report surveys generally did so with a smaller number of general population prisoners, excluding those prisoners housed in segregation units and protective custody (Zamble

& Porporino, 1988; Sappington, 1996, Harreveld, et al., 2007). As Liebling (1999) pointed out in her study of prisoner suicide, excluding these types of prisoners results in an inherent bias since prisoners in these types of units are often most at-risk for aberrant behavior. There have been a few studies that have examined coping and or prison misconduct in special populations of prisoners—those undergoing correctional treatment, those who are deemed mentally ill, and those who are suicidal, bullied, or under protective custody (McDonald, 2006; Toch, & Adams,

1986; Toch, Adams, & Grant, 1989; Toch, 1975; Liebling, 1999; Biggam, 1999 respectively).

Toch and his colleagues (1989) estimated from their study of maladaptation that three percent of prisoners were chronically disruptive, accounting for twelve percent of adjudicated disciplinary infractions, 13 percent of time in segregation, and nine percent of time in special housing disciplinary units (p. 33). When the semi-chronic disruptive prisoners were included, then 12.5 percent of the prisoner population accounted for 34 percent of disciplinary infractions, 35 percent of time in segregation and 32 percent of time in special housing disciplinary units.

QUANTITATIVE SAMPLE The research used a stratified random sample of male RIDOC prisoners from the population of prisoners housed in the HSC, the maximum security facility, and the two medium security facilities. Prisoners in the Intake Service Center were excluded because most have not been incarcerated for long enough to have been involved in serious misconduct. Those in minimum security were excluded for three reasons. First, prisoners who seriously misbehave are most likely to be living in either maximum or medium security facilities and were thus drawn from those facilities. Second, many prisoners in minimum security have sentences of less than one year—the equivalent of county jail populations in other states. Finally, inclusion of those in minimum security would have been problematic because their sentences are so short that most chosen in the sample would have been released by the time self-report surveys were administered. The current study also excluded those who had not yet been convicted but were, for whatever reason, residing in these facilities. The target population was all of those male prisoners who were residing in HSC, maximum security, and the two medium security facilities in the RIDOC on May 19, 2009. Prisoners under the age of 18 years old were excluded since they cannot provide legal consent for themselves. There were no exclusions based on race, ethnicity, or any other researcher-chosen criteria.

The target population (prisoners residing in RIDOC‘s medium and maximum facilities) was stratified into two groups, based on whether or not they had received a serious dreport during 2008. Seriousness of the dreports was based on the RIDOC‘s four levels of dreports:

Class 1 – Highest; Class 2 – High; Class 3 – Moderate; and Class 4 – Low Moderate. The study used the Class I Highest Predatory and Class I Highest Non-Predatory dreports for its sampling frame. (A more in-depth explanation of the levels of dreports was outlined above in the section

65 on the Disciplinary Process.) The target population‘s two groups included those prisoners who had not received a Class I Highest dreport and those prisoners who had received that level of dreport during 2008. Since the prisoners who had received a Class I Highest Predatory or Non-

Predatory dreport made up a much smaller proportion (440) of the overall population (1888), that group was oversampled. Once the population was stratified into these two groups, a random sample was extracted from each group. Toch and his colleagues similarly oversampled prisoners who were chronically or semi-chronically disruptive (1989, p. 5). Oversampling is usually carried out when a study is focusing on a group of people that makes up a very small proportion of a population. The oversampling deliberately samples a much higher proportion of this rare group than the rest of the population. This is done to obtain reasonable estimates about the characteristics of the group. A random sample of the entire population would not yield sufficient cases. In this case, prisoners with serious dreports make up a small proportion of the prisoner population, so oversampling is warranted.

Analysis of an adequate sample size for multivariate regression was completed using both sample size calculators from the web and guidelines for selecting sample sizes (Cohen, 1988).

Two simplistic calculators3 yielded a sample of 319, using the population size of 1888 prisoners in HSC, and the medium and maximum facilities, a significance criterion (ɑ) of .05, and a confidence interval of five percent. However, Cohen (1988) suggests using more criteria to calculate sample size: a significance criterion, an effect size, power level, and the number of predictors. More control variables were significant in the bivariate analysis than anticipated and thus the multivariate analysis had up to 29 predictors. Using a sophisticated web sample

3http://www.surveysystem.com/sscalc.htm; http://www.custominsight.com/articles/random-sample-calculator.asp

66 calculator4 with a significance criterion (ɑ) of .05, a power of .80 (the convention proposed for general use), a medium effect size of .15, and the use of 29 predictors yields a necessary sample of 184. Of course, this minimum sample size is the required size of the sample for analysis and was easily met. However, at the time that the sample was extracted the more conservative estimate of about 300 was used. But it did not take into account sample attrition due to prisoners being released from the RIDOC, transferred to other states or to minimum, and refusals to take part in the research. It was assumed that if there was a 40 percent response rate and that other types of attrition (left facility, too dangerous to interview) reached another 10 percent then the sample should be about 600. Therefore a random sample of 300 prisoners was taken from the population of 1448 residing in the aforementioned RIDOC facilities on May 19, 2009 who had not received a Class I Highest Predatory or Non-Predatory dreport during 2008. Similarly a random sample of 300 was taken from the group of 440 prisoners residing in the same facilities who had received a Class I Highest dreport during 2008.

Prisoners who moved from one facility to another were not dropped unless they were transferred to minimum security, to the Intake Service Center, or to another state. Thus if a prisoner was residing in maximum security at the time the sample was drawn but was no longer there by the time survey administration was underway at maximum, then he would be followed to whatever facility he was transferred to be surveyed. A few prisoners were transferred multiple times between several facilities within a four to six week period. Fortunately access to the

RIDOC INFACTS database facilitated the process of finding and following transferred prisoners.

About a dozen prisoners were transferred continually during the interviewing process and could

4 http://www.danielsoper.com/statcalc/calc01.aspx

67 not be contacted despite multiple visits at various times of day. These were placed in the refused/no show category. There were also between one and three prisoners in the sample in each of the four facilities (Moran, Price, Maximum, and the HSC) whom staff believed to be inappropriate for surveying, based on the prisoner‘s dangerousness or mental health status.

There were also less than five prisoners who were in the prison‘s hospital unit who were not surveyed. Both of these types of non-surveyed prisoners are also included in the refused/no show category below.

Below Table 4 breaks down the survey sample by facility where the prisoner was residing at the time of the interview as well as the reasons for those who were not interviewed. Overall a smaller percentage of prisoners refused participation than anticipated at the planning stage and a larger percentage transferred out of the facility or was released. As can be seen 34 percent of the overall sample refused to participate in the survey or did not show up at the interview site when they were called. This included those prisoners, mentioned above, who had been transferred and were at another facility when interviews were taking place, those in the hospital, and those whom staff believed to be too dangerous or ill to be surveyed. The refusal rate ranged from a low of 24 percent at Price to a high of 42 percent at the Maximum Security Facility (not shown). Only two percent were transferred to the minimum security facility, the Intake Service Center within the

RIDOC, or to an out-of-state facility. One prisoner in this category died from natural causes.

Thirteen percent completed their sentence and were either released to the street or released into the custody of another state or to the Immigration and Naturalization Service (INS) or were paroled home. Finally 312 prisoners (52 percent) completed the survey. Overall, 14 percent of the prisoners surveyed were from Price, 50 percent were from Moran, 30 percent were from

Maximum Security, and six percent were from the HSC (not shown). The table differentiates

68 between those prisoners who had received a Class 1 dreport in 2008 (D) and those who did not

(NoD).

Table 4 - Response Rate of Sample by Facility Refused/ Transferred/ Paroled/ Participated Facility No Show Other Released in Survey Total

D NoD D NoD D NoD D NoD

n n n n n n n n n Total (%) (%) (%) (%) (%) (%) (%) (%) (%) %

Price 9 12 1 1 10 10 20 23 86 (14) (9.2) (11.5) (20.0) (25.0) (27.0) (25.0) (12.3) (15.3)

Moran 28 62 2 2 16 18 72 84 284 (47) (28.6) (59.6) (40.0) (50.0) (43.2) (45.0) (44.4) (56.0)

Maximum 54 29 1 0 9 12 53 40 198 (33) (55.1) (27.9) (20.0) (0.0) (24.3) (30.0) (32.7) (26.7)

HSC 7 1 1 1 2 0 17 3 32 (5) (7.8) (1.0) (20.0) (25.0) (5.4) (0.0) (10.5) (2.0)

Sub-total 98 104 5 4 37 40 162 150 600 (100) (16.3) (17.3) (0.8) (0.7) (6.2) (6.7) (27.0) (25.0) (100)

Total 202 9 77 312 600 (34%) (2) (13%) (52%) (100)

Since the data were stratified into two groups and one of the groups (those involved in serious disciplinary misconduct in 2008) was oversampled, the dataset needed to be weighted to make up for the oversampling. Weighting the data requires determining the weight by which each of the two groups should be multiplied in order to approximate the target population. There were 1448 prisoners with no Class I dreports during 2008 and 440 who did have Class I dreports during that year. Since 1448 is 76.7% of the total population of prisoners in medium, maximum, and the high security facilities (1888), the percentage of the sample that have no dreports should have been similar or 239 prisoners. In fact, 151 prisoners with no such dreports were surveyed.

Therefore, 239 was divided by 151 to derive the necessary weight of 1.58. The 440 prisoners with 2008 Class I dreports made up 23.3% of the prison population and should make up a similar percentage of the sample or 73 prisoners. The actual number of prisoners in the sample with

Class I dreports during 2008 was 161 prisoners. Therefore, 73 was divided by 161 to derive the necessary weight of .45. A weight variable was then created which gave the weight of 1.58 to the data of those prisoners who had not received Class I dreports in 2008 and .45 to the data of those prisoners had received such dreports. All of the multivariate analysis was conducted and reported on the weighted data.

QUALITATIVE SAMPLING The qualitative research was conducted in the two medium security facilities, the maximum security facility and the HSC. This included observation and in-depth interviews with both prisoners and staff. In-depth interviews were conducted with 25 prisoners who were reported to have been involved in serious prison misconduct and/or violence during 2008 and who indicated on their survey a willingness to be interviewed at a later date. A question was placed at the end of the survey asking all prisoners if they would be interested in participating in an in-depth interview. Though no such interviews were planned for those without a serious dreport during 2008, 91 of 150 of these prisoners agreed to an interview (61 percent). Of the 162 prisoners with a serious dreport during 2008 that completed the survey, 100 prisoners (62 percent) agreed to an in-depth interview. This resulted in a non-probability sample that while not random, was at least purposive. Care was taken to interview prisoners at HSC, Moran, and

Maximum where almost all of these prisoners resided and to ensure that various age groups and races/ethnicities were represented. After initial perusal of survey results, observation of disciplinary hearings, and many staff interviews, it was decided to try to choose prisoners who

70 also made up certain categories that were associated with high disciplinary problems. These included young prisoners, those belonging to a street gang, those who had previously been adjudicated delinquent and committed to a juvenile correctional facility, those with prior mental health problems, and those who had abused drugs. Clearly many of these categories overlapped, but attempts were made to interview four to five prisoners from each of these categories while also being representative of age, race/ethnicity, and facility.

In-depth interviews were also conducted with 26 staff members from the two medium facilities, from maximum, and from the HSC. Staff members were chosen who had dealings with prisoners with high levels of serious misconduct and violence. The interviewees included most of the deputy wardens, both wardens, lieutenants who conducted disciplinary boards, correctional counselors, and various uniformed staff at the HSC.

OPERATIONALIZATION OF THE VARIABLES

VARIABLES IN THE QUANTITATIVE RESEARCH DEPENDENT VARIABLE The dependent variable for this study was serious prison misconduct or violence. The majority of studies on prison misconduct have operationalized prison misconduct by examining the number of disciplinary infractions committed by prisoners. Thus researchers have collected data on the number of dreports for which prisoners were found guilty. The studies have either examined misconduct in general (Zink, 1958; Johnson, 1966; Forst & Brady, 1983; Wooldredge, et al., 2001) or aggressive misconduct such as assaults (Ellis, et al., 1974; Gaes & McGuire,

1985; Harer & Langan, 2001). A few have studied two or more specific types of infractions, such as assaults and substance-abuse related infractions (Harer & Steffensmeier, 1996) or infractions broken down into several categories including violence, property, drug, security-

71 related offenses (interfering with daily operations), accountability (escapes), and other misconduct (Camp, et al., 2003). Gaes and McGuire (1985) took into consideration other factors, namely whether the person assaulted was staff or another prisoner and whether or not a weapon was used, to create four categories of assaultive behavior.

There is agreement among researchers in this field that reported incidents of misconduct and violence resulting in dreports represent only a fraction of actual misconduct and violence in prisons (McCorkle, 1992; Bottoms, 1999). Bottoms discusses two ―filtering points‖ in the reporting of prisoner misconduct: prisoners who refrain from reporting most victimization and correctional staff who may or may not officially document prisoner misconduct. Much of the research has discussed the prisoner code or culture that prevents prisoners from reporting victimization and other misconduct to prison authorities. In their description of the disciplinary process leading to lock-up in a segregation unit, Rocheleau, Forcier, and Jackson (1998) discuss the various discretionary decision points in which prison staff might decide to deal with the misconduct informally. These include the decision to write up a dreport by prison staff who first learn of the incident; the decision by prison administrators to have the incident remain in the formal disciplinary process; and the disciplinary hearing where the report can be dismissed or the prisoner found guilty or innocent. However, most researchers while acknowledging the weaknesses of using dreports, especially their underestimation of prison misconduct, have concluded that it is the best proxy measurement for prison misconduct. One alternative measure of prison misconduct is the use of self-report data elicited through surveys or in-depth interviews

(Sieverdes & Bartollas, 1986; Smith, 1988). However, this method is not as cost-effective and relies on prisoners‘ memory to recall potentially many separate incidents, resulting in questions about validity.

Very recently, two studies have examined those prisoners who have been classified to administrative segregation for disciplinary issues (Lovell, et al., 2000; Peck, Jr., 2004). They used the classification of prisoners to long-term segregation units or ‗supermax‘ units as a proxy for involvement in serious prison misconduct and violence. One potential weakness of this construct is that it relies on the discretionary decision-making of staff to determine who is labeled as disruptive and dangerous. However, Lovell and his colleagues (2000) found that prisoners in the Intensive Management Units, not surprisingly, had committed more major prison infractions (7.7 on average) than prisoners in the general population (.9 on average). Of course the usefulness of this classification measure as a reliable construct depends on the factors taken into account by prison staff in their classification decisions.

The dependent variable in this study was operationalized into three measures in order to be able to take into account various dimensions of serious misbehavior, including quantity, seriousness, and whether it was violent. One continuous measure (annual misconduct score) takes into account both the frequency and seriousness of most dreports. Another continuous measure restricts the count to only dreports that are violent since it is possible that prisoners who are violent may have differing characteristics than those that are involved in serious misconduct in general. The third measure divides the sample into two groups (those with a high annual misconduct score and those with a low one) for use in the bivariate analysis. All three measures used prisoners‘ disciplinary records to count the number of dreports for which prisoners were found guilty. The first measure hereinafter referred to as the ―annual misconduct score,‖ took into account both the frequency and seriousness of various offenses by assigning points to certain categories of dreports received during 2008 and 2009, adding in points for offenses categorized

73 as ―violent,‖ summing the points, and then taking the annualized rate for the two-year period.

Thus:

Annual Misconduct Score= (((# highest predatory offenses * 6) + (# highest non-predatory offenses * 5) + (# high offenses * 3) + (# violent offenses against staff or prisoners * 3) + (# other violent offenses * 1)) / # days incarcerated during 2008 and 2009) * 365.

The second measure of the dependent variable is a dichotomous variable that divides the annual misconduct score into two groups: the low-level disciplinary group and the high-level disciplinary group. Those in the low-level group had an annual misconduct score ranging from zero to three, while those in the high-level group had an annual misconduct score of four or more. This cut point resulted in two groups that were roughly equal in size and ensured that none of those in the low-level category had received a Class 1 Highest dreport.

The third measure of the dependent variable was the annual rate of violent dreports.

Thus:

Annual Rate of Violent Dreports = (# violent Dreports received in 2008 and 2009 / # days incarcerated during 2008 and 2009) * 365.

The annual misconduct score and the annual rate of violent dreports were both used in the multivariate analysis. The dichotomous low- vs. high-level disciplinary group variable was used in the bivariate analysis. There were no missing data for any of these dependent variables.

INDEPENDENT VARIABLES – MAIN VARIABLES The current study tested whether or not a prisoner‘s possession of positive coping skills or negative ways of coping affected his involvement in serious prison misconduct or violence.

Thus the main independent variables needed to measure the various aspects of prisoners‘ coping

74 skills. In addition the research incorporated three trait emotions—anger, depression, and anxiety—into the model. These are the three emotions that are most often mentioned in the coping literature. They are discussed below along with the instruments used to measure them.

Coping Skills

The study tested whether a prisoner‘s coping skills were a good predictor of involvement in serious prison misconduct or violence. The literature on coping by various sub-groups and in various environments is quite extensive (for example Folkman & Lazarus, 1985; Sarason,

Shearing, Pierce, & Sarason, 1987; Carver, Scheier, & Weintraub, 1989; Amirkhan, 1990;

Jackson, 1992; D‘Zurilla, & Chang, 1995; Sorlie & Sexton, 2001). A number of different instruments have been used in both psychological and criminological research to measure coping skills5 and most of them contain sub-scales that measure various dimensions of coping including positive or negative problem-solving, avoidance, use of social support, social withdrawal, and emotional coping, among others. Instruments to measure coping can be broken into two types: those instruments that ask the respondent to think of an event or incident that required the respondent to cope and then ask questions related to that event; or a series of questions about various dimensions of how a person copes in general. Carver, Scheier, & Weintraub (1994) have labeled the first type situational coping and the latter dispositional coping. Situational coping has been used in prison studies to measure how prisoners cope with specific types of stresses and to compare prisoners‘ coping abilities before and after participation in some kind of correctional

5 In addition to the proposed instruments, others include: 1) Ways of Coping – Folkman & Lazarus, 1985; 2) Construction Thinking Inventory - Epstein, S., & Meier, P. (1989). Constructive thinking: A broad coping variable with specific components. Journal of Personality and Social Psychology, 57, 332-350; 3) Coping Strategies Inventory – Tobin, D. L., Holroyd, K. A., Reynolds, R. V., & Wigal, J. K. (1989). The hierarchical factor structure of the Coping Strategies Inventory. Cognitive Therapy and Research, 13, 343-361.

75 treatment program such as a boot camp or substance abuse treatment (Soderstrom, et al., 2001;

McDonald, 2006). Indeed Soderstrom and her colleagues used the Coping Strategies Inventory, an instrument that measures situational coping abilities, with prisoners prior to and after completion of a boot camp program with great success. They also used a dispositional instrument, the Social Problem Solving Inventory, with the same prisoners. While they believed that the latter instrument was too long and the questions somewhat repetitive, they concluded that researchers could use either type of instruments, situational or dispositional, with a prison population.

This study used the COPE Inventory, a multidimensional coping inventory developed by

Carver, Scheier, and Weintraub (1989) based on various elements and principles of coping theory. Depending on how questions and responses are worded, it can be used as a dispositional or situational instrument, although most studies have used it to measure the general coping disposition of individuals. Carver et al. (1989) wanted to expand on the work of Folkman and

Lazarus (1980 and 1985) to elicit more specifics about emotion-focused and problem-focused coping. The original COPE Inventory contained 13 subscales; two others were added after the original instrument was developed for a total of 15 subscales.

Several of the subscales measure elements of emotion-focused coping that take place during the primary appraisal. These include positive reinterpretation and growth, denial, acceptance, and use of social support for emotional reasons. Positive reinterpretation and growth is a tendency to manage one‘s initial emotional reactions to a stressor in a positive manner before actually dealing with the stressor itself. How one construes the situation will ultimately impact further problem-focused coping. The COPE Inventory operationalizes denial as the ―refusal to believe that the stressor exists or of trying to act as though the stressor is not real‖ (Carver, 1989,

76 p. 270). Acceptance can be conceived as a coping response, but it can also take place during primary appraisal when a person accepts the reality of their stressor or situation. Many of the coping instruments include a subscale for seeking social support. The COPE Inventory created two subscales of social support—one that focuses on the use of emotional social support and another on the use of instrumental social support. The former, use of emotional social support, usually takes place during primary appraisal, although it could continue throughout the coping process. When people seek out emotional social support, they are looking for sympathy, understanding or ―moral support.‖

Many of the subscales measure various types or elements of problem-focused coping skills that take place during the secondary appraisal. These include planning, suppression of competing activities, restraint coping, use of instrumental social support, focusing on and venting emotions, behavioral disengagement, and mental disengagement. Planning is thinking about how to cope with a stressor, ―what steps to take and how to handle a problem.‖ Another subscale measures suppression of competing activities—that is, putting aside other issues, activities, worries, or projects that might impede or distract from the stressor at hand. Thus a person who suppresses competing activities would prioritize addressing the problem at hand. One of the more interesting subscales that might be relevant for prisoners is restraint coping. Carver et al. define this as ―waiting until an appropriate opportunity to act presents itself, holding one-self back, and not acting prematurely. This is an active coping skill in the sense that the person‘s behavior is focused on dealing effectively with the stressor, but it is also a passive strategy in the sense that using restraint means not acting‖ (1989, p. 269). The other social support subscale, the use of instrumental social support refers to the seeking of advice, help, or information to address one‘s problem. Three of the subscales measure problem-focused coping that is

―maladaptive.‖ The first, focusing on and venting of emotions, occurs when a person focuses more on one‘s feelings about a problem than on addressing the problem at hand. While Carver and his associates acknowledge that sometimes focusing on the distress itself can be functional, it is more likely that it impedes progress and distracts people from active coping. The two other scales that he labeled as maladaptive are behavioral disengagement and mental disengagement.

Behavioral disengagement is ―reducing one‘s effort to deal with the stressor, even giving up the attempt to attain goals with which the stressor is interfering‖ (1989, p. 269). It is associated with helplessness and often occurs when a person expects a poor coping outcome. Mental disengagement is similar to behavioral disengagement, but it usually occurs when behavioral disengagement is untenable. It refers to a wide variety of activities and tactics that serve to distract the person from thinking about the problem at hand. These include using alternative activities to take one‘s mind off a problem, daydreaming, sleeping, or excessive television viewing. Thus the former involves disengaging one‘s behavior from doing anything about the problem at hand, while the latter involves actually becoming involved in other types of behavior that will also take one‘s mind off of the problem.

Finally, the COPE Inventory includes active coping, which is very similar to Lazarus and

Folkman‘s general problem-focused coping and refers to the ―process of taking active steps to try to remove or circumvent the stressor or to ameliorate its effects...(and) includes initiating direct action, increasing one‘s efforts, and trying to execute a coping attempt in stepwise fashion‖

(1984, p. 268). This subscale refers to behavior that takes place during the actual coping process itself. Another subscale, religious coping assesses whether a person turns to ―religion in times of stress.‖ It is a single scale that measures whether people turn to religion in general rather than determining whether religious-coping is emotional or instrumental. The two added subscales

78 measure whether a person uses humor to aid them during stressful times and whether they engage in substance use— either alcohol or drugs—during times of stress.

Carver and his colleagues (1989) tested the internal consistency of the 13-subscale original version of the COPE with two samples of college students. The Cronbach‘s alpha reliability coefficients ranged widely from .45 to .92. The lowest Cronbach‘s alpha was for mental disengagement (.45). This subscale includes various activities that one might do to take one‘s mind off of the situation at hand so the authors were not surprised that they did not hang together as a scale. Although four of the subscales had Cronbach alpha reliability coefficients

(.70 - .79) that were in the acceptable range (Nunally & Bernstein, 1994) and three were very good (.80 - .92), five other scales were under .70. However most researchers using the COPE or other instruments that measure coping tend to use factor analyses that result in distinct factors such as problem-focused coping, emotion-focused coping, and negative/denial coping or simply positive coping and negative coping strategies (Folkman & Lazarus, 1985; Carver, et al., 1989;

Soderstrom, et al., 2001; Fritch, 2006; Hyman, Paliwal, and Sinha, 2007). The COPE does not have a global score as it contains subscales of coping strategies that are both adaptive and maladaptive. Instead it is meant to reflect the plethora of coping abilities (or disabilities) that people turn to in times of stress. Carver and his associates (1989) have noted that many of the positive coping skills are highly correlated, as are the negative ones. Other ones might be correlated to both or neither. For instance, seeking social support can be positive especially if one is seeking advice on how to address a problem; but might be negative if one is dependent on others to address the problem for them.

Although the COPE, like other coping instruments has been used mostly within the health field, it has been used on populations of offenders and substance abusers. Fritch (2006)

79 successfully used the Brief COPE (an abbreviated version of the COPE Inventory) with one hundred incarcerated women in a western state prison to assess the relationships among trauma, coping, and mental and physical health. Using 14 of the subscales (mental disengagement was excluded) she conducted a principal components analysis which yielded four separate factors: adaptive coping, maladaptive coping, substance use, and use of religious coping. She eventually settled on using just the adaptive (alpha = .85) and maladaptive (alpha = .65) factors to test her hypothesized relationships. More recently Hyman, Paliwal, and Sinha (2007) used the full

COPE Inventory to study the associations between a personal history of child abuse and neglect, and the current perceived stress and coping styles of both male and female cocaine dependent adults in treatment. Hyman and his colleagues divided the COPE‘s original 13 subscales into three categories: problem-focused coping (active, planning, suppression of competing activities, restraint, and use of instrumental support); emotion-focused coping (use of emotional social support, positive reinterpretation, acceptance, denial, religious); and avoidant coping (focus on and venting of emotions, behavioral disengagement, and mental disengagement). In their sample the Cronbach‘s alpha reliability coefficients were high for each of the three categories— problem-focused coping (alpha = .89), emotional-focused coping (alpha = .83), and avoidant coping (alpha = .70).

The dispositional version of the inventory asks respondents to think about what they usually do when they experience a stressful event. Respondent choices include: 1) I usually don‘t do this at all; 2) I usually do this a little bit; 3) I usually do this a medium amount; and 4) I usually do this a lot. The current study used the items from the original 13 subscales, plus the new subscale on humor but not the subscale on alcohol use. Since alcohol is considered to be

80 contraband in prisons, it is either unlikely to be a coping option, or at least unlikely to be reported as a coping option since possession of it could result in a dreport.

As mentioned in developing the COPE Carver and his colleagues (1989) created thirteen separate subscales. However his factor analysis resulted in eleven loadings because two sets of items loaded onto the same factor. Although he had separated emotional social support from instrumental social support, they loaded on the same factor. Similarly, planning and active coping also loaded on the same factor. His second-order factor analysis based on the scores of the thirteen subscales yielded four factors that he labeled task, emotion, avoidance, and cognitive coping. In his description of the COPE, Carver explains that there is no global score but suggests that if researchers would like to narrow down the number of subscales that they create second-order factors, as he did, from among the scales using their own data. However, subsequent analysis of the dispositional items of the COPE by other researchers (Ingledew,

Hardy, Cooper, & Jamal, 1996; Lyne & Roger, 2000) have concluded that those using the COPE should factor analyze all of the individual items rather than assuming that the original COPE subscales are the best solution for one‘s data.

Table 5 presents the results from a series of Principle Components analyses using

Varimax Rotation. This factor analysis yielded eight separate scales. Five of these scales, acceptance, turning to religion, focus on and venting of emotions, denial, and use of humor mimic Carver‘s COPE scales (1989) and each contained four items. However, one of the items of the focus on and venting of emotions scale (hereinafter referred to as ‗venting emotions‘ in the text) loaded by itself on its own factor but was combined with the other three items that had loaded together because it was included in Carver‘s COPE scales, and also because reliability

Table 5 - COPE Scales for this Study: Items Listed with Loadings on their Factor Scale Name and Items Loading Scale Name and Items Loading

Seeking emotional and instrumental social Active coping activities support I ask people who have had similar .59 I take additional action to get rid of .58 experiences what they did. the problem. I try to get advice from someone about .57 I concentrate my efforts on doing .45 what to do. something about it. I talk to someone to find out more .56 I do what has to be done, one step at a .60 about the situation. time. I talk to someone who could do something .50 I take direct action to get around the .44 concrete about the problem. problem. I talk to someone about how I feel. .78 I make a plan of action. .55 I try to get emotional support from .60 I try to come up with a strategy about .61 friends or relatives. what to do. I discuss my feelings with someone. .78 I think hard about what steps to take. .71 I get sympathy and understanding from .49 I think about how I might best handle .72 someone. the problem. Turning to religion I put aside other activities in order to .52 concentrate on this. I seek God‘s help. .91 I focus on dealing with this problem, .59 and … let other things slide a little. I put my trust in God. .85 I keep myself from getting distracted by .49 other thoughts or activities. I try to find comfort in my religion. .86 I try hard to prevent other things from .73 interfering with my efforts …... I pray more than usual. .84 I force myself to wait for the right .57 time to do something. Humor I make sure not to make matters worse by .59 acting too soon. I laugh about the situation. .73 I restrain myself from doing anything .49 too quickly. I make jokes about it. .84 I look for something good in what is .62 happening. I kid around about it. .80 I try to see it in a different light, to .52 make it seem more positive. I make fun of the situation. .85 I learn something from the experience. .43 I try to grow as a person as a result of .46 the experience. Acceptance Focus on and venting of emotions

I learn to live with it. .71 I get upset and let my emotions out. .76 I accept that this has happened and that it .48 I let my feelings out. .67 can‘t be changed. I get used to the idea that it happened. .65 I feel … emotional distress & find .86* myself expressing those feelings a lot. I accept the reality of the fact that it .79 I get upset, and am really aware of it. .80 happened. Mental and behavioral disengagement Denial

I watch TV or movies, to think about it .70* I refuse to believe that it has .72 less. happened. I day dream about things other than .71* I pretend that it hasn‘t really .77 this. happened. I admit to myself that I can‘t deal with it, .62 I act as though it hasn‘t even .76 and quit trying. happened. I reduce the amount of effort I‘m putting .56 I say to myself ―this isn‘t real.‖ .66 into solving the problem. I sleep more than usual. .42 I give up the attempt to get what I want. .76 I just give up trying to reach my goal. .75

*Loaded on a separate factor but was combined with main factor.

82 analysis of that scale demonstrated that the additional item strengthened the overall scale. The next scale, seeking emotional and instrumental social support, also mimicked Carver‘s COPE scales in that all eight items loaded onto one factor, even though he eventually separated them into their emotional and instrumental components. However, this study will use them as one scale and they will hereinafter be referred to as ‗getting social support.‘ The items on the behavioral and mental disengagement scale loaded onto two separate factors, five on one and three on another. However, they were combined, again because they were very similar and two of them (watching TV and daydreaming) strengthened the scale according to the reliability analysis. The other, (I turn to work on other substitute activities to take my mind off things) weakened the scale and was thus eliminated from it. The final coping scale for this study, labeled as active coping activities and referred to in the text as ‗active coping,‘ is made up of the items in five of Carver‘s subscales. These include his subscales of active coping, planning, suppression of competing activities, restraint coping, and positive reinterpretation and growth.

The first four are all part of what Carver referred to as the problem-solving phase, while positive reinterpretation and growth takes part during emotion-focused coping. All of the items in these five subscales of his loaded on this 19-item factor except for one item from his restraint coping scale (I hold off doing anything about it until the situation permits) which was eliminated.

It is not surprising that there were some missing data in these coping scales given that prisoners had to answer 56 individual questions about coping. Although they appeared in the survey immediately following the initial questions on prison hardships and although extra spaces were put between every four questions to make it easier to answer them, some questions were not answered. There is no way to know whether questions were purposely not answered or randomly skipped. Overall for each of the eight subscales, if there were any missing data, it was

83 usually one or two items. However, list-wise deletion of each of the created scales would have resulted in the loss of too many cases and subsequent imputation of the whole missing scale might have resulted in coping data that were skewed or biased. Instead imputation of data was conducted separately for each of the items in the eight subscales. First, if more than half of the items of any subscale was missing for an individual, then the subscale was made missing for that individual. Only five individuals did not answer most or any of the coping questions and thus their coping scales were made missing. However, for the remaining cases that had any missing values, most of the scales were only missing one individual item with a few missing two if they were four-item scales and a few missing up to three if they were made up of seven to nineteen items. Imputation was conducted for those cases with little missing data by taking the average of those items of a scale that were not missing for each case. Thus for the denial scale, if the item ‗I refused to believe that it has happened‘ was missing for prisoner X, the average of his remaining three items would be calculated and that number imputed for the missing item. Therefore if prisoner X scored low on the other three items, his score for this missing item would also be low.

Table 6 - Reliability Results for Coping Scales Coping Scale Cronbach’s Alpha

Seeking emotional and instrumental social support .83

Turning to religion .91

Use of humor .84

Active coping activities .70

Denial .78

Mental and behavioral disengagement .74

Acceptance .63

Focus on and venting of emotions .64

As mentioned, reliability analysis was conducted to ensure that each of the items that were combined helped to strengthen the overall scale and that the items hung together (Table 6 above). The analysis was conducted on the weighted sample (described previously in the

Research Methodology section) with the Cronbach‘s alpha reported for each of this study‘s scales. Six of the eight coping scales in this study were in the acceptable to strong range, and only two had a Cronbach‘s alpha that was under the acceptable range (.70).

Negative Emotions

As mentioned previously, researchers who have studied offenders‘ coping skills have also studied their trait emotions to understand how they interact in the coping process (Maitland &

Sluder, 1998; Biggam & Power, 1999; Soderstrom, et al., 2001; McDonald, 2006; Harreveld, et al., 2007). While some have studied situational coping and state emotions; others have studied general dispositional coping and trait emotions of offenders. One goal of this study is to determine whether the general traits of anger, depression, and anxiety are associated with the likelihood of engaging in serious prison misconduct and violence. Another goal is to explore the relationship between the trait emotions and the various aspects of coping.

The current research used the State-Trait Personality Inventory (STPI), a self-report survey that can measure both situational and dispositional anger, anxiety, curiosity, and depression in adults (Spielberger, 1995; Spielberger, Ritterband, Sydeman, Reheiser, & Unger,

1995). The STPI is a comprehensive instrument that was initially based on Spielberger‘s State-

Trait Anxiety Inventory (STAI), the State-Trait Curiosity Inventory (STCI), and the State-Trait

Anger Inventory (STAXI). The original version of the STPI (often referred to as STPI Form X) did not contain the depression subscale. The current version of the STPI (often referred to as

STPI Form Y) with the depression subscale was published in 1995. The overall STPI consists of

85 eight 10-item subscales: state and trait anxiety, state and trait anger, state and trait curiosity, and state and trait depression. As mentioned previously there is a distinction between state and trait emotions. Using anxiety as an example of this difference, ―(s)tate anxiety is defined as an unpleasant emotional arousal in the face of threatening demands or dangers. A cognitive appraisal of threat is a prerequisite for the experience of this emotion. Trait anxiety, on the other hand, reflects the existence of stable individual differences in the tendency to respond with state anxiety in the anticipation of threatening situations‖ (Stewart, 1999). Since the self-administered survey is divided between the state and trait emotions and the current research questions focus on trait emotion, only the half of the survey with the four trait emotion subscales was used.

The research used the 40 items associated with the four trait emotion subscales. Even though curiosity is not a part of the current study, the curiosity questions are woven within the other questions and might affect comparison to the validated samples if those specific items were removed. Respondents were asked to evaluate how they ―generally feel‖ regarding various statements, like ―I am a steady person‖ (anxiety); ―I am quick-tempered‖ (anger); ―I feel inquisitive‖ (curiosity); and ―I feel gloomy‖ (depression) to which they must choose ―almost never,‖ ―sometimes,‖ ―often,‖ and ―almost always.‖ The STPI has been tested on male and female navy recruits, college students, and working adults (Spielberger, 1995). Cronbach‘s alpha reliability coefficients for the anger, curiosity, and depression subscales ranged from .81 to

.96, with the coefficients being highest in the working adults test group. In addition, trait subscales correlate highly with the trait subscales of the State-Trait Anger Inventory for males and females (.97 and .95 respectively), the State-Trait Curiosity Inventory (.95 and .96 respectively), and the State-Trait Anxiety Inventory (.93 and .96 respectively), all of which have been extensively used and validated within the field of psychological testing (Stewart, 1999). In

86 addition, the depression subscales correlate significantly with other depression measures

(Spielberger, 1995). In addition, the STPI has been used successfully with incarcerated populations (Klein, 1987; Silverman & Vega, 1990; Bowman, 1997).

Similarly to the coping scales, there were some missing data in the items that made up the three trait emotion scales. However, there were nine cases where either all or a majority of items for these scales were missing. About a half dozen prisoners either chose not to fill out these 40 trait emotion questions or did not have time to finish them in the allotted time. Again as with the coping data, if a scale had more than half of the items missing, the entire scale variable was made missing. Thus the trait emotion scales were made missing for those nine individuals. The remainder of the cases that had some missing trait emotion items had those missing items imputed by averaging the score for each of the items for the scale. Thus if prisoner X had two depression items missing, the remaining eight of the ten depression items were added together and divided by eight, and that depression average was imputed into the two missing scores.

Table 7 - Reliability Results for Trait Emotions Scales Trait Emotions Scale Cronbach’s Alpha

Anxiety .78

Anger .73

Depression .83

The reliability analysis (Table 7) conducted on the three trait emotions revealed all to be within the acceptable range (over .70).

INDEPENDENT VARIABLES - CONTROL VARIABLES In addition to collecting data on coping, and the auxiliary variables of anger, anxiety, and depression, the researcher also collected data on other independent variables for the purpose of description and as control variables for the model. These include two general categories, prison strains and individual characteristics.

Prison Strains

The current research design assumes that all prisoners experience various strains and stresses related to adjusting to and being in prison and that these may affect one‘s involvement in serious prison misconduct and violence. It is also hypothesized that prisoner‘s coping abilities are stronger predictors than the presence of prison strains or stresses. The research explored the types of prison strains experienced by prisoners to be able to describe the various types of prisons strains that RIDOC prisoners are most likely to experience and to determine their prevalence. This would provide prison administrators with information which they might use to minimize or address some prisons strains; and help them direct future coping skill training programs on those prison strains and stresses that prisoners must regularly address.

There have been a few studies that have included variables that measure prison strains and stresses (MacKenzie & Goodstein, 1986; Zamble & Porporino, 1988; Zamble, 1992;

Maitland & Sluder; 1998; Gover, et al., 2000). These studies have either fit into the deprivation/importation research category or have focused on coping strategies in prison. They have measured various conditions of confinement in myriad ways including the creation of prison stress or problem scales (e.g. miss family, conflicts with prisoners); fear of victimization scales; conflict with other prisoner and conflict with correctional staff scales; and various perceptual measures of the conditions of confinement. In their longitudinal study of prisoner

88 coping, Zamble and Porporino (1988) found that prisoners were consistent in their delineation of their prison problems both at the beginning and over a year into their prison sentence.

The current research included three quantitative measures of prison strain—general prison stresses, fear of victimization, and actual prison victimization. Each of these three measures of prison strains was used successfully in Maitland and Sluder‘s 1998 study of victimization in which these and other measures were administered to a non-probability sample of 111 prisoners in a Midwestern prison. Cesaroni and Peterson-Badali (2005) also successfully administered these three measures in their study of 113 incarcerated male youths. In the present research, these measures of strain were incorporated into the self-report surveys. In addition, data were collected from the RIDOC database on the number of enemies that a prisoner had at the time of data collection. Information on enemies is usually reported by prisoners to prison staff at intake or during the course of their incarceration and might also include those with whom they have had serious altercations.

GENERAL PRISON STRESSES

The first scale was based upon Maitland and Sluder‘s (1998) 11-item scale that lists various problems that prisoners typically encounter during their stay in prison and asks them how hard each was for him, with responses from ‗not hard‘ to ‗very hard.‘ Their analysis of reliability demonstrated internal consistency—Cronbach‘s alpha was .78 (p. 61). The problems included: missing family or friends; missing certain activities; conflicts with prisoners; regrets about the past; concerns about the future; missing personal possessions; boredom; lack of privacy; excessive noise; lack of proper facilities; and missing freedom. Additional questions were added to this scale based on the researcher‘s prior experience in prison, input from others, and feedback from prisoners after conducting a pilot of the survey during September, 2008 at the

Rhode Island House of Correction. These additional questions addressed prisoners‘ difficulties in dealing with their living conditions and their inability to make autonomous decisions.

Principal components analysis yielded only one factor that made intuitive sense and reliability analysis of all other factors resulted in alphas that were below the acceptable range (.70).

However, a global prison hardships variable was created that included all nineteen of the individual prison hardships. The Cronbach‘s alpha was .83 indicating that it is certainly an acceptable and reliable scale.

Finally there was very little missing data among these nineteen questions that asked about the strains and stresses of prison. Most questions had between one and three missing answers. A few had none and only one (regrets about the past) had six missing answers. Because of so little missing data, no imputation was done. However, creation of the global prison hardships variable used listwise deletion and resulted in 32 missing cases. Given that this variable will only be used for exploratory purposes, missing data were not imputed.

FEAR OF VICTIMIZATION AND ACTUAL VICTIMIZATION

Maitland and Sluder (1998) also developed a 3-item additive index to measure fear of victimization—including prisoners‘ perceived safety in prison and their fear or likelihood of being attacked during their sentence. It had internal consistency—Cronbach‘s alpha was .72 (p.

61). Each question is used individually and as an additive index. There were missing data for only four cases for these variables.

The victimization scale was a 14-item scale, initially developed by McCorkle (1993) but slightly altered by Maitland and Sluder (1998) for their study on the victimization experienced by prisoners. Their analysis of the victimization scale found it to have an acceptable level (.78)

90 of internal consistency (p. 61). In this scale, prisoners are asked if they have encountered a range of victimization experiences including being hit, kicked, punched, slapped, pushed, shoved, hurt, threatened, had money stolen, etc. to which prisoners answered ‗yes‘ or ‗no.‘ Maitland and

Sluder did not differentiate between victimization by staff of victimization by other prisoners. In the current research this scale was altered by asking prisoners to differentiate whether they had been victimized by other prisoners or by staff. Therefore for each item, respondents chose between ‗yes‘ or ‗no‘ responses regarding victimization by other prisoners, then by staff.

Unfortunately, this might have not been a good strategy since respondents were more reluctant to answer staff victimization questions resulting in more missing data in those questions than in the prisoner victimization questions. Unlike the coping or trait emotions, it would not be appropriate to impute individual items of this victimization scale, because they are all different behaviors.

For example, one should not impute that a prisoner was hurt by staff just because one knows that the staff insulted or threatened him. Instead the questions were grouped into four categories of victimizing behavior including being threatened, having their property stolen, being verbally insulted, or being physically hurt. There was also one question asking whether they had been forced sexually. Instead of adding them together to make a scale or index, if the prisoner answered ‗yes‘ to any of the questions within a category, the category variable was coded as

‗yes.‘ If the prisoner answered ‗no‘ to all of the questions within a category, the category variable was coded as ‗no.‘ Cases where more than half of the questions in a category were missing were coded as ‗missing.‘ The five categories of victimization by other prisoners had missing data for less than a dozen cases. However, as mentioned previously, prisoners were less likely to answer questions about victimization by staff and missing cases ranged from 33 missing

91 cases in the staff threatened variable to 19 missing cases in the staff verbally insulted variable.

There was only one missing case in the number of enemies variable extracted from INFACTS.

Other Individual Characteristics of Prisoners

The other independent variables that were included were those that previous researchers have found to affect prison misconduct: age, prior criminal history, prior history of prison misconduct, race, gang affiliation, ‗commitment to convention‘ variables including marital status, education, and employment prior to incarceration, time incarcerated, and level of security.

Inclusion of these variables will improve the conceptual model by controlling those factors that might affect involvement in serious prison misconduct or violence. The research will also include other variables that might shed light on the population including current sentence, current commitment offense, level of social support, involvement in programming, and involvement in work. These other independent variables have been categorized into four groupings: 1) sample characteristics; 2) criminal history, substance abuse history, and current sentencing status; and 3) psychiatric treatment; 4) incarceration experience. Table 8 below specifies each of the variables, their data collection source and whether they are a dependent, main independent, or other type of independent variable.

VARIABLES IN THE QUALITATIVE RESEARCH The field research focused on gaining a more in-depth picture of the nature of serious prison misconduct and violence. The field research in general and the in-depth interviews in particular honed in on the causes, processes, and consequences of prison violence and serious misconduct that lead to prisoners being downgraded in a classification hearing to a higher security facility. In particular, prisoners were asked their reasons for being involved in prison

Table 8 - Variables in Study by Use and Data Source Variables Use Data Source

INFACTS Survey Interview Observe

Serious misconduct and violence

Annual misconduct score DV X

Annual rate violent dreports DV X

Low- vs. high-level disc. group DV X X X

Sample characteristics

Race/ethnicity CV X

Nativity/residence CV X

Age, marital status CV X

Prior serious relationship CV X

High grade completed CV X

Prior employment CV X

Criminal history, substance abuse history, & current sentencing status

Criminal history

Prior gang involvement CV X X

Juvenile correctional CV X X

Prior incarcerations CV X X

Substance abuse history CV X

Current sentencing

Most serious charge category CV X

Total sentence length CV X

Time incarcerated CV X

Psychiatric treatment CV X X

DV= Dependent Variables IV= Main Independent Variables CV= Control Independent Variables

Table 8 - Variables in Study by Use and Data Source Continued Variables Use Data Source

INFACTS Survey Interview Observe

Incarceration experience

Level of social support

#Visits, letters receive monthly CV X X

# Friends in prison CV X X

% of week spent doing:

Work, visits, TV, idle, read, etc CV X

Prior Dreport history CV X

Disposition coping scales IV X X

Trait emotion scales IV X

Prison hardships

Prison hardship variables CV X X

Fear of victimization scale CV X X

Actual victimization CV X X

Evidence of enemies/PC status CV X

DV= Dependent Variables IV= Main Independent Variables CV= Control Independent Variables

misconduct in general. Staff members were also asked their opinions on the causes of serious prison misconduct and violence. These causes may or may not relate to prison strains, but prison strains were further explored during the field research. Both prisoners and staff were also asked about the various components of the disciplinary process. Observation of both the disciplinary and classification processes was carried out to further understand how those processes deal with serious misconduct and violence. Finally both staff and prisoners reported in the interviews

94 about the consequences of prison violence. In addition, much information about the various consequences was learned during observation of disciplinary and classification hearings including personal consequences for prisoners, consequences for staff, consequences for a prisoner‘s family, and consequences for the prison in general. The qualitative research explored the same independent variable—these prisoners‘ coping strategies. Prisoners were asked a series of questions about the hardships they experienced in prison and how they coped with them.

DATA COLLECTION STRATEGY

QUANTITATIVE DATA COLLECTION The quantitative data collection had two components—a self-report survey that all prisoners in the sample were asked to fill out and data extracted from the RIDOC prisoner database for the same sample of prisoners.

SELF-REPORT SURVEYS As mentioned previously, the researcher conducted a pilot of the self-report survey with a small group of prisoners from minimum security in order to learn what questions or instructions, if any, were problematic and needed to be reworded or changed and to ensure that the survey did not take more than 45 minutes to complete.6 As a result of the pilot, it was determined that the researcher‘s description of the research and the informed consent form were clear and comprehensive; the survey was clear and easy to fill out; that for the most part, prisoners were not upset by the survey questions because they dealt with issues that many of them think about

6 Self-report surveys have varied in length from 30 minutes to over one hour. However, most researchers believe that surveys that exceed one hour are too long and that quality of answers potentially deteriorate after that amount of time (MacKenzie & Goodstein, 1986; MacKenzie & Souryal, 1994; Maitland & Sluder, 1996; Sappington, 1996; Gover, et al., 2000; Soderstrom, et al., 2001; Cesaroni & Peterson-Badali, 2005; McDonald, 2006; Dhami, et al., 2007).

95 on a regular basis; and that the survey could be completed within the allotted time. A few of the individual items in the self-report survey itself needed clarification and prisoners gave some suggestions for time intervals for such issues as visits. Subsequently, Northeastern University‘s

Institutional Review Board (IRB) and the RIDOC gave approval to the project and the researcher met with RIDOC staff from the Planning and Research Division to discuss extraction of data to use for sample selection. The self-report survey can be found in Appendix 2.

A number of researchers have successfully conducted prison surveys by administering them to small groups of prisoners in common areas designated by prison administrators

(MacKenzie & Goodstein, 1986; MacKenzie & Souryal, 1994; Maitland & Sluder, 1996;

Sappington, 1996, 1998; Gover, et al., 2000; Soderstrom, et al., 2001; Cesaroni & Peterson-

Badali, 2005; McDonald, 2006; Dhami, et al., 2007). Once the sample was extracted, the researcher worked with the Deputy Superintendents of each facility to create a schedule for administering the self-report surveys. Survey administration began on July 1, 2009 and was completed by the end of November, 2009. For the majority of prisoners who were residing in medium or maximum security facilities, and those residing in B-Mod of the HSC, the self-report surveys were administered by the researcher to groups of 15 to 25 prisoners at a time in a public area of the prison like the cafeteria, a school classroom, program room, or the chapel as designated by the prison administrators. Prisoners were either called down to the common area or a notice was posted for them to go to the common area. The research and consent form were explained to them in depth and they were given time to ask any questions they may have had.

Those who did not want to participate in the survey left the room immediately. Survey participants then read and signed the consent form and subsequently filled out the survey.

Approximations of the time participants took to complete surveys ranged from 20 minutes to

96 almost an hour and a half to complete. However, the average survey participant completed the survey in approximately 35 to 45 minutes. Younger prisoners completed it faster; elderly prisoners and those whose primary language was not English took longer. The survey and consent form were translated into Spanish and prisoners could choose whether they wanted to use the English or Spanish versions. A handful of prisoners could not read or did not have reading glasses. The researcher offered to read the questions to the prisoner and most of the prisoners were able to follow along and answer the questions themselves once the questions and potential responses were read to them. In the two medium security facilities, in maximum security, and in a couple of units in the HSC, prisoners who were serving segregation time or who were in certain HSC units were not allowed to come out of segregation or their units to participate in the group survey. In the medium security facilities and in some of the HSC housing units, some of these surveys were conducted on a one-on-one basis. Otherwise the researcher went to the cell doors in the segregation unit and explained the survey and consent form. If the prisoner agreed to participate, he was given the survey to fill out on his own within a one- to two-hour time span. The researcher subsequently was again escorted around to the segregation cells to pick up the completed surveys. In the maximum security facility and in some other units in the HSC, the researcher was not allowed to address the prisoners at all. In these few instances (about a dozen prisoners altogether) packets were made up with a letter describing the survey, the consent form, and the survey itself. They were delivered in maximum security by the Deputy Superintendent and by the Lieutenant in the HSC and were picked up by staff about an hour to two hours later and then delivered to the researcher immediately.

Prisoners were given the opportunity to seal the envelope with the completed survey with either tape or labels. As mentioned above, 312 prisoners completed the surveys.

EXTRACTION FROM RIDOC PRISONER DATABASE The Rhode Island Department of Correction maintains the INFACTS database that contains socio-demographic, criminal history, current sentencing information, current prison information, disciplinary history and associated sanctions, work, and movement. The extraction of data was completed in phases and included extraction of socio-demographics, current sentencing, and current incarceration variables for all prisoners residing in the medium and maximum facilities and the HSC on May 19, 2009. RIDOC Research Division staff then downloaded all of the dreports (event data) received by prisoners during 2008. Once the datasets were merged and the population was divided into those who had received a Class I dreport in

2008 and those who had not, a random sample was taken of each group, as mentioned previously. After surveys were completed, several other datasets were extracted for the 312 prisoners who were surveyed. These included all dreports from 2005 through 2009, program participation in 2008 and 2009, classification data from 2005 through 2009, and release data.

Data was not extracted prior to 2005 due to upgrades in the RIDOC database but also because a large proportion of the sample would have been lost if any variable was created with data from prior to 2005. They were extracted at the beginning of 2010 so that all relevant 2009 data could be included. All of these additional datasets were extracted separately and were event-based datasets, meaning that each case was an event (a disciplinary report or a classification hearing as opposed to being prisoner-oriented). This required further analysis and extraction of parts of the data in order to be able to merge it with the prisoner-oriented dataset. Prior incarceration data were extracted manually by the researcher from viewing the INFACTS system in the Research

Division offices because it was determined that extraction of that particular variable was too difficult and would take an inordinate amount of steps by the RIDOC Research Division. Table

8 above lists the variables that were collected via the RIDOC INFACTS system.

QUALITATIVE DATA COLLECTION The qualitative data collection also had two components—non-participant observation and in-depth interviews with prisoners who had been involved in serious misconduct and violence and with staff at each of the four facilities.

NON-PARTICIPANT OBSERVATION Non-participant observation is a qualitative research methodology that aids the researcher in understanding how an environment works and provides rich information that the researcher can use to help interpret and make sense of the quantitative data. In the current research study, non-participant observation refers to two activities: casual conversations with prison administrators and staff about why prisoners are involved in serious prison misconduct and violence and attending classification and disciplinary hearings at each of the four facilities.

Casual conversations with staff at each of the facilities helped to build relationships with staff that resulted in greater staff participation in the interviews, gave the researcher ideas to pursue in both staff and prisoner interviews, and allowed the researcher to find out about other unanticipated issues regarding misconduct. Observing classification and disciplinary hearings was essential to understanding the processes by which prisoners are sanctioned and possibly classified to higher security at either maximum security or the HSC. Both the observation of hearings and the casual conversations helped to shed light on the underlying assumptions that are made about prisoners who misbehave and about the sanctions associated with such behavior.

Observation of classification hearings was scheduled with Central Office classification staff and took place from August, 2009 through May, 2010. If only a small number of prisoners were classified in a set of hearings, then the researcher would observe another set of hearings at that facility. Classification hearings typically ran from an hour to two hours. Observation of

99 disciplinary hearings was scheduled with the individual lieutenants at each facility whose job it was to conduct disciplinary hearings. Two to three sets of disciplinary hearings were observed at each of the medium facilities, at maximum, and at the HSC. Anywhere from five to 30 dreports were heard during each set of hearings, depending on the number of dreports received by prisoners the few days prior in that facility. Observation of disciplinary boards happened while the researcher was at each of the facilities, from July, 2009 through July, 2010.

IN-DEPTH INTERVIEWS WITH PRISONERS AND STAFF In-depth interviews were conducted with 25 prisoners who at the end of the self-report survey affirmatively answered a question asking if they would be willing to take part in an in- depth interview and who had received a Class I Highest dreport during 2008. This sampling method obviously resulted in a non-probability sample, but was purposive with the goal of interviewing a diverse group of prisoners as mentioned previously. Interviews took place during

May and June, 2010 and took from one hour to two hours, depending on the amount of time the interviewee took to answer questions. The interview schedule is attached in Appendix 3. Prison administrators were consulted as to the appropriateness of candidates for these interviews. A couple of prisoners in Moran could not be interviewed because they were either on visits or were involved in other activities from which they could not leave. Two prisoners at Maximum were in segregation and were not made available for interviews. All prisoners from the HSC, whether in segregation or not, were allowed to be interviewed. However, two prisoners at HSC declined to be interviewed. Individual interviews were conducted in a classroom in Moran, in the cafeteria in Maximum, and in either the dining area or an interview room in the housing units at the HSC.

100

As mentioned previously, various staff members at the four facilities were also interviewed. In all 26 staff interviews were conducted in private offices or in rooms where classification and disciplinary hearings take place. Staff interviews took from 30 minutes to about 90 minutes depending on the length of the staff responses. The staff interview schedule can be found in Appendix 4.

ANALYTIC STRATEGY

Prior to any analysis, a number of variables had to be created from two or more pieces of data. For example, the dreport variables combined disciplinary information for 2008 and 2009, with date committed to the RIDOC, and date released from the RIDOC (if any), in order to create an annual rate. Another example, time incarcerated, used both the commitment and release dates to calculate the amount of time that the prisoner had actually served. Next variables that were to form a scale were examined. This included the coping and negative emotion variables, as well as others such as prison strain and victimization. Each scale was factor-analyzed using Principal Components analysis for scale validation. In addition, the internal reliability of each scale was examined using the coefficient alpha. Depending on the factor analysis and reliability, scales were created and finalized. Once the bulk of the variables were created, the preliminary set of analyses examined all of the frequencies of the entire set of variables to check them for missing data, skewness, kurtosis, and extreme outliers. Missing data were dealt with as outlined above in the section describing each set of variables. Table 9 below details the number of cases missing for each variable and also whether data were imputed.

Analysis began with a thorough examination of all of the dreport variables. After their description, two were chosen for further multivariate analysis (annual misconduct score and

101

Table 9 - Missing and Imputed Data Variable Cases Missing Imputed?* Variable Cases Missing Imputed?*

Disciplinary Reports Level of Social Support Low- high-disciplinary grp 0 0 # visits received monthly 7 0 Annual misconduct score 0 0 # letters received monthly 5 0 Rate of violent dreports 0 0 # friends in prison 14 0 Sample Characteristics % Week Spent Doing: Race/ethnicity 0 0 Work 69 0 Age 0 0 School 43 0 Marital status 2 0 Socializing 78 0 Highest grade completed 3 0 Treatment programming 28 0 Prior employment 5 0 TV/Radio 87 0 Criminal History Working out/sports 69 0 Prior gang involvement 6 0 Reading 81 0 Juvenile incarceration 9 0 Idle 55 0 Prior incarceration 0 0 Prior Dreport History 0 (85 N/A) 0 Substance Abuse History Prison Hardships 0-3 0 Alcohol 6 0 Fear of Victimization 4 0 Tobacco 7 0 Victimized by Prisoners 4-12 0 Marijuana 6 0 Victimized by Staff 19-33 0 Cocaine/crack 9 0 Trait Emotions Heroin 15 0 Anxiety 21 12 Inhalants 14 0 Anger 32 23 Other 12 0 Depression 35 26 # drugs used 19 0 Coping Current Sentencing Get support 15 13 Most serious charge 0 0 Religion 5 2 Total sentence length 35 0 Humor 10 6 Time incarcerated 30 0 Denial 11 8 Psychiatric Treatment Disengagement 22 20 Before incarceration 7-8 0 Acceptance 14 11 During incarceration 0-9 0 Active coping 31 27 Venting emotion 13 11

* At least one item in scale was imputed. If more than half missing, case was made missing for scale.

annual rate of violent dreports) and one, (the dichotomous low- vs. high level disciplinary groups) was chosen for bivariate analysis. Next bivariate analysis was conducted on those independent variables considered to be either control variables (those that other researchers have found previously to be associated with serious misconduct and violence) or other variables that were chosen because they might affect serious misconduct and violence. These independent variables could be categorized as 1) sample characteristics; 2) criminal history, substance abuse history, and current sentencing status; 3) psychiatric treatment, and 4) incarceration experience variables. Both groups (low- vs. high-level disciplinary group) were described in terms of the

102 above independent variables and correlations were highlighted. Next the prison hardships of the sample, including stresses, fear of victimization, and actual victimization were explored and the differences between those in the high- and low-level disciplinary groups were noted. The bivariate analysis continued next with the examination of each of the trait emotion scales and whether they were related to the dichotomous serious prison misconduct and violence variable.

Finally, the bivariate relationships between the eight coping scales created in this study and serious prison misconduct and violence were investigated.

The dependent variables to be used for the multivariate analyses were examined as mentioned above, and there were no missing data. However, some outliers were revealed in histograms of the data as well as scatterplots of the dependent variables against the main independent variables. Further investigation of these outliers revealed that they were not errors in data collection or creation of the variables, but instead represented the actual behavior of those prisoners who were involved in extremely high rates of disciplinary activity. It was decided that these outliers should not be excluded because excluding them would be the equivalent of excluding the small percentage of prisoners whom Toch (Toch, et al., 1989) labeled as the

‗chronic disrupters‘ who make up a very small percentage of the prisoner population but who also commit a significant percentage of the serious misconduct and violence and take up the largest proportion of energy and resources expended by correctional staff and the administration to control misbehavior. Further examination of the dependent variables revealed that they were highly skewed in that over one third of the sample had a count of zero dreports. Because of this skewness and the fact that the number of dreports cannot fall below zero, linear regression should not be used because it violates the assumption of a continuous normally distributed dependent variable (Long, 1997). Using this study‘s skewed dependent count variables with

103 linear regression could lead to parameter estimates that would be biased, inconsistent, and inefficient. One alternative to linear regression is the Poisson regression model that can be used when the dependent variable is measured by a nonnegative integer (starts at zero and no fractions) and thus can be used with count data, such as the number of dreports. The Poisson model has two additional assumptions—that there can be no over- or underdispersion and that predicted counts of the dependent variable are independent of each other (Osgood, 2000; Berk &

MacDonald, 2008). Overdispersion happens when data have greater variance than expected and can be determined by examining the mean and variance of the dependent variables (Kremelberg,

2011). In this study the variance is greater than the mean of both dependent variables, the annual misconduct score (variance= 254.32; mean=5.59) and the annual rate of violent dreports

(variance=1.01; mean=.37), violating the first assumption. Misconduct data also violate the second assumption that instances of the dependent variable, in this case events of serious misconduct and violence, are independent. In reality, acts of misconduct often occur in clusters, where acting out elicits an official response from staff or other prisoners, which in many cases triggers further responses by the prisoner. Thus the dependent variable is not independent, violating the second assumption of the Poisson model.

Many criminologists have acknowledged the inappropriateness of using the standard

Poisson regression with criminal justice rates and event data and instead have successfully used the negative binomial regression model (Paternoster, Brame, Bachman, & Sherman, 1997;

Braga, 2003; Lattimore, MacDonald, Piquero, Linster, & Visher, 2004; and Bottcher & Ezell,

2005). The negative binomial regression is an extension of the Poisson regression that combines the Poisson distribution of event counts with the inclusion of a dispersion parameter that allows the variance of the dependent variable to be greater than its mean (Long, 1997; Osgood, 2000).

104

Berk and MacDonald (2008) have warned against the automatic usage of the negative binomial regression whenever one has dependent count data that are ‗overdispersed.‘ They warn that researchers must differentiate between the systematic (predictable) portion of one‘s variance and the stochastic (random) portion that can be addressed by the negative binomial regression parameter. They explain that if the overdispersion is due to systematic misspecification, like omitted variables and functional forms that are inappropriately specified, then the negative binomial regression model would be inappropriate since it cannot fix those problems.

Fortunately, there have been over five decades of criminal justice research on the issue of prison misconduct and violence, dating back to Zink‘s first study in 1958. As mentioned previously, many of the factors associated with prison misconduct are known and have been incorporated into this study. In addition, many other variables were collected and tested against the study‘s dependent variables to determine correlation of additional variables. Therefore while one cannot ensure that every single variable associated with serious prison misconduct and violence is included in the model, it does incorporate most of the variables known to researchers of this subject up to the present time and used the experience and expertise of those researchers to ensure that measurement was appropriate.

Negative binomial regression is one of several generalized linear models (GZLM) that compared to the general linear models (GLM) do not ―assume a normally distributed dependent variable,…nor linearity between the predictors and the dependent, nor homogeneity of variance for the range of the dependent variable‖ (Garson, 2009, p. 1). However, it does have some assumptions. One assumption is that a link function must be selected to ensure a linear relation between the linear predictor and the link function of the dependent variable. In the case of the negative binomial, the SPSS default is the log link function (log(x/x + k – 1)) where x is the

105 dependent variable and k is the ancillary parameter that allows for stochastic variance. Another assumption of the negative binomial regression is that using this model is significantly different from using the Poisson model. The Lagrange Multiplier Test tests the hypothesis that k (the ancillary dispersion parameter) = 0, which would indicate that the Poisson regression model would be preferred for parsimony. However, the Lagrange Multiplier Test using the coping, feelings, and general predictors with the dependent variable in this study was significant,

(p=.019) indicating that k is indeed different from 0 and that use of the negative binomial regression model is preferable.

While it was determined to keep the outliers at the univariate and bivariate levels as discussed previously, further analysis of outliers at the multivariate level was carried out.

Scatterplots of Pearson residuals by the predicted values of the dependent variables revealed no serious outliers. The highest Leverage test for outliers (.35) was lower than the recommended cut-off (.50) (Belsey, et al., 1980). The Cook‘s D measure of the overall influence of cases ranged from 0 to .08, clearly less than the suggested value of one (Belsey, et al., 1980). Thus no outliers were removed.

Another assumption of generalized linear models is that observations of the predictor variables are independent. An OLS regression on the unweighted data using all of the independent variables was conducted and the Durbin-Watson statistic examined. This statistic determines whether observations are independent and whether there is correlation among adjacent residuals (Norusis, 2003). If variables are not highly correlated with one another, the

Durbin-Watson statistic should be between 1.5 and 2.5—as close to 2 as possible. In this study, the Durbin-Watson statistic was 1.98, indicating that observations were indeed independent.

106

Another similar assumption is the absence of multicollinearity (Garson, 2009). The

OLS regression mentioned above yielded Variance Inflation Factors (VIFs). These were examined and found to be less than the cut-off of 4, indicating that the level of variance in the independent variables that was explained by the other independent variables was not inappropriately high. The OLS regression was rerun with the weighted data and all VIFs were still under 4, except for the trait emotions of anxiety and depression (VIF=4.20). Indeed in the examination of the weighted correlation matrix of independent variables, the highest correlation was between anxiety and depression (r=.78). One solution is to drop anxiety and depression in the final model and instead use their cross-product (Garson, 2009). Correlations among the other variables were within an acceptable range and will be discussed in the section on the multivariate results.

Finally a great deal of missing data can lead to biased coefficients and inaccurate results.

As mentioned previously there were no missing data in the majority of the data extracted from the RIDOC INFACTS system, including no missing data in the dependent variables. The most missing data was in the main independent variables (the coping scales and the trait emotions) collected via the self-report surveys. Also as mentioned, the missing data in individual items of those scales were imputed based upon the average of the remaining scale‘s items. After imputation, less than a dozen cases had missing data in those two sets of scales. Two other sets of variables had missing data that might limit their use in the multivariate analysis. One was the amount of dreports that prisoners received in the two years prior to the current time frame.

Research has shown that receiving dreports in the past is not surprisingly a predictor of receiving future dreports (Harer & Langan, 2001; Gaes, et al., 2002; and Camp, et al., 2003).

Unfortunately, while there is no real missing data for prior dreports in this study, the fact is that

107

85 of the prisoners in the sample had not begun their incarceration prior to 2008 and thus could not have an annual prior rate of dreports. Therefore while prior dreports was included in the bivariate analysis, it was excluded from the multivariate analysis in order to keep those 85 mostly young prisoners in the sample. The set of questions that asked prisoners what percentage of their waking hours in a week were spent on various activities also had a lot of missing data.

Instead of putting in the number of hours, as directed, many of the prisoners wrote in responses like ―a little‖ or ―often‖ which could not be used, unlike a response of ―not at all‖ which was interpreted as zero. These variables were not included in the multivariate analysis.

Those independent variables that were control variables (personal characteristics, criminal history or current offense, or incarceration experience) that were not correlated with the dependent variables in the bivariate analysis were dropped from the multivariate analysis in order to maximize the power of the statistical models with the exception of race/ethnicity. Thus the number of independent variables included in the bivariate analysis was over 60; but that number was reduced to 29 in the multivariate analysis. Finally preliminary analyses were conducted to determine whether nominal variables should be used in their categorical form

(which is allowed in the negative binomial regression) or in their recoded dummy variable, dichotomous form. The goodness of fit tests revealed that the use of dummy variables was preferable for further analysis.

108

V. DISCIPLINARY REPORTS, DISCIPLINARY PROCESS, AND SANCTIONS

A previous section described the RIDOC disciplinary policy and briefly discussed the prevalence of dreports and use of segregation during 2005. This chapter provides the figures on dreports for the sample during 2008 and 2009 and then discusses staff and prisoner feedback on the disciplinary process collected in the in-depth interviews. The statistics on the dreports for

2008 and 2009 for the sample provide the backdrop for all further quantitative analyses. Also provided are other indicators of severe disciplinary problems including certain classification decisions. Although the study‘s main focus is the reasons why prisoners are involved with serious prison misconduct and violence, it is important to understand how staff think about bookings and carry out disciplinary boards, and how both staff and prisoners view the process and end results.

SAMPLE DISCIPLINARY REPORTS FOR 2008 AND 2009

During 2008 and 2009 prisoners in the sample incurred 864 dreports—the smallest portion was in the ―highest predatory‖ category (seven percent) and the largest portion (46 percent) in the ―highest non-predatory‖ category (Table 10). It is assumed that prisoners were involved in a lot more incidents than is evident in the combined ―low‖ and ―moderate‖ category

(22%), but that staff had most likely offered and many prisoners had most likely accepted the option of completing a one- or two-nighter in lieu of receiving a dreport for a Class 4 offense.

The mean number of dreports was 2.54 per inmate per year—however 26 percent of the sample did not receive any dreports during the 2008 or 2009 timeframe.

If one only examines the 458 dreports considered to be serious (i.e. those in the ‗highest predatory‘ and ‗highest non-predatory‘ categories), 39 percent of the prisoners did not receive

109

Table 10 - Annual Rate of DReports in 2008-2009 by Level of Seriousness & Violence Type

Type of Disciplinary Report N (%) Mean SD Range

Level of Seriousness Highest Predatory 63 (7) .20 .59 0 – 5 Highest Non-Predatory 395 (46) 1.27 2.78 0 – 28 High 216 (25) .69 1.29 0 – 10 Low-Moderate 190 (22) .61 1.18 0 – 9 Total 864 (100)

Types of Violence Violence against Staff 19 (8) .06 .40 0 – 5 Violence against Prisoners/Fights 66 (29) .21 .50 0 – 3 Riots/Weapon/Threats to Staff 72 (31) .23 .64 0 – 5 Threats to Prisoners 73 (32) .23 .59 0 – 6 Total 230 (100)

N=312; Missing Data=0

any serious dreport in either 2008 or 2009 but 24 percent received at least one serious dreport in both 2008 and 2009. The mean number of serious dreports was 1.43 for the sample. An examination of just the 230 violent offenses revealed that less than one-tenth of the violent dreports were for violence against staff. The remaining violent dreports were split evenly between violence against or fights between prisoners, threats against prisoners, and other types of violence (possession of a weapon, threats against staff, mayhem, involvement in riots, or other types of assaults). Almost two-thirds (64 percent) of the sample did not receive a violent dreport in either 2008 or 2009 and only 11 percent of the sample received at least one dreport for violence in both years. The overall mean for violent dreports was less than one per year.

Next the serious dreports were combined with those violent dreports in the ‗high‘ category (mostly threats against prisoners) to create the serious and violent combined category which included the dreports in the ‗highest predatory‘ category (all of which are violent), the violent and non-violent offenses in the ‗highest non-predatory category,‘ and only the violent

110 behavior from the ‗high‘ category. The mean annual rate of serious and violent dreports was

1.64 and 37 percent of the prisoners still had no dreports in this category. As mentioned in the

Research Methodology section, a composite score was created in order to take into account both the number and seriousness of dreports, as well as their level of violence. The mean annual misconduct score for the sample was 9.61 and ranged from 0 to 214. This composite score was used in any analysis of serious and violent misbehavior requiring a continuous variable.

Table 11 - Annual Rate of DReports in 2008-2009 by Various DReport Groupings

Disciplinary Report Groupings Mean SD % w/ 0 Dreports

All 2.54 5.14 26%

Serious 1.43 3.14 39%

Violent .64 1.36 64%

Serious and Violent Combined 1.61 3.44 37%

Annual Misconduct Score 9.61 21.63 N/A N=312; Missing Data=0

The sample was also divided into two groups—those with a low annual misconduct score of 0 thru 3 (49 percent) and those with a high annual misconduct score of 4 or more (51 percent).

This version of the dependent variable was used to differentiate prisoners with no or a low level of serious and violent misbehavior and those with a higher level.

A re-examination of the mean annual rates of dreports in the various groupings of categories demonstrates, as expected, that the two groups of prisoners—low-level misbehavior and high-level misbehavior—do indeed vary from one another. For example, the mean annual rate of all dreports is .47 for the low-level misbehavior group but 4.49 for the high-level misbehavior group. In fact the mean annual rate for all other types of dreport groupings is less

111 than one per year for the low-level misbehavior group but more for the high-level group. The mean annual misconduct score was .88 for the low-level misbehavior prisoners and 17.91 for the high-level misbehavior prisoners.

Table 12 - Mean Annual Rates of DReports during 2008 and 2009 by Prisoner Groupings

Prisoner Grouping Low-Level Misbehavior High-Level Misbehavior Mean SD Mean SD

Disciplinary Report Groupings All .47 .53 4.49 6.59

Serious .24 .43 2.57 4.06

Violent .03 .16 1.23 1.69

Serious & violent combined .26 .44 4.79 7.84

Annual misconduct score .88 1.33 17.91 27.76

Ever failed in lower security? .28 .64 .69 .96

Moves to higher security? .01 .11 .21 .47

Classification Score 11.97 5.35 18.07 4.98 % %

Classified to B/C status 08-09? 3.3% 18.8%

Security Level High Security Center 1.3 11.2

Maximum 25.0 34.4

Moran 61.2 39.4

Price 12.5 15.0

% Designated Security Risk Group 15.8 25.0

N=312; Missing Data=0

112

In addition to sanctions, the disciplinary board might recommend that prisoners found guilty of serious or multiple disciplinary offenses see a classification board for a possible downgrade in security, meaning that they could be classified to a higher security facility. The

RIDOC classifies prisoners in two ways, with a point-based classification score and whether or not they have a Security Risk Group (SRG) designation. The classification scores for the sample ranged from a low of one to a high of 34; the mean for the sample was 15 points (not shown).

The mean classification score for the high-discipline group was 18 compared to 12 for the low- discipline group indicating that those in the high-discipline group had more indicators of being a management problem than those in the low-discipline group. As for SRG, more prisoners with disciplinary problems had an SRG status compared to prisoners with little or few disciplinary problems.

Prisoners in the low-level misbehavior group experienced fewer failures in lower security than did those in the high-level misbehavior group. The number of failures in lower security ranged from zero to four. Eighty (80) percent of the prisoners in the low-level misbehavior group had never failed in lower security, compared with only 54 percent of those in the high- level misbehavior group. In fact, within the latter group, 14 percent had failed between two and four times. Low-level misbehavior prisoners were less likely to be moved to the HSC than high- level misbehavior prisoners, though the majority of prisoners in both groups (99 percent and 81 percent respectively) had not been placed in the HSC in 2008 or 2009. In fact only two of the low-level group and 30 of the high-level group were transferred to the HSC7. Similarly, only 11

7 While transfer to the HSC is most often due to disciplinary and management issues, some prisoners are moved to the HSC for participation in the OSU (observation unit for those with mental health problems) and others are moved there for protective custody reasons.

113 percent of the sample had been classified to either B or C status during 2008 or 2009. However a much larger percent of high-level misbehavior prisoners were so classified.

Finally, Table 12 above includes information on security level of prisoners at the time of the interview. Within the sample, almost one-third were living in Maximum and most of the remaining prisoners in the sample were living in either Moran or Price. Thus only six percent were residing in the HSC. However, once divided by the two disciplinary groups, those in the high-level group were more apt to live in the HSC or in Maximum compared to those in the low- level group.

*****

In summary, sample analysis of dreports received during 2008 and 2009 revealed a wide range of disciplinary behavior among prisoners in the sample. One must keep in mind that this sample over-represents those involved in serious disciplinary issues in 2008 and 2009 and that the true rate of disciplinary behavior for the population is smaller. However, even with the oversampling of the high misbehavior population, still one-fourth of the sample had not received any dreports during 2008 and 2009. Moreover, over one-third had not received any serious dreports and almost two-thirds had not received any violent dreports. Only one-fourth of the sample had received a serious dreport in both 2008 and 2009. Even fewer (one-tenth) had been found guilty of a violent disciplinary offense during each of those years.

The number of dreports received per year for the low-level group was less than one per year and only 4.5 per year for the high level group. Thus even among the half of the sample with high disciplinary problems, the mean number of dreports received was relatively low—an average of one dreport (in all categories) every three months. An examination of other indicators

114 of misbehavior including classification score, SRG status, transfer to the HSC, and classification to B or C status shows that a little more than one-tenth of those in the high disciplinary group were deemed to present sufficiently serious management problems to warrant the RIDOC‘s most serious designations. Thus in this sample more than half of the prisoners had little to no sanctioned disciplinary activity and two-fifths had moderate sanctioned disciplinary activity during 2008 and 2009. This leaves about one-tenth of this sample who received the greatest percentages of dreports, who consistently failed at lower security, and who ended up being classified to B or C status and transferred to the HSC.

STAFF AND PRISONER FEEDBACK ON BOOKINGS, BOARDS, AND SANCTIONS

Almost all of the staff and prisoners who were interviewed provided feedback on the new disciplinary policy, the issuance of bookings, and the disciplinary process, including investigation, the board, appeals, and sanctions. Not surprisingly, there was a marked difference between staff responses (more often positive) and those of prisoners (more often negative). Staff were specifically asked, ―What do you think of the current disciplinary process? Of the current sanctioning process? Are there other sanctions/solutions that you would like to see implemented?‖ Of course, since there was a relatively recent overhaul of the disciplinary policy, most staff members started their comments with that fact, whether they liked the new policy, how it differed from the old policy, and some explained the process. Twenty-six staff members were interviewed from the two medium facilities, from maximum, and from the HSC. Staff members were chosen who had dealings with prisoners with high levels of serious misconduct and violence. The interviewees included most of the deputy wardens, both wardens, lieutenants who conducted disciplinary boards at each facility, correctional counselors at each facility, and various uniformed staff at the HSC. Responses among staff are categorized by respondent rank

115 only when patterns emerged that had to do with rank (e.g. most of the correctional officers felt one way but the lieutenants thought another way). To maintain the anonymity of responses, the responses of the two wardens who were interviewed are referred to as deputies. Similarly, the responses of the two captains who were interviewed are referred to as lieutenants. Members of the central office classification staff are referred to as correctional counselors. All females are referred to as ―he‖ to maintain the anonymity of the four females who were interviewed.

Prisoners were asked, ―Have you been booked in the last year? In general do you think the bookings issued are fair? Did you go through a disciplinary process? What do you think of the disciplinary process? Is it fair? Are the sanctions fair?‖ and what changes if any, they would like to see implemented. Responses among prisoners will only be categorized by place of interview/status when patterns emerged that had to do with where one was housed at the time of the interview. As mentioned in the Research Methodology, 25 prisoners were interviewed who had received a Class I dreport in 2008 and who noted willingness to be interviewed at the end of the survey. Thus they are not representative of the entire RIDOC population and probably are not even representative of all prisoners with serious misconduct. However, it is hoped that feedback from this group will shed light on their views about RIDOC policy and procedures and subsequently, on their stresses, prison hardships, ways of coping, and opinions about curbing serious misconduct and violence. Finally, no prisoners were interviewed who resided in the

Price building as none of them were in the high-disciplinary category. Thus when prisoners are discussing disciplinary procedures, none are referring to processes at Price.

Staff members‘ overall comments about the disciplinary process were categorized as favorable, unfavorable, or mixed. A little more than half of the staff comments were mostly favorable, one citing the change in policy as a ―remarkable improvement over the old one‖ and

116 another saying that ―when (the policy) was revamped, it was for the better.‖ Most staff believed that the policy has had a deterrent effect on prisoner misconduct, and that it has provided more flexibility to correctional officers in choosing a booking level. However, a little less than half of the staff gave the disciplinary process mixed reviews and three staff members disliked it. These staff most often said that the policy itself was very good, but that the implementation of the policy was problematic. Overall, many of the staff, especially the deputies, felt the policy was fair and that it provided a system of ―checks and balances.‖ Several staff commented on the fact that bookings, hearings proceedings, sanctions, and appeals results are recorded in the INFACTS database; one said it simplified the process and was ‗user-friendly.‖ Another noted that

INFACTS allowed administrators and those running disciplinary and classification boards to easily scan an inmate‘s disciplinary history. However, a few said that recording bookings on the computer had made giving out a dreport more difficult and more complicated, sometimes reducing the amount of detail that a correctional officer (CO) would include in the report.

DISCIPLINARY PROCESS BOOKINGS Staff gave both positive and negative feedback on the dreport options—that is the categories and levels of individual charges that officers choose from when writing a dreport. The previous policy only offered a few dozen booking options, whereas, there are over 160 different charges in the new policy. Deputies commented that it provided officers with a ―menu of infractions to choose from‖ and that it contained a ―lot of very specific infractions‖ to cover many more situations than the prior policy covered. However, one of the lieutenants felt that the codes were either ―too generic or too specific, (that it was sometimes) hard to fit dreports into them,‖ and that many officers were not sure what to do with the discretion. He said, and indeed

117 a few officers acknowledged, that it is a common practice to book the misbehavior into a higher class than warranted because they thought it would be lowered anyway. Another lieutenant thought that the class levels of some of the codes were too high and that others were too low— that some staff who thought the charge warranted a higher or lower sanction would try to fit the booking into a category that matched the sanction they thought appropriate. One lieutenant suggested that there needed to be more education about the codes for both inmates and staff. He felt that inmates should be apprised of the behaviors that they could be booked for and that staff should be given more training on how to choose categories for bookings. Indeed one prisoner said that since one can now receive a year in segregation for predatory behavior, that the administration needed to explain more what ‗predatory‘ behavior entailed. He wondered if it included his aggressively arguing with staff or if he made a stupid comment in the heat of the moment. The lieutenant suggested that in the future whenever there were changes in the policy that staff and prisoners are notified, that they are given the opportunity to read it, think about it, and then have time to ask questions of administrators to ensure that they understand it.

More than two-thirds of the prisoners interviewed had been booked in the past year and almost all had had a serious booking either in the past year or the year prior. When asked if the bookings were fair, two of the 23 prisoners responded affirmatively. Of the remainder half said

―no‖ and the other half said that it depended—that sometimes bookings were fair and sometimes they were not. Many prisoners acknowledged their responsibility for their bookings. For example, one responded that he thought the bookings he had received were fair because he

―caused them.‖ Another gave an example of what he thought was a fair booking. He had been booked for fishing (moving property down the hallway from one cell to another); ―that‘s against

118 the policy, but you might get five days loss of privileges—but it‘s policy and you‘re breaking the rules‖ so he deemed it as fair.

However, many of the prisoners thought that many of the bookings were unfair. There were three reasons that prisoners gave for why they thought bookings were unfair: ―pettiness‖ of bookings; bookings for behavior they did not commit; and inconsistent bookings among staff.

By far the most prevalent complaints were that officers gave out bookings for minor misconduct.

One prisoner complained, ―There are certain bookings that are petty…the stuff they put you in segregation for…if you tell us to tuck in our shirts, we will, if you stick the mirror out the doorway, they book you…there are too many items. I could see if I had a weapon or was threatening somebody.‖ Prisoners called many of the charges ―frivolous‖ and gave as examples being booked for eating slow, praying, and not tucking in your shirt. A prisoner from HSC explained that before the policy change, there were fewer types of infractions but now prisoners can get booked for petty things. He said, ―The warden made it sound like it was for our benefit, but now because of all of the categories, there are actually more bookings.‖ Another prisoner from Maximum noted the consequences of petty bookings, ―When you confine a man, you deprive him and dehumanize him. When you‘re in seg, you lose a bit of yourself…don‘t hold a guy in seg for 30 days because he has ketchup.‖ He said he understands that the guy should not have ketchup but that the punishment is too severe for the offense. Another said that some prisoners had been booked for having braided hair, ―that‘s not in the rule book…cops have their own rules and pick on whoever they want to pick on.‖ A few prisoners also noted that rookie officers are the most likely to give out ―petty‖ bookings ―as notches on their belt.‖

While prisoners mostly complained about ―petty‖ bookings, the dozens of examples they gave of unfair bookings were most often those they denied doing. Many prisoners claimed that

119 bookings had been fabricated by officers who disliked them. For example, one gave a typical scenario, ―say you‘re sleeping and it‘s count time. If the officer doesn‘t like you, he‘ll book you for not standing for count. You can get 30 days and a downgrade. He‘ll write that he banged on your door, when really he didn‘t even try to wake you up.‖ Many mentioned, as one said, that ―it is always your word against theirs.‖ Another prisoner recounted that once when an officer grabbed him, he pushed him off but was written up as if he had assaulted the officer, ―I never had proof that he hit me because the tapes (in the camera) were not on…I just did my seg time.‖

A final example of how prisoners were unfairly booked involved a recent raid in the Moran housing units that turned up contraband items. However, some of the prisoners booked had been incarcerated for a long time and had purchased many of the items from the prison commissary before they were banned.

Finally, prisoners complain about inconsistencies in booking: some officers will let certain behaviors go while other officers will book prisoners at first sight of the same behavior.

For example, a prisoner recounted being booked for having extra food and given a two-nighter but explained that the officers on the second shift usually allowed it. One prisoner said that he had been booked 27 times in the past year, 24 times by the same officer. He said that other officers had come by to tell him that he should not have been booked. He believed that the officer who had so often booked him was being overzealous or possibly did not like him since other officers would not have booked him for the same behavior. In addition prisoners reported that some inmates who have been incarcerated a long time are often cut more slack.

DISCIPLINARY PROCESS As mentioned above, staff overwhelmingly approved of the relatively new disciplinary policy, but several felt that the implementation of the policy needed to be improved upon.

120

Specifically, staff brought up four areas where they believed the process could be improved: investigation, the disciplinary hearing, use of confidential informants, and the processing of prisoners with mental health problems. Prisoners also had criticisms regarding those four areas but also were critical of other aspects such as lack of evidentiary material, the lack of an advocate, and the preponderance of guilty findings.

INVESTIGATION AND CONFIDENTIAL INFORMANTS As mentioned in the description of the policy, a warden‘s designee, usually another lieutenant, is supposed to review a disciplinary report, investigate it if it is appropriate, and determine if it is properly written and whether the charge matches the behavior discussed. One lieutenant said if proper investigations are not carried out, ―it opens up the door for mistreatment by staff.‖ He recommended that the actual process of investigation should be re-examined to ensure that violations are coded properly and that the inmate actually did something wrong. He explained that persons investigating a booking should speak to both the reporting officer and the inmate and should look into any questions that the inmate might raise. Some staff believed that dreports where the inmate was truly not guilty should be thrown out by the investigating officer before getting to a hearing. One example given by a correctional counselor involved an inmate who was booked for having excess property because he had multiple pillows on his bed. He was put in segregation, even after explaining that it was his job to hand out pillows and that he had put them on his bed while delivering them. The counselor did not believe the matter should have reached the disciplinary board since the inmate‘s story could easily have been corroborated beforehand. A few lieutenants running disciplinary boards complained about the lack of investigation, about the lack of available evidence, and that dreports triggered by Special

Investigative Unit (SIU) staff lacked specific details. As mentioned previously, some prisoners

121 felt that they had been unfairly booked and several specified the lack of investigation as a problem. As one grumbled, ―they need to investigate more instead of all believing what the cops said. Because we‘re in jail, (they think) we‘re all bad, that we‘re all liars.‖ Similarly, several prisoners complained that they were not allowed witnesses even though they are supposed to have them by policy. They said that the disciplinary officer often tells them that they should not bother, that they are not needed.

Closely associated with the lack of investigation is the use of confidential informants.

One staff person acknowledged that when a booking is the result of a confidential informant

(CI), that the disciplinary board is a formality because he cannot have the informant appear before the board. He said that only the warden can question an informant. A lieutenant complained about the overuse of CIs and that they are usually the same ones all of the time. He complained that they are supposed to be trustworthy and give good information, but wondered how anyone really knows if they are and what their motivations are for stepping forward. He said that bookings by SIU with CIs are usually two or three lines and there is nothing to do at the hearing but find the prisoner guilty. He gave as an example one prisoner charged as a mule for bringing in drugs. The prisoner had had no prior criminal history of drug use or drug sales or institutional history of drug possession or trafficking and he adamantly denied the charge, but the CI‘s complaint was deemed sufficient and he had to find him guilty, which he thought was unfair to the inmate. One prisoner explained that he tried to tell the board that if another prisoner does not like him, ―all they need to do is go to a CO and say I‘m doing something.‖ Several prisoners agreed that if a CI comes forward with false information, one is always found guilty.

122

DISCIPLINARY BOARD MAKE-UP Several staff, many of them correctional counselors, did not like the change in the policy that reduced the number of people on a disciplinary board from three to one. As one counselor put it, ―I don‘t like how the d-boards are run. I like the old way, an officer, a counselor and the lieutenant and each had an equal vote. Now it‘s left to one person to be the judge and jury. If you have a lieutenant who is not particularly fair or who does not have favorable views of the inmates, it‘s not a fair process.‖ Another explained that correctional counselors often know an inmate better than the lieutenant running the disciplinary board and that if they have a say, they can point out situations that do not make sense and can explain circumstances. He explained that he once sat on a d-board where an inmate had been booked for not standing for count. He knew this inmate always stood for count and that he was about to go before the Parole Board. Despite being adamant that he did stand for count, he was found guilty. The counselor pursued it, finally showing a picture of the inmate to the rookie officer who had written the report who realized he had booked the wrong inmate. The counselor reported that, had he been on the board, he would have insisted on more investigation before finding the inmate guilty. While many staff credited the change in personnel on the d-board with making the d-boards more consistent, several said that there was little consistency among the lieutenants doing the disciplinary boards anyway and that some let their opinions of inmates affect their decisions. Several of the prisoners explained that whether the disciplinary process was fair or not depended on which lieutenant was conducting it and what his mood was that day. One prisoner also would like to see more people on the board saying, ―they need more people to agree and disagree and to see if there really is booking material. No one is on our side. We‘re always guilty.‖

123

MENTAL HEALTH The third area of concern is the process by which mental health workers review dreports to determine if any of them are related to prisoners‘ mental health status. As one lieutenant explained it, the process allows a mental health clinician to request that dreports be altered for inmates with mental health issues. However, he said that the process had never been put to the test as far as he knew. He said that he did not think that mental health workers regularly examined dreports—especially given their high caseload— and felt that correctional officers should be helping to identify relevant ones for the counselors to examine. One deputy agreed but said that sometimes the hearing officer picks up on possible mental health problems and asks a counselor to weigh in on how much it was a factor in a given dreport and that other times an administrator picks up a potential problem in the appeals process. Another lieutenant said that when mental health issues came to his attention, he would check into it and consider it, but does not think that prisoners should use mental health as a crutch to get out of dreports. This is a real dilemma. One prisoner said that he had told staff he should not get booked because of his mental health status and they had replied that he could not use that as an excuse. The inmate felt that since he had been labeled ―psychotic‖ he should be in a hospital, not in segregation.

OTHER PRISONER FEEDBACK ON THE DISCIPLINARY PROCESS Overall, out of the 22 prisoners who responded to the question about the fairness of the disciplinary process, 16 thought it unfair, three said it was fair, and three others said that it was sometimes fair. Most prisoners felt that if they had done something wrong, they should be held accountable. Some were dramatic in their criticisms calling disciplinary boards ―kangaroo courts‖ or run like ―Hitler‘s court.‖ Many said that going to the disciplinary board was a waste of time, since one is nearly always found guilty. As one put it ―if you get booked, 99 percent of the time that booking is sticking. It‘s rare to get thrown out. That‘s in every building, (because

124 it is) your word against the cops.‖ Prisoners often spoke about officers sticking together and one explained, ―if you‘re a lieutenant and an officer books somebody, you‘re going to take his side.

You‘re his brother. It‘s different if it was a deputy doing it.‖ Many complained that in addition to the lack of evidence, sometimes the hearing officer does not take into account extenuating circumstances. They also grumbled that there was nobody to advocate for them. While some acknowledged that they could request a counselor if they wanted, they explained that often the disciplinary officer discouraged it, by saying, ―do you really want us to get ____?‖ and often the prisoner decided that it was probably not worth it. One explained the importance of a counselor‘s presence to ―represent you, give you advice on how to deal with the problem, how to address it at the board, what the scenario might be. Somebody to give us advice. You want to do the right thing; you don‘t want to dig yourself deeper in a hole.‖

SANCTIONS SEVERITY OF SANCTIONS Many staff hailed the changes in the policy that increased the potential time in segregation from a maximum of 30 days to a new maximum of one year in order to address the most serious forms of misconduct. One correctional counselor said the change in policy had had a big deterrent effect because for inmates there was a ―big difference between 30 days and 365 days in segregation.‖ One deputy felt the change had been necessary because, under the old policy, no matter how horrendous an offense had been, they could only sanction inmates with 30 days of segregation. He gave two examples of inmates in the past who had killed officers in the line of duty and had been segregated for 30 days as punishment. However, he further explained that since there is an automatic review of sanctions that exceed 90 days, many disciplinary board officers keep the sanctions for most bookings under 90 days. He also indicated that predatory

125 infractions were not as prevalent as administrators had thought. He said he had been afraid that staff would give out too much segregation time, but that they had not. He acknowledged that inmates ―still accrue years in segregation, but that‘s because they get a lot of dreports.‖

A few of the staff interviewed felt that sometimes the sanctions given were too harsh.

One disciplinary hearings officer felt that he and his colleagues should be given more latitude regarding lengths of sanctions. He felt that more consideration should be given regarding inmates‘ past behavior, especially in those instances where it is the inmate‘s first disciplinary report. He said that in those circumstances, sometimes the booking did not warrant segregation and that the sanction ranges were not progressive enough. A few administrators said that sometimes staff tended to sanction on the harsh side and that they needed to ―keep an eye on it cause people can get heavy-handed.‖ One lieutenant felt it important to bring to the attention of both staff and inmates the seriousness of loss of good time so that COs would charge appropriately, disciplinary officers would sanction fairly, and inmates would hopefully try harder to avoid being involved in misconduct if they realized how much longer they would have to serve as a result of the good-time lost.

Only a few of the prisoners interviewed thought the level of sanctions given was fair.

One said, ―(It‘s) all right, if you‘re wrong, you‘re wrong‖ and another said that ―people don‘t usually say they got more than they deserved.‖ Another few prisoners said fairness depended again on who was conducting the disciplinary board—that ―some are fair and some are not.‖

However, two-thirds of the prisoners interviewed said that sanctions were unfair, but interestingly, not one mentioned the new guidelines that can go up to a year as too long. Most of the complaints were that disciplinary officers gave the maximum penalty, that segregation lengths were too long for many smaller offenses, and that in Maximum and the HSC, one could

126 get ―buried in seg time.‖ A number of prisoners protested that some of the lieutenants conducting the boards always gave the maximum and always gave out segregation time. They described the amount of time as being ―too much,‖ ―too punitive,‖ and ―too drastic.‖ One prisoner complained that ―even if you tell you did it and you show remorse, he still gives you the max, no matter what.‖ Another griped that he had gotten 15 days confinement to his cell for having an extra apple. One prisoner could not produce a urine sample on demand, was booked, and received 26 days of segregation and loss of good-time, six months loss of visits, 90 days on

B status, and loss of his job. He said he really could not urinate and had not been given a second chance before being booked and thought the punishment too harsh. Some prisoners thought that alternatives to segregation could be used, like taking away radio or TV or other privileges.

SEGREGATION Both staff and prisoners talked about that small percentage of prisoners who were ―buried in seg.‖ Staff working at HSC pointed to two difficulties that they were having with prisoners: keeping track of their segregation time and whether further segregation time was effective. The first one was simply a difficulty of keeping straight the length of time a prisoner should be in segregation if he had incurred multiple dreports over a length of time. The half dozen staff at

HSC that brought up the effectiveness issue said that prisoners who accumulate a lot of time in segregation either want to live in segregation or have so much segregation time built up that they have nothing to lose. Of the first scenario, one correction officer articulated that in order ―for discipline to work, you have to take something away that they want. (But) to some of them,

(being in the HSC is) a better life. They don‘t have to cope with the prison society.‖ He concluded that sanctions are only good if they ―matter.‖ A lieutenant related that often when he asks inmates why they keep getting booked, they say ―I want to do my time in segregation.‖

127

Regarding efficacy another CO said, ―Then you have the inmates with so much time racked up.

The general opinion of people is that more of the same is insufficient.‖ Another lieutenant said,

―Some are doing so much time, in their head, it‘s doesn‘t matter. Some flatten (have their sentence expire) before their segregation time is done.‖ A few prisoners interviewed reported that they were or had been ‗buried in seg‖ but only one of those said it had been his choice; he admitted bizarre behavior in Maximum in order to be transferred to the HSC. However, there were many prisoners in the HSC who declined participation in either the survey or the interview and it is completely plausible that some of them prefer being housed in segregation than in general population. Of those interviewed, many discussed the difficulty of getting out of segregation. One said, ―If you are working hard to be on the right path and get booked for something small, like 20 days seg for borrowing someone‘s magazine, it makes me feel like,

‗what am I being good for? I may as well be bad.‘‖ Some prisoners agreed that after a while in segregation that it became easier, that one did grow used to it. For example one explained,

―here, you keep getting seg time, but once you get used to not having visits and you don‘t have any money to buy stuff, it doesn‘t matter anymore if you get more seg time. (People) get to the point where they have nothing to lose, nothing to do.‖ However, not all prisoners found it easy to spend lengths of time in segregation. For example, one said, ―It messes a lot of people‘s head up. I‘m going on two years now, I think I‘m going to be worse off than I was before…going back into population with so many people. Right now when I go out, I‘m in a cage with only one other person. So eventually when I do get out of seg and go to max, I‘ll be with a hundred people (out in the yard). You get to the point you don‘t care no more, you lose your patience.‖

He continued that sometimes he would start having ―bad thinking…then you have COs who

128 know how to tip you off…you start kicking your door and flooding your cell. You don‘t care anymore cause you‘re buried doing time in seg.‖

However, one deputy explained that inmates in long-term segregation in both the HSC and in Maximum are reviewed by the warden every 90 days, who by policy, has the ability to suspend segregation days. The warden uses the review to empower officers, telling inmates that if the officers report that he is doing well, he will suspend some segregation time. One counselor said that once inmates realize they cannot beat the system, they often take advantage of an offer like ―You show me you‘re good for the next 60 days, and I‘ll cut you a break.‖

SANCTION ALTERNATIVES In addition to segregation time, disciplinary officers can recommend that the prisoner go before a disciplinary board for a possible downgrade, although classification staff said that downgrades were not common unless the prisoner was deemed unmanageable. Disciplinary officers can also recommend that the prisoners pay restitution when state property is damaged, but not all disciplinary officers use this option. Another tool that correctional staff can use to minimize serious misconduct is to bring criminal complaints against the inmates. As one lieutenant said, ―they don‘t prosecute fights, most are handled in house. (But) on serious assaults, they are criminally charged. We‘ll secure the crime scene and the SIU will come in and investigate and bring in internal affairs.‖ Then if deemed appropriate, state troopers are brought in to determine whether it should go to the state Attorney General‘s office for prosecution. One correctional counselor explained the change in Rhode Island state law8 regarding good time and that well-behaved prisoners could earn up to ten days per month of good-time, in addition to

8 Rhode Island General Law (RIGL §42-56-24).

129 possible good-time for jobs (two days), programs and education (up to five days), and 30 days for completion of a program. Taken together many staff felt that the new policy in combination with the new good-time law provided flexibility to the staff and a deterrence to misbehavior.

*******

In summary, almost all RIDOC staff members interviewed were extremely favorable towards the relatively new disciplinary policy calling it flexible and comprehensive. About half believed that the implementation of the disciplinary policy was at times problematic and needed further attention. Regarding bookings, most of the staff liked the new categories and levels of bookings, claiming they offered staff more discretion and that the ranges allowed discipline to be progressive. Prisoners did not comment on the categories for bookings but on their fairness.

Prisoners labeled bookings as ―petty‖ if they were falsely accused or had reasonable explanations for their behavior, and if staff were inconsistent in their bookings.

Staff brought up four aspects of the disciplinary process that they believed needed improvement: investigation, use of confidential informants, one-man disciplinary boards, and dealing with bookings involving prisoners with mental health problems. In addition, more than two-thirds of the prisoners interviewed believed the disciplinary process to be unfair, in that they felt they were always found guilty, there was often a lack of evidence, and that extenuating circumstances were often not taken into count.

Most of the staff complimented the sanction structure of the new policy especially the increase in the maximum potential sanction in segregation from 30 days to 365 days, crediting it with having a deterrent effect on prison violence. Some staff thought that sanctions for less serious violations were too harsh, and that the range did not give disciplinary officers enough

130 leeway to implement progressive discipline. Two-thirds of prisoners thought the sanctions were unfair: that they were almost always given the maximum in some facilities, that sanctions for lesser offenses were too harsh, and that some prisoners in HSC and Maximum ended up being

―buried in seg‖ for long periods of time; the latter also considered problematic by staff.

131

VI. INDIVIDUAL PREDICTORS OF SERIOUS MISCONDUCT AND VIOLENCE

This chapter includes a description of the sample and the results of all of the bivariate analyses of the predictors of prison misconduct as well as the qualitative data about certain of these variables. The first four sections present the results of the bivariate analysis of prison misconduct with prisoners‘ personal characteristics, their criminal history and current sentencing, prisoners‘ psychiatric treatment, and their incarceration experiences. The next section analyzes the relationship between prison misconduct and the prison stresses reported in the survey and details these experiences as reported by prisoners in the in-depth interviews. The subsequent section discusses the findings of the bivariate relationship between trait emotions and prison misconduct. It is followed by the bivariate analysis of coping and prison misconduct along with prisoners‘ and staff members‘ feedback on coping in prison.

SAMPLE CHARACTERISTICS

This section describes the socio-demographics of the sample. Most of the socio- demographic data were provided by the RIDOC, though employment and relationship data were gathered through the self-report survey. It is reported for the entire sample as well as for both the low-level and high-level misbehavior groups

Starting first with race and ethnicity, slightly less than half of the sample was white, about one third was black, and 19 percent were categorized as Hispanic (Table 13). Prisoners in the ―other‖ category included two Asians and two American Indians. There were no appreciable differences between the group regarding race; the differences tested by the chi-square goodness- of-fit test were not statistically significant (χ2=4.15, ρ=.246). Similarly, no significant difference

132 was found for race/ethnicity when the sample was divided into a white and nonwhite dichotomy

(χ2 =1.76, with Fisher‘s Exact Test ρ=.212 not shown).

Although citizenship data were not available for almost half of the sample, place of nativity was available (not shown in table). Overall, about 87 percent of the prisoners had been born on the mainland of the United States, seven percent in Puerto Rico, and the remaining six percent outside of the United States—in the Dominican Republic, Cape Verde, Guatemala,

Liberia, Portugal, Great Britain, Jamaica, Mexico, and Panama. Forty-three percent (43 percent) of the sample resided in Providence prior to incarceration (not shown in table). Another 31 percent came from other Rhode Island cities, while 19 percent came from smaller Rhode Island towns. Seven percent (7 percent) lived out-of-state prior to their incarceration.

Turning to age, a little over one-third of the sample was under 30 years old and almost half were between the ages of 30 and 50; only 16 percent were over the age 50. At either ends of the age spectrum were seven prisoners (two percent) who were under 20 and twelve (four percent) who were over 60 years of age. The average age for the sample was just under 37 years of age. A much greater percentage of prisoners in the high-level misbehavior group were under

30 years old compared to the low-level group. Consistent with this finding, an examination of the older three categories of offenders shows that as age increases the percentage of offenders exhibiting high-level misbehavior decreases. Indeed, a t-test comparing the mean ages of those in the low-level misbehavior group with those in the high-level misbehavior group revealed that the differences were significant (t=4.34, ρ<.001).

Most of the prisoners in the sample were not married with two-thirds (67 percent) being single and 14 percent having been divorced or separated. A greater percentage of prisoners

133

Table 13 - Sample Characteristics by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total Mean % Mean % Mean % (SD) (SD) (SD)

Race & Ethnicity χ2(3, N=312)=4.15, ρ=.246 Black 28.9 38.1 33.7 White 50.0 42.5 46.2 Hispanic 20.4 17.5 18.9 Other 0.7 1.9 1.3

Age*** 40 34 37 t (310, N=312)=4.34,ρ<.001 (12.08) (12.14) (12.45) Under 30 23.0 47.5 35.5 30 - 39 30.3 21.2 25.6 40-49 26.3 18.1 22.1 50+ 20.4 13.1 16.6

Marital Status** χ2(2, N=310)=10.44, ρ=.005 Single 60.0 73.8 67.1 Married/Widowed 26.0 11.9 18.7 Divorced/Separated 14.0 14.4 14.2

Grade 11.2 11.1 11.1 t (307, N=309)=.40, ρ=.693 (1.93) (1.55) (1.74) Less than 12th grade 48.0 52.2 50.1 High school graduate 37.3 39.6 38.5 Some college 14.7 8.2 11.3

Previously Employed χ2(1, N=307)=1.87, ρ=.207 No 41.9 49.7 45.9 Yes 58.1 50.3 54.1

* ρ<.05 ** ρ<.01 *** ρ<.001 involved in misbehavior were single compared to those not so involved. Married or widowed men were less likely to have high levels of dreports. The chi-square analysis did reveal a significant relationship between marital status and involvement in serious disciplinary behavior

(χ2=10.44, ρ<.005).

134

Prisoners were also asked in the survey if they had been in a serious relationship during the three months prior to their incarceration. Almost two-thirds of the prisoners reported being in a serious relationship (not shown). Only slight differences between the groups were found for prior relationship. Chi-square analysis revealed that the Fisher‘s Exact Test was not statistically significant (χ2=4.15, with Fisher‘s Exact Test ρ=.051).

Half of the prisoners in the sample had not completed high school (10 percent of whom had not even reached high school at all) and only 11 percent had received education beyond high school. The mean grade for the entire sample was the 11th grade. Those in the high-level disciplinary category were slightly more likely not to have graduated from high school or to only be high school graduates compared to those in the low-level disciplinary category. Since the mean grade completed was the same in both categories (11th grade) it is not surprising that no statistical differences were found by the t-test analysis (t=.40, ρ=.693).

Almost half of the prisoners in the sample were unemployed during the three months prior to incarceration. The percentage of unemployed was somewhat greater in the high-level misbehavior category as compared to the low-level disciplinary category. However, the differences were not statistically significant (χ2=1.87, with Fisher‘s Exact Test ρ=.207).

*****

In summary, initial bivariate analysis of socio-demographic characteristics found a statistical relationship between level of prison misbehavior and only age and marital status.

There were no statistically significant relationships between race, educational level, or employment status prior to incarceration and serious disciplinary behavior in the current sample.

135

CRIMINAL HISTORY, SUBSTANCE ABUSE HISTORY, AND CURRENT SENTENCING

Criminal history, substance abuse history, and sentencing information came from both the RIDOC database and from prisoner surveys. In the survey prisoners were asked about being in a gang, juvenile adjudications resulting in residential placement, and whether they had been previously incarcerated. Information on prior incarceration in Rhode Island was also gathered manually using the INFACTS system. They were also asked detailed questions about their own substance abuse during the six months prior to their incarceration. The RIDOC provided data on prisoners‘ most serious charge and their current sentence length.

CRIMINAL HISTORY Only one-tenth of the sample reported that they had been in a street gang prior to their incarceration (Table 14). Informal feedback from prisoners about the survey revealed that this question and the one asking about gang affiliation while in prison were the two most sensitive questions in the survey and thus the ones that prisoners were less likely to answer at all or answer truthfully. However, there was not more missing data for this question than others about prior criminal history so one can only wonder about the validity of responses for this question.

Prisoners in the high disciplinary group were almost twice as likely to report having been in a gang on the street compared to those in the low discipline group (seven percent); however, the numbers are still relatively small and the differences were not statistically significant (χ2=2.41, with Fisher‘s Exact Test ρ=.133).

Overall 39 percent of prisoners reported that they had been adjudicated delinquent as a youth and committed to a secure juvenile correctional facility. Almost two-thirds of the sample reported a previous adult incarceration. The slightly higher percentage of prisoners reporting prior incarcerations compared to the percentage of officially-recorded prior incarcerations was

136 probably due to prisoners including incarcerations in states other than Rhode Island. Overall 38 percent had been previously incarcerated for a violent offense in Rhode Island.

Table 14 - Criminal History by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total % % %

% Reported Having Been 7.4 12.7 10.1 in a Street Gang χ2(1, N=306)=2.41, ρ=.133

% Reported Placement 29.1 48.4 38.9 Juvenile Sec. Facility*** χ2(1, N=303)=11.90, ρ<.001

% Reported Prior 52.3 66.2 59.4 Incarcerations* χ2 (1, N=308)=6.19, ρ=.015

% With Actual Prior 51.3 63.8 57.7 Incarcerations in RI* χ2 (1, N=312)=4.94, ρ=.017

% Prior Incarcerations for 36.2 40.0 38.1 Violent Crime in RI χ2 (1, N=312)=.48, ρ=.282

* ρ<.05 ** ρ<.01 *** ρ<.001

Prisoners in the more serious disciplinary group were more likely to report having been adjudicated delinquent and committed to a secure juvenile correctional facility than prisoners in the less serious disciplinary group. The differences were statistically significant (χ2=11.90, with

Fisher‘s Exact Test ρ<.001). Similarly, those in the high disciplinary group were also more likely to have been previously incarcerated compared to those in the low disciplinary group, the differences again being statistically significant for reported prior incarcerations (χ2=6.19, with

Fisher‘s Exact Test ρ=.015) and for actual prior incarcerations in Rhode Island (χ2=4.94, with

Fisher‘s Exact Test ρ=.017). However there was no statistically significant difference in the

137 percentage of high-disciplinary prisoners who had previously been incarcerated for a violent crime compared to low-disciplinary prisoners (χ2=.48, with Fisher‘s Exact Test ρ=.282).

SUBSTANCE USE Prisoners were asked to report their substance use during the six months prior to incarceration (Table 15). Almost two-thirds of the entire sample reported using tobacco and almost three-fourths reported drinking alcohol. Two-thirds of the prisoners responded affirmatively to using any drug including pot, cocaine, crack, heroin, inhalants, or any other type of drug. Almost the same proportion reported smoking pot. Smaller percentages of prisoners reported using harder drugs such as cocaine or crack (30 percent), heroin (14 percent), or inhalants (2 percent). The use of other drugs included mostly methamphetamines and some hallucinogens and the abuse of over-the-counter prescription drugs such as Ritalin. Interestingly, one-third of the sample (33 percent) had used no drugs in the six months prior to incarceration, one-third (33 percent) had used one drug, 23 percent had used two drugs, and the remaining 11 percent had used three or more drugs during that period.

There were few differences in substance abuse use between the low- and high-level disciplinary groups. There were no statistically significant differences between the groups when any drug use was measured as a ‗yes‘ or ‗no‘ question (χ2(=3.18, with Fisher‘s Exact Test

ρ=.082). However drug use was also measured as a count variable ranging from using zero to five different types of drugs in the six months prior to incarceration. A t-test comparing the mean number of drugs used found high-level prisoners used more; the differences were statistically significant (t =-2.27, ρ=.024). There were no statistically significant differences between the two groups regarding the specific use of tobacco, alcohol, or any type of drug

138

Table 15 - Substance Abuse Six Months Prior to Incarceration by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total % % %

% Used tobacco 57.4 66.9 62.3 χ2(1, N=305)=2.89, ρ=.099

% Used Alcohol 70.7 75.0 72.9 χ2(1, N=306)=.73, ρ=.441

% Used Any Drugs 62.3 72.1 67.2 χ2(1, N=293 =3.18, ρ=.082

% Used Pot** 52.0 68.6 60.5 χ2 (1, N=306)=8.80, ρ=.003

% Used Cocaine / Crack 28.2 31.8 30.0 χ2(1, N=303)=.47, ρ=.532

% Used Heroin 10.2 18.0 14.1 χ2(1, N=297)=3.72, ρ=.067

% Used Inhalants 1.4 2.0 1.7 χ2(1, N=298)=.18, ρ=1.000

% Used Other Drugs 13.6 17.6 15.7 χ2(1, N=300)=.93, ρ=.346

Mean (SD) Mean (SD) Mean (SD)

Count of Drug Types Used 1.03 (1.04) 1.32 (1.16) 1.17 (1.11) t (291, N=293) =-2.27,ρ=.024

* ρ<.05 ** ρ<.01 *** ρ<.001 with the exception of marijuana. A higher percentage of prisoners in the high-level misbehavior group smoked pot compared to those in the low-level group. The chi-square analysis revealed the relationship to be statistically significant (χ2=8.80, with Fisher‘s Exact Test ρ=.003).

CURRENT SENTENCING STATUS The most serious charge resulting in each prisoner‘s current sentence was categorized into several offense types (Table 16). Over half of the prisoners in the sample were incarcerated

139 for a violent offense. One-fifth were incarcerated for a sex offense, about 10 percent for

Breaking and Entering (B & E), 12 percent for drug sales and possession, and another five percent for either weapons charges or a nonviolent crime.

About the same percentage of prisoners in the high disciplinary category were incarcerated for violent offenses as in the low disciplinary category. Similarly about the same percentage in each disciplinary category were charged with either weapons or nonviolent crimes.

However, the groups differed dramatically on other offenses. Only 12.5 percent of those in the high disciplinary group, but 30 percent in the low group were sex offenders. Conversely, 15 percent of the prisoners in the high disciplinary group had committed B & Es compared to only four percent in the low disciplinary group. A similar discrepancy was found regarding drug charges, with higher percentages of those in the high disciplinary group charged with drug crimes compared to the low-level group. The chi-square analysis of the various categories by misconduct revealed the differences to be statistically significant (χ2=24.99, ρ<.001).

About 10 percent of the prisoners in the sample had received life or life without parole.

The average sentence for the rest of the sample was 11 years; the range was six months to 85 years. About a fifth of the prisoners were serving a total sentence of four years or less.

The two disciplinary groups differed in total sentence length with those in the high disciplinary group having shorter sentences than those in the low disciplinary group.

Specifically about two-thirds of those in the high disciplinary group were serving eight years or less, compared to 41 percent of prisoners in the low disciplinary group. Conversely, a greater percentage of those in the low disciplinary group were serving more than 12 years and 14 percent were serving life sentences. The chi-square analysis revealed the differences to be statistically

140 significant (χ2=22.82, ρ<.001). The t-test comparing the means of the two groups‘ sentences

(excluding the lifers) revealed the mean total sentence for the high disciplinary group to be 8.6 years compared to 13.8 years for the low disciplinary group. The differences were also significant (t=3.88 ρ<.001).

Table 16 - Most Serious Offense Type and Total Sentence Length by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total Mean % Mean % Mean % (SD) (SD) (SD)

Most Serious Offense Type*** χ2(6, N=312)=24.99, ρ<.001 Violent 53.3 52.5 52.9 Sex 30.3 12.5 21.2 B & E 3.9 15.0 9.6 Drug Sales & Possession 7.9 15.0 11.6 Weapons 1.3 1.2 1.3 Nonviolent 3.3 3.8 3.5

Total Sentence Length*** 5029 days 3145 days 4009 days t (275, N=277)=3.88, ρ<.001 13.8 years 8.6 years 11.0 years (4791.4) (3248.4) (4130.6) χ2(4, N=306)=22.82, ρ<.001 Less than 4 years 12.2 25.3 19.0 4 – 8 years 29.1 37.3 33.3 8 – 12 years 10.1 13.3 11.8 More than 12 years 34.5 19.0 26.5 Life with/without parole 14.2 5.1 9.5

Time Incarcerated*** 2118 days 1340 days 1719 days t (310, N=312)=4.66, ρ<.001 5.8 years 3.7 years 4.7 years (1684.84) (1239.75) (1521.84)

* ρ<.05 ** ρ<.01 *** ρ<.001

The average time-incarcerated for the sample was 4.7 years, with 28 percent having served less than two years, and 30 percent having served more than five years of their current sentence (not shown). Prisoners with serious disciplinary problems had been incarcerated for less than two years (37.5 percent) compared to those with fewer disciplinary problems (18

141 percent). Conversely, 43 percent of those in the low-disciplinary group had been incarcerated for more than five years, compared to 18 percent of the high-disciplinary group. The mean time- incarcerated for the high-disciplinary group was 3.7 years compared to 5.8 years for the low- disciplinary group; the differences were statistically significant (t=4.66, ρ<.001).

*****

In summary, prisoners in the high-level misbehavior category were much more likely to have had a more extensive criminal history—including prior adjudication as a juvenile and prior incarceration as an adult—compared to those in the low-level misbehavior category. Regarding substance abuse, only the number of drugs used in the six months prior to incarceration and the use of marijuana correlated with serious prison disciplinary behavior. The differences between the offenses of the two groups were statistically significant, although slightly more than half of both disciplinary groupings were in for a violent offense, or for the less frequent weapons or non-violent charges. However, a much greater percentage of the low-level misbehavior group was made up sex offenders, compared to those in the high-level misbehavior group. Conversely, a much greater percentage of the high-level misbehavior group was made up of prisoners involved in B & Es and drug charges, types of charges that often go hand-in-hand. Given the most prevalent type of serious offense for each group, it is not surprising that high-level misbehavior prisoners have, on the average, shorter sentences than low-level misbehavior prisoners and that those in the former group are less likely to be serving life with or without parole. Finally, prisoners in the high disciplinary group had been incarcerated for less time than those in the low disciplinary group.

142

PSYCHIATRIC TREATMENT

This section presents information gathered in the self-report surveys on whether prisoners had been treated for psychiatric or emotional problems before and during their incarceration.

Prisoners were asked about being seen individually or in groups, having been transported to a psychiatric hospital or unit, or being on medication for a mental health or emotional problem either prior to or during their current incarceration.

Regarding treatment prior to incarceration, slightly over one-third of the sample reported having been treated for psychiatric or emotional problems, either by meeting individually or in group meetings with psychiatric staff, by being transferred to a psychiatric hospital or unit, or by taking medication for psychiatric or emotional problems (Table 17).

When the sample was divided by disciplinary group, 42 percent of those in the high disciplinary group had been previously treated for psychiatric or emotional problems compared to 29.5 percent of the low disciplinary group; the results were statistically significant (χ2=6.28, with Fisher‘s Exact Test ρ=.016). There were only slight differences between the two groups regarding being seen individually or in groups by psychiatric staff, or being transferred to a psychiatric hospital or unit—however, these differences were not statistically significant. The widest gap between the two groups was for taking medication for psychiatric or emotional problems. Thirty-two percent of prisoners in the high disciplinary group had taken such medications prior to their incarceration compared to only 17 percent in the low group. The differences were statistically significant (χ2=9.66, with Fisher‘s Exact Test ρ=.002).

Regarding treatment in prison, forty-two percent of the sample reported being in some type of psychiatric treatment while in prison (Table 18). Over one-fourth were on medication for

143

Table 17 – Psychiatric Treatment Prior to Incarceration by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total % % %

% Treated Before* 28.5 42.4 35.4 χ2(1, N=305)=6.28, ρ=.016

% Seen by Psych Staff 17.9 20.1 19.0 Before χ2(1, N=305)=.25, ρ=.663

% in Psych Group Before 7.9 11.0 9.5 χ2(1, N=305)=.85, ρ=.436

% Transferred to Psych 11.0 16.2 14.1 Hospital Before χ2 (1, N=305)=1.17, ρ=.325

% Taking Psych Medication 16.6 31.8 24.3 Before** χ2(1, N=305)=9.66, ρ=.002

* ρ<.05 ** ρ<.01 *** ρ<.001 psychiatric and emotional problems and one-fourth had seen prison mental health staff.

Overall a greater percentage of prisoners in the high-level misbehavior group had been treated in prison for psychiatric or mental health issues compared to those in the low-level misbehavior group. The differences were statistically significant (χ2=4.48, with Fisher‘s Exact

Test ρ=.037). Similarly, eight percent of those in the high-disciplinary group but only two percent of those in the low-disciplinary group had been transferred to a psychiatric unit or facility while incarcerated. Again the differences were statistically significant (χ2=5.14, with

Fisher‘s Exact Test ρ=.032). There was little difference between the groups regarding being seen by prison mental health staff or taking medications and virtually no difference in the percentages of prisoners involved in group treatment in the prison.

144

Table 18 – Psychiatric Treatment in Prison by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total % % %

% Treated in Prison* 36.1 48.1 42.2 χ2(1, N=303)=4.48 ρ=.037

% Seen by MH Staff in 22.4 30.1 26.4 Prison χ2(1, N=303)=2.30, ρ=.152

% in Prison Psych Group 2.7 2.6 2.6 χ2(1, N=303)=.01, ρ=1.000

% Transferred to Psych 2.0 7.7 5.0 Hospital while in Prison* χ2 (1, N=303)=5.14, ρ=.032

% Taking Psych Medication 23.1 32.7 28.1 in Prison χ2(1, N=303)=3.43, ρ=.074

% Treated Before & During χ2(1, N=147)=25.63, χ2(1, N=152)=53.98, Prison *** ρ<.001 ρ<.001

Of those treated before: % Treated in Prison 68.3 81.2 76.2 % Not Treated in Prison 31.7 18.8 23.8

Of those not treated prior: % Treated in Prison 23.6 21.6 22.7 % Not Treated in Prison 76.4 78.4 77.3

* ρ<.05 ** ρ<.01 *** ρ<.001

There was a high correlation between whether prisoners had been treated prior to incarceration and whether they were treated during incarceration: the relationship was statistically significant for both the low- and the high-level misbehavior groups. Interestingly, a greater percentage of prisoners from the high-level group (81 percent) who had been previously treated were also treated in prison, compared to the percentage in the low-level group (68 percent). Still 19 percent of the high- and 32 percent of the low-level misbehavior groups had

145 been in treatment on the street but were not receiving it in prison. On the other hand, about one- fifth of each group that had not received treatment previously began treatment once in prison.

*****

Those prisoners in the high disciplinary group were more likely to have reported some type of treatment for psychiatric or emotional problems both before and during their incarceration compared to prisoners in the low disciplinary group. Prisoners with disciplinary problems were also more likely to have been on psychiatric medication prior to prison and to have been transferred to a psychiatric unit or facility while in prison. Prisoners in both groups had been treated prior to entering prison but were not being treated in prison and conversely others who had not received treatment on the street were being treated in prison.

INCARCERATION EXPERIENCE

This section describes prisoners‘ incarceration experiences including their level of outside social support, how they spent their time in prison, and their prior disciplinary history. In the self-report survey, prisoners were asked questions about the frequency of visits and letters received monthly as well as the number of prisoners they considered their friends. They were asked a series of questions about how they spent their days in prison. Prior disciplinary data was culled from the RIDOC database for those prisoners who were incarcerated from 2003 to 2007.

LEVEL OF SOCIAL SUPPORT Prisoners‘ level of social support while in prison was examined by three measures: the number of reported visits and letters received per month, and prisoners‘ reported number of friends within the facility (Table 19). Prisoners reported receiving a little over an average of three visits from outsiders per month. In addition, on average, prisoners said they received a

146 little more than four letters per month. Finally, when asked how many friends they had in prison, prisoners reported an average of four friends in prison at the time of the interview.

There were few differences between prisoners in the high-level and low-level disciplinary groups regarding their level of social support. When examining the level of visits reported per month, 33 percent of the low-level group and 41 percent of the high-level group reported receiving no visits during a typical month‘s time (not shown in table). The average number of visits reported by each group was three per month; thus there was no real difference between the groups (t=3.43, ρ=.732). There was a small difference between the two groups in the number of letters that prisoners reported receiving per month. High discipline prisoners received about five letters per month compared to an average of a little more than three letters per month for the low discipline group; the differences were statistically significant (t=-2.53, ρ=.012). Again, this is not surprising because high-level prisoners are more apt to have lost their visiting privileges while in segregation or in the HSC and may resort instead to letter-writing.

Table 19 - Levels of Social Support by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total Mean Mean Mean Social Support Types (SD) (SD) (SD)

# Visits You Receive in a Month? 3.47 3.30 3.38 t (303, N=305)=3.43, ρ=.732 (4.55) (4.44) (4.49)

# Letters You Receive in a Month?* 3.57 5.23 4.43 t (305, N=307)=-2.53, ρ=.012 (4.04) (6.95) (5.78)

# Friends You Have in Prison? 5.53 2.59 4.03 t (296, N=298)=1.08, ρ=.281 (33.09) (5.52) (23.50)

* ρ<.05 ** ρ<.01 *** ρ<.001

147

The final indicator of social support while in prison was the number of friends that prisoners reported they had at the time of the interview. Thirty percent of prisoners in the low and 37 percent in the high misbehavior group reported that they had no friends in prison.

Similarly, prisoners in the high-level group reported an average of a little more than two friends, while those in the low-level group reported an average of a little more than five friends.

However, the differences were not statistically significant (t=1.08, ρ=.281).

HOW PRISONERS OCCUPY THEIR TIME In the self-report survey, prisoners were asked to estimate how many hours per week they spent doing various activities. They were asked not to include the number of hours spent sleeping. As mentioned previously, some prisoners were diligent about figuring out the number of waking hours per week and ensuring they counted for each of those. Others guessed at hours with the result of underestimating or overestimating their waking hours. Because of this, each prisoner‘s hours were totaled and the percentage of each person‘s total activity hours was calculated for each activity, thereby standardizing the separate activities across prisoners.

The activities that took up the greatest percentages of prisoners‘ time overall was watching television or listening to the radio (M=23 percent) and socializing with other prisoners

(M=14 percent), together making up more than one-third of prisoners‘ waking hours (Table 20).

Prisoners spent 11 percent of their time reading and 10 percent of their time playing sports or working out. Only about 11 percent of prisoners‘ time was spent working. This contrasts with the 40 hour work week, which is approximately 36 percent of most people‘s waking hours.

Being idle in one‘s cell averaged about eight percent of total activity hours per week. Finally, going to school accounted for four percent and attending treatment programming accounted for less than one percent of prisoners‘ total activity hours within a week.

148

There were few differences between the prisoners in the low-level misbehavior group and those in the high-level misbehavior group regarding the percentage of time spent in most activities. Both groups spent an equivalent amount of time watching television, listening to the radio, socializing, doing sports, working out, reading, and attending school and treatment programming. Prisoners in the high-disciplinary group were less likely to spend time working and more likely to be idle in their cell than those in the low-level group—the differences were statistically significant (t=4.80, ρ<.001). This is not surprising since a larger percentage of prisoners in the high-level misbehavior group resided in the HSC or were in segregation where there is little chance for activity outside one‘s cell. Thus, the 22 prisoners housed in the most restricted section of the HSC or who were in segregation at the time of the interview were removed and the t-tests were rerun on all of the activities. The results were similar with those in the high-level group still reporting that work took up only seven percent of their time, while it took up 17 percent of prisoners‘ time in the low-level group—the differences were again statistically significant (t=-4.03, ρ<.001, not shown). Of those who reported working, the actual number of hours worked ranged from one hour per week (11 percent) to 68 hours per week.

Fifty percent of the prisoners worked 10 hours per week or less. They often worked as ―porters‖ who were responsible for cleaning an area, either in a housing unit or in another building in the prison. Many of these jobs took between 15 minutes to one hour per day. Overall, only six percent of workers worked 35 hours or more, usually in the Prison Industries or in the kitchen.

There were also pronounced differences between the two groups regarding being idle — the differences were statistically significant (t=-3.26, ρ<.001). With those in segregation removed again, idle time accounted for 10 percent of time for those in the high-level group and five percent of time for those in the low-level group—the differences were again statistically

149 significant (t=-3.06, ρ=.002, not shown). Thus even those prisoners in the high-level misbehavior group who are in population are less likely to be working and more likely to be idle than prisoners in the low-level disciplinary group.

Table 20 - Mean Percentage of Weekly Activity Hours by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total Mean % Mean % Mean % Type of Activity (SD) (SD) (SD)

Work*** 16.4 6.4 11.2 t (241, N=243)=4.80, ρ<.001 (19.37) (12.43) (16.81)

School 3.8 3.4 3.6 t (267, N=269)=.45, ρ=.651 (6.00) (6.60) (6.31)

Socializing 14.8 13.3 14.0 t (232, N=234)=.93, ρ=.354 (13.33) (10.85) (12.10)

Treatment Programming .71 .96 .84 t (282, N=284)=-7.44, ρ=.457 (2.26) (3.25) (2.81)

Television/Radio 22.9 22.8 22.9 t (223, N=225)=.02, ρ=.980 (16.52) (18.68) (17.65)

Sports/Working Out 8.5 10.4 9.5 t (241, N=243)=-1.63, ρ=.104 (8.07) (10.48) (9.42)

Reading 10.4 12.3 11.4 t (229, N=231)=-1.08, ρ=.280 (11.62) (14.84) (13.40)

Time Idle in Cell*** 4.8 10.7 7.7 t (255, N=257)=-3.26, ρ<.001 (11.86) (16.76) (14.8)

* ρ<.05 ** ρ<.01 *** ρ<.001

Two indices of participation were created to summarize some of the activities (not shown). Prisoner-initiated activities added the percent of time involved in socializing, visiting, and sports and ranged from zero to 100 percent. Though low-level misbehavior prisoners spent

150 slightly less time in prisoner-led activities (mean=29.6) than high-level misbehavior prisoners

(mean=32.1), the differences were not statistically significant (t=-1.02; p=.308). The other index was prison-led activities and included working, school, and treatment programming and ranged from zero to 100 percent again. Again there were differences between the two disciplinary groups with low-level prisoners spending a greater percentage of their week in prison-led activities (mean=22.1) and high-level prisoners spending less time in these constructive activities

(mean=11.6). The differences were statistically significant (t=4.25; p<.001).

PRIOR DISCIPLINARY HISTORY

Prisoners‘ disciplinary data were extracted from the RIDOC database for 2003 thru 2007 to determine if having a prior period of dreports was correlated with acquiring them in 2008 and

2009 (Table 21). Eighty-five prisoners in the sample had not begun their incarceration prior to

2008 and therefore the sample for this analysis was 227 prisoners. The mean annual rate of all types of dreports for the entire sample was just under two dreports per year. For only serious dreports (predatory and non-predatory dreports in the highest category), the average was less than one per year during the 2003 to 2007 time period.

Table 21 - Annual Rate of DReports 2003 to 2007 by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total

Mean Mean Mean 2003-2007 Disciplinary Reports (SD) (SD) (SD)

All*** .85 3.12 1.88 t (225, N=227)=-3.29, ρ<.001 (3.78) (6.45) (5.28)

Serious* .46 1.44 .90 t (225, N=227)=-2.09, ρ=.038 (3.02) (4.03) (3.54)

* ρ<.05 ** ρ<.01 *** ρ<.001

151

*****

There were no appreciable differences between the two groups regarding two indicators of social support (visits and number of friends) in prison. Prisoners involved in serious disciplinary issues received more letters than those not so involved. Prisoners in both the high- and low-level disciplinary groups spent the greatest proportion of their time watching television, listening to the radio, and socializing with friends and acquaintances. Both groups spent much less time in constructive activities such as reading, going to school, working out or involved in sporting events, and participating in programs. The activities that clearly differentiated the two disciplinary groups was working and being idle in one‘s cells. Overall prisoners involved in prison misconduct were more likely to spend time in prisoner-initiated activities and less likely to spend their time in prison-led constructive activities. Finally prisoners with serious disciplinary issues were much more likely to have prior histories of disciplinary problems.

PRISON HARDSHIPS AND VICTIMIZATION

This section includes information about prison hardships and victimization as reported by prisoners in the self-report survey and the in-depth interviews. These include prisoners‘ survey responses to a series of questions about how hard it was to deal with 19 various issues in prison as well as questions regarding fear of victimization and actual victimization. Survey data were supplemented by data from the RIDOC on number of enemies and protective custody status. In- depth interview questions asked prisoners to identify and explain in detail the prison hardships that were most difficult for them. In addition prisoners were asked about problems from the past or outside prison problems that contributed to the difficulty of their incarceration. Finally interview questions focused on the existence of physical victimization and disrespect by staff and/or prisoners. The first section presents the survey data on prison hardships and victimization

152 data from the survey on the entire sample. The second section expounds on the same data by dreport grouping and presents prisoner responses from the in-depth interviews.

SAMPLE FEEDBACK ON PRISON HARDSHIPS AND VICTIMIZATION PRISON HARDSHIPS In the surveys prisoners were asked to assess the difficulty of a number of issues that they must confront and deal with in prison. Responses ranged from ―not hard at all‖ to ―very hard.‖

Based on the percentage of prisoners who responded with ―very hard,‖ the three most difficult hardships in prison were missing freedom (76 percent), missing family (68 percent), and lack of privacy (56 percent) (Table 22). Over 40 percent of prisoners felt that the quality of medical care and the quality of food were ―very hard‖ to deal with. Almost half of the prisoners also ranked regrets about the past and concerns about the future as ―very hard.‖ Lack of autonomy, missing activities, excessive noise, and cleanliness of the facility were rated as ―very hard‖ by just over one-third of the respondents. Dealing with boredom was assessed as ―very hard‖ by 29 percent

Table 22 - Prison Hardships: % Rated “Very Hard” by Survey Participants Type of Activity % Type of Activity %

Missing Freedom 76.2 Missing Activities 33.5

Missing Family 68.3 Boredom 28.8

Lack of Privacy 56.1 Eating Environment 26.5

Regrets about the Past 46.7 Missing Possessions 23.2

Concerns about the Future 45.6 Conflicts with Staff 21.9

Quality of Medical Care 43.9 Overcrowded Conditions 21.5

Quality of Food 41.9 Concerns about Safety 14.0

Lack of Autonomy 39.4 Following Prison Rules 12.6

Excessive Noise 36.5 Conflicts w/ Prisoners 10.4

153 of the prisoners. Less difficult to deal with for most but still ―very hard‖ for one-fifth of prisoners were missing possessions, conflicts with staff, the eating environment, and overcrowded conditions. Finally there were a few issues that were only viewed as ―very hard‖ by a small percentage of prisoners. These included conflicts with prisoners, concerns about safety, and following prison rules.

FEAR OF VICTIMIZATION AND ACTUAL VICTIMIZATION The RIDOC provided information about the existence of enemies and protective custody status for prisoners in the sample. In addition survey participants were asked about their fear of victimization and answered questions about actual victimization by other prisoners and by staff for a number of physical and verbal types of victimization. Overall about one-third (35 percent) of the prisoners said they either did not feel very safe in their prison or did not feel safe at all (not shown). Fifteen percent were worried a great deal about being attacked in prison and 16 percent felt that the chance of being attacked in their prison was high. This means that a full two-thirds of the sample felt safe or very safe and that 85 percent were not very worried about being attacked and thought the chances of it were low. Only six prisoners in the sample had been designated by the RIDOC as needing protective custody status. Regarding enemies, 53 percent of all of the prisoners in the sample did not have any enemies listed according to RIDOC records.

The number of enemies ranged from none to 14 enemies.

Prisoners were asked a number of questions about their actual victimization by other prisoners and their responses were summarized into four types of victimizations: threatened by other prisoners; property stolen by other prisoners; verbally insulted by other prisoners; or physically hurt by other prisoners. There was also one question asking whether they had been forced sexually by other prisoners. Overall a little less than one-quarter of those surveyed

154 reported being threatened by other prisoners, but almost two-thirds reported being verbally insulted in some way by other prisoners. One-third reported having property stolen or taken by other prisoners. Forty-four percent had been physically hurt in some way by other prisoners.

However, fewer than three percent reported having been sexually assaulted by another prisoner.

Table 23 - Reported Prisoner and Staff Victimization

Types of Victimization By a Prisoner By a Staff Member % % % Threatened 23.7 31.5 % Property Stolen 33.2 24.8 % Verbally Insulted 63.6 70.0 % Physically Hurt 43.6 37.4 % Forced Sexually 2.6 1.8

Overall there were about a dozen prisoners who did not answer any of the questions about staff victimization and several that did not answer some of those questions. Of those that did respond, one-third of prisoners reported having been threatened by staff and 70 percent reported having been insulted by staff. About one-fourth reported having their property stolen by staff and 37 percent reported being physically hurt in some way by staff. Fewer than two percent reported being forced sexually by staff.

PRISON HARDSHIPS AND VICTIMIZATION BY DISCIPLINARY GROUPINGS PRISON HARDSHIPS Measures of association were used to assess the relationship between prison hardships and membership in the high-level or low-level misbehavior groups. Prisoners in the high-level misconduct category were asked several questions about prison hardships in in-depth interviews.

They were asked, ―What are the most difficult parts about being imprisoned? Do you have issues or problems from your past or from the outside that you think made your incarceration

155 more difficult?‖ Asking these questions in this way put the onus on prisoners to indicate which problems bothered them most instead of choosing from a list of problems as in the survey.

Hardships Rated Similarly by Low- and High-Level Disciplinary Groups

Prisoners in the low-level misbehavior group responded similarly to those in the high- level misbehavior group for eight of the nineteen hardships listed in the survey. These included missing family, missing activities, regrets about the past, quality of food, quality of medical care, missing freedom, the eating environment, overcrowded conditions, boredom, and a lack of privacy. Over half of the interviewees said that missing their families was the most difficult and stressful part of being in prison. When those who rated it as second or third most difficult were added, two-thirds of those interviewed said that missing their family was one of the top three difficulties of being in prison. As one prisoner put it, ―Not seeing my family regularly, it weighs on you. You can‘t help them out with the struggles they‘re going through.‖ Many other prisoners were concerned that their incarceration meant that they were not ―able to support them physically, financially, and emotionally.‖ One prisoner who recently got married and was very concerned about his wife and two children, said, ―for her to be by herself is stressful for me.‖

The only other prison hardship mentioned by more than one person as being the most difficult was lack of opportunity. Three young prisoners said that the most difficult part about being in prison was the lack of opportunity to grow and to educate themselves. Overall four prisoners mentioned this as a difficulty.

Missing freedom had been a choice in the survey and had been assessed as ―very difficult‖ by 76 percent of the sample with no differences between those in the low- vs. high- level disciplinary category. It was mentioned by about one-third of those interviewed as one of

156 the three most difficult hardships. Many said that missing freedom was like a catch-all for missing many other aspects of being on the street. Food was mentioned by about one-fifth of those prisoners interviewed as being one of the three most difficult aspects of living in prison.

Prisoners complained mostly about the quantity of food, as well as its quality and several said it was necessary to supplement meals with food from the canteen. The only other prison hardship that had been assessed similarly by the two disciplinary groups that was mentioned by two interviewees was regrets about the past. Several other hardships that had been similarly assessed by both groups were not brought up as one of the three most difficult to deal with, including: missing activities, quality of medical care, the eating environment, and overcrowded conditions.

Hardships Rated Differently by Low- and High-Level Disciplinary Groups

Not surprisingly, surveyed prisoners in the high-level misbehavior group were much more likely to report as difficult those issues associated with misbehavior compared to those in the low-level group. Specifically, high-level disciplinary prisoners who took the survey were more likely to rate conflicts with prisoners, and conflicts with staff as ―very hard‖ compared to prisoners in the low-level misbehavior group. The differences were statistically significant for conflicts with prisoners (τ=.23, ρ<.001) and for conflicts with staff (τ=.28, ρ<.001). Prisoners with more dreports were also more likely to rate following prison rules as ―very hard‖ compared to prisoners with fewer dreports—again the differences were statistically significant (τ=.24,

ρ<.001). Not only did they find conflicts and rules more difficult than their peers, but prisoners in the high-level misbehavior group were also much more concerned about their own safety and thus more likely to rate concerns about their safety as ―very hard‖ compared to those in the low- level misbehavior group. The differences were statistically significant (τ=.17, ρ<.001).

157

Table 24 - Percentage of “Very Hard” Responses by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total Type of Activity % % % Missing Family ( N=312), τ=.03, ρ=.550 67.1 69.4 68.3 Missing Activities ( N=310), τ=.01, ρ=.821 31.8 35.2 33.5 Conflicts w/ Prisoners*** ( N=309), τ=.23, ρ<.001 6.0 14.6 10.4 Regrets about the Past ( N=306), τ=-.02, ρ=.661 49.3 44.3 46.7 Concerns about the Future*** ( N=309), τ=.17, ρ<.001 37.6 53.1 45.6 Missing Possessions* ( N=311), τ=.12, ρ=.013 17.9 28.1 23.2 Boredom ( N=309), τ=.06, ρ=.255 23.5 33.8 28.8 Lack of Privacy ( N=310), τ=.09, ρ=.091 51.7 60.4 56.1 Excessive Noise* ( N=307), τ=.11, ρ=.026 29.3 43.1 36.5 Quality of Medical Care ( N=310), τ=.03, ρ=.554 38.7 48.8 43.9 Missing Freedom ( N=311), τ=.03, ρ=.530 74.8 77.5 76.2 Conflicts with Staff*** ( N=310), τ=.28, ρ<.001 12.7 30.6 21.9 Lack of Autonomy* ( N=307), τ=.13, ρ=.013 31.5 46.8 39.4 Quality of Food ( N=308), τ=.03, ρ=.589 39.6 44.0 41.9 Eating Environment ( N=309), τ=.03, ρ=.560 24.8 28.1 26.5 Cleanliness of Facility** ( N=309), τ=.14, ρ=.006 25.2 41.8 33.7 Following Prison Rules*** ( N=309), τ=.24, ρ<.001 6.0 18.9 12.6 Overcrowded Conditions ( N=307), τ=.06, ρ=.243 18.8 24.1 21.5 Concerns about Safety*** ( N=308), τ=.17, ρ<.001 8.1 19.5 14.0

* ρ<.05 ** ρ<.01 *** ρ<.001

In the interviews, none of the prisoners discussed conflicts with prisoners specifically as one of their top three difficulties of being in prison; however, three mentioned harassment by

158 other prisoners and six mentioned that dealing with all of the various personalities were among the top three hardships of being imprisoned. One prisoner said he had come to learn that he should not ―take everything so personally. But it‘s hard, sometimes I do better than other times….all inmates are going through the same things. Sometimes we have disagreements over little things….or a cop will have a bad day. They‘re human like you‘re human too.‖ Next to missing one‘s family, more prisoners chose ―dealing with officers‖ as one of the three most difficult parts of being imprisoned. As one prisoner answered, ―Dealing with officers and staff.

They think we‘re less than they are. They talk to us in a degrading way. If we talked to them like they talk to us, they‘d put us in seg.‖ Another lamented, ―they treat you like crap. If they don‘t like you, they‘ll toss your cell; maybe you‘ll lose your job. If you do something minor, they‘ll turn it into something major.‖ While high-level misbehaving prisoners who took the survey had been more likely to assess following rules and concerns about safety as ‗very hard,‘ none of the interviewees mentioned these difficult parts of being imprisoned.

There was another group of hardships in the survey that differentiated those who were in the high-level misbehavior group from those in the low-level one. These included concerns about the future, missing possessions, lack of autonomy, excessive noise, and cleanliness of the facility. Specifically, prisoners with higher levels of misbehavior were more likely to rate as

―very hard‖ concerns about the future, missing possessions, and lack of autonomy. The differences between the groups on all three of these hardships were statistically significant

(τ=.17, ρ<.001 for concerns about the future; τ=.12, ρ=.013 for missing possessions, and τ=.13,

ρ=.013 for lack of autonomy). One might expect that those in the high-level misbehavior group would rate some of these hardships as more difficult because more of them were locked up in a segregation unit or housed in the HSC when surveyed. However, even when the approximately

159 two dozen who were in segregation or the HSC were removed from the sample, there were still similar differences between those in the high vs. low misbehavior groups. Interestingly, prisoners with serious disciplinary records were also more likely to rate as ―very hard‖ the excessive noise in prison and the cleanliness of the facility compared to those with less serious disciplinary records. The differences were statistically significant for excessive noise (τ=.11,

ρ=.026) and for cleanliness of the facility (τ=.14, ρ=.006). However, out of all of these hardships that mostly have to do with the conditions of confinement only two were brought up in the interviews: lack of autonomy and cleanliness of the facility. In response to the question about the most difficult parts of prison, a number of prisoners interviewed brought up issues that are unique to the HSC9. These included always being locked up, sensory deprivation, lack of privacy, and trouble concentrating.

An index that grouped together some of the hardships and a global hardships scale were created (not shown). The index mimicked Sykes‘ deprivations of prison, namely, missing possessions (lack of goods and services), lack of autonomy, concerns about safety, missing freedom (lack of liberty), and missing family (absence of heterosexual relationships). This index score ranged from 5 through 25. While the difference between the means of both groups was only -1.43, the differences were statistically different (t=-3.58, ρ<.001). Finally a global prison hardships scale was created that used all 19 of the individual prison hardships. Reliability analysis proved it to be a strong scale (alpha=.83). A comparison of the means of this global prison hardships scale that ranged from 19 to 95 revealed an average score of 65 for those

9 No prisoners were interviewed in segregation in either Maximum or Moran.

160 prisoners in the low-level disciplinary group and 70 for those in the high-level disciplinary group. The difference between the two groups was statistically significant (t=-3.64, ρ<.001).

Hardships Outside of Prison

In addition to the hardships in prison, prisoners were also asked in the interviews whether there were any issues or problems from the past that made their incarceration more difficult.

More than half said they had no such issues. Of the nine prisoners who said they did have such issues, four mentioned gang or neighborhood turf issues. A couple of prisoners mentioned difficult childhoods of abuse and molestation. Another was concerned that relatives of his victim were also in the prison system or knew people in there.

Prisoners were also asked if they had any issues or problems from the outside that made their incarceration more difficult. Only six of the 25 prisoners said they had no such problems.

The most prevalent problem that prisoners had on the outside also involved their families. A couple of prisoners had family members who were ill and one‘s family was struggling financially. One prisoner lamented, ―Family issues makes it hard for me, when I hear my grandmother is sick…I‘m locked up and doing all this time and I can‘t help them. It messes me up knowing I got myself in this situation and I can‘t help nobody.‖ Another was scared of losing his father while in prison. Several were just worried about their family‘s wellbeing or were unhappy because they had never seen nieces and nephews and even a son. One prisoner was stressed because he had not heard from his family. He said, ―I get stressed when I don‘t hear from them. I‘m not allowed to get phone calls…they‘re too busy to write me…When I was younger, they would say, ‗give me a call so we don‘t worry.‘ Now I worry about them and don‘t hear from them.‖

161

FEAR OF VICTIMIZATION AND ACTUAL VICTIMIZATION Turning to fear of and actual victimization, survey results also revealed differences between the low-level misbehavior group and the high-level misbehavior group, the latter more likely to feel less safe. Though prisoners in the high misbehavior group were more likely to report not feeling very safe compared to the low misbehavior group, the differences were not statistically significant (τ=.09, ρ=.083). This group was also more likely to worry about being attacked a great deal (21 percent) compared to those in the low level group (9 percent). High misbehaving prisoners were also more likely to rate their chance of being attacked as high (22 percent) compared to low misbehaving prisoners (10 percent). The differences were statistically significant for being worried a great deal about attack (τ=.14, ρ=.011) and thinking chances of being attacked are high (τ=.18, ρ<.001). These three questions about fear of victimization were combined to create a fear of victimization scale ranging from 3 to 10 (Cronbach‘s alpha=.81).

As expected, a comparison of the means for the two groups (5.13 for the low group vs. 5.85 for the high group) was statistically significant (t=-3.02, ρ=.003).

Table 25 - Fear of Victimization by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total % % %

% Prison Not Very or at All Safe 29.1 41.4 35.4 (N=308) τ =.09, ρ=.083

% Worry Great Deal - Attack* 8.6 21.0 14.9 (N=308) τ =.14, ρ=.011

% Chance of Attack is High*** 9.9 21.7 15.9 (N=308) τ =.18, ρ<.001 Mean Mean Mean (SD) (SD) (SD)

Fear of Victimization Score** 5.13 5.85 5.49 t (306, N=308)=-3.02, ρ=.003 (1.89) (2.27) (2.12)

* ρ<.05 ** ρ<.01 *** ρ<.001

162

Only six prisoners in the sample had been designated by the RIDOC as needing protective custody status. Two were in the low misbehaving group and four were in the high misbehaving group. Regarding enemies, the mean number of enemies was 1.7 for both the low and high misbehaving groups (not shown). When number of enemies was converted into either having enemies or not, high misbehaving prisoners were slightly less likely to have an enemy compared to those in the low misbehaving group but the difference between groups were not statistically significant (χ2=.51, with Fisher‘s Exact Test ρ=.497).

Table 26 - Reported Prisoner Victimization by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total % % %

% Have an Enemy 48.7 44.7 46.6 χ2(1, N=311)=.51, ρ=.497

% Threatened by a Prisoner 25.9 21.6 23.7 χ2(1, N=300)=.76, ρ=.416

% Property Stolen by a Prisoner 35.8 30.7 33.2 χ2(1, N=301)=.88, ρ=.392

% Verbally Insulted by a 60.7 66.5 63.6 Prisoner χ2(1, N=308)=1.11, ρ=.343

% Physically Hurt by a Prisoner 42.7 44.5 43.6 χ2 (1, N=305)=.11, ρ=.817

% Forced Sexually by a Prisoner 3.4 1.9 2.6 χ2(1, N=304)=.63, ρ=.492

* ρ<.05 ** ρ<.01 *** ρ<.001

There were no statistically significant differences between the two groups in rates of victimization by other prisoners. In-depth interviews also asked prisoners about whether they had ever been disrespected by other prisoners. Almost three-fourths of the respondents said they had been disrespected by other prisoners. One prisoner said that when being disrespected, pride

163 comes into play and often results in fighting. He noted, ―If you don‘t deal with it though, you‘re going to be dealing with it your whole bit.‖ He explained that being disrespected in prison was worse than on the streets because ―if you‘re in West Providence and someone disrespects you, then you can go to East Providence and they won‘t know about that. But here, if you get disrespected in the yard, the whole building knows about it.‖ Several commented on the high levels of testosterone among prisoners, one saying that ―everybody in here has to poke their chest out.‖ When asked if he had ever been disrespected by other prisoners, one prisoner in protective custody (PC) status answered ―Yes, every day. For everybody in PC, every single inmate hates

PC inmates.‖ He likened being put in segregation as a PC prisoner with ―cruel and unusual punishment‖ and ―psychological torture‖ because ―they know I‘m a PC and they scream at me all of the time.‖ He said he had no choice but to sit in his cell and just deal with it. Conversely, of the six prisoners who said they were not disrespected by other prisoners, most had been in a gang on the street or were actively in a gang in prison. When asked, several opined that it helped that they had some gang association and thus, were less likely to be disrespected.

Only one-third of those interviewed said they had been physically hurt by other prisoners.

Although several said they had been in fights, most said they had not been injured by them.

Those who had been physically hurt were more likely to be protective custody prisoners and prisoners with mental health problems who were living in the OSU, compared to prisoners residing in other units in the HSC or the other two facilities. Whereas one admitted that he was as much the cause for the fight in which he got hurt as the other, others who got hurt mostly talked about being victimized by others. One said he felt he had ―become desensitized to violence. It‘s unfortunate. It changes you.‖

164

Turning to victimization by staff, there were also differences between the low-level misbehavior and high-level misbehavior groups with the latter group consistently more likely to report having been victimized by staff than the low-level group. Prisoners with higher levels of misbehavior were more likely to report having been threatened by staff and insulted by staff compared to those with lower levels of misbehavior. Similarly they were more much likely to report having had their property stolen by staff and to have been physically hurt by staff compared to prisoners in the low misbehaving group. The differences between these four types of victimizations were all statistically significant—for threatened by staff (χ2=6.59, ρ=.014); for having had property stolen by staff (χ2=9.97, ρ=.002); for having been verbally insulted by staff

(χ2=7.81, ρ=.007); and for having been physically hurt by staff (χ2=23.04, ρ<.001). There was no difference between the two groups in having been sexually assaulted by staff (χ2=.11,

ρ=1.000)—both were extremely low (two percent or lower).

Table 27 - Reported Staff Victimization by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total % % %

% Threatened by Staff* 24.1 38.4 31.5 χ2(1, N=279)=6.59, ρ=.014

% Had Property Stolen by Staff** 16.3 32.5 24.8 χ2(1, N=286)=9.97, ρ=.002

% Verbally Insulted by Staff** 62.1 77.1 70.0 χ2(1, N=293)=7.81, ρ=.007

% Physically Hurt by Staff*** 23.2 50.7 37.4 χ2(1, N=286)=23.04, ρ<.001

% Forced Sexually by Staff 1.5 2.0 1.8 χ2(1, N=285)=.11, ρ=1.000

* ρ<.05 ** ρ<.01 *** ρ<.001

165

Three-fourths of those interviewed said they had been disrespected by staff, one responding, ―every day. They talk to you like you‘re some little kid and you‘re not,‖ and another responding, ―It happens every day. They disrespect everybody.‖ One prisoner lamented that he had been working hard to change for the positive but said that officers would not recognize the change. ―They degrade you all the time. I try to have an educated conversation and it always stems back to being in prison. People don‘t have any faith in me, it‘s hard to hear.‖ He said that staff in most conversations always go back to the fact that he was in prison and did not deserve things and that they never ―show faith‖ that he can change. He lamented that it was very depressing to hear and that he often went back to his cell and dwelled on it. Prisoners responded in different ways to staff disrespect. One said ―a lot of people take advantage of their position.

You‘ve got to bite your tongue and swallow your pride‖ but said he saw it as ―helping me in the future‖ because it made him stronger. Others did not respond as positively. One said ―When they disrespect you, it puts an individual where you have to do something back and they bury you (in segregation) or press charges…I argue a little and disrespect them back, yelling, screaming, kicking the door and flooding (out) the cell.‖ Another averred that if staff came after him, he would have to retaliate by going after them. Yet another said the disrespect was mutual, saying, ―I ask them to do something, they won‘t do it, I‘ll give them some disrespect and they give it back.‖ Another prisoner discussed how it was hard to change his attitude toward staff because of his past reputation. ―When I was younger, they knew if they disrespected me I would do something (to them), but now they see the way I‘m changing and (sometimes) they try to tempt me to (see what I will do).‖ Finally, one prisoner explained the differences between the staff at various facilities with regards to disrespecting prisoners. He (and several other prisoners) said that the officers at the Intake Service Center were the worst, that they constantly put down

166 prisoners and made known prisoners‘ charges. He said the officers at the HSC were less likely to disrespect prisoners or argue with them but more likely to spray them when there was conflict.

Officers at Maximum were more disrespectful in general but there were a handful who ―acts like that on an everyday basis.‖ Finally he said most staff at Moran were not disrespectful in general, with the exception of a few officers. Of the five prisoners who said that staff had not disrespected them, all were residing at either Maximum or Moran at the time of the interview. A couple said they were not disrespected but received ―smart comments‖ from staff all of the time.

One prisoner advised that prisoners in Maximum could minimize disrespect from staff, saying that ―in here, you (need to) pull up your pants and tuck your shirt in. Otherwise, everyday they‘ll be on you if you don‘t. You can‘t have people telling you all the time what to do,‖ especially if one has a long sentence to serve as he does.

Over one-third of those interviewed said they had been physically hurt by staff. Among them were two prisoners with mental health problems in the OSU and a protective custody inmate. Several said that they had been hurt by Intake Service Center staff and a couple mentioned being hurt at HSC or Maximum. Of those who said they had never been hurt by staff, most of them were residing in Maximum or Moran. As with the surveys, a couple of prisoners did not want to answer questions about being hurt by staff at all.

*****

In summary, the most difficult prison hardships for prisoners in the sample to endure were missing their freedom, missing their family, and the lack of privacy. The least difficult prison hardships for prisoners in the sample to endure were concerns about their safety, conflicts with prisoners, and following prison rules. These are a testament to the relative perceived safety

167 of the prisons in Rhode Island. Two-thirds of the sample either felt safe or very safe in the prison where they were residing and 85 percent were not worried about being attacked and felt the chance of attack was not high. Nevertheless almost half of the sample had listed enemies in the RIDOC database and over 40 percent had been physically hurt in some way by other prisoners. In addition more than one-third of the prisoners reported having been physically hurt by staff and 70 percent reported having been verbally insulted by staff.

There were no differences between the low-level and the high-level disciplinary groupings regarding the difficulty of most prison hardships. In the in-depth interviews, over two-thirds of prisoners said that missing family was the most difficult hardship. Four prison hardships—concerns with safety, conflicts with prisoners, conflicts with staff, and following rules—differentiated the low-level from the high-level misbehavior groups with those in the high group much more likely to have rated then as being very difficult. Next to missing family, dealing with the treatment of officers was deemed most difficult and a number of prisoners rated the difficulty of dealing with so many personalities in the prison as being particularly hard.

Prisoners in the high-level disciplinary group were much more likely to fear victimization and believed their chances of being attacked were much higher than did those in the lower disciplinary group. High misbehaving prisoners were significantly more likely than low misbehaving ones to report having been victimized by staff. Three-fourths of the prisoners interviewed said that they had been disrespected by both other prisoners and by staff. Those who had not experienced disrespect from other prisoners were mostly either current or former gang members.

168

TRAIT EMOTIONS

The trait emotions of depression, anger, and anxiety were measured through the use of the State-Trait Personality Inventory (Spielberger, 1995) and included in the self-report surveys

(see Research Methodology section). Table 28 presents the means and standard deviations for the trait depression, anger and anxiety scales for the current incarcerated sample in comparison with Spielberger‘s results from other male samples, one a general male sample, one a sample of male college students, and another of male Navy recruits (1972, 1980, 1983). As can be seen, the prisoners in the current study were more likely to live in a constant state of depression, anger, and anxiety as compared to males in the general population.

Table 28 - Comparison of Trait Emotions: Prison Sample with Other Spielberger Samples

Current Sample of Males Sample of Sample of Navy Incarcerated College Students Recruits Sample Feeling Subscale Mean % (SD) Mean % (SD) Mean % (SD) Mean % (SD)

Depression 23.7 (6.17) 17.6 (6.57) -- --

Anger 22.8 (7.17) -- 18.6 (2.06) 20.9 (5.73)

Anxiety 22.4 (5.62) -- 17.9 (4.47) 19.2 (5.14)

Next intercorrelations among the trait emotions were determined by the Pearson correlation for the entire incarcerated sample (Table 29). This analysis revealed a particularly high correlation between depression and anxiety (r=.78; ρ<.01) and smaller correlations between anger and depression (r=.33; ρ<.01) and anger and anxiety (r=.48; ρ<.01). All the correlations were statistically significant.

169

Table 29 - Intercorrelations among the Trait Emotions for Current Sample

Depression Anger Anxiety

Depression 1 Anger .333** 1 Anxiety .785** .485** 1

* ρ<.05 ** ρ<.01 *** ρ<.001

Each of the trait emotions was then compared for the high- vs. low-level disciplinary groups. For each of the three subscales, prisoners in the high-level misbehavior category had higher scores than those prisoners in the low-level misbehavior category. The mean score for depression was 23.2 for the low-level group and 24.3 for the high-level group. The differences were not statistically significant (t=-1.52, ρ=.130). The widest difference between the groups was for anger; high disciplinary prisoners were more likely to be angry (mean=24.6) than were low disciplinary prisoners (mean=21.0); the differences were statistically significant (t=-4.58,

ρ<.001). Finally, prisoners with more severe disciplinary records were also more likely to be

Table 30 - Trait Emotions for Sample by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total Mean % Mean % Mean % Trait Emotion (SD) (SD) (SD)

Depression 23.2 24.3 23.7 Range=4 - 16 (6.07) (6.23) (6.17) t (301, N=303)=-1.52, ρ=.130

Anger*** 21.0 24.6 22.8 Range=8 - 32 (6.67) (7.19) (7.17) t (301, N=303)=-4.58, ρ<.001

Anxiety 21.8 23.0 22.4 Range=4 - 16 (5.78) (5.42) (5.62) t (301, N=303)=-1.82, ρ=.070

170

anxious (mean=23) than those with minimal or no disciplinary record (mean=21.8). As with depression, the differences between the group means for anxiety were not statistically significant

(t=-1.82, ρ=.070).

PRISONER COPING

The primary independent variable for this research was coping and included various measures of prisoners‘ coping skills. Ultimately the research will test whether a prisoner‘s coping skills are a good predictor of involvement in serious prison misconduct or violence. As mentioned previously, the self-report survey contained the dispositional version of the COPE

Inventory, a multidimensional coping inventory developed by Carver, Scheier, and Weintraub

(1989) based on various elements and principles of coping theory. Prisoners were asked to think about what they usually do when they experience a stressful event. The first section below outlines the coping scales, presents the results from the bivariate analysis, and ends with staff‘s assessment of prisoners‘ coping skills. The second section covers prisoner responses in the in- depth interviews to questions about coping. Prisoners were asked how they coped with the specific strains they mentioned and other questions about how they coped in general. They were also asked questions about the extent of emotional and instrumental support they received from family and friends outside, from prisoners, and from staff.

PRISONER COPING AND SERIOUS MISCONDUCT AND VIOLENCE The factor analysis and reliability analysis (as described in the Research Methodology section) resulted in eight different coping scales (Table 31). As mentioned, a number of the subscales were identical to Carver‘s subscales including venting emotions, denial of the problem, acceptance of the situation, reliance on religion, and use of humor all with scores ranging from 4

171 to 16. A low score is indicative that prisoners did not use this coping mechanism very much, if at all; a high score indicates they used it a great deal.

A comparison of the means of each coping subscale was carried out for both the low- level misbehavior and the high-level misbehavior groups. There were differences between the two groups in five of the eight coping subscales. Prisoners in the high-level group were more likely to vent their emotions when confronted with problems compared to those in the low-level group. The differences were statistically significant (t=-2.77, ρ=.006). However, those in the high disciplinary group were less likely to get emotional or instrumental support from others to help them deal with their problems compared to those in the low disciplinary group. The differences were again statistically significant (t=2.29, ρ=.023). Prisoners with a more severe discipline record were more likely to use humor or joke about their problems than those with few or no dreports. Similarly they were more likely to use strategies of both behavioral and mental disengagement compared to those in the low disciplinary group. The differences were statistically significant for both the use of humor (t=-2.83, ρ=.005) and for disengagement (t=-

2.54, ρ=.012).

Prisoners with a more severe disciplinary record were also a little more likely to be in denial about their problems than prisoners with few or no dreports. Again the differences were statistically significant (t=-2.09, ρ=.038). Interestingly, there was little difference between the two groups in acceptance of the problem (t=.87, ρ=.383). Both denial of the problem and acceptance (acknowledgment) of the problem take place during the primary appraisal of a stressor or problem. However, acceptance can also be considered a coping response itself—that is, not only can one acknowledge that one has a problem (the opposite of denial), but also one can accept the situation as being one‘s current reality. For example, one might both

172 acknowledge the death of a loved one and also accept it, since one cannot change that stressful event. Therefore while prisoners in the high disciplinary group were much more likely to deny their problems, there was no clear cut differentiation in the more complex acceptance scale.

Table 31 - Coping Subscales for Sample by Prisoner Groupings Low-Level High-Level Prisoner Grouping Misbehavior Misbehavior Total Mean % Mean % Mean % Coping Subscale (SD) (SD) (SD)

Venting Emotions** 8.9 9.8 9.3 Range=4 - 16 (2.71) (3.00) (2.90) t (308, N=310)=-2.77, ρ=.006

Getting Support* 19.3 17.9 18.6 Range=8 - 32 (5.68) (5.30) (5.52) t (308, N=310)=2.29, ρ=.023

Denial of Problem* 6.4 7.1 6.8 Range=4 - 16 (2.92) (2.84) (2.89) t (307, N=309)=-2.09, ρ=.038

Acceptance of Problem 11.9 11.6 11.8 Range=4 - 16 (2.68) (3.07) (2.89) t (307, N=309)=.87, ρ=.383

Reliance on Religion 10.7 10.3 10.5 Range=4 - 16 (4.26) (4.48) (4.37) t (307, N=309)=.84, ρ=.401

Use of Humor** 7.6 8.8 8.2 Range=4 - 16 (3.37) (3.62) (3.54) t (306, N=308)=-2.83, ρ=.005

Disengagement* 14.1 15.3 14.7 Range=7 - 27 (3.97) (4.24) (4.15) t (308, N=310)=-2.54, ρ=.012

Active Coping Strategies 55.1 54.2 54.6 Range=1 – 16 (10.16) (10.38) (10.27) t (306, N=308)=.81, ρ=.416

* ρ<.05 ** ρ<.01 *** ρ<.001

173

There were no statistically significant differences between the two groups in their reliance on religion as a coping strategy (t=.84, ρ=.401) or in the last coping subscale, active coping strategies, (t=.81, ρ=.416).

Not surprisingly, the initial examination of the crosstabs between each individual question of the subscales and the prisoner disciplinary groupings mimicked, for the most part, the bivariate analysis of its larger subscale and the prisoner disciplinary groupings. Thus, none of the questions that made up the acceptance of the problem or the reliance on religion subscales were associated with the prison disciplinary groupings. However, that was not the case for the much larger subscale named active coping strategies. Despite being grouped together by the principle components analysis and having a decent alpha of .70, the individual variables that made up this subscale differed from one another, especially regarding responses that had been categorized by Carver as planning or active coping. According to Carver, planning is thinking about how to cope with a stressor, ―what steps to take and how to handle a problem.‖ Active coping refers to the ―process of taking active steps to try to remove or circumvent the stressor or to ameliorate its effects...(and) includes initiating direct action, increasing one‘s efforts, and trying to execute a coping attempt in stepwise fashion‖ (Carver, et al, 1989, p. 268). Out of the eight questions that make up these two Carver subscales, there was only one, ―I make a plan of action,‖ where there were no differences at all between those prisoners in high- vs. low-level disciplinary categories (not shown). In four of these questions, high disciplinary prisoners were more likely to engage in these strategies than low disciplinary prisoners. These included: ―I concentrate my efforts on doing something about it; ―I take additional action to try to get rid of the problem;‖ ―I try to come up with a strategy about what to do;‖ and ―I take direct action to get around the problem.‖ However, high-disciplinary prisoners were less likely to engage in three of

174 these other strategies, compared to low-disciplinary prisoners. These questions included: ―I think about how I might best handle the problem;‖ ―I think hard about what steps to take; and ―I do what has to be done one step at a time.‖ The first four questions describe taking direct action and the latter three involve more thinking, reflection, and the taking of sequential steps. Perhaps then, when the dependent variable is taking a negative action vs. a positive action, as it is in this study, Carver‘s scales do not capture the subtleties of the differences between coping strategies that might result in positive behavior and those that might result in either positive or negative behavior. Instead, there seems to be a difference between action that is impulsive (and possibly negative) and action that is well-thought out (and therefore more likely to be positive).

Interestingly, the only other individual question within the positive coping strategies subscale that differentiated between the two groups, ―I restrain myself from doing anything too quickly,‖ also represents thinking through a problem sequentially. Again, those in the high disciplinary category were less likely to restrain themselves from acting too quickly compared to those in the low discipline category.

In the in-depth interviews, staff were asked whether inmates who were involved in serious prison misconduct and violence had poorer, better, or similar abilities to cope with their stresses than other inmates. All but two of the staff said that prisoners with severe disciplinary problems had poorer coping skills. As one deputy put it, ―They can‘t cope with certain circumstances and their answer is to respond with violence. They do have poorer coping skills and we don‘t find out (about) the root causes of outbursts.‖ He explained that a fight after a sports game could be the result of the inmate not liking to lose; disrespect shown; or a gambling debt. Two lieutenants said that most of those involved in serious misconduct and violence did have poorer coping skills with the exception of one group—gang members. They believed that

175 gang members do have better coping skills compared to others involved in misconduct, one saying that they ―have a tremendous amount of coping skills;‖ that they are used to ―adjusting to situations all the time for survival;‖ and that they have their own support group. The other said it was their gang activities that got them in trouble and that these gang members did have the ability to turn it around if they wanted to. He said that as they grow older, they learn that there is an easy way and a hard way to do time and that eventually they ―get sick of the hard way of doing things—getting lugged down in segregation and getting in fights.‖

Most of the staff said that inmates did not know how to act, that they had not learned proper coping skills, and that they reacted to stress by using violence. One correctional counselor said that when inmates had proper coping skills, they could face a problem, figure out if it can be solved, and if not, ―deal with it‖ (accept that there was nothing they could do about it). However those with disciplinary issues could not properly assess the problem or deal with it properly. One officer said their poorer coping skills is what got them in the HSC, explaining,

―they never grew up with discipline. (They were) not guided in how polite society conducts themselves. Nobody ever taught them and they just act on ego.‖ A handful of both deputies and correction officers also noted that ego often guided their interactions—that these inmates had not learned how to walk away but lived by the ‗survival of the fittest‘ philosophy. Others noted that some prisoners with serious misconduct and violence were not able to ―take no for an answer‖ while others were impulsive. A few staff noted that these inmates often lacked communication skills, which combined with anger, resulted in misperceiving others‘ comments as threats. As one deputy put it, they have a ―lack of communication skills. (They have a) lack of ability to process what somebody is saying and what they really mean. They perceive it as a threat or attack and escalate it. They don‘t have the ability to de-escalate. They view all negative

176 language as a threat.‖ A few staff also noted that some inmates with discipline problems also had mental health issues that caused them to act irrationally. Overall, staff believed that prisoners with disciplinary problems had poorer coping skills than other inmates.

PRISONER FEEDBACK ON PRISONERS‘ WAYS OF COPING When people are confronted with a stressful problem or situation, they usually go through two processes before actually choosing their way of coping. These processes include the more emotion-focused primary appraisal drawing on one‘s experiential system and the more rational secondary appraisal involving a problem-focused process where one is choosing how to cope.

This involves defining the problem, deriving possible solutions, and then weighing the costs and benefits of those possible solutions.

USE OF EMOTIONAL AND INSTRUMENTAL SUPPORT One of the most important ways of coping is to elicit emotional and/or instrumental support from others in times of stress and crisis. Eliciting emotional support would take place during the primary appraisal of a problem; eliciting instrumental support would take place during the secondary appraisal process as one is planning how to cope with a problem or stressor.

Prisoners were asked a series of questions about getting support, starting with asking who prisoners talked to in general in order to get emotional support when they had problems and who they talked to get advice in how to solve problems (instrumental support).

When prisoners were asked who they talked to for emotional support, they gave four answers: ―nobody,‖ family, friends in prison, and staff. They gave the same four responses when asked about who they talked to for advice on addressing problems. About one-third said that they did not have anybody to go to for emotional support. Some seemed embarrassed by that, some said they were concerned about trusting others, and some did not want to stress out their

177 loved ones. As one put it, ―If you rely on your family (and tell them your problems) you‘re putting them down. They worry too much.‖ One prisoner talked about the anger he felt at not having anyone to confide in. ―Not really cause nobody helps you. That‘s when the anger builds up inside of you. You misbehave or some people do suicide or hurt themselves.‖ Fewer prisoners said they had ―nobody‖ to discuss how to solve problems in prison. Thus more prisoners were able to rely on others for advice on solving problems than for emotional support.

Reliance on Family and Outside Friends

A little more than one-third said they spoke to family members for emotional support when they had problems. A few said they were reluctant to burden their family but as one said,

―I guess if it gets really bad, my family…(but) more often than not, I take it on the chin.‖ Others said they do not keep their emotions from their family or that their family members were the only ones who understood them. However, prisoners were less likely to ask their family members about how to solve problems.

Almost all of the prisoners responded affirmatively when asked specifically about whether they relied on their family or friends on the outside to help them through their prison experience. Two did not: one could not receive visits in segregation and another was out of touch with his family. Most of the prisoners explained it was family, not friends on the outside, that they relied on to help them through the prison experience. Most relied on visits, letters, photographs, and calls from family members to cheer them up. When asked how family members helped, one prisoner replied, ―Sticking by my side, accepting phone calls, writing me, once in a while if I need something, putting money in my account.‖ As mentioned, some were reluctant to rely on family as they did not want to burden them or worry them too much. One noted that he relied on them ―as far as getting pictures and mainly stuff to brighten up the day,‖

178 but that he did not turn to them for other things because he did not want them to worry. He ended, ―They live their life, I can‘t expect them to jump and do everything I want.‖ Another said he enjoyed their visits, but did not want to put any stress on them. He explained, ―there‘s a point of maturity that you reach (when you just say) ‗I put myself in here‘ so I have to deal with it.‖

Still another said, ―My family had invested a lot in me…(so) I minimize the situation (here) to the extreme. I try to make it seem like a summer camp, especially for my mother and my girlfriend. They worry about me.‖ Several prisoners said they relied heavily on their family at the beginning of their incarceration, but that as time passed, that need faded away.

Reliance on Prison Friends

Less than five interviewees said they confided in prison friends, rather than family or staff, for emotional support. One said, ―Sometimes I talk with my friends, instead of getting in a bad mood and swinging at someone. That helps me out a lot.‖ About a handful said they spoke with prison friends, rather than others about how to solve problems; mostly going to older prisoners who shared their wisdom about the best way to do their time and deal with problems.

Prisoners were asked if there were other prisoners that they considered to be their friends.

A little more than half said they did have friends who were prisoners. Those who had friends tended to be living in either Maximum or Moran. About a handful said they had friends who were older than they were. A little less than half said they did not have friends—―associates and casual acquaintances, yes, but friends, no.‖ Several were wary of having friends in prison. One said that friends had gotten him in trouble. Another warned that there were ―a lot of fake friendships in here—a lot of people say they‘re your friends, but they‘re not.‖ One likened friends in prison to ―wolves in sheep‘s clothing‖ and said that he did not trust anybody. One prisoner said ―we make alliances disguised as friendship.‖ He said that once two people

179 disagree, then they let those alliances go and make new alliances; that over time, one makes

―millions of friends, but they‘re not really friends.‖ Many had changed their attitude over time about the wisdom of having friends in prison. One prisoner aptly summarized the sentiment, ―I used to call a lot of them friends, but not now…ain‘t nobody you can call a friend.‖

While more than half of the prisoners said they had friends in prison and the rest were wary of calling their acquaintances friends, nevertheless two-thirds of the prisoners said that other prisoners had helped them deal with stressful situations or helped them solve their problems. However, it is not surprising given the pervasive mistrust of friendship that fewer prisoners relied on other prisoners for emotional support but instead relied on them more for instrumental support—help with everyday prison problems. Those few who did seek emotional support from other prisoners usually discussed: venting about one‘s problems; reliance on other prisoners to help calm them down; or friends cheering them up when they are down. One explained, ―Sometimes when I get angry, the two or three that know me, the way that I am, they tell me to think about my family before I do anything (bad).‖ Another explained, ―You know each other‘s moods and ways. If they see me with my head down, they‘ll ask about it.‖

Most of the help that prisoners got from each other could be defined as instrumental support—advice on how to address prison problems. Thus other prisoners had helped them to learn the ropes of prison, had explained the rules, had advised them what to do and what not to do, and had advised them of the steps they should take to prepare for parole. One explained,

―When you first get here, you don‘t know about how to deal with officers. Others sit you down and explain. Also when you have a pending (court) case, they give you an idea of what might happen.‖ A couple of prisoners mentioned soliciting help with legal issues and found that there were a number of prisoners who were knowledgeable in that area. Prisoners also mentioned

180 talking to other prisoners when dealing with girlfriend or wife problems. A few discussed again their willingness to learn from prisoners who were older. As one said, ―I‘m the type if somebody talked to me, I‘d rather it be somebody older.‖ He said he did have somebody older advise him one or two times, that it was ―rare, (but) you‘ve got to cherish it.‖

Reliance on Staff

When asked who they approached for emotional support, only two prisoners said they spoke with staff members, either a correctional counselor or a social worker. Several prisoners said they would choose to go to staff for help with concrete problems and this included not only correctional counselors or social workers, but other types of staff. Yet when prisoners were asked how much they relied on RIDOC staff to help them through their prison experience, almost two-thirds said they did not rely on staff, and if so, only to obtain a document or other type of paperwork. Some prisoners were bitter in their answer saying things like, ―they don‘t help us,‖ or ―they‘re here for a paycheck…not here to help us.‖ One started his explanation vehemently stating that he got no help but then thought of a few staff who had been helpful. He said, ―I hate these people….well counselors help a little bit…and the cop at the desk helps a little.‖ Another admitted, ―some staff are all right.‖ He said a lot of them knew him when he was young and that they understand he was being a ―knucklehead‖ then. He said there are some staff he can really talk to and others with whom he talks only about a ball game.

About one-third of the prisoners said they relied on staff for help dealing with prison.

However given the negative attitude toward staff, it is not surprising that prisoners relied on staff more for instrumental support, rather than emotional support. That is, help solving prison-related problems; programs like anger management and AVATAR: and help with coping. One prisoner explained how the social worker ―helps me cope with my anger, gives me tools and ideas.‖

181

Another related, ―One of the COs, he‘s cool. If he sees me get into it with another, (he takes me aside) and tells me to keep cool.‖ Prisoners in HSC seemed to rely on staff for other help: two protective custody prisoners said they relied on staff to keep them safe from other prisoners and one discussed how he relied on staff as did others in segregation for maintaining his hygiene, getting a shower, being fed, and having access to a supervisor if need be.

EXPLORING SECONDARY APPRAISALS In the interviews, prisoners were asked questions about their secondary appraisals, including how they delineated what types of problems could and could not be solved and their typical choices of solutions. Specifically, they were asked, ―Are you more likely to try to solve problems or stressful situations or try to put them out of your mind and ignore them? What type of problem can be solved? What type of problem cannot be solved?‖ Thus prisoners were asked how they determined whether they could solve problems or chose to ignore them. The latter could be rephrased as whether or not they accepted the problem as one that could not be solved.

Besides using the ―ignore‖ terminology, some prisoners referred to this choice as ―doing nothing‖ while others referred to it as ―dealing with it.‖

About half of the prisoners said that whether or not they tried to solve or ignore a problem depended on the type of problem it was. One Moran prisoner said that it depended on the seriousness of the problem; if not serious he might be able to brush it off. As mentioned above, within active coping there was a differentiation between being reflective and almost blindly addressing the problem, whether in a positive or negative manner. About a third of the prisoners say they always tried to resolve their problems if possible. One assessed problem- solving as a positive skill, ―At the point I‘m at, I like to solve them. If you put a situation out of your mind, when it comes back around, you don‘t have the experience to solve it.‖ However a

182 number of prisoners discussed how in the past they had not paused to reflect on how best to solve their problems but had instead just reacted, most often in a negative fashion. One prisoner said,

―I like to take action. (However) I most often did it in a violent way. That‘s why I was in seg all of the time. (But) now I‘m trying to do better. If a person has a problem, I tell them to come to me and talk about it. I‘m more mature.‖ While some prisoners have learned to be reflective in their secondary appraisals, a handful of others have learned to ignore problems. Most of these said they used to try to solve their problems in prison with violence. One prisoner explained, ―I grew up in a violent project, so if a guy came in front of me and disrespected me, I would deal with it (in a negative fashion). I have learned to leave it and walk away.‖

Prisoners were then asked what types of problems could be solved and which could not be solved and must be accepted. Two prisoners were optimistic saying that almost any kind of problem could be solved. But three were pessimistic saying that prisoners really could not solve any problems. About half of the prisoners believed that disagreements with other prisoners could be solved. However, there were only a few other types of problems that some prisoners said could be resolved. Two protective custody prisoners said that daily prison issues, like getting mail could be solved. A few HSC prisoners said they had finally come to realize that they could solve some of their segregation problems, like working their way out of ―B‖ and ―C‖ status, controlling how they act, and minimizing disagreements with staff. Others said they could work on their court case or address some of their family issues. One prisoner was somewhat pessimistic when responding to the question about what could be solved. He answered ―things that are just on you, little things like your mail. You could write the warden.

But to (really) take control of something in here? No.‖ But another prisoner was more optimistic about the possibility of solving problems but said that many prisoners were too

183 incompetent to do that. He gave an example of someone who got mail with pictures who thought some of the pictures were missing. He said typically inmates would come up with a whole scenario of how staff must be messing with their mail and would often start arguing with the officer who delivered the mail. He said that prisoners often don‘t think out the situation to determine who they should speak with about the mail, like the mail officer. He concluded, ―It‘s the same thing every day in prison, lots of problems and a lot of incompetent advice passed around from inmate to inmate so everyone‘s dealing with situations incompetently.‖

When asked what types of problems could not be solved, one-third of the prisoners pointed to problems with officers. Several responses were similar to this one by a Moran prisoner, ―The way cops are. They‘re not going to change. You get some that are respectful, but a lot will get right in your face and provoke you.‖ A similar response by another, ―the way the

COs talk to us. You‘re still wrong cause you‘re a convict.‖ Although half of the prisoners interviewed had said problems with other prisoners could generally be solved (whether negatively or positively is another matter), there were three prisoners who could not solve their own problems with other prisoners. They were protective custody prisoners who felt they had no control whatsoever over how other prisoners treated them, but instead relied on staff to protect them. Other types of unsolvable problems included disagreements with classification or parole decisions, getting booked and doing seg time, the quality of the food, the lack of opportunities to grow, and the lack of freedom. One maximum security prisoner said that outside of working on his court case, ―most everything else‖ could not be resolved. He said, ―I mostly ignore everything. I just worry about doing my time. I just want to go home and be with my family.‖

When one Moran prisoner was asked what could not be solved, he lamented, ―A lot. The list is long. Every day is a problem; you just can‘t let it get to you. You really don‘t have any alone

184 time. You‘re going to have bad days and (those) lead to more and more issues. You have to stay to yourself when you‘re in one of those moods.‖

COPING WITH PRISON HARDSHIPS In the in-depth interviews, prisoners were asked how they coped with the three specific prison hardships that they found most difficult and stressful to deal with in prison. The responses to these questions were summarized and are reported as general ways of coping. Then they are reported individually by the specific type of hardship with which they were associated.

Many of those interviewed admitted that they could do nothing about a lot of the prison hardships that they faced and so instead got involved in other activities. This can be categorized as behavioral disengagement—that is, participating in other activities to keep one‘s mind off and possibly to avoid having to deal with other problems. Another popular way of coping mentioned by almost half of those interviewed was acceptance—what prisoners called ―ignoring it,‖ ―doing nothing,‖ or ―dealing with it.‖ A handful of prisoners said they turned to prayer, spirituality, or meditation to help them cope with prison difficulties in general.

Coping with Missing Family

Missing family was one of the prison hardships that surveyed prisoners found most difficult to deal with and it was named as the most difficult by those who were interviewed.

Most of those interviewed tried to mitigate this hardship by staying in touch with their family through visits, phone calls, and letters. Thus most dealt with missing their families by communicating with them as much as possible, a form of problem-solving or active coping available to them. For a variety of reasons, others did not get visits or make phone calls. To deal with missing family, some of these prisoners used behavioral or mental disengagement (e.g.

185 journal-writing, working on self-improvement), and a couple had no other option but acceptance saying that there was nothing they could do but to ―just deal with it.‖

Coping with Interpersonal Problems with Staff

There was immense variation in how prisoners dealt with interpersonal problems, especially problems with staff. Many had come to learn that there was little they could do to change the demeaning ways that some correctional officers spoke to them. As mentioned, many had mentioned problems with staff as the type of problem one should ignore. As one prisoner expressed, ―There are more consequences with pressing the issue and you say to yourself ‗it is what it is‘ and put it out of mind.‖ A few prisoners (now in medium or maximum security) said they had learned to be polite, to smile, or just do what they were told. Others avoided officers as much as possible. Some (especially those in the HSC) responded with negative behaviors such as kicking their door and yelling, which sometimes yielded more time in segregation. Some said they often spoke to a family member or a friend in prison about difficulties. Still others turned to other activities such as working out or reading to take their minds off the problem. Thus prisoners most often coped with interpersonal problems with staff through acceptance, disengagement, obtaining emotional support, and positive and negative problem-solving.

Coping with Interpersonal Problems with Other Prisoners

There was less variation in how prisoners addressed interpersonal problems with each other but the type of coping depended more on the nature of the problem. As mentioned previously, prisoners felt that problems with prisoners could be more easily addressed compared to problems with staff. Many discussed how they had always or had recently learned to tackle small problems with other inmates. As one prisoner put it, ―if it‘s a problem that could make me

186 stay longer here or makes me irrational, (I‘ll solve it). Like the phone, some will fight for the phone. I tell them, ‗we‘re going to go to seg and lose 30 days of good time for the phone?‘‖

Another discussed his approach to some altercations by giving an example, ―Summer league basketball. Some inmates get caught up in the moment, won‘t like a call, start getting into it. I‘ll tell the guy, ‗it‘s just a basketball game.‘ Try to bring them back to reality that the guy is just doing his job making a call and if we don‘t have him making a call, we can‘t play basketball.‖

One prisoner felt there were two positive choices that prisoners could make when faced with disagreements with other prisoners: ―you end up talking it out or ignoring them. No point in fighting cause if you do, you‘re going to end up in the box.‖

Almost all prisoners brought up the issue of disrespect by other prisoners during their interviews. Some had learned to ignore all disrespect by others; some tried to gauge when it was appropriate to ignore it and when it must be addressed; while others believed it always needed to be addressed. One who gauged the situation said, ―(But) there are other situations that you can‘t ignore. You have to solve them—certain forms of disrespect. You can‘t let them slide because

(prisoners will) think they can get away with it. If you let somebody disrespect you, this is an environment where everybody is trying to get ahead. It‘s a ‗strong survive‘ sort of place. The form of disrespect is going to escalate to the point where you might have to get in a physical altercation which is a lose-lose situation but in the long run helps avoid being disrespected more.‖ He reflected on the fact that he had acted on impulse a lot in the past, but is getting older now and is trying to think about the whole situation before he acts.

Several said that if someone comes after them physically, they have no choice but to choose to fight back. One prisoner with enemy issues said when ―encountering someone who might be an enemy on the streets, (like) someone from another side of town, there needs to be

187 either a positive or negative solution. I always try for the positive because I have an end game.

Parole is my outcome so I seek a positive solution now.‖ However almost every prisoner believed their survival in prison demanded that they fight if physically threatened or attacked.

All in all, prisoners were more likely to try to actively solve their interpersonal problems with other prisoners compared to their interpersonal problems with staff.

Coping with Other Prison Hardships

There was one prison hardship—missing freedom— that prisoners said they could do little about. As one put it, ―When you‘re in your cell, honestly (you) don‘t know how to get over that (being in prison).‖ Some said that they focused on life in prison, one saying, ―I keep my head out of the streets and focus on what‘s going on here. I don‘t think about what‘s going on there. (I do things) based on what is in here….(though my) girlfriend does visit.‖ These coping strategies demonstrate a mixture of acceptance and mental disengagement—accepting that they do not have their freedom, cannot change that, and mentally disengaging from the street.

Finally, those locked up in segregation in the HSC discussed the various means by which they handled being locked up continually. One who claimed to have been physically harmed by staff had turned to active problem-solving by filing a number of lawsuits against the department.

Besides acceptance of the situation, others chose either positive problem-solving strategies

(joining programs, pacing) or negative problem-solving strategies (hunger strike, beating on the door, flooding out) to handle the frustration of segregation. Many vacillated between acceptance and positive or negative problem-solving. A typical negative response to frustration with officers was explained by one prisoner. ―Sometimes officers come in with a bad attitude and want to deprive us of our needs, things we need like toilet paper or a pen. We‘re in segregation so we know we can‘t get other things, (but when) some deprive us of what we need, it pushes us

188 over the edge…Sometimes I kick the door and a supervisor comes down. Sometimes I get the thing or if he‘s in a bad mood, you don‘t.‖ Though such behavior might initially be rewarded, it also might result in a dreport and 20 to 30 days of additional segregation time.

*****

In summary, the initial principal components and reliability analysis of the coping questions resulted in eight subscales. Differences were found in the bivariate analysis between the two disciplinary groupings for five of the eight coping subscales. Prisoners with serious prison disciplinary records were much more likely to vent their emotions when confronted with problems and to use humor or joke about their problems. At the same time, they were much more likely to be in denial about their problems and to use both mental and behavioral disengagement to distance themselves from the problem. Finally, those with serious disciplinary records were less likely to solicit either emotional or instrumental support from others to help them with their problems. The bivariate analysis revealed no appreciable differences between the two disciplinary groupings in three of the eight subscales, including religion, acceptance, or active coping.

All of the staff believed that prisoners involved in serious misconduct and violence had poorer coping skills compared to other inmates. A couple of staff felt that this did not pertain to gang members whom they believed received support from other gang members and used their coping skills in their gang activities. Most staff felt that inmates who got in constant trouble had not been taught coping skills and lacked the education in assessing problems and planning positive strategies to deal with problems. Instead of relying on coping skills, staff said these

189 prisoners acted impulsively, out of ego, misperceiving others‘ communication and often lacking the communication skills needed to address problems in a positive fashion.

Prisoners varied in their responses regarding their secondary appraisals of their prison problems. Overall prisoners most often pointed to problems with other prisoners as solvable.

Over half said they could not solve the problem of how correctional officers treated them. All in all, there were few problems that prisoners appraised as being solvable on their own.

About one-third of the prisoners said that they had nobody to talk to for either emotional support or for help in solving their problems. The other two-thirds either went to family, friends in prison, or staff for emotional and instrumental support. The in-depth interviews revealed prisoners most often solicited emotional support, engaged in positive- or negative problem- solving, practiced mental and behavioral disengagement, or used acceptance when dealing with their most difficult prison hardships. Prisoners varied on how they coped with interpersonal problems, depending on whether it was staff or other prisoners. Most prisoners had come to the realization that acceptance combined with disengagement or support from others was the best way to handle conflicts or everyday interaction with staff. Prisoners alternated between acceptance and positive or negative problem-solving when choosing how to address minor altercations with and disrespect by other prisoners. There were two types of problems that most prisoners agreed could not be ignored: disrespect and physical attack by other prisoners. Most agreed that ignoring these problems would result in their escalation, though some said they had learned to ignore mild forms of disrespect. There was wide agreement that negative problem- solving (fighting) was the only way to handle being attacked or physically threatened. Those with a prolonged period in segregation also used a variety of positive and negative problem- solving strategies and acceptance, often vacillating between each of these ways of coping.

190

VII. MULTIVARIATE ANALYSIS OF PRISON MISCONDUCT

This chapter includes all of the multivariate analysis. It addresses the four hypotheses previously proposed, as well as the additional exploratory analyses of coping, trait emotions, and other predictors of serious prison misconduct and violence. Finally, it explores the relationship between prison hardships and serious prison misconduct and violence. Multivariate analyses were conducted on the weighted dataset.

COPING AND SERIOUS PRISON MISCONDUCT AND VIOLENCE

Table 32 is a correlation matrix of both the coping scales and trait emotion scales. An examination of the correlations among the coping scales reveals that more than half of the coping scales‘ correlations were statistically significant. However, the correlations themselves were relatively low (all under .50). The two largest correlations among the coping scales were between getting emotional/instrumental support and active coping (r=.49) and between denial and mental/behavioral disengagement (r=.49). Interestingly, most of the statistically significant relationships between the coping scales were positive, even though some of the coping scales have generally been deemed adaptive (getting support, active coping, acceptance) while others have generally been deemed as maladaptive (denial, disengagement, focus on and venting of emotions), and still others may be thought of as neutral or being both positive and negative (use of religion, humor). Finally one of the coping scales, venting emotions, had statistically significant correlations with all the other seven coping scales though the correlations were relatively modest.

191

Table 32 - Correlations among Coping Scales and Trait Emotions

Get Religion Humor Active Denial Disengage Accptance Venting Anxiety Anger Deprssion support coping

Coping Scales

Get support 1.00

Religion .20** 1.00

Humor .09 -.19** 1.00

Active coping .49** .19** .07 1.00

Denial .20** .16** .31* .20 1.00

Disengagement .018 .06 .34** -.00 .49** 1.00

Acceptance .11 -.01 .21** .35** .020 .21** 1.00

Venting emotions .34** .20** .14** .22** .18** .30** .12** 1.00

Trait Emotions

Anxiety .01 .07 .01 -.12* .20** .36** .07 .28** 1.00

Anger -.01 -.05 .28** .02 .09 .27** .12* .43** .48** 1.00

Depression -.14* .04 -.10 -.21** .11 .26** -.04 .22** .78** .33** 1.00

* ρ<.05 ** ρ<.01 *** ρ<.001

The first two of this study‘s hypotheses are:

Hypothesis 1: The greater the possession and use of positive coping skills by prisoners, the lesser will be their

involvement in serious misconduct and violence.

192

Hypothesis 2: The greater the possession and use of negative ways of coping

by prisoners, the greater will be their involvement in serious prison misconduct and

violence.

Since not all of Carver‘s original 13 subscales were considered positive coping skills, these two hypotheses were worded to differentiate between those adaptive ways of coping that might help prisoners deal with their stresses and result in less involvement in prison misconduct and those ways of coping that are generally thought of as maladaptive that may result in more involvement in prison misconduct. Of the eight subscales in this study, three would have been designated as positive in the coping literature (getting support, acceptance, active coping); three would have been designated as maladaptive

(denial, disengagement, venting emotions) and Carver‘s more recently developed subscales were not deemed to be positive or negative (religion, humor) (Folkman & Lazarus, 1985;

Carver, et al., 1989; Amirkhan, 1990; D‘Zurilla, & Chang, 1995; Sorlie & Sexton, 2001).

The negative binomial regression model predicting the annual misconduct score from the eight coping scales was statistically significant with likelihood ratio chi-square =

78.31, df=8, ρ<.001 (Table 33, Model 1). The likelihood ratio chi square examines the

―difference in likelihood values between the researcher‘s model and the model with the intercept only (null model). If significant, the researcher concludes that the coefficients in the model are different from 0 and the model is accepted‖ (Garson, 2009, p. 21) or technically rejects the null hypothesis that there is no difference between the models.

Unfortunately, unlike OLS regression, there is no R2 statistic that gives the percentage of variance in the dependent variable that is explained by independent variables in the model.

Instead the log likelihood statistic can be used to compare various models of independent

193 variables using the same dependent variable, with ―the larger the likelihood variable, the better the fit of the model to the data‖ (Garson, 2009, p. 21).

Seven of the eight coping scale parameters in Model 1 were statistically significant.

Only religion was not related to serious misconduct. The parameters for the independent variables were expressed as both the B coefficients and as incidence rate ratios (i.e., exponentiated coefficients). Incidence rate ratios are interpreted as the rate at which things occur and were calculated by hand by raising the mathematical constant e to the power of the unstandardized b coefficient (Kremelberg, 2011). For example,

INCIDENCE RATE RATIO FOR GET SUPPORT = E -.039 = 2.71828 -.039 = 0.962

The incidence rate ratio ―is interpreted in the following way: With one additional unit (of the independent variable), the average value of the dependent variable increases by a factor

(of the incident rate ratio value). With incidence rate ratios, a value of 1 indicates no relationship, while a value of greater than 1 indicates a positive or direct relationship, and a value of less than 1 indicates a negative or inverse relationship‖ (Kremelberg, 2011, p.

498). Thus for every one-unit increase in the get support scale, there was a four percent decrease (1-.962=.038) in the misconduct score, controlling for the variables in the model.

As mentioned previously, get support was one of three scales that were assumed to be positive coping skills and were hypothesized to result in lower levels of serious prison misconduct. According to this negative binomial model, get support was associated with less serious prison misconduct. That is, prisoners who were willing to solicit both emotional and instrumental support from family and friends (or other sources) were less likely to be involved in serious disciplinary behavior. Acceptance, another positive coping

194 skill, also was associated with lower levels of serious misconduct. Specifically, for every one unit increase in the acceptance scale, there was a 15 percent decrease in the annual misconduct score, controlling for all variables in the model. However, the third scale that was assumed to be positive, active coping had a statistically significant correlation with serious prison misconduct, but not in the anticipated direction. Instead of resulting in less serious prison misconduct, as expected, a one-unit increase in the active coping scale resulted in a four percent increase in serious prison misconduct.

Of the three coping scales deemed to be negative and hypothesized to be associated with greater levels of serious prison misconduct, the results were also mixed. The strongest of all the predictors was focusing on and venting of emotions. For each one-unit increase in the venting emotions scales, there was a 17 percent increase in serious prison misconduct.

A one-unit increase in the denial scale did result in an eleven percent increase in serious prison misconduct, controlling for all other variables in the model. However, mental and behavioral disengagement had an effect opposite to what was expected. That is, a one-unit increase in the disengagement scale was associated with a six percent decrease in serious prison misconduct. Of the two coping scales deemed to be more neutral, religion was found to have no relationship with serious misconduct. Interestingly, a one-degree increase in the humor scale was associated with a ten percent increase in serious prison misconduct.

The negative binomial regression model predicting the annual rate of violent dreports from the eight coping scales was statistically significant with likelihood ratio chi- square = 26.47, df=8, ρ<.01 (Table 33, Model 2). However, only four of the eight coping predictors had parameters that were statistically significant. Again religion was not associated with violent dreports. But denial, disengagement, and venting one‘s emotions

195 were also not related to whether one was involved in violent misbehavior. The strongest predictor of violent misbehavior was acceptance. For every one-unit increase in the acceptance coping scale, there was a 20 percent decrease in the annual level of violent behavior, controlling for the remaining predictors. The second strongest coping predictor was humor—that is for every one-unit increase on the humor coping scale, violent behavior

Table 33 - Negative Binomial Regression of DReport Rates: Coping

Variable Description Annual Misconduct Score Annual Rate Violent DReports Model 1 Model 2

B (SE) IRR B (SE) IRR

Intercept -.68 (.55) -3.52** (1.169)

Coping Scales Get support -.04* (.02) 0.96 -.084* (.04) 0.92

Religion -.03 (.02) na -.028 (.04) na

Humor .10** (.03) 1.10 .176** (.06) 1.19

Active coping .04*** (.01) 1.04 .071** (.02) 1.07

Denial .10** (.03) 1.11 .047 (.07) na

Disengagement -.06* (.03) 0.94 -.039 (.06) na

Acceptance -.16*** (.04) 0.85 -.234** (.08) 0.79

Venting emotions .16*** (.04) 1.17 .104 (.07) na

Likelihood ratio chi-square 78.3*** (N=294; df=8) 26.47** (N=294; df=8)

Log Likelihood -.557.34 -121.51

* ρ<.05 ** ρ<.01 *** ρ<.001 na=not applicable since relationship not statistically significant increased by 19 percent. Lastly, the impact of the active coping scale was again not as expected since a one-unit increase in active coping skills resulted in a seven percent increase in level of violence, holding all other predictors constant.

*****

196

In summary, the negative binomial analysis mostly confirmed the first two hypotheses of the current research. Both Models 1 (annual misconduct score) and 2 (annual violent dreports), which included only the eight coping scales, were statistically significant, indicating that they were significantly different from the null model with the intercept alone. Seven of the eight coping scale parameters in Model 1 were statistically significant, though not all in the anticipated direction. Specifically, it was anticipated that active coping would be associated with lowering serious prison misconduct, but instead it was associated with a greater level of misconduct. In contrast, it was anticipated that disengagement would be associated with higher levels of serious disciplinary behavior, but instead it was associated with lower levels of misconduct. In fact this is not surprising given what prisoners in the in-depth interviews said about the need to ignore and focus on other activities (i.e. acceptance and disengagement) when faced with some unsolvable problems in prison (e.g. missing family, missing freedom, demeaning feedback from staff). Only four of the coping scales‘ parameters were statistically significant with the dependent variable of violent behavior, with active coping again not in the anticipated direction.

Nevertheless both models were statistically significant and at least four or more of the coping scales had a statistically significant relationship with the serious prison misconduct and violence variables.

TRAIT EMOTIONS, COPING, AND SERIOUS MISCONDUCT AND VIOLENCE

A re-examination of the correlation matrix above (Table 34) revealed some statistically significant correlations between coping and trait emotions. Specifically there was a significant inverse relationship between anxiety and active coping and a positive relationship between anxiety and denial, disengagement, and venting emotions. There was

197 also a positive and statistically significant relationship between anger and the use of humor, disengagement, acceptance, and venting emotions, the highest correlation being for the latter. Depression held a significant inverse relationship with two coping scales, getting support and active coping. Depression had a positive relationship with disengagement and venting of emotions. Examining it from another angle, those who reported that they often vented their emotions did so regardless of whether those emotions were anxiety, anger, or depression. Interestingly, there was also a statistically significant positive relationship between prisoners‘ tendency to disengage mentally and behaviorally and anxiety, anger, and depression. Finally, as mentioned in the research methodology section, the three trait emotions were all positively related. There was a strong correlation between depression and anxiety, and moderate relationships between anxiety and anger and anger and depression.

The next hypothesis was in regards to the relationship between these trait emotions and serious prison misconduct and violence.

Hypothesis 3: Prisoners with stronger levels of negative trait emotions will

be involved in greater levels of serious misconduct.

The negative binomial regression model predicting the annual misconduct score from the three trait emotion scales was statistically significant with likelihood ratio chi- square = 36.05, df=3, ρ<.001 (Table 34, Model 3). However, the log likelihood of the trait- emotion only model (LL= -580.62) was slightly smaller than the log likelihood of the coping only model (LL= -557.34) indicating that the coping model was a better fit for the data. The parameters for each of the trait emotions were statistically significant. Though

198 negative trait emotions were hypothesized to be associated with greater levels of prisoners‘ involvement in prison misconduct, there was a negative relationship between anxiety and the annual misconduct score in that a one-unit increase in the anxiety scale was associated with a seven percent (not shown) decrease in serious prison misconduct, holding other predictors constant. Although anxiety and depression were highly correlated, depression had the opposite effect (positive relationship) on serious prison misconduct in that a one- unit increase in the depression scale resulted in a six percent (not shown) increase in serious prison misconduct. Since anxiety and depression were highly correlated, two separate models (not shown) were run on the annual misconduct score, one with only anger and anxiety, and another with only anger and depression in order to determine whether the inclusion of two highly-correlated variables in one model changed the direction of their relationship to prison misconduct. However, the direction of the relationship between these two trait emotions remained the same, even without the influence of the other. In addition, a third model of the cross-product of the two was run (not shown), but did not significantly change the results. Finally, not surprisingly, those prisoners who were generally angry were more likely to be involved in serious prison misconduct in that a one-unit increase in the anger scale correlated with an eight percent (not shown) increase in the annual misconduct score, again controlling for the other trait emotion predictors.

The negative binomial regression model predicting the annual misconduct score from the three trait emotion scales and the eight coping scales was also statistically significant with likelihood ratio chi-square = 99.19, df=11, ρ<.001 (Table 34, Model 4). An examination of the log likelihood statistic demonstrates that this model is an even better fit of the data (LL= -544.50) than the two prior models. While all three negative trait emotion

199 scales retain a statistically significant and similar relationship with the annual misconduct score as in Model 3, two of the coping scales, get support and venting emotions, no longer have a statistically significant relationship with the dependent variable. It is not surprising at all that the relationship between venting emotions and the annual misconduct score disappeared with the addition of the three negative trait emotions, given the significant correlations between it and all three negative trait emotions revealed in the correlation matrix (Table 34). The disappearance of a statistically significant relationship between getting support and serious prison misconduct with the introduction of the trait emotions is less easy to understand, except that it was the weakest correlation among the seven coping scales in Model 1. Of the coping scales with statistically significant parameters in Model 4,

(humor, active coping, denial, disengagement, and acceptance) the coefficients changed slightly but the direction and general magnitude of the association was similar to Model 1.

Specifically humor, active coping, and denial were still associated with a higher annual misconduct score, while disengagement and acceptance were associated with a lower score.

Turning once again to the annual rate of violent dreports, the negative binomial regression model predicting violence from the three trait emotion scales was statistically significant with likelihood ratio chi-square = 11.54, df=3, ρ<.01 (Table 34, Model 5). The model log likelihood was slightly less for the trait-emotion only model (LL= -128.98) compared to the coping only model (LL= -121.51), indicating a better fit with the latter.

Unlike Model 3 for the annual misconduct score where all three trait emotion parameters were statistically significant, only the trait emotion of anger revealed a statistically significant association with violent behavior in Model 5. Specifically, a one-unit increase

200

Table 34 - Negative Binomial Regression of DReport Rates: Coping & Trait Emotions

Variable Description Annual Misconduct Score Annual Rate Violent Dreports Model 3 Model 4 Model 5 Model 6

B (SE) B (SE) B (SE) B (SE)

Intercept -.52 (.36) -2.4*** (.69) -2.7*** (.74) -4.7** (1.57)

Trait Emotions Anxiety -.08*** (.02) -.07** (.02) -.10 (.05) -.10 (.06)

Anger .07*** (.01) .05** (.02) .09** (.03) .06 (.03)

Depression .06** (.02) .09*** (.02) .05 (.04) .08 (.05)

Coping Scales Get support -.02 (.019) -.07 (.04)

Religion -.03 (.02) -.02 (.04)

Humor .09* (.03) .16* (.07)

Active coping .05*** (.01) .08** (.02)

Denial .13*** (.03) .07 (.07)

Disengagement -.08** (.03) -.05 (.06)

Acceptance -.12** (.04) -.22** (.08)

Venting .07 (.05) .05 (.09)

Likelihood ratio chi-sq. 36.05*** 99.19*** 11.54** 31.44** (N=294;df=3) (N=292; df=11) (N=294; df=3) (N=292; df=1)

Log Likelihood -580.6 -544.5 -128.9 -118.7

* ρ<.05 ** ρ<.01 *** ρ<.001 in the anger scale was associated with a nine percent increase in annual violent dreports

(not shown).

Adding the coping scales to the three trait emotions, the negative binomial regression of Model 6, which predicts violence, was statistically significant with likelihood ratio chi-square = 31.44, df=11, ρ<.01 (Table 34). However, none of the parameters of the three trait emotions demonstrated a statistically significant relationship with violence; their

201 p values were all between .05 and 1.0. Also only three of the coping scales had statistically significant relationships with violent behavior in this model. The relationship between get support and violent behavior was no longer statistically significant with the introduction of the trait emotion variables. However, prisoners who used humor and active coping were significantly more likely to exhibit more violent behavior and those who used acceptance were significantly less likely to exhibit violent behavior.

*****

In summary, the negative binomial analyses revealed mixed results regarding

Hypothesis 3—that prisoners with stronger levels of negative trait emotions will be involved in greater levels of serious misconduct. Two of the trait emotions, anger and depression, were associated with greater levels of serious prison misconduct as measured by the annual misconduct score. However, the third trait emotion, anxiety, was associated with lesser levels of serious prison misconduct. Toch and Adams (1986) study of pathology and disruptiveness among prisoners also found that among the disturbed those with anxiety disorders had an unexpectedly low violation rate. Only one of the trait emotions, anger, was associated with greater prison violence as measured by the annual rate of violent dreports.

Exploratory analysis of the relationship between the trait emotions, coping, and misconduct also revealed differences between the models for serious general misconduct

(Model 4) and those for misconduct that was strictly violent (Model 6). The statistically significant relationship between the trait emotions and the annual misconduct score were retained when coping skills were introduced into Model 4. However, the statistically

202 significant relationship between anger and violent behavior disappeared with the inclusion of the coping skills in Model 6. Finally, the makeup of the relationship between the ways of coping and misconduct changed slightly with the inclusion of trait emotions in both

Models 4 and 6. The coping skill of seeking support, which had been statistically significant in both of the coping only models (Models 1 and 2), was no longer significant with the inclusion of the trait emotions. Similarly, the parameter for venting emotions and serious misconduct was no longer statistically significant once trait emotions were included in the model. Overall though, the inclusion of the trait emotions improved the fit of the models for both measures of prison misconduct, leading one to conclude that ways of coping and trait emotions are both associated with involvement in serious prison misconduct and violence.

ADDING OTHER PRISONER CHARACTERISTICS TO THE MODEL

The next set of analyses involves those prisoner-related variables that were found to be associated with serious prison misconduct and violence in prior research studies or in this study‘s bivariate analyses. Specifically, there were 15 variables among the many independent variables that had a statistically significant positive or negative relationship with the dichotomous low- vs. high-level disciplinary group variable. These variables have been included in the next set of multivariate analyses of serious prison misconduct and violence along with the three dummy variables of race and ethnicity. An examination of the correlation matrix of just these predictors (Table 35) reveals that there were several dozen statistically significant correlations among these predictors, however the majority of them were in the low to modest range (-.11 to .35). There were three high correlations of

203

Table 35 - Correlations among Prisoner Characteristics Predictors Predictors Age Blk His Oth Sin Div Gng Del PrP B&E Sex Sent Tim Psy Drg HSC Max Let

Age 1.00

Race/ethnicity

Black -.09 1.00

Hispanic -.22** -.34** 1.00

Other -.05 -.06 -.05 1.00

Marital status

Single -.51** .06 .20** .07 1.00

Divorced .32** -.04 -.18** -.04 -.53** 1.00

Crim history

St gang -.26** -.01 .13 * -.03 .04 -.07 1.00

Adj delin -.27** .03 .00 -.01 .17** -.07 .22** 1.00

Prior pris .04 .18** -.14* -.01 .02 -.00 -.06 .08 1.00

Offense type

B&E/drugs -.06 .02 -.02 -.05 .10 -.05 .01 .14* .28** 1.00

Sex offense .31** -.05 .19** .07 -.16** .04 -.01 -.21** -.24** -.27** 1.00

Total sentence .16** -.01 .06 -.03 -.13* .09 -.07 -.08 -.35** -.32** .03 1.00

Time in prison .28** -.03 .01 -.05 -.13* .07 -.04 -.05 -.33** -.29** .16** .74** 1.00

* ρ<.05 ** ρ<.01 *** ρ<.001

204

Table 35 - Correlations among Predictors Continued Predictors Age Blk His Oth Sin Div Gng Del PrP B&E Sex Sent Tim Psy Drg HSC Max Let

Psych treatmnt .08 -.18** -.11* -.06 -.06 .09 .04 -.00 .12* -.04 .03 -.07 -.07 1.00

# Prior drugs -.22** -.08 .05 -.06 .07 -.11 .16** .15** .16** .14* -.19** -.11** -.24** .14* 1.00

Facility at

HSC -.08 .04 .06 -.02 .03 -.06 .07 .15** -.01 -.03 .00 .07 .07 .15** -.03 1.00

Max -.07 -.02 .08 -.02 -.01 -.02 .02 .08 -.07 -.16** -.20** .35** .09 .02 .03 -.12* 1.00

# Letters/mo -.27** .18** .01 -.02 .25 ** -.15* .12* .10 .03 .15* -.14* -.12* -.13* -.17** .05 .06 .04 1.00

* ρ<.05 ** ρ<.01 *** ρ<.001 note; all were necessarily related. Age and being single were inversely correlated as were being single and being divorced. There was also a very high positive correlation between time incarcerated and sentence length. All of these variables were further monitored as the multivariate analysis progressed.

A series of negative binomial regression analyses began with just the inclusion of these 18 prisoner characteristic variables and the dependent variable, annual misconduct score. Table 36 above lays out the results of this analysis (Model 7) as well as the original

Model 1 with coping variables only and subsequent models with prisoner characteristics and coping (Model 8) and prisoner characteristics, coping, and trait emotions (Model 9).

205

Table 36 - Negative Binomial Regression of DReports: Predictors, Coping & Emotions Annual Misconduct Score Model 1 Model 7 Model 8 Model 9 B (SE) B (SE) B (SE) B (SE) Intercept -.68 (.55) -20.80*** (1.30) -1.99 (1.61) .20 (1.75) Predictors Age -.02* (.01) -.04** (.01) -.06*** (.01)

Race/ethnicity - .02 (.27) -.25 (.31) .28 (.36) Black Hispanic -.67* (.29) -1.10** (.33) -.90** (.33) Other race ------NA NA

Marital status Single .61* (.28) .81* (.34) .56 (.34) Divorced .31 (.42) -.75 (.51) -1.10* (.53)

Criminal history Street gang - .85 (.48) -1.18* (.56) -.63 (.57) Adj delinquent .05 (.218) -.08 (.25) .24 (.28) Prior prison .60** (.25) .95** (.32) .98** (.33)

Offense type B&E/drugs .80** (.29) 1.11*** (.31) 1.46*** (.33) Sex offender - .54 (.32) -1.49*** (.40) -1.6*** (.43)

Total sentence -8.3E-5* (4.1E-5) -2.7E-5 (4.8E-5) -4.31E-5 (5.10E-5)

Time in prison .000 (.00) -3.114E-5 (.00) .00 (.00)

Psych. Treatment .36 (.25) .28 (.26) .42 (.33)

# Drugs prior .19* (.08) .123 (.09) .26** (.10)

Facility at: HSC 1.97** (.73) 1.64* (.76) 1.89* (.76) Max .40 (.25) -.028 (.29) -.11 (.31)

# Letters /month .03 (.02) .00 (.02) -.00 (.02) Coping Scales Get support -.04* (.02) -.09*** (.02) -.11*** (.03) Religion -.03 (.02) .14*** (.03) .15*** (.04) Humor .10** (.03) .01 (.04) .07 (.05) Active coping .04*** (.01) .02 (.02) .02 (.02) Denial .10** (.03) .00 (.06) -.044 (.06) Disengagemnt -.06* (.03) .022 (.04) .044 (.05) Acceptance -.16*** (.04) -.005 (.05) -.055 (.06) Venting emot. .16*** (.04) .13* (.06) .25*** (.07) Trait Emotions Anxiety .12** (.04) Anger -.13*** (.03) Depression -.05 (.03) Liklihood ratio chi-sq 78.31*** 217.27*** 275.18*** 292.82*** (N=294; df=8) (N=264; df=18) (N=260; df=25) (N=260; df=28)

Log Likelihood -.557.34 -372.61 -336.89 -328.07

* ρ<.05 ** ρ<.01 *** ρ<.001 - - - Number too small to calculate

206

The negative binomial regression of Model 7 predicting the annual misconduct score from the prisoner characteristics was statistically significant with likelihood ratio chi- square = 217.27, df=18, ρ<.001 (Table 36). The log likelihood statistic increased dramatically for Model 7 (LL= -372.61), compared to Model 1 with just the coping scales

(LL= -557.34) or even Model 4 with coping and trait emotions (LL= -544.50), indicating that the model is a better fit for the data.

An examination of the socio-demographics in the model reveals that age has an inverse relationship with the annual misconduct score, meaning that older prisoners are less likely to be involved in prisoner misconduct compared to younger prisoners. Out of the three race/ethnicity dichotomous variables, only being Hispanic was significant. In short,

Hispanics are less likely to be involved in serious prison misconduct, holding all other variables constant. Being single had a positive relationship with being involved in serious prison misconduct. There was no relationship between being divorced and serious prison misconduct. Of the three criminal history variables, street gang involvement, prior juvenile adjudication, and prior incarcerations, only the latter had a statistically significant relationship with serious prison misconduct. Thus prisoners with prior incarcerations were more likely to be involved in serious prison misconduct. Those with charges of breaking and entering or drug offenses (but not sex offenses) had a strong and statistically significant positive relationship with annual misconduct score, controlling for other variables in the model. Although the correlation matrix revealed a high correlation between sentence length and time incarcerated, only sentence length had a statistically significant parameter in this model, indicating that those with shorter sentence lengths were more likely to be involved in serious prison misconduct.

207

The psychological treatment variable that was created for the multivariate analysis was a combination of receiving psychological treatment prior to incarceration and movement to a psychological unit/hospital while incarcerated. However, there was no demonstrable relationship between this psychiatric treatment variable and the annual misconduct score. There was a significant relationship between number of drugs used and annual misconduct score. As the number of drugs that a prisoner had used prior to his incarceration increased, so did his chances of being involved in serious prison misconduct.

Turning to the facility where prisoners were residing at the time of the survey, those in

Maximum were not more or less likely to have high annual misconduct scores compared to those in the HSC and medium facilities. However, those in the HSC were much more likely to have high annual misconduct scores compared to those in maximum and medium; the relationship was statistically significant holding all other variables constant. Finally, while the number of letters prisoners received per month was related to serious prison misconduct in the bivariate analysis, it was not so related in this negative binomial model.

Table 37 below is the correlation matrix with prisoner characteristics across the top and the coping and trait emotion scales on the left hand side. Out of the 198 correlations between prisoner characteristics and the coping scales, less than two dozen were statistically significant and all of them were very low (r= .115 to .203). With the exception of the humor scale, all of the other coping scales correlated significantly with at most one or two prisoner characteristics. However, prisoners who reported that they coped using humor and joking were less likely to be divorced, but more likely to be single, to have a B&E/drug offense as their most serious crime, to have a lower sentence, to have used more drugs prior to incarceration, and to receive more letters while in prison. Thus an image begins to

208 emerge of the type of prisoner who uses joking as a means of coping—young offenders who have committed ―street‖ crimes, who use drugs, and who have shorter sentences.

Almost half of the correlations between prisoner characteristics and trait emotions were statistically significant. The majority of the correlations were in the low to moderate range (r=.12 to .31); however there were two correlations that were higher. Both anxiety and depression were highly correlated with the psychiatric treatment variable, as might be expected. Anger was highly correlated with more of the prisoner characteristics, compared to anxiety or depression. Specifically, prisoners with trait anger were more likely to be young, single, to have been a gang member, to have been adjudicated delinquent as a teen, to have been previously incarcerated, to have a shorter sentence, to have served less time in prison at the time of the survey, to have been treated for psychiatric problems, and to be receiving more letters from the outside. Since many of the characteristics that were associated with humor are also associated with anger, it is not surprising that the two were indeed correlated (Table 37). Model 8 (Table 36 above) adds the coping variables to the prisoner characteristics. The fourth and final hypothesis of the study has to do with these individual level factors.

Hypothesis 4: That the effect of individual level factors (e.g. age, criminal

history, race) on serious prison misconduct and violence will be at least partially

mediated by prisoners‘ possession and use of positive and negative ways of coping.

209

Table 37 - Correlations between Coping and Trait Emotion Scales and Other Predictors

Predictors→ Age Blk Hisp Oth Sing Divr Gang Delin PrPr B&E Sex Sent Time Psyc Drug HSC Max Lettr ↓ Coping

Get support -.04 .01 .06 -.06 -.05 -.12* -.01 -.01 -.02 .03 .02 -.04 .01 -.06 -.02 -.04 -.08 .07

Religion .16** .20** .05 -.03 -.14* .07 -.05 -.11 .01 -.06 .05 .02 -.03 -.01 -.08 .03 .04 .02

Humor -.09 -.06 -.00 -.02 .14* -.11* .08 .03 -.09 .14* -.06 -.18* -.11 -.03 .12* .04 -.07 .15*

Act. Coping -.03 .12* -.01 -.02 -.02 -.13* .03 .08 .07 -.04 -.01 .00 .03 -.07 .11 -.02 .03 .08

Denial .08 .09 .03 -.03 .05 -.06 .07 .05 .10 .05 .01 -.09 -.09 -.02 -.01 -.06 .07 .16**

Disengage -.09 -.06 .04 -.02 .04 -.03 .10 .09 .05 .05 -.01 -.15* -.20** .04 .14* .00 .04 .10

Acceptance .04 -.05 -.01 .03 .06 -.16** .08 .02 .08 -.02 .00 -.08 -.04 .05 .15* -.03 -.06 -.05

Venting -.15** .057 -.01 -.05 .05 -.07 .14* .11 .02 .09 -.01 -.13* -.07 .06 .10 .10 -.01 .20**

Emotions

Anxiety .01 -.22** -.08 -.02 -.03 .11 .04 .06 .13* .09 -.10 -.13* -.17** .43** .10 -.05 -.01 -.03

Anger -.27** -.05 -.05 .06 .13* -.04 .22** .31** .12* .21** -.23** -.19** -.12* .23** .30** .10 -.07 .15**

Depression .01 -.15** -.14* -.04 -.04 .10 -.03 -.00 .16** .03 -.07 -.04 -.13* .45** .03 -.02 .06 -.02

* ρ<.05 ** ρ<.01 *** ρ<.001

210

When one variable mediates the relationship between another independent variable and the dependent variable, several conditions must occur (Baron & Kenny, 1986). First, there must be a statistically significant relationship between the personal predictors and misconduct. Second, there must be a statistically significant relationship between the personal predictors and ways of coping. Third, there must be a statistically significant relationship between ways of coping and misconduct. Finally, in the full model, the relationship between the personal predictors and misconduct would be decreased if there was partial mediation or disappear if there was full mediation Unfortunately, negative binomial regression, while a form of generalized linear regression, cannot be used for path analysis, as can OLS regression. However, one might expect if this hypothesis is true, that various parameters that were statistically significant when just the individual level factors were in the model would lose their statistical significance upon the introduction of the coping scales. However, for the most part, this did not happen. Six of the personal characteristics (age, not being Hispanic, being single, having a prior incarceration, having a

B&E/drug offense, and being in the HSC) that were statistically significant in Model 7 were also so in Model 8. Only two personal traits (number of drugs used previously and total sentence) lost their statistical significance once the coping scales were added to Model 8.

However, two other personal traits (being in a street gang and being a sex offender) that had not been statistically significant in the individual-factors-only model, became statistically significant once the ways of coping were added.

Overall the negative binomial regression of the annual misconduct score using both the personal prisoner characteristics and the coping scales was statistically significant with likelihood ratio chi-square = 275.17, df=25, ρ<.001 (Table 36, Model 8). In addition the

211 log likelihood statistic (LL= -.336.89) demonstrates that the inclusion of the coping scales improves the model from the one using only the prisoner characteristics (LL= -.372.61).

This is so even though only three of the eight coping scales—getting support, religion, and venting emotions—were statistically significant at this point. Model 9 adds in the trait emotions and is also statistically significant with likelihood ratio chi-square=292.82, df=28, p<.001 (Table 36, Model 9). Again the model is improved as the log likelihood statistic is further increased (LL= -328.07). Most of the same personal characteristics and coping scales continued to be statistically significant in this model. However, the direction of the relationship between the trait emotions and annual misconduct score flipped. In Model 4, prisoners who reported being anxious were less likely to be involved in serious prison misconduct. In Model 9 prisoners who reported being anxious were more likely to have serious dreports. The direction of the relationship between anger and annual misconduct score was also reversed from Model 4. The number of statistically significant predictors was somewhat reduced in this final model. The personal predictors of the annual misconduct score included being young, not being Hispanic, not being divorced, having a prior incarceration, being incarcerated for a B&E/drug offense but not for a sex offense, and being more apt to use drugs. Statistically significant ways of coping were reduced from seven to three, including getting support (negative association), reliance on religion, and venting one‘s emotions.

Turning next to violence, another series of negative binomial regression analyses began with the inclusion of the 18 prisoner characteristic variables and the dependent variable, annual rate of violent dreports. Table 38 below lays out the results of this analysis (Model

10) as well as the original Model 2 with coping variables only and subsequent models with

212 prisoner characteristics and coping (Model 11) and prisoner characteristics, coping, and trait emotions (Model 12). Unfortunately, the negative binomial regression analyses of the variables in these three models were not optimal10. Since the incidence of violent behavior is even lower than the overall incidence of serious prison misconduct and violence, it is no wonder that all three models were somewhat unstable. The negative binomial regression of

Model 10 predicting the annual rate of violent dreports from the prisoner characteristics was statistically significant with likelihood ratio chi-square = 47.305, df=18, ρ<.001 (Table

38). The log likelihood statistic increased a great deal for Model 10 (LL= -77.74), compared to Model 2 with just the coping variables (LL= -121.51). Fewer of the personal characteristics had a statistically significant relationship with violent behavior compared to

Model 7 with personal characteristics and the annual misconduct score. Age was inversely related to violent behavior and those who had been previously incarcerated were more likely to have received violent dreports. While these two relationships were consistent with prior analyses, the relationship between residing in the HSC and violence was not. The relationship was reversed, but one can see that in the next two models (Model 11 and 12), the model was not even able to calculate the statistics for this variable. This is probably because there are so few prisoners who were residing in the HSC and thus the numbers were too small to calculate.

10 The warning given by the SPSS 16.0 for each of these negative binomial regressions was that ―the Hessian matrix is singular. Some convergence criteria are not satisfied. The GENLIN procedure continues despite the above warning(s). Subsequent results shown are based on the last iteration. Validity of the model fit is uncertain.‖

213

Table 38 - Negative Binomial Regression of Violence: Predictors, Coping & Emotions Annual Rate Violent Dreports Model 2 Model 10 Model 11 Model 12 B (SE) B SE B SE Intercept -3.52** 1.17 -46.78*** 1.47 -33.05*** 4.13 -28.88*** 5.14 Predictors Age -.08** .03 -.11* .05 -.14* .06

Race/ethnicity Black -.06 .36 -.20 .84 -.42 .98 Hispanic -.83 .61 -.60 1.17 -.86 1.18 Other race ------N/A N/A N/A N/A

Marital status Single -.78 .55 -1.18 .83 -2.24* 1.08 Divorced -.29 .61 -1.34 1.25 -2.43 1.51

Criminal history Street gang .83 .72 .65 1.03 1.56 1.25 Adj delinquent .41 .47 .81 .61 1.04 .64 Prior prison 1.20** .42 2.35* 1.137 3.149* 1.34

Offense type B&E/drugs .27 .35 .88 .81 2.08 1.14 Sex offender -1.26 .89 -1.50 1.165 -1.85 1.37

Total sentence 3.40E-5 7.00E-5 .00 .000 .00 .00

Time incarcerated 3.36E-5 .00 .00 .00 .00 .00

Psych treatment .77 .40 1.64 .85 2.59* 1.21

# Drugs prior -.35 .23 -.84* .36 -.97 .42

Facility at: HSC -24.01*** .8114 ------Max -1.71 .90 -1.88 .99 -2.22 1.15

# Letters /month .04 .02 .02 .05 -.01 .05 Coping Scales Get support -.08* .04 -.20** .07 -.30** .10 Religion -.03 .04 .078 .0796 .084 .09 Humor .18** .0603 .20 .12 .36* .17 Active coping .07** .02 .13** .04 .14* .06 Denial .05 .07 -.04 .15 -.12 .16 Disengagement -.04 .06 .004 .10 .04 .11 Acceptance -.23** .08 -.03 .13 -.14 .14 Venting emotion .10 .07 .11 .14 .38 .23 Trait Emotions Anxiety .07 .12 Anger -.20* .10 Depression -.09 .11 Likelihood ratio chi-sq 26.47** 47.31*** 67.94*** 73.26*** (N=294; df=8) (N=264; df=18) (N=260; df=25) (N=260;df=28)

Log Likelihood -121.51 -77.74 -66.961 -64.30

* ρ<.05 ** ρ<.01 *** ρ<.001 - - - Number too small to calculate

214

Hypothesis 4, that the addition of coping variables to the model would at least partially mediate the relationship between the personal characteristics and serious prison misconduct and violence, was also not borne out by Model 11 that examined the effects of those two sets of variables on violent behavior. Two of the three personal characteristics were still significant (age and prior incarceration) once the ways of coping were added. But as mentioned, the negative binomial could not be calculated for residence in the HSC. The parameter for one variable—number of drugs used previously—became statistically significant with the introduction of the coping scales, however, in the opposite direction from what it was in the models for the annual misconduct score.

Nevertheless the negative binomial regression for Model 11 was statistically significant (likelihood ratio chi-square = 67.94, df=25, ρ<.001) and the model was an improvement (LL= -66.96) over the model that included personal characteristics alone

(LL= -77.74). This was true despite the fact that only two of the coping skills—getting support and active coping—were statistically significant. The final model (Model 12) introduced the three trait emotions along with coping and the personal characteristics. Age and prior incarceration were still significantly related to violence. But with the introduction of these trait emotions, being in psychiatric treatment became significantly correlated with involvement in violent behavior. The other variable that became correlated with violent behavior with the addition of the trait emotions was being single—but like a number of other variables, the direction changed from all previous models. In the coping only model of violence, four ways of coping had been statistically significant. However, in the final model only three ways of coping were correlated with violence, getting support (negative

215 association), humor, and active coping. Acceptance was no longer statistically significant despite having the strongest association with being violent in the coping only model.

*****

In summary, aside from the coping and trait emotions variables, many of the same variables that were predictors in prior research were also statistically significant predictors of prison misconduct in this study. For the annual misconduct score, which measures the quantity and seriousness of dreports, the predictors were age, being single, not being

Hispanic, having been previously incarcerated, serving a short sentence, having a B&E or drug offense for one‘s most serious charge, using drugs prior to incarceration, and being housed in the HSC. But far fewer predictors were statistically significant when the dreports were restricted to violent offenses. Only age, having been previously incarcerated, and being housed in the HSC were predictors of the annual rate of violent dreports.

The correlation matrix of the personal predictors against the ways of coping and the trait emotions revealed that there were no high statistically significant correlations among the two sets of variables. Thus ways of coping and trait emotions are more about one‘s personality and probably, personal upbringing, than related to personal characteristics.

Although the correlations were still low, one way of coping and one trait emotion were significantly related to many of the personal characteristics and to each other. Using humor and having anger as a trait emotion were both positively correlated with being single, using drugs, having B&E/drug charges for one‘s offense, having a shorter sentence, and in the case of anger, being young, in a gang, having been adjudicated a juvenile delinquent, and having been previously incarcerated. Thus, an image emerges of young angry offenders,

216 possibly gang related, who are highly involved in street level crime, though not necessarily violent crime who are repeat offenders and possibly also ‗state-raised convicts.‘

Hypothesis 4, that ways of coping partially mediated the effect of the personal predictors, was not supported by the analysis. Instead, the introduction of the personal predictors caused the statistical relationship between some ways of coping and misbehavior to vanish. However, more analysis would have to be conducted to determine the reasons for their disappearance. Specifically for the annual misconduct score, the introduction of the personal predictors resulted in five ways of coping losing their statistical significance— humor, active coping, denial, disengagement, and acceptance. Getting support still had an inverse relationship and venting had a positive relationship with the serious misconduct. In addition, religion became statistically significant with the addition of the personal predictors. Specifically, prisoners who relied on religion were more likely to be involved in serious misconduct. Anger, though positively associated with serious prison misconduct in the models with just trait emotions or just trait emotions and coping, was inversely associated with serious prison misconduct in this final model. Similarly, the relationship between anxiety and serious misconduct flipped from an inverse relationship to a positive relationship with the introduction of the prisoner characteristics. Depression lost its statistical significance completely.

Regarding violence, the addition of the personal predictors resulted in humor and acceptance no longer being statistically significant, though humor became statistically significant again with the addition of the trait emotions in the final model. As happened with the model for serious misconduct, the relationship between anger and violence was

217 again reversed from a positive to a negative relationship with the addition of the prisoner characteristics.

PREDICTING WAYS OF COPING AND TRAIT EMOTIONS

One of the assumptions made at the beginning of this research was that ways of coping were related to those factors that are the strongest predictors of misbehavior: age, prior criminal history, mental illness, and substance abuse. It was assumed, for example, that young people, substance abusers, and those who are mentally ill are all more likely to be involved in prison misbehavior because they lack positive coping skills and instead use negative ways of coping. The correlation matrix between individual predictors and coping resulted in few correlations with the exception of humor and several of the other predictors

(Table 37). However, in the negative binomial analysis there were a number of instances where significance of either coping predictors or other individual predictors was eliminated at the inclusion of the other. Many RIDOC staff interviewed linked the lack of coping skills among the young and mentally ill, but not among gang members, with prison misbehavior and violence.

To help explain these findings, a series of linear regressions using each of the coping scales and trait emotions as the dependent variable and the personal predictors as the independent variables were carried out. Forward stepwise regression was used to select the variables. This procedure allowed variables to be entered sequentially, at each step including the variable that had the largest positive or negative association with the dependent variable and continuing until no other variables met the entry criteria (p=.05).

218

An initial scan of Table 39 below leads one to conclude that there is not much correlation between the coping variables and the other predictors but more correlation between the predictors and trait emotions. Two of the ways of coping, get support and active coping, were not associated with any of the predictors; none of them loaded on the forward stepwise regression. Five of the ways of coping—humor, denial, disengagement, venting emotions, and acceptance—had statistically significant relationships with between one and four of the traditional predictors; but the amount of the variance explained by these ways of coping was very low, ranging from two to seven percent. Five of the predictors— being older, Black, Hispanic, not single, and fearful—were associated with the religion coping variable, explaining 13 percent of its variance. Interestingly, religion was not a predictor of either serious misconduct or violence in any of the models until the prisoner characteristics were entered into the serious misconduct model; then religion had a positive relationship with serious misconduct. The way of coping with the most predictors, aside from religion, was humor. Specifically, those who used humor or joked about their stresses were more likely to be young, not previously imprisoned, to be in for a B&E/drug offense, and to have used drugs. Thus it is not surprising that the introduction of the personal predictors into the analyses of annual misconduct score eliminated the effect of humor on the model (Models 1, 8 and 9). While humor might be one factor that links these predictors, there are probably others like being impulsive and lack of self-control that together makes these predictors stronger than the humor predictor alone.

Age, being black, being Hispanic, having been adjudicated delinquent, and being mentally ill were not predictors of ways of coping. Outside of the predictors of humor mentioned above, being in a street gang was predictive of disengagement and venting

219 emotions. It is assumed that gang members might be more likely to divert their attention away to other thoughts and behaviors when confronted with problems, but then at some breaking point vent their emotions and react negatively. Other relationships make some intuitive sense: those previously imprisoned and those who were fearful were more likely to be in denial of their problems; time incarcerated had an inverse relationship to disengagement; and prior use of drugs was more likely to be predictive of acceptance.

While the relationship between ways of coping and the personal predictors was weak, the relationship between the trait emotions and those same predictors proved to be much stronger. Each of the trait emotions—anger, anxiety, and depression—had between four and six predictors that accounted for between 27 and 30 percent of the variation in the trait emotion. Not being black, less time incarcerated, fear, and having been treated for mental illness were all predictors of being anxious. Anxiety had had a statistically significant inverse relationship with the annual misconduct score until the addition of the personal characteristics flipped the relationship. Not being black or Hispanic, less time incarcerated, fear, and having been treated for mental illness were all predictors of depression, which had had a statistically significant positive relationship with annual misconduct score in the models without the personal predictors. Anger had had a statistically significant positive relationship with both the annual misconduct score and violence in preliminary models (Models 4 and 6), but in the final full model it had an inverse relationship with annual misconduct score alone (Model 12). Being young, having been in a street gang, having a B&E/drug offense, having been treated for mental illness, and using drugs were all predictors of anger. Thus it is not surprising given the strong associations between trait emotions and other personal predictors that the effect of the trait

220

Table 39 - Are Predictors of Misbehavior also Predictors of Coping and Trait Emotions? Predictors→ Age Blk Hisp Sing Gang Delin PrPr B&E/D Time Fear Psyc Drug Coping & Emotions Get support -No variables entered Act. Coping -No variables entered Religion*** F(5, 283)=9.28, * *** ** - * * ρ<.001, adj R2=.13 Humor *** F(4, 283)=6.61, * - ** ** * ρ<.001, adj R2=.07 Disengage*** F(2, 285)=10.37, * - *** ρ<.001, adj R2=.06 Denial*** F(2, 283)=8.41, * ** ρ<.001,adj R2=.05 Venting** F(1, 285)=8.92, ** ρ=.003, adj R2=.03 Acceptance* F(1, 283)=5.76, * ρ=.017, adj R2=.02 Depression*** F(5, 283)=25.70, - * - ** - ** *** *** ρ<.001, adj R2=.30 Anger*** F(6, 283)=19.35, - ** * ** ** *** ** ρ<.001, adj R2=.28 Anxiety*** F(4, 283)=27.24, - * - *** ** *** ρ<.001, adj R2=.27 * ρ<.05 ** ρ<.01 *** ρ<.001

221

emotions either disappeared or changed when the personal predictors were added to the equation. For example, age, B&E/drug offense, and substance abuse were all predictors of anger that also ended up as predictors in the final full model that examined annual misconduct score (Model 9). It is not surprising that once the variance of these factors is controlled for, that the relationship between anger and misbehavior might change. Thus the anger of the young, street-involved, drug user might lead to prison misbehavior, while possibly older, more seasoned prisoners might be able to control their anger, resulting in less misbehavior.

One cannot help but notice that having been treated for mental illness was a statistically significant predictor of all three trait emotions (Table 39). When just personal predictors or personal predictors and ways of coping were in the model examining violence, having been treated for mental illness was not statistically significant (Models 10 and 11).

However, once the trait emotions were controlled for, having been treated for mental illness became a statistically significant predictor of violence (Model 12). This makes sense given that treatment for mental illness is correlated with all three trait emotions but that two of these emotions, depression and anger, lead to misbehavior, while the other, anxiety, results in lowering the chances for misbehavior. Thus when all three trait emotions are controlled, the relationship between having been treated for mental illness and misbehavior is unmasked. It is not surprising, given the multi-dimensionality of mental illness that some prisoners who have been treated for mental illness are unlikely to become involved in violence, while others are more inclined.

222

*****

In summary, analyses were conducted to further clarify the relationship between ways of coping, trait emotions, and traditional predictors of prison misbehavior and violence that involved a series of forward step-wise regressions using each of the ways of coping and the trait emotions as the dependent variable and the traditional predictors of misbehavior as the independent variables. Although it was assumed at the beginning of this study that the traditional predictors were related to ways of coping (e.g. young prisoners might have poorer coping skills), with few exceptions this was not the case. Ways of coping do not mediate the effect of traditional predictors but instead exert their own influence on prison misbehavior. The only way of coping that was correlated to several traditional predictors was humor; young prisoners in for a B&E/drug offense who had not been previously imprisoned but who had used drugs were more likely to cope in prison by joking about their circumstances and putting on a mask of bravado and humor.

The traditional predictors of prison misbehavior were more predictive of trait emotions. Indeed between 27 and 30 percent of the variance in each of the three trait emotions (anxiety, depression, and anger) was explained by the predictors. Interestingly, the predictors of anxiety and depression were similar (less time incarcerated; fear; having been treated for mental illness; and not being black or Hispanic for depression) even though anxiety correlated with a decrease in misbehavior while depression did the opposite. Anger had some of the same predictors as humor: age; street gang affiliation; B&E/drug offense; drug use; and having been treated for mental illness further solidifying the image of angry young, possibly gang-involved, street offenders who enter the prison using humor and bravado as their way of coping with prison stresses.

223

PRISON HARDSHIPS, VICTIMIZATION, AND SERIOUS MISCONDUCT AND VIOLENCE

The next set of multivariate analyses using the negative binomial regression is exploratory and examines prison hardships, fear of victimization, and the three self-reported victimization variables that the bivariate analysis found to be related to serious prison misconduct—being hurt, threatened, and insulted by staff. The first set of analyses is presented in Table 40 and involves all 19 of the prison hardships that prisoners rated for level of difficulty in the survey, along with their relationship to the annual misconduct score and then to the annual rate of violent dreports. The negative binomial regression of Model

A predicting the annual misconduct score from the prison hardships was statistically significant with likelihood ratio chi-square = 247.91, df=19, ρ<.001. Only those prison hardships with statistically significant relationships to the dependent variables are presented in the table. Those not related include: missing activities, conflicts with other prisoners, regrets about the past, concerns about the future, lack of privacy, excessive noise, quality of medical care, lack of autonomy, quality of food, eating environment, and following rules.

Eight hardships—missing family, missing possessions, boredom, missing freedom, conflicts with staff, cleanliness of the facility, overcrowded conditions, and concerns about one‘s safety—were significantly related to serious prison misconduct. Those prisoners who reported that missing freedom was difficult or very difficult for them were more likely to be involved in misconduct. Specifically a one-unit increase in the missing freedom variable was associated with a 137 percent increase in one‘s annual misconduct score. Similarly, prisoners who reported missing their families very much were also likely to be involved in serious prison misconduct. A one-unit increase in the missing family question was associated with 54 percent increase in the annual misconduct score. Those prisoners who rated conflicts with staff as difficult were also more likely to have higher annual

224 misconduct scores (a one-unit increase in conflicts with staff item resulted in a 53 percent increase in the annual misconduct score). The bivariate analysis revealed that prisoners in the high-level disciplinary group were less likely to report having a job and were more likely to report spending a good percentage of their weekly time idle in their cell.

Unfortunately, these variables could not be included in the multivariate analysis because of their high level of missing data. However, boredom can definitely be considered a proxy measure for idleness and not working. It is therefore not surprising that those who rated being bored as difficult or very difficult were also those who had high levels of serious prison misconduct. Specifically, a one-unit increase in the difficulty of boredom was associated with a 38 percent increase in one‘s annual misconduct score. Concerns about one‘s safety were also related to higher disciplinary records. That is, a one-unit increase in the concerns about safety variable was associated with a 37 percent increase in the annual misconduct score. An interesting association was found between missing possessions and annual misconduct score. Again the more one found missing possessions to be difficult, the more likely they were to be involved in high prison misconduct unless it is a proxy measure for younger prisoners who might be more likely to be missing material things like their cell phone, electronics, etc… Alternatively, older prisoners are more likely to have been incarcerated for longer periods of time and may have gotten used to living without material possessions; younger prisoners had, for the most part, shorter sentences and have not gotten used to missing their possessions. Finally two prison hardships had inverse relationships with serious prison misconduct. Those who found it very difficult to deal with the lack of cleanliness of the facility and the overcrowding were less likely to be involved in prison misconduct. Regarding the concern about cleanliness, perhaps it reflects that a person

225

Table 40 - Negative Binomial Regression of DReport Rates: Prison Hardships+

Variable Description Annual Misconduct Score Annual Rate Violent Dreports Model A Model B B (SE) IRR B (SE) IRR Intercept -8.18*** (1.24) -118.99*** (1.85)

Prison Hardships Missing family .43** (.15) 1.54

Missing possessions .23* (.09) 1.26

Boredom .32*** (.09) 1.38 .48* (.23) 1.6

Medical care quality -.47** (.17) 0.62

Missing freedom .86*** (.20) 2.37

Conflicts with staff .43*** (.08) 1.53

Cleanliness - .27** (.10) 0.76

Overcrowded conditions -.20* (.08) 0.82

Concerns about safety .31** (.09) 1.36 .62** (.228) 1.86

Likelihood ratio chi-square 247.91*** (N=264; df=19) 65.81*** (N=264; df=19)

Log Likelihood -.429.52 -93.41

* ρ<.05 ** ρ<.01 *** ρ<.001 + All prison hardships in the analysis, but table only included where the analysis revealed statistical significance for either dependent variable.

worried about cleanliness probably feels relatively safe and worries about a somewhat more mundane, though still legitimate concern. Those worried about overcrowding could also be those who might spend less time in prison areas that are crowded and where trouble most often occurs, like the yard, the gym, and the cafeteria and thus are less likely to become involved in disputes that so often take place in those areas.

In the next set of analyses, the negative binomial regression of Model B predicting the annual rate of violent dreports from the prison hardships was statistically significant

226 with likelihood ratio chi-square = 65.81, df=19, ρ<.001.11 Only three hardships had parameters with statistical significance—boredom, concerns about the quality of medical care, and concerns about safety. Those prisoners who were very concerned about their safety had annual misconduct scores that were 86 percent higher for each unit increase in that variable. Prisoners who rated boredom as being very difficult for them also were more likely to be involved in prison violence. Specifically, a one-unit increase in one‘s difficulty with dealing with boredom was associated with a 61 percent increase in one‘s annual rate of violent dreports. The third statistically significant variable was concern about medical care.

It is assumed that this is a proxy for prisoners who are older, in that being very concerned with the quality of medical care would probably be true for those who generally have more medical issues—those who are older. Thus the more difficult prisoners rated their concern over medical care, the less likely they were to be involved in prison violence.

The last set of analyses used the fear of victimization scale and the three self- reported victimization variables that were found to be related to prison misconduct in the bivariate analysis (Table 41). An examination of Model C below reveals that only two of the four variables, hurt by staff and fear of victimization, bore a statistically significant relationship with the annual misconduct score. Still the negative binomial analysis of annual misconduct score using these four variables was statistically significant (likelihood ratio chi-square=99.91, df=4, ρ<.001). Prisoners who reported having been hurt by staff were four times more likely than other prisoners to be involved in serious prison misconduct. Also a one-unit increase in the fear of victimization scale was associated with

11 The SPSS 16.0 again warned that the validity of this model was uncertain, as with prior analyses of the annual violent dreports.

227 a 15 percent increase in one‘s annual misconduct score. Model D examining the relationship between violent behavior and victimization was not statistically different from the intercept only model. Thus fear of victimization and being insulted, threatened, and hurt by staff were not associated with violent behavior.

Table 41 - Negative Binomial Regression of Dreport: Fear & Victimization

Variable Description Annual Misconduct Score Annual Rate Violent Dreports Model C Model D B (SE) IRR B (SE) IRR Intercept -.81* (.39) -1.76* (.72)

Hurt by staff 1.47*** (.29) 4.34 .66 (.49)

Threatened by staff -.32 (.29) -.18 (.49)

Verbally insulted by staff .042 (.1932) -.12 (.41)

Fear of victimization scale .14** (.05) 1.15 .06 (.09)

Likelihood ratio chi-square 99.91*** (N=262; df=4) 3.09 (N=262; df=4)

Log Likelihood -490.15 -124.46

* ρ<.05 ** ρ<.01 *** ρ<.001

*****

In summary, the exploratory analyses of the prison hardships, fear of victimization, and victimization by staff variables yielded some interesting results. A number of prison hardships were positively associated with serious prison misconduct. That is, the higher the prisoners‘ ratings of these concerns as difficult, the greater were their levels of serious prison misconduct. Specifically, serious prison misconduct was associated with prisoners who found it very difficult to miss their family, their possessions, and their freedom. Next,

228 it is not surprising that those prisoners who rated conflicts with staff and concerns about their safety as being very difficult to deal with were also more likely to have higher annual misconduct scores. Finally difficulty dealing with boredom was also positively associated with prison misconduct. Turning to fear of victimization and actual victimization by prisoners or staff, only fear of victimization and specifically having been previously hurt by staff were associated with a higher annual misconduct score. Specifically those who had a higher fear of victimization and those who reported having been hurt in the past by staff were both more likely to be involved in serious misconduct.

Fewer of the prison hardships were related to prison violence alone. Only those prisoners who rated concerns about their safety and boredom as being very difficult to deal with were more likely to be involved in violence. Those who found the quality of medical care difficult to deal with, most likely older prisoners, were less likely to be involved in violence. Neither the fear of victimization nor any actual victimization by staff was associated with being violent. Overall, these findings are exploratory and should be interpreted with caution, especially since neither prisoner characteristics nor coping skills were included in the model. It is entirely probable that prisoners who find it more difficult to deal with some conditions of confinement and with other prisoners and staff are also lacking positive coping skills and are instead using negative ways of coping.

229

VIII. STAFF AND PRISONER FEEDBACK ON PRISON MISCONDUCT

This chapter presents the results of the in-depth interviews with staff and prisoners regarding reasons why prisoners are involved in serious misconduct and violence as well as the current barriers and their proposed solutions to such misconduct. Some of these qualitative findings are associated with coping, while others are not.

STAFF AND PRISONER THOUGHTS ON SERIOUS MISCONDUCT

This section focuses on a set of questions put to staff and prisoners in the in-depth interviews—why prisoners are involved in serious prison misconduct and violence. The first sub-section presents data from questions about why there is misconduct and violence and other supporting information on mental health, gangs, and reliance on other prisoners that might shed light on the issue. The second sub-section presents information on the staff‘s categorization of prisoners that become involved in serious prison misconduct and violence and the categorization of interviewed prisoners into these groupings.

In the process of answering questions, many staff and prisoners expounded on their views on levels of prison violence. Prisoners‘ opinions on the level of violence tended to vary with the prisons in which they resided. Most interviewed in Moran said that there was not really any violence, ―just aggravation‖ and fallout from people knowing other people‘s charges. In contrast, prisoners disagreed about the level of violence in Maximum Security, with some describing it as very violent and full of conflict and others saying that there was little violence, especially in contrast to maximum security prisons in other states. Because of the level of security, most said there was little violence in High Security. Staff members who commented about the level of violence also had varied opinions. Some deputies and

230 lieutenants noted that the level of violence had decreased dramatically since the 1990s with the change in administration and changes in prison policy. However, others felt that prison violence, especially that associated with gangs, had increased recently.

REASONS FOR SERIOUS MISCONDUCT AND VIOLENCE Staff were asked at the beginning of their interviews, ―In general, what do you think are the main reasons prisoners are involved in serious prison misconduct and violence?‖ Prisoners were asked the same question but closer to the end of their interview.

In addition, towards the end of the interview, staff were asked ―Are there any prison strains, stresses, or circumstances that contribute to inmates‘ involvement in serious prison misconduct?‖ The responses to each of these questions were summarized into four categories: individual prisoner characteristics; certain group characteristics; the prison environment; and other prison stresses. Table 42 below presents the overall results from these questions. Each category will be discussed in detail below.

Staff were also asked whether they thought that ―prisoners‘ involvement in serious prison misconduct (was) volitional or circumstantial or a combination of both.‖ Over ten of the 25 staff said they believed it to be equal between misbehavior that was volitional or planned and behavior that was circumstantial or more spontaneous. One deputy propounded that inmates choose to misbehave but not consciously. He thought that most of the misbehavior was in response to daily circumstances but that inmates made a conscious choice to respond to the circumstances with violence. Another deputy said that even the same person can sometimes misbehave intentionally but at other times act in response to a situation. About the same number of staff (ten) believed that misbehavior and violence are more often circumstantial than planned; that they are spontaneous and rooted in emotion.

231

As one deputy clarified, ―in the prison environment, circumstances indicate actions. If an inmate feels in danger, (he‘ll) act out. There are not a lot who plan it out; most of it is a knee-jerk reaction.‖ Staff discussed those situations where prisoners had to respond because they could not get away and felt backed into a corner. One lieutenant said, ―Most that happens is (a result of) the situation, reacting to a situation. In the gym or yards, two inmates have words, because they‘re in front of other inmates they have to act out, they can‘t back down.‖ One correctional counselor gave examples in her response. ―It‘s more circumstantial…being impulsive when faced with another inmate who wants to take the phone away from you…or the officer whose taking them to account for something.

They‘re not trying to get a discipline, but it happens with circumstances.‖ One officer said that most serious misbehavior was circumstantial but that a lot of inmates ―prepare for the circumstance,‖ keeping a weapon (the ―just in case shank‖), being ―hypervigilant,‖ and always being prepared for the possible confrontations. Only three staff members thought serious misbehavior was more volitional than circumstantial, citing the fact that many other prisoners do not get dreports or that ―sometimes a guy just wants to be bad.‖

PRISONER CHARACTERISTICS When asked what contributed to serious prison misconduct and violence, almost one-third of the staff and one-quarter of the prisoners brought up various personal characteristics of the inmates. Interestingly, most of these personal characteristics had to do with prisoners‘ personalities or attitudes, rather than socio-demographic characteristics. As one correctional counselor noted, ―a lot of the attitudes that brought them here are still with them when they get here.‖ The largest sub-category of responses from staff and prisoners were that prisoners‘ poor upbringing resulted in their lack of life skills, not knowing how to

232 act, and not knowing how to obey. Of their upbringing, one deputy commented, ―they didn‘t have the nurturing environment where they were taught how to cope with everyday stresses, how to act. A lot of individuals have suffered from abuse in their own lives so they think that‘s the way to act.‖ One deputy lamented that some inmates were enabled as children and teens; he discussed how some offenders have been enabled by their families to the point where they never had consequences for their actions until they ended up in prison.

Another said, ―It‘s acceptable behavior in the culture they‘re in, as far as what‘s right and what‘s wrong. (Most regular) people don‘t get violent and (instead) think things out.‖

Indeed almost half of the prisoners interviewed had been involved in serious delinquency resulting in their spending their teen years in and out of juvenile correctional facilities or group homes. While others had been placed on probation or had been detained for short periods of time, these eleven prisoners had spent considerable time in mostly locked facilities, sometimes going in and out over a number of years.

Several staff mentioned that prisoners involved in misbehavior were impulsive, lacking self-control; some said they could not cope. Several staff and one prisoner said that inmates who get into disciplinary problems were often angry or as that prisoner described it, ―it‘s a lot of hate in their heart.‖ Some staff just concluded that these inmates were more deviant, sinful, selfish, jealous, and some were sociopaths. But prisoners also characterized those in trouble as sometimes just having a bad day or as needing help of some kind.

PRISONER GROUPINGS IN INTERACTION WITH THE PRISON ENVIRONMENT The next category of reasons that staff and prisoners give for misconduct involves some categories of people (gangs, the young, the mentally ill, the weak) but those categories alone do not explain the misbehavior. Instead there are environmental factors in

233 the prison that make these groups particularly susceptible to prison misbehavior. For example, weak inmates in and of themselves are not necessarily violent, but putting them in a situation where they might have to fight for their lives results in what some might call

―necessary‖ misbehavior. Although the discussion is divided into four distinct sections for readability: youth/peer pressure/disrespect; gangs; weak prisoners; and mental health problems, they are so interconnected that it is nearly impossible to discuss each one without mention of the others. As can be seen the four sub-categories are various types of people but the youth sub-category entails the issues of peer pressure, disrespect and reputation and the ―weak‖ sub-category entails a discussion of prisoners relying on others or helping others ensure safety; issues that affect nearly all of the other sub-categories.

Youth, Peer Pressure, Disrespect, and Reputation

Both the bivariate and multivariate analysis found age to be one of the strongest predictors of serious prison misconduct and violence in this study. That is, the younger the prisoner, the more likely he is to be involved in serious prison misconduct and violence.

Over a dozen staff and half a dozen prisoners brought up the fact that younger inmates were more likely to be involved in prisoner misconduct, compared to older ones; that peer pressure among youth was a major factor in misconduct; that much of the violence stemmed from incidents of disrespect among the young, or that young prisoners were violent to make or uphold their reputations. One correctional counselor summarized many of these points. ―The majority of violent inmates are young guys. They have to make

234

Table 42 - What Contributes to Serious Prison Misconduct and Violence?

Reasons Staff – Why Staff – Prisoners – Misconduct? Strains Why Leading to Misconduct? Misconduct?

Prisoner Characteristics 23 0 8 Prisoners‘ upbringing 7 4 Lack of impulse control 4 Cannot cope 4 Angry 3 1 Other characteristics 5 3

Prisoner Groupings 36 13 17 Youth/peer pressure/disrespect/reputation 13 2 6 Gangs 8 5 9 Ensuring safety/weak prisoners/protective custody 10 5 1 Mental health problems 5 1 1

Prison Environment 9 13 8 Different personalities/ cannot relax 1 5 5 Gambling/strong-arming 4 1 Boredom/boredom in isolation 4 7 3 Staff Interaction 3 4 1

Other Prison Stresses 0 22 1 Family issues 14 Lack of autonomy/freedom 4 Other 4 1

a name for themselves so they won‘t be messed with. Some are the ―gap‖ kids (those incarcerated before age 18) who were housed in HiSi (HSC) before they aged out (into population). (There‘s a) bunch of them now trying to prove themselves. Some are gangsters, but many are not. Mostly they‘re in their own cliques, not necessarily a gang, but usually people (they know) from the same neighborhood.‖ A correctional officer believed that ―most of (the violence) comes out of the fact that they want to be respected and if they‘re dissed….well.‖ Another classification counselor said he heard a lot of

235 discussion about respect and being disrespectful at boards. He acknowledged, ―They‘re willing to engage in serious conflict if they think they‘re being disrespected.‖ Though only six prisoners brought up the issue of disrespect as being one of the main reasons for prisoner misconduct and violence, this topic was brought up again and again throughout the interviews by many of the prisoners. One young prisoner said ―A lot of it happens over disrespect. In here there are a lot of inmates that will get spit on and they won‘t do anything about it so some inmates think they can do that to anybody. Then they try and there are fights. They can get in a fight or even take it to the next level and get a shank.‖

Another prisoner from the HSC explained, ―It all comes down to your pride, you can see people get disrespected. If you get disrespected and there‘s only three people to see it, you can let it go. But if there is a crowd of people, you have to do it.‖ He also discussed how disrespect is used by others to instigate conflict between two prisoners, explaining that there are inmates who will make up stories about what one inmate said about another and then egg them on and say ―are you going to let him disrespect you like that?‖ Another prisoner similarly explained, ―The environment in prison, it‘s a culture of everybody being macho.

You can‘t let anything go. You can‘t let anything pass. Everybody is putting their chest out and nobody is trying to solve issues. A lot of people feed off of it on both sides

(inmates and staff) whether it be an inmate having a problem with another inmate or an inmate with a cop.‖ He said that there are many instigators who use disrespect to stir up conflict knowing that most people resort to ―confrontation first.‖ Inherent in the issue of disrespect and instigation of conflict is the issue of peer pressure; an issue brought up by both staff and prisoners as a source of strain for prisoners and a contributor to misconduct.

One young prisoner lamented his weariness with the issue saying, ―The peer pressure, it

236 never ends.‖ The issue of reputation is one that also overlaps with youth, peer pressure, and disrespect but also with being a weak prisoner or a gang member. Staff and prisoners alternatively discussed prisoners who were trying to ―make a name for themselves,‖ or those trying to prevent themselves from getting a reputation as weak or an easy target. As one prisoner in HSC clarified, prisoners ―want a reputation. They‘re building their reputations so that Joe Blow won‘t do what John Doe did to them. They want people to be able to say about themselves, ‗don‘t mess with that guy.‘‖ One older prisoner said of those young prisoners coming into Maximum, ―This is the most safe prison in the world, like a camp. But the kids comin‘ in here are scared, (yet) in a couple of days, they think they‘re superman (because nothing has happened to them). Then…the way they talk (gets them in trouble.) They want to prove something to survive, to impress; they want to make a name.

I used to act that same way. I used to not talk and just act. Now I talk.‖

Gang Members

When asked about the main reasons for serious prison misconduct and violence about one-third of the staff mentioned gang involvement. It was also mentioned by a couple of staff when discussing what prison strains contributed to inmate misconduct. A few staff discussed the pressure that gang members put on each other to ―prove themselves.‖ As one lieutenant explained, ―Gang members (use violence) to perpetuate their power. They have no choice. They might have to do more here than outside (on the street) because people are in their area (all the time).‖ Some staff believed the presence of gangs was growing in population, one explaining that they had ―a captive audience to recruit from.‖ Another officer concurred that violence had increased because of an increase in gang membership, increasing peer pressure to join gangs, and gang initiations that

237 involved either punching or stabbing somebody. But other staff, while acknowledging the presence of some gang members, questioned whether violence could be attributed mostly to gangs. As one lieutenant said, ―the real gang members usually stay out of trouble. But when they are violent, it‘s big time,‖ giving the example of five who had gathered around one prisoner stabbing him. There was a lot of discussion among staff about gang wannebes, those prisoners who tried to emulate the toughness of gang members. As one deputy said, ―gangs can be an issue. (But) in Rhode Island, it‘s the neighborhood and family relationships that can cause the rift. We haven‘t gotten to that level yet of seriousness of organized gangs.‖ Another deputy concurred and said most gang members did not act out too much, preferring to wait until they are back on the street where they are less likely to be caught. He explained how general disputes on the street spill over into prison saying, ―If you had beefs with someone on the street, you might run into them here.

They‘re not necessarily involved in a gang.‖

Only seven of the 25 prisoners interviewed said they either were or had been in a real gang. These seven had been in the Crips, Bloods, and NETA or in a gang in

Massachusetts or in another location outside of Rhode Island. Several of these prisoners had initially said they were not presently in a gang, but when asked about their time on the street, relented that they had been in a gang before. One said, ―No, not right now, but I was on the street. When you come into prison, you‘re expected to do certain things.‖ He said this had led to a lot of fights. Several of those who had been in gangs said that being in the gang had led them into more problems or trouble and only one said that the gang had sometimes helped him on the street. ―I never had a father figure in my life. They would take you in as a son or brother and help you out.‖ He said that the gang had done positive

238 things for the community also, but that in prison, it was not helpful, and ―it‘s a different environment.‖ Thus several of those who had admitted to being in a gang said they had decided to drop out of it, a couple having notified people on the street, telling them they could no longer be gang members. One HSC prisoner said he was ―putting my priorities in order‖ and ―that‘s why I backed away from the gang.‖ When asked how he went about that he said he sent letters to gang members on the street saying he was done and that he explained that it had gotten him in a lot of trouble in prison and was not worth it. However, prisoners said that it was difficult to renounce gang membership—that both prisoners and staff would often not accept that one was no longer in a gang. One gang member said that after being moved from Maximum to the HSC he had ―decided to let it (gang membership) go. (But) some people tend to hold grudges, and some don‘t want to forget it.‖ Several complained that even though they had not been involved in any gang activities in prison, their gang label affected their classification score and they still had an SRG status, both of which affected the level of security in which they resided. However, some admitted that having a gang label kept other inmates from harassing or disrespecting them. One ex-gang member said having been in a gang contributed to his safety in prison, unless the people he had shot at on the street or their associates were to come into the prison system, at which point he would be far less safe compared to other prisoners.

Three prisoners said they were not in a ―gang‖ per se, but that they were involved in turf issues that had carried over from the street. One said, ―I was never a fan of gangs‖ but said that there were turf issues depending on where you lived, especially if you were from

Providence. He said he ended up in HSC because he had gotten involved in a number of assault incidents because he felt like he had to stick with the people from his neighborhood.

239

As one prisoner explained, ―A lot of problems come from the street and carry on in here.

That‘s where most of my problems came from...people I got in fights with before on the street, but they come in here and it carries on.‖

Fifteen prisoners said they were not currently, nor had they ever been in a gang. A few said that they had been labeled a gang member by RIDOC staff because they associated with a cousin or former friend who was a known gang member. One prisoner said that staff ―ask you when you come if you‘re in a gang. If you say ‗no‘ at entry, they watch you and see who you hang around with. (So SRG level) depends who you with. When I kept talking to my co-defendant…months later, they put an SRG on me.‖ Only four prisoners admitting to gang membership or turf issues gave gang association as one of the main reasons for serious misconduct and violence; behind disrespect. However, others discussed their source of trouble in the same language as other inmates—peer pressure, respect, building a reputation as somebody not to mess with, and problems with staff. One of them concluded, ―(Staff in Maximum) say everything is gang-related, but it‘s really not.‖

Another explained that there used to be more violence during the 1990s in Maximum that could have been attributed to gang members. But he continued, it was ―different then.

Now the majority of the population want to do their time and leave prison.‖

Ensuring Safety, Weak Prisoners, and Protective Custody

Prisoners were asked if they ever relied on other prisoners to help ensure their safety or if they had helped to ensure the safety of other prisoners. An overwhelming majority of prisoners (20) said they did not rely on others to keep them safe. Many made comments like, ―I‘m me; I‘m by myself,‖ or ―I don‘t rely on nobody. I can‘t trust nobody, not even a friend.‖ One did say that a few of his associates had told him that if he had any problems to

240 let them know but he had not chosen to go to them. Another said that over time he came to realize that relying on others was not feasible. He said, ―The more you experience in here, the more you see that everyone is basically for themselves. They all have problems, problems and worries. You can‘t expect others to help you out. They have issues too.‖ Of the few who said they could rely on others to ensure their safety, one explained how ―in max, everybody cliqued up‖ but that it had not turned out to be really necessary. Another responded about his reliance on others by answering, ―Yes and no. Yes, I know they‘ll be there for me. No, I know how to protect myself.‖

Interestingly, while most prisoners did not rely on others to ensure their safety, half of them said they would get involved in incidents to ensure the safety of others. As one put it, ―I don‘t expect help, but would give it.‖ Several of these prisoners were in a gang or had close associates from their neighborhood. One said, when you‘re in a gang, ―you guys stay together and protect yourself from others.‖ Another said ―Yes, I just got into an altercation in the yard, attacked someone because they punched my boy in the face. I went to seg and they put me on B status.‖ Another prisoner serving a long sentence explained his thinking on helping out his friends. ―If one of my boys was about to get in a problem, I‘d help out so he wouldn‘t go to seg. He might have something good going for him, might be seeing parole.‖ He explained that he had gotten involved in another person‘s issues, especially when that person was about to be paroled because he did not want that person to lose that.

He stated that he himself had so much time left to do that he would not be seeing the parole board for a long time. When pressed further about his willingness to help others, but not to have others help him, he explained, ―I look at it that I ain‘t got nothing to lose. If I can maintain that seg bit, lose a few visits, then they could go home and see their kids. I would

241 only do that for the people I went to training school with or knew from growing up with, not just anybody.‖ Others said they had helped out relatives, like a cousin or a friend from the street. Another said he helped out others, but only in the case of their being bullied.

―Yes, I don‘t like bullies so I try to help someone when I see a situation get out of hand.

I‘ll jump in and put my two cents in just so they won‘t beat down this guy.‖

While many replied that they had in fact ensured the safety of others, several said that they try to not get involved in others‘ disputes any longer. As one maximum security prisoner put it, ―Yes, too many times. I used to do that, but not anymore. It wasn‘t my problem. I can‘t do that anymore. Some people have told me, ‗you‘ve changed too much,‘ but I tell them that I have to change for my benefit. I‘m proud of who I am now.‖ Another replied, ―[I‘ve helped somebody stay safe] a few times, it‘s changed recently. I think I would be more likely to do it if the person couldn‘t help themselves, if the person was going to be in big trouble,‖ which he clarified to mean getting really hurt. One prisoner said he had often inserted himself in others‘ disputes and was lucky he had not gotten in more trouble than he actually did. As he explained, ―Yes I was sent to HiSi (HSC) for assaulting somebody else. It saved me from going to HiSi for something more than that.‖

He clarified that he had stepped in between two people who were having a dispute over the phone. ―I‘ve inserted myself into many incidents, but that one was the only one that erupted in violence.‖ Another acknowledged that there was a line between helping and fighting. This older prisoner tried to help younger prisoners. ―Yes, I see how they are. I‘m older than them. I believe on passing on knowledge.‖ He said he tried to ensure safety by talking to them, but that he never had to react physically.

242

While half of the prisoners recounted how they had helped ensure the physical safety of other prisoners, about half said they never inserted themselves into others‘ disputes. One prisoner‘s comments summarized most of these prisoners‘ sentiments when he said, ―I try to avoid problems and keep myself out of trouble.‖

Of course while most staff would agree that prisoners should not be intervening in other prisoners‘ disputes or disagreements, the flip side is that failure to do so leaves some prisoners defenseless. When asked about the main reasons for violence, a number of staff

(less than one-third) mentioned that weak inmates needed to defend themselves from other prisoners or they would be targeted throughout their entire incarceration. Similarly five staff members mentioned that being preyed upon by other prisoners was a stressor that often contributed to serious misconduct and violence. Some of these are simply weak prisoners, others are targeted for their crime whether it was a sex offense or a high-profile crime, and others are targeted because they turned state‘s evidence against their codefendant or another prisoner. Staff believed that many weak inmates are just potential victims who will do a pre-emptive strike against others to save face, to be moved, or to prevent further victimization. One lieutenant said that some prisoners become involved in violence because they are being extorted and feel that they have to stand up for themselves or continue to be extorted for the rest of their incarceration. A correctional officer noted that the enemy list had grown so long in the RIDOC, that prisoners use misconduct in order to be put into segregation as a way to ―PC themselves‖ or put themselves into protective custody. As another lieutenant explained, ―(It‘ll be) an inmate who is somewhat weak, who has no associates, who will sometimes do a serious booking to get himself into segregation.‖ He said some weak prisoners believe it is better to attack someone and end

243 up in segregation rather than population. Several staff at the HSC mentioned that there were some prisoners who never wanted to leave the facility, one articulating, ―We have individuals with chronic problems and others who don‘t want to leave. They feel safe.‖

Prisoners with Mental Health Problems

Five staff members, all deputies and lieutenants, brought up mental health issues as a main cause of serious misconduct and violence. One deputy said that a ―primary issue is mental illness, (that there were inmates with) serious behavioral needs. They are afforded treatment, but they‘re still in a prison environment. They‘re managed with medication; the environment does not afford them the capability to develop their pro-social skills.‖ Others talked about how some of the prisoners would vacillate between misbehavior and good behavior, depending on whether they were taking their medication at all or whether their medication had been changed and was then working properly. Only one prisoner mentioned mental illness as a major cause of prisoner misconduct and violence.

Staff were also specifically asked to assess what percentage of the prisoners involved in serious prison misconduct also had mental health problems. Responses ranged from five percent of all inmates to 90 percent of those at the HSC. Staff at the HSC believed the prevalence of mental illness among prisoners to be much higher than staff at other facilities. Several commented that the deinstitutionalization of the mentally ill in

Rhode Island had caused an increase in the mentally ill in prison. A few though were skeptical that mental health problems existed (―not taking responsibility, not sure if that is mental illness or improper schooling‖); or that mental illness could affect misbehavior

(―some use it as a crutch to avoid discipline‖); or that prisoners needed mental health treatment (―the department over-medicates, some are manipulative to get medication‖ or ―a

244 lot will go that route to get into OSU because they get their privileges back in OSU‖).

Several staff said that the mentally ill can be found among many different types of inmates, even gang members and gang wannebes and that ―we say all the time (in boards) that ‗this person belongs in Mental Health.‘‖ However other staff averred that mentally ill prisoners are often at Maximum or the HSC as both have specific units that house those prisoners and it is difficult for some prisoners with mental illness to succeed in a medium security facility.

One deputy tried to characterize those with mental health problems who are also behavioral problems. ―Guys that kill people don‘t have mental health problems. Those with mental health problems get numerous infractions, harm themselves, and harm others.‖ Many staff commented that these ―special management inmates‖ needed extra care and stricter monitoring. Staff said mentally ill prisoners were monitored more closely at Maximum and the HSC than at medium-security facilities.

Of those who become especially out of control, one lieutenant said, ―I‘m not fond of using force on the mentally ill. Lots of police departments have training on it. For us,

(there is) minimal training. Some young officers see a nut job; they treat him like he‘s just another inmate. You can‘t go in with the same approach for those with mental health problems.‖ Some staff suggested there was a need for more mental health workers or social workers and separate units in each facility where they could have their issues addressed.

Prisoners were asked in the in-depth interviews whether or not they currently or previously had mental health issues. Ten of the 25 prisoners acknowledged that they did have mental health issues. Three of the prisoners were residing in the OSU unit in the

HSC, two others were also in HSC, three were in Maximum, and two were in Moran.

Some had been in treatment or on medication prior to their incarceration but were not

245 currently getting any treatment. Others were in treatment both before and during incarceration. A couple said that they had not been treated prior to incarceration but that being incarcerated had given them anxiety or ―gave me mental health problems cause of the stress‖ and were now being heavily treated. Mental health problems mentioned included

ADHD, anxiety, paranoia, post-traumatic stress syndrome (PTSD), though some prisoners did not specify the exact nature of their mental health problem, some seeming not to really have understood it. One prisoner said his anxiety had started at Maximum, where he ―got into it with a lot of people‖ and experienced paranoia, anxiety, and uncontrollable rage.

Two described problems with PTSD that had resulted from emotional, physical, and sexual violence when they were young. Prisoners had serious concerns about the efficacy and side effects of medications as well as positive and negative staff treatment.

PRISON ENVIRONMENT When asked about the strains or stresses that contributed to misconduct, one-third of staff mentioned the prison environment. One lieutenant commented: ―The walls are hard; they break down people after years. You have to be on different states of alert at all times except in your cell. Prison is prison. If you put the 1000 best Oxford students and put them in here, they will change after a while. It‘s a tight, enclosed environment.‖ Several staff discussed the fact that from time to time even model inmates pick up serious bookings for a type of misconduct that many called ―doing business.‖ This occurs when an inmate is going about his daily business and somebody tries to either take something from him, or hurt him, or attack him in front of other people. One lieutenant gave an example of a youthful offender who tried to take the phone away from an older, long-term offender.

After the older offender refused and then got off the phone later, the younger inmate

246 followed him out to the library and started pounding on him. The older inmate beat him.

When the officers responded, the older inmate stopped and put out his hands to be cuffed.

At the disciplinary hearing, he apologized for the incident but said he had had no choice.

As the lieutenant explained, ―older guys don‘t want to do this. They just want to go home but sometimes they can‘t let these things go.‖

One prisoner described ―instigators‖ as inmates who are not necessarily young, but immature. ―Young people are more rowdy and confrontational, but once they get older

(some) go from active participants to instigators… (the older ones say) ‗Don‘t take that‘ and they step back and watch. (There are) some guys… who try to calm things down‖ but he said they were not a part of the dominant prison culture. Other prisoners discussed the difficulty of dealing with all of the ―personalities,‖ how ―everyone‘s different,‖ and how

―people get on each other‘s nerves.‖ One prisoner in Moran described having a roommate as ―more aggravation, like living with a sister or brother. After a while you just fight.‖

Several staff mentioned that gambling contributes to misconduct and violence.

Prisoners who cannot pay their gambling debts are liable to be assaulted or to commit a serious offense to ―PC themselves into segregation‖ to avoid reprisals.

Another prison environment problem brought up by both staff and prisoners was boredom. One lieutenant thought it was the most salient factor in serious misconduct. He felt that inmates should be outside, growing their own food, or working in some other capacity, but because of budget cuts, there were fewer activities for them to do. He believed that inmates needed to be working and kept busy and that privileges should be commensurate with work. Several staff noted that there were not enough jobs to go around.

247

The last environmental factor brought up by several staff as contributing to serious misconduct is the way that some staff address prisoners. Although it was a complaint issued by nearly all prisoners, only one prisoner blamed serious prison misconduct on the way staff treated them. One correctional counselor stated that every once in a while there is an officer who does not know how to talk to inmates and theorized that officers might be a big part of the problem. Another said that he had heard a lot of inmates complain at disciplinary boards about officers being disrespectful. He was not sure if it was true but had heard it so often from so many different inmates that he suspected that there were some officers who did not treat inmates well. He lamented that ―rules are rules and security comes first but you have to respect someone as a human being and respect their dignity, but sometimes there are officers who look at inmates as less than human and treat them that way.‖ One lieutenant desired more training to educate both staff and inmates on proper behavior. He said staff needed to learn to step back for the sake of maintaining a healthy prison climate, rather than escalating the problem—the worst thing an employee could do.

OTHER STRESSES CONTRIBUTING TO SERIOUS MISCONDUCT When staff and prisoners were asked what the main reasons were for serious inmate misconduct, most identified a characteristic of the prisoner, a characteristic of the prison environment, or the interaction between the two. However, when staff were asked whether there were prison stresses that contributed to serious misconduct and violence, other responses emerged that had not come out previously. Only two staff did believe there were not any strains that led to misconduct and violence, saying that misconduct was caused by other issues not prison strains. A couple of staff discussed the loss of freedom and the loss of autonomy. As one deputy explained, ―you have COs telling you all the time what to do.

248

There‘s always that authority figure. Especially for new inmates, it‘s difficult.‖ He also said that the loss of freedom for first-time offenders often resulted in ―shell-shock‖ and sometimes bookings. A similar stressor is the lack of privacy about which one officer said,

―We have to be able to go in and search for contraband that is dangerous, but it‘s difficult for them. We even scan their letters and that takes away their privacy. They have to deal with it.‖ A few staff said that substance abuse addiction was also stressful for some inmates, leading them to misconduct associated with drug use. However, the stressor brought up by the most staff (14) involved issues with prisoners‘ families. Staff said that inmates often worried about their families. One deputy thought that prisoners had too much access to their families via visits and phones, resulting in inmates being very much aware of problems going on at home but being helpless to do anything about them. That is, inmates have the ―burden of their families‖ but are unable to act on it, furthering their stress. Some staff noted that it is particularly difficult on inmates when there is a death in their family or they lose their partner, either through a divorce or a break-up. One lieutenant said that ―life goes on outside and they lose total control of that (outside) life.‖ When they find out their wife is divorcing them, or they get a Dear John letter from their girlfriend, they take it out on somebody in the prison. Finally one correctional counselor said some inmates are stressed by the lack of contact with their family members: they sometimes accuse staff of tampering with their mail, when in fact the families have stopped writing.

CATEGORIZING PRISONERS INVOLVED IN SERIOUS PRISON MISCONDUCT After conducting a couple of staff interviews, a new question was added to the staff interview asking them if they could categorize those prisoners who were involved in serious prison misconduct and violence. The thinking behind adding this question was that

249 identifying the type of prisoners would help in creating solutions to reduce misconduct.

This question was asked after the general question of why prisoners are involved in serious misconduct and violence. Thus some of the responses mimic those given in that question.

According to staff responses, there were roughly six categories of people involved in misbehavior, four of which have already been discussed including young inmates, gang members, weak inmates, and those inmates with mental health problems. One of the two categories that staff had not mentioned previously were those prisoners, who staff deemed as being ―disruptive for their own entertainment,‖ who did not like authority, and who had nothing to lose. Often these prisoners were described as having long sentences to serve, as extremely assaultive and violent, and sometimes as sociopaths. The final category was not really a category of offenders at all but a category of the incidents themselves. This category, normally referred to by staff as ―doing business‖ involved those instances where prisoners were drawn into serious prison misconduct and violence unwillingly by other prisoners who were trying to prey on them or get something from them. Most staff brought up this category sometime during the interview, thinking it unfortunate, but a part of ―being in prison.‖ Some said that prisoners caught up in bookings in this manner were sometimes sanctioned less seriously, and not usually downgraded as their booking was attributed to an unavoidable fact of life in prison.

Given these six categories, efforts were made to categorize the 25 prisoners interviewed in these categories. Each of the 25 prisoners interviewed easily fit into one of these six categories. Unfortunately, many of them fit in multiple categories. There were certainly a half dozen prisoners interviewed who reported mental health problems, who were young, and who were either gang members, gang wannebes, or who were considered

250 weak or a protective custody inmate. Several of them were also substance abusers.

Although one interviewed might have fit into the category of ―being disruptive for their own entertainment,‖ some staff had referred to him as having mental health problems, and he could also be categorized as a ―rebel.‖ This last category coined by this researcher refers to those few prisoners who choose not to stop their rebellion against the system, pretty much considering themselves ‗at war‘ with staff due to their perceptions of unrighteousness and injustice. Unfortunately, many prisoners fit into multiple categories and are, as a result, possibly more at risk for involvement in serious prison misconduct and violence than those who fit into only one category.

*****

In summary, staff and prisoners‘ thoughts on serious misconduct and violence fell into several patterns. Only a few staff believed that most prisoner misconduct was volitional, planned, and purposeful. Half of the others believed it was a combination of volitional and circumstantial (sparked by events and circumstances within the prison) and the other half believed it was more circumstantial, though they also noted that inmates were hypervigilant, on alert, and prepared for possible altercations. The questions regarding why prisoners were involved in serious misconduct and violence and whether there were stresses that contributed to misconduct yielded responses that fell into four categories: prisoner characteristics; prisoner groupings that interacted with the prison environment; the prison environment; and other prison stresses not included in the first three categories.

One third of staff and one-quarter of prisoners pointed to prisoners‘ personalities, inner qualities, and attitudes. Many supposed that these prisoners had poor upbringing and

251 lacked the necessary life skills to get along with others. They believed those involved in misconduct were often impulsive, lacked self-control, and lacked coping skills. Not surprisingly, almost half of the prisoners interviewed had been incarcerated since their early teens. Some staff said inmates were angry; but only a few characterized them as simply bad. Some prisoners said that they needed help and sometimes were just having a bad day.

Four groupings of prisoners were mentioned, all of which interacted with difficulties in the prison culture. Half of the staff and prisoners said most of those who got involved in serious misconduct were young. They noted that peer pressure and gaining a reputation as being able to protect one‘s self were issues that were particularly relevant for young prisoners. However, the most salient factor for youth was the notion of ―disrespect.‖ Most of the young prisoners believed that one could not walk away from situations where they had been disrespected, because the disrespect would continue and escalate. The second grouping was gangs, which one-third of staff and a few prisoners brought up as being a major reason for misbehavior. However, there was some disagreement among staff about the prevalence of real gangs vs. neighborhood turf issues, and about the prevalence of gang involvement in violence. Only 10 of 25 prisoners interviewed said they had been in a gang on the street, in prison, or had been engaged in turf issues. Though their comments did confirm the role of gangs in some of the prison violence, most of the issues these prisoners discussed had to do with peer pressure, building and maintaining their reputation, confrontations with staff, and their quest for respect. The third grouping was weak prisoners; about a third of the staff noted that weak prisoners were likely to be involved in misconduct in order to either have themselves segregated or to defend themselves. Finally mostly deputies and lieutenants identified mental illness as contributing to serious prison

252 misconduct and sometimes violence. When asked what proportion of those with serious disciplinary problems had mental illness, the percentage ranged from five percent to 33 percent of all prisoners department-wide to 50 to 90 percent of those housed in the HSC.

Indeed 10 of the 25 prisoners interviewed admitted to having mental health problems.

When prisoners were asked if they relied on other prisoners for help in ensuring their safety, almost all said they did not. However half of the prisoners reported that they had at one time or another helped to ensure the safety of others, either friends, fellow gang members, or relatives. Many had gotten bookings and some had come to the conclusion that they could no longer get involved in other people‘s disputes. Yet aspects of the prison environment were cited by one-fourth of the prisoners and one-sixth of the staff as being a main reason for prison misconduct. One of the most difficult parts of the prison environment was dealing with all of the different ―personalities,‖ which often resulted in prisoners not being able to relax in prison. Gambling and the debts incurred were another factor in misbehavior. Both prisoners and staff brought up boredom as a factor in disciplinary issues, saying that prisoners had too much time to think about, plan, and carry out misbehavior and not enough time involved in constructive activities that might keep their focus away from it. The last prison environmental stressor brought up by mostly staff as a reason for misbehavior was staff interaction with prisoners that was disrespectful, confrontative, and inappropriate. Interestingly, while almost every prisoner brought that up as a problem sometime during the interview, only one had given it as a reason for misbehavior. Lastly, some staff believed that some stressors set off people who might not normally be involved in misbehavior and contributed to the increased misbehavior of others. These stresses mostly revolved around prisoners‘ families and loved ones; whether

253 it was because they were aware of but could not help solve problems at home, had lost contact with family members, or were left by them (divorce or break-up), or were concerned about their health and sad when they died. Other general stressors that staff mentioned were the lack of autonomy, freedom, and privacy.

Finally, staff members were asked to categorize prisoners involved in serious prison misconduct and violence into their own groupings. Six categories were mentioned. In order of most to least frequently mentioned, they included prisoners with mental health problems, young prisoners, weak prisoners, gang members, inmates who choose to be bad, and those prisoners, usually more well-behaved, who fought back when victimized by others. After trying to categorize the 25 prisoners who were interviewed, it was determined that they all fit into at least one of the six categories; however, two-thirds of them fit into multiple categories indicating the complexities of addressing serious misconduct and violence.

CURRENT BARRIERS AND PROPOSED SOLUTIONS TO SERIOUS MISCONDUCT

This section expounds on staff members‘ and prisoners‘ thoughts on solutions already in place and suggestions for addressing serious misconduct and violence in the future. Responses were again derived from the staff and prisoner in-depth interviews.

CURRENT BARRIERS TO SERIOUS MISCONDUCT AND VIOLENCE Both staff and prisoners were asked what RIDOC policies and practices were helping to keep the violence and misconduct down and without which the misbehavior and violence would increase. In the interviews staff articulated the extent and depth of the policies and practices that served as barriers to serious prison misconduct and violence. It is not surprising that prisoners, the recipients of those policies, were less likely to discuss

254 those practices possibly because of their reluctance to acknowledge the capability of staff or their opposition to or ignorance of some of the policies. Four prisoners went so far as to say that there were no policies in place in the RIDOC that acted as barriers to prison misconduct. However, three-fourths of prisoners were able to think of at least one type of barrier. However, staff members were able to articulate many policies and practices that have been instituted that prevent misconduct and violence. The various policies, procedures, and practices can be divided roughly into six categories: direct supervision/vigilance, the disciplinary process, programs, classification/SRG, addressing prisoner issues/good communication, and other types of practices (Table 43 below).

DIRECT SUPERVISION AND VIGILANCE OF RIDOC STAFF When asked what policies and practices already in place served to reduce misconduct and violence, almost all of the staff interviewed cited the vigilance of staff and the RIDOC‘s ability and practice to provide direct supervision to inmates. Vigilance included staff‘s observation of prisoner behavior and any changes in it as well as standard correctional practices to reduce misconduct such as random cell searches and monitoring of mail. One correctional officer expounded on officer practices that reduced misconduct,

―officers noting behavioral changes, cliques forming and changing; knowing who you are watching and who is watching you. Most of it stems from vigilance; it‘s all about knowing.

You can get info on the computer. You need to know who the sexual offenders are because they need protection.‖ One lieutenant discussed the importance of the constant monitoring, giving an example of one area, prisoner employment, that can result in misconduct. He said

―When we become aware that inmates are using their job‘s freedom of movement to do bad things, we address it right away by firing them from that job, thus restricting their

255 movement. Officers do the screening for jobs. They know when an inmate is trying to get over on them.‖ Another lieutenant discussed the importance of vigilance and of building a rapport with prisoners to help gain further information about misconduct. ―Staff are the eyes and ears for everything; (they are) unsung heroes.‖ He explained that when staff members hear about something from inmates, it goes up the channel to whoever will deal with it or investigate it. ―(Officers) develop a rapport (with inmates). Inmates want to do quiet time. Often an inmate will get the word out to us that there‘s a troublemaker so that officially we (proceed to) do a thorough search, conduct an investigation of all involved.‖

One administrator at Maximum reviewed both the institutional and individual practices of vigilance that helped to reduce prison misconduct and violence at that facility.

Institutional practices included high officer presence; routine searches; a newsletter and regular communication to announce changes in policies or procedures to staff and inmates alike; programs; and the elimination of the inmate hierarchy (―if they stick their head out, they‘re gone‖). Individual practices included not allowing inmates in other inmates‘ cells; refraining from bullying inmates; protecting weaker inmates; and keeping inmates away from their cells when doing searches (watching officers search had resulted in many inmates getting upset and causing a scene). Several staff credited officer training and experience for the vigilance. As one correctional officer explained, ―You can walk into an area and know there‘s tension. It‘s just basic experience that you get over time.‖ Several prisoners did credit staff‘s vigilance with reducing prison misconduct and violence. As one observed, ―They analyze the surroundings and pay attention to what‘s going on. If five guys or less bundle up in a group setting, they want to know what‘s going on. They‘re watchful.‖ In addition, one noted that staff members make an effort to protect sex

256 offenders and another said that staff‘s vigilance and willingness to ―try to nip it in the bud‖ served to reduce violence. One deputy averred that ―RIDOC has done a good job stemming the violence and violent activity. Serious stabbings and slashes are few and far between.

Most assaults involve a couple of punches and then staff intervenes.‖

Table 43 - What Contributes to Keeping Prison Misconduct and Violence Down? Barriers to Serious Misconduct Staff Prisoners

Direct supervision &vigilance of staff 17 3 Disciplinary policy & good-time incentives 15 11 Programs 11 4 Classification, SRG & keeping enemies separate 8 1 Addressing prisoners’ issues & positive communication 5 3 Other 17 2 None 0 4

In addition to the vigilance, RIDOC staff practice direct supervision, monitoring prisoners in the block, so that ―staff are not behind the cage all of the time but are physically visible,‖ communicating directly with inmates. As one deputy explained it,

―Rhode Island is a small system. Staff do a good job getting to know inmates in their unit.

(There‘s) lots of close personal contact. They tend to start to know people who move through the system. Getting to know inmates plays a big role in keeping violence down.‖

When asked how that worked he elucidated ―It helps because you have a rapport. If there‘s a fight and an officer steps in, that‘s usually the end. A lot of these guys, if they know you, feel comfortable with you, they‘ll just stop….If you‘re in with thousands, it‘s just a numbers game. We‘re unique. Visitors from other systems come here and don‘t know what to make of it.‖

257

Finally along with vigilance and direct supervision, several staff singled out the importance of consistency. As one deputy noted, ―Just staying on it. Line staff has to be there and handle it every day. We‘re fortunate that our line staff work together; when staff work well together, it‘s a deterrent to inmates. Another officer at HSC concurred,

―Consistency and everybody trying to maintain the rules of order. If inmates know what to expect, they will do it. It‘s more about consistency. I am the most strict (officer) but inmates know what I expect.‖

DISCIPLINARY POLICY AND GOOD-TIME INCENTIVES Three-fifths of the staff and almost half of the prisoners credited the disciplinary and good-time policies with reducing serious prisoner misconduct and violence. One deputy called it a ―solid system in place with consequences and sanctions.‖ Others noted that the system used progressive discipline, providing for limited sanctions for first-time disciplinary offenders but increasing the punishment for subsequent and repetitive misbehavior. About the progressive nature of the policy, one lieutenant said, ―We try to be fair and do justice.‖ Several prisoners acknowledged that the potential consequences they faced for misbehavior was a deterrent. As one acknowledged ―With all of the petty bookings, you get so much time (in segregation), you may as well not do the violent stuff.

There are violent people but they‘re not people who are going home.‖ A few staff and prisoners believed that the RIDOC‘s relatively recent policy of state-charging prisoners for serious assaults (especially for those where someone has been hurt) has also been responsible for stemming violence in the prison system. As one lieutenant stated, ―before we did not state charge, now we do. So they‘re worried about picking up extra charges.‖

One prisoner acknowledged its deterrent effect, saying that the department‘s policy of

258 instilling fear helped reduce serious misconduct and violence. When asked to clarify, he explained that prisoners refrained from misconduct ―so people don‘t get a state charge or get sprayed (with gas)...Not a good way to stop it, but it does.‖

However, in addition to the sanctions one faces when involved in serious misconduct and violence, there are also incentives for good behavior, especially the good- time policy. As mentioned previously, prisoners accrue ten days of good time for every month in which they are not found guilty for a booking. In addition, if they are found guilty of a booking, they lose the same amount of good time as the number of days they spend in segregation. Therefore, if a prisoner is found guilty of a booking and receives 20 days in segregation, he loses 30 days of good time—20 days plus the ten that he will not accrue that month. One prisoner pointed to good-time as reducing misconduct when he said, ―That‘s helped a lot for people who don‘t want to give it up.‖ He explained that ―if you‘re doing short time‖ or only have a short amount of time left to do, the good-time is incentive not to get involved in disciplinary action. He had calculated that prisoners can basically do a year sentence in seven and half to eight months if they do not lose good-time. ―A lot of people have started calming down; it can almost cut your time in half.‖ Many of the staff concurred, though some said that there were inmates, especially those who were young, who had not yet begun to realize the impact of misconduct on the length of their sentence.

PROGRAMS, EDUCATION, AND JOBS The presence of treatment programs, educational classes, and jobs was mentioned by almost half of the staff and about one-sixth of prisoners as helping to stave off prison misconduct and violence. As one lieutenant in Maximum Security reasoned, ―you have to occupy their time. The more time they have to do nothing, the higher their violence.

259

Programming, vocational education, and industries helps keep them busy. They get into trouble when idle.‖ Several of the staff brought up the various cognitive behavioral therapy programs within the prisons as particularly effective. One deputy averred that cognitive programs like anger management, Crimanon, and substance abuse treatment run by outside agencies helped prisoners recognize their anger issues, take responsibility for their actions, and kept them occupied. He acknowledged that ―initially they go for the good time, but by osmosis, they‘re picking up better attitudes. They don‘t even realize how much change is taking place. Programs are important unless we just want to recycle them back (into the prison system). They haven‘t had any guidance. How they grew up impacts how they are now. Some are on the family plan, with generations having come here.‖ Another deputy praised the cognitive learning at the HSC, including the 12 to 16 week AVATAR program and Think First, the follow-up program. He credited these programs with ―teach(ing) inmates to slow down, think, learn basic things like (that) what they say affects others.

Some didn‘t have it growing up. Some simple skills they missed, like communication. The anger management program uses self-directed workbooks with weekly counseling and is an evidence-based program.‖ He gave an example of the success of these programs at HSC.

There had been an inmate who could not go a day without a booking. He did not want to participate in programming. However, he eventually agreed to do the self-directed anger- management workbooks and got a nice certificate twice. Next he was asked to go in front of the AVATAR board to determine whether he was appropriate for admittance into that program. At the first meeting he was so disruptive that the staff could not get through an explanation of AVATAR. However, officers were asked to work on doing behavioral modification with him. One officer went out of his way to help the inmate. Eventually, the

260 inmate completed AVATAR, wanted to get to B status, and when he did then signed up for

Think First. The deputy said it took three years but eventually he went nine months without a booking. He concluded, ―the change was phenomenal.‖

Other staff members were less convinced about the efficacy of programs to prevent recidivism, but many were in favor of them anyway. As one lieutenant said, ―I can‘t say for certain if they pay off, if they‘re effective, but I‘ve seen evidence of these programs

(Spectrum substance abuse program and AA/NA) helping. They are beneficial—(they) put you in touch with a mentor and a sponsor.‖ Throughout the interviews prisoners discussed programs that were helpful to them and some of them were able to link what they learned from the programs to a decrease in their own misconduct and violence. Though one questioned whether other prisoners were taking to heart what they learned in the programs, another talked about how it was beneficial to prisoners in HSC. He stated, ―They created the AVATAR program here and at max. It gives (prisoners) hope. Otherwise if people don‘t have hope they just say, well ―I don‘t have anything to lose‖ (in segregation).

Although only one-sixth of prisoners linked participation in programs with lower levels of prison misconduct, three-fifths of the prisoners interviewed said they had participated in programs and the majority of those who had participated had positive comments to make about most of the programs. Only one prisoner who had been involved in several programs was critical of them. Several other prisoners who had not been able to get off the waiting list and into a program were also critical of either the way people were chosen for programs or the fact that there were not enough program slots. Two of the prisoners residing in the OSU in the HSC said there were no programs in which they could participate, although they would have liked to have done so.

261

The general comments about programs below demonstrate the various perceived benefits of the programs and prisoner motivations to join. One older prisoner in Maximum admitted that he had been incarcerated for four years in the early 2000s and ―did a lot of programs but didn‘t really take it seriously. I took every program they had, (including) anger management. Not saying it didn‘t help, I just was focused on getting out.‖ Another prisoner in Moran said he was in the Spectrum program and was ―doing it for the good-time and it keeps me out of trouble, keeps me busy. I don‘t have too much time left (in prison).

And I‘m living in a block where everyone wants to not be kicked out of. For the most part

I‘m just trying to stay out of trouble (until I get out).‖ Another prisoner at Maximum found his program involvement helpful to a degree. ―It helped me get away from certain areas (of the prison) cause I take the class. (However) there are always going to be people who haven‘t taken the class in population.‖ Thus he said while he has learned to control himself and communicate better, there were still many prisoners in population whom he had to deal with who had not learned those lessons. Another Maximum prisoner was delighted by the programs in which he participated saying, ―I learned a lot; I made progress‖ and another said that overall he had ―learned some very good information.‖

Although it is not possible to detail each program at the four prisons or provide feedback on them, needless to say there were many different programs in which these 25 prisoners had participated at the three facilities, HSC, Maximum, and Moran. They included AVATAR, Think First, substance abuse programs, anger management, Peaceful

Solutions, victim impact, non-violence, Community College of Rhode Island, G.E.D. classes, meditation group, health programs, Crimanon, a men‘s trauma group, NA and AA, parenting classes, and domestic violence.

262

Several prisoners from HSC who had or were participating in the AVATAR program said that it had taught them how to cope. One explained that it ―helped me express myself, open up to my counselors, helped me with my anger.‖ A few had been or were participating in the Think First program, a follow-up to AVATAR. They described it as a little bit of substance abuse, how to deal with emotional abuse, and as trying to help prisoners further along, giving them breathing exercises and other ways to relieve stress.

One prisoner expressed his frustration that he had not been able to get into the program. He said, ―I applied but they denied me three times cause they say I‘ve gotten into too much trouble. I‘m trying to get in next month when it starts again.‖ There were also some mixed feelings about the initial anger management class, a course that was workbook oriented.

Some found it helpful; others would have liked it to be more hands-on.

Of the substance abuse programs, one prisoner said he ―learned how to deal with anger, how to deal with cravings.‖ Several said that it was helpful in learning how to deal with life without resorting to substance abuse and that it was run by ―good people.‖ About

Crimanon, one Moran prisoner said that it ―teaches you about being clean, being respectful to others and yourself and justification—about happiness, communication, and personal integrity. The communication part taught me how to communicate with someone, how to say the words right. I (can) talk right when it comes to the right person.‖ One prisoner said he learned a lot in the victim impact class. ―I learned about the crimes that I did, (how) it affects more people than me being here. I‘m hurting my family.‖ Finally, two prisoners credited their program involvement with helping them calm down. One had started doing meditation two months prior and he said, ―it relieves stuff. When you meditate, it relieves everything. People ask why I‘m so calm, I tell them I just want to do my time and get out.‖

263

He added ―when you meditate, you forget about everything.‖ The other prisoner claimed that ―Peaceful Solutions is what got me thru this bid, to be peaceful. I used to spaz out at

ISC.‖ While not all of the comments were as effusive or life-changing as some of these were, with the exception of one prisoner who was critical, most prisoners said that programs had been helpful in some way; that they had learned from them; and were appreciative of them.

CLASSIFICATION AND THE SRG PROCESS About one-third of the staff and one prisoner credited the classification and SRG processes with keeping down misconduct and violence. Staff discussed both officers‘ and classification staff‘s efforts to keep enemies separate. In addition, the SRG process identifies prisoners who have been management problems, had high-profile crimes, were escape risks, but mostly served to identify those prisoners with gang affiliations. One deputy asserted ―we have a good handle on gang violence (as a result of) the SRG process for gang members or gang leaders.‖ He further explained that ―a lot of the violence is drawn from gang-related issues or the drug trade. It‘s rare to have an individual working outside of a gang affiliation selling drugs in prison.‖

ADDRESSING PRISONERS’ CONCERNS AND POSITIVE COMMUNICATION A handful of staff and a few prisoners believed that addressing prisoners‘ concerns and positive communication with prisoners helped reduce prisoner misconduct and violence. A lieutenant at Moran noted that if concerns are brought forward, staff try to resolve them. He noted, ―Before we didn‘t do that. If they had a problem, too bad. But everything is addressed now. Ninety percent have legit issues or concerns, for example about their good-time, personal problems, family problems, furloughs.‖ Mostly the

264 captains and lieutenants deal with their problems.‖ Some of the facilities have what is known as ―happy hour‖ where prisoners coming out of lunch have the chance to talk to prison administrators about their issues and problems. As another lieutenant said, ―At happy hour, they can come in and ask us questions. I do what I can to help them. We make ourselves available. I look at them like children and we‘re the parents. They have someone to go to; maybe they didn‘t have that at home. We tell them when they have problems to let us know (because) we have people to talk to (them) about it. Their lives don‘t end when they get here.‖ A deputy said that even at HSC, ―if a guy has a problem and you can address that, interacting with them, getting out there and talking to guys goes a long way with the inmate population. Making yourself visible and talking to people plays a role in lessening violence. They appreciate it.‖ One prisoner in HSC concurred saying, ―in this building we have a counselor, you can request to see her. Just to be able to talk to her, to get (things) off your chest. We have religious people and they will talk to you and give you that positive energy. (There are) some officers in here that will give you positive feedback.

Some will try to help you, give you that advice.‖ Another long-term prisoner at Maximum said he believed that staff had become more ―compassionate‖ compared to years past, saying, ―They try to tell them how to do their time right and give advice to young people.

However, after three strikes, you‘re out.‖ One lieutenant concluded, ―Sometimes plain old communication works, so sometimes just letting them vent calms (them) down.‖

OTHER PRACTICES Staff and prisoners suggested that a number of other policies and practices helped reduce prison misconduct and violence. One that a few staff brought up was the monitoring of prisoners with mental health problems, especially being cognizant that they appropriately

265 took their medication. At Maximum one staff member said that they ensure that those with mental health problems come out of their cell from time to time to get some daily interaction with other people and not lose themselves in their cell.

Several staff brought up the importance of professionalism among staff along with their being good role models to inmates. When asked what reduced misconduct and violence, one deputy answered, ―Staff training and professionalism. The communication skills that staff have to step in and prevent situations from happening…We see if someone is having a difficult time or is about to get into a situation with someone and we try to prevent that.‖ A lieutenant at Moran said ―we remain professional and treat them with respect and dignity. It does influence them. Some leave with a little thankfulness and less of a negative image of authority.‖ Another explained that the department had increased its expectation of officers and that if they used too much force, they would be held accountable for it.‖ He believed that the staff ―worked well together,‖ and that it provided inmates

―positive reinforcement if staff were professional.‖ One deputy further explained, ―the communication skills of officers are critical for inmates who have a propensity for misconduct. A good CO/staff can serve as role models for inmates. It can make or break the inmate‘s outlook on his incarceration. They can be encouraged or coaxed into program participation.‖ He said that unfortunately the reverse was also true, that he has seen inmates make forward strides through programming that ―were annihilated with staff hostilities.‖ Finally, a deputy explained the RIDOC‘s practices of focusing on small issues to affect the overall climate of the institutions. He explained, ―when (Moran) first opened, we had officers telling inmates what to do and found that staying on inmates (e.g. ―tuck that shirt in; make that bed‖), that those little things put them in the direction of better behavior.

266

Some relish being told what to do; they never had people tell them what to do before and they need structure.‖

Finally a few staff brought up other practices that they thought had reduced serious misconduct including the camera system, single cells at Maximum, and family visits, which inmates did not want to lose. At HSC, one staff said that such practices as having prisoners in outside cages rather than the full yard for recreation, the use of pepper spray, installation of bars in the windows, and instituting the practice of not opening doors until inmates had backed up, had all reduced the violence in that facility. A few prisoners in HSC alternatively pointed to the fact that staff restored TV and radio privileges to B status prisoners, saying it ―reduces the violence; it gives people another outlet.‖ About giving those in B block their TVs, a protective custody prisoner said, ―That‘s going to calm people down so much. People want to see that stuff. Spend time watching shows, watching commercials, seeing what‘s going on with the world. It takes people‘s mind off of things.‖

PROPOSED FUTURE SOLUTIONS Staff members were asked their recommendations for reducing misconduct and helping prisoners cope: ―What type of correctional practices or policies would you recommend to decrease the likelihood of prison violence? Of serious disciplinary misconduct? Do you think that the prison system could help inmates cope better? How?‖

Prisoners were asked similar questions: ―What would you recommend that this prison or the system do to decrease the likelihood of prison violence? Of other types of serious disciplinary misconduct? Are there things that the DOC could do to help you not get further bookings? What could the DOC do to help prisoners address their problems and

267 stressful situations?‖ The summarized responses are divided by solutions for coping and solutions for reducing serious misconduct by staff and by prisoners (Table 44).

With the exception of one or two people, most staff and prisoners had lots of ideas of how to reduce serious prison misconduct and most prisoners had ideas about increasing prisoners‘ abilities to cope. However, almost half of the staff were not sure that there was anything more the RIDOC could afford to do to help prisoners cope. One lieutenant acknowledged that the RIDOC could probably give help to prisoners to cope, but he continued, ―Probably could be lots to be done but at what costs? Corrections is a necessary evil but it costs a lot of money. Even volunteer programs cost money for staff to supervise.‖ Other officers either felt that prisoners with serious discipline problems did not want to change or could not. One at HSC answered, ―A lot of them don‘t want to cope, it‘s my way or no way. They‘re living with the bare minimum here, why would they do that?

When time gets less, they act like bigger jerks. I know there‘s nothing we can do to them.

Many get lots of segregation time just before they leave.‖ Another officer at HSC was equally pessimistic, ―Only purpose they serve is to keep me employed. They add nothing to society. We can‘t fix them, we can only contain them.‖

More than half of the staff did believe that the prison system could do a better job of helping prisoners cope. Their suggestions fell roughly into two categories: more programs and activities and more staff to help prisoners address issues. Interestingly, those two categories were also the two largest categories for how to reduce prison misconduct and violence. There was also another category, correctional officer treatment of prisoners that both staff and prisoners pointed to as needing improvement. There were some solutions that only a few staff or a few prisoners thought might decrease serious misconduct. Finally,

268 when asked about how the department could help them achieve decreased bookings, a handful of prisoners said there was little more that the department could do, saying either

―it‘s up to me‖ or ―it‘s all on me. I got to do what I got to do so I won‘t get booked again.‖

Table 44 - Potential Solutions to Serious Misconduct and Violence Solutions How help prisoners How reduce cope? misconduct? Staff Prisoners Staff Prisoners Unsure if can help further 11 1 1 7 Programs, education, jobs 13 12 22 22 More staff; more communication 9 10 8 2 End staff disrespect 1 3 6 12 Other 0 2 3 9

TREATMENT PROGRAMS, EDUCATION, JOBS, AND CONSTRUCTIVE ACTIVITIES As mentioned above, the need for programs and constructive activities topped the list of suggestions to improve prisoners‘ coping and to reduce serious misconduct. Twenty- two staff members and 22 prisoners said that there needed to be more programs and other constructive activities like education, jobs, training, and organized pro-social activities to reduce misconduct. Of the need for more treatment programs to help prisoners cope, one lieutenant said of inmates in HSC, ―They need to be less warehoused. Need more treatment, no way they‘re getting the treatment they need. We need more time with this, more attention paid to it…we have manpower issues.‖ A deputy declared, ―At some point, we‘re going to have to deal with it. It‘s going to take a commitment involving resources we don‘t have currently. It‘s almost too late for some inmates. I wonder if they had (coping skills) on the street, how many wouldn‘t be here.‖

Several staff members recommended increased programs for rehabilitation purposes. As one correctional counselor noted, ―The more programming, the more chance

269 you have on changing violent and negative thinking, I think the better the inmates will be.‖

He also explained that a lot of the programs in place were excellent but that there needed to be more slots for the programs so that more prisoners could get involved in them. One deputy was striving for a more holistic approach to treatment, somewhat like a therapeutic community with the need to include officers as well as staff. He said this would require the buy-in of the staff and a redefined mission statement, centered on impacting violent and high risk offenders. It would also require the implementation of unit management and training of all of the staff. One correctional counselor recommended that the department needed to ―individually look at someone and make a recommendation. If he‘s in here for drugs and alcohol consistently, he needs drug counseling.‖ He did say that prisoners who wanted to deal with anger issues and other types of issues could eventually get the needed programming.

Several staff commented on the specific need for treatment programming for prisoners with serious mental health and substance abuse problems. One officer at High

Security pointed to the lack of programming and activity in OSU for prisoners with mental health problems. He said, ―They have more time to sit there and think; it‘s not good. If the time was spent doing something, it would be better. They‘d have less time to think and get paranoid.‖ One of the HSC lieutenants agreed and believed it was essential to expand the

‗mental health piece.‘ He bemoaned the fact that in the 1970s Rhode Island had closed many of its mental health facilities, leaving those with mental health problems to roam the streets and eventually end up in prison, where they cannot function. He suggested keeping them in population and treating them. He said corrections in general ―always treats problems with action-reaction, action-reaction.‖ He criticized staff saying they do not often

270 ask prisoners why they are misbehaving. He suggested that there be more case management, where you could have all the necessary people in a room and where you tried to figure out how to help an inmate. While it is done for some inmates in High Security, he said everyone (with serious disciplinary problems) should have their own plan and that it would require staff from mental health, security, and counselors, although he admitted that currently there were not enough staff to do that. Without it, he said, their behavior can be so detrimental to themselves and to correctional officers. He referred to the team-building problem-solving exercises—―forming, storming, norming, and performing12‖— that might be helpful in addressing special management inmates.

Several staff pointed to the more practical aspects of creating programs, such as keeping prisoners busy and giving them the skills that might help translate into employment on the street. One lieutenant acknowledged that he was ―not a big believer in programs‖ but thought they were important, ―but just so they can take part in something.‖ He explained he did not think programs could rehabilitate prisoners but that they do keep them constructively busy. One deputy would like to see the department offer more vocational training programs and have prisoners work in jobs at fair wages while in prison so they could also begin to pay their room and board. He explained, ―I would like inmates to be responsible for their actions, for what they did on the street. They have the financial responsibility to pay back to taxpayers in addition to paying back their victims.‖ He said that prisoners now earn from fifty cents to three dollars a day working, but that if they could be tied into local industries, they could get paid well by the hour, get a going wage, pay

12 This team-building approach was developed by Dr. Bruce Tuckman. Tuckman, B. (1965). Developmental sequence in small groups. Psychological Bulletin, 63, 384-399.

271 room and board, and learn to manage their money. As he concluded, ―Get them in sync so when they hit the street, they know what to expect.‖ Another deputy agreed that the prison system needed to offer ―More vocational courses, carpentry and welding, the types of jobs that will be readily available to them.‖ He said, ―They‘re not going to get jobs in banks or insurance, because they don‘t have their integrity.‖ He also believed that the system should be using inmate labor to repair the prison. He said that inmate labor could also be used to help them pay for their incarceration. He lamented that there is no longer a farm on the property, saying the state had gotten away from it. He believed the cost of prisons had gone way up because the department was spending money on vendors and thought that instead, inmate labor should be used to offset the cost of incarceration.

Prisoners also pointed to programs and increased programmatic activities and education as solutions for how to improve coping skills and decrease serious prison misconduct. Many of them wanted more constructive activities to decrease the time they are idle and bored and to take their mind off their worries. As one prisoner in Moran said,

―Give people more options, programs, classes…and programs all year long….The (waiting) lists are so long to get into programming.‖

Three prisoners in HSC said there was especially a need for constructive activities to break up the monotony of isolation and help them cope with the stress of segregation. A few prisoners wanted more recreation, one saying, ―We only get one hour of rec a day. If there was a basketball game (or some other kind of activity), everyone could look forward to it.‖ Another HSC prisoner said there needed to be ―more activities that people can do to take their minds off of stressful things.‖ In addition to being a diversion, yet another wanted to see more programs for rehabilitation purposes in HSC. He said, ―Instead of

272 everything being a punishment, they could try to rehabilitate people, bring in more programs. That would help us out a lot.‖

A number of prisoners from HSC admitted that when they were residing in

Maximum they and other prisoners did not get involved in programs or that when they did, they often did not take them seriously. One HSC prisoner suggested that the prison run a

―Scared Straight (sort of) program when you first come into the prison; not run by the COs but by some people like me who have been in for a long time. It would teach them how to move through the prison system; how they needed to do programs and what parole expects of them.‖ He believed younger prisoners coming in might be more willing to listen to inmates who had similarly gotten in trouble and learned the ―hard way‖ how to do time. An older Maximum prisoner had a similar idea, ―start a young kids group; keep young guys in sessions…Motivate them to do constructive things in prison. Give them nutrition classes and mental health classes; they still have negative energy out in the yard.‖ An HSC prisoner admitted that he had not been motivated to get into programs when he first got to

Maximum. ―Now I see a lot that they‘re worth it. I was thinking about myself instead of my family.‖ He said that since he had come to HSC that he had come ―a long way, all the things I was doing, if I had stayed where I was at, I would have ended getting state- charged.‖ He added ―good thing I came over here.‖ He continued to explain that prisoners often are not motivated and will not change ―unless they want to change their life, like me, I want to better myself. I don‘t want this to be my career.‖

Prisoners interviewed at Maximum also wanted to see more programs and more constructive activities. One said he played as many sports as he could, but he wanted more sports and more games, ―something to do to help me stop thinking on what is going on.‖

273

Another older prisoner said he wished the system ―would allow inmates to spend more time outside of their cell; if they‘re not working, they spend most of their day in their cell.‖

Another former HSC prisoner now in Maximum said he wanted to be able to pursue interests to keep himself busy. He said, ―give me some programs; let me learn. I‘ll pay for my own tapes, CDs; let me get it. If I want to learn Chinese or Spanish, why not let me? If

I want to learn about business, why not?‖ He complained that the barbershop was the only real program that mattered at Maximum; it was the only one that prisoners could learn from and use in real life once they left the prison.

Several prisoners would like to see increased programs and classes that could help prisoners avoid drug use when released and help prisoners secure employment. One prisoner in medium said prisoners needed programs that would ―Help people when they get out. It seems that everybody reoffends. They keep coming back. (There needs to be programs that) help people from coming back. There are no halfway houses. No job training. I hear about so many people who say they don‘t want to use drugs, but they‘re so exposed to it.‖ Another agreed saying there was a ―Need for more programs to deal with addiction and transitional housing to deal with your drug problems. Offer more (options) for a (vocational) trade. Send me somewhere where I can get into a trade. There‘s not enough here.‖ A third prisoner felt that in order to succeed on the street and in prison there needed to be programs that taught prisoners how to communicate. He would like to see

―more programs in terms of communication, how to express yourself in a non-violent manner. Teach (us) how to talk to someone, like for a job interview. The majority of the population is from low-income (neighborhoods) and lacking in that area. They don‘t see why they shouldn‘t react. They say ‗this was the way I was brought up.‘ They don‘t

274 understand that (the way they talk is) like ‗putting gas on the fire.‘ They don‘t know the type of communication skills they should be using.‖

Other suggestions about programs from prisoners included creating programs for people doing life, since ―they have problems too;‖ creating parenting and special visiting programs for fathers to interact with their children at Maximum; classes to learn stress management techniques; and constructive activities like board games, creative writing groups, or even presentations by inmates who might know something about a topic like history. A number of prisoners complained that actually getting into a program once one signed up for it was very difficult. They said waits of three to eight months were typical.

One complained that he would see other people who signed up after him get in a program before he did. He felt it was not fair and that it caused ―people to get ripped off for good time.‖ Staff explained that prisoners who are closer to release are given priority for programming over long-term prisoners and admitted that the policy often caused confusion and resentment among the prisoners.

MORE STAFF TO PROBLEM-SOLVE WITH PRISONERS Next to the need for programs and constructive activities, the most popular solution for combating serious misconduct and enabling coping was more staff to work directly with prisoners to address their problems. Almost half of the prisoners and one-third of staff said more staff were needed to help prisoners learn to cope and a third of staff similarly believed that more staff were needed to help reduce misconduct. There was both the need to help motivated inmates resolve their problems and to identify and prod other inmates into programming. As one correctional counselor noted ―someone can come to jail and not work and not come to class and skate through the system.‖ He said the system was so big

275 and ―that there were not enough staff to monitor‖ each inmate to ensure he got what he needed to improve his chances of not recidivating. To reduce prison misconduct another correctional counselor suggested that the department use a form of triage. He suggested putting ―away those who are the worse;‖ having ―fatherly conversations or lectures with the least problematic;‖ and ―putting your resources into those in the middle.‖ An officer at

HSC said, ―We need more social workers and psychiatrists. We‘re going to put them in the street and they‘ll come back because we don‘t deal with their issues. We need to break the cycle… We pass them on instead of trying to resolve the problem; we put them on meds.

People do something for a reason and we should deal with it. More social workers are needed at the beginning of their bit and not now when they‘re acting out.‖

Prisoners acknowledged that there just were not enough staff to address all of the issues that prisoners had or enough to help them in a more in-depth manner. The majority of prisoners wanted more correctional counselors or social workers to help inmates deal with problems. Several said that correctional counselors had so much to address, they often did not have a chance to address one‘s problem until it was too late or after the prisoner had already dealt with it on his own. A few of the prisoners in HSC wanted to see a correctional counselor or social worker and believed that there needed to be somebody who kept track of them, and to help advocate for their movement to lower security or out of state. Finally some said that staff needed to find out why prisoners are involved in serious misconduct. As one prisoner articulated, ―ask (prisoners) what‘s wrong and why. Nobody seems to ask ‗why.‘ Everybody does something wrong for a reason… They don‘t ask us why (we‘re) doing it.‖ He noted that another inmate had tried to commit suicide and that staff had never asked him why he had done it—only whether he planned on doing it again.

276

Another prisoner concluded, ―For me, we‘ve been focusing on the consequences part. We need to focus on why it is, why we get in trouble.‖

CORRECTIONAL OFFICER TREATMENT OF PRISONERS When asked what the DOC should do to reduce serious prison misconduct, almost half of the prisoners and one-fourth of the staff said that there needed to be an improvement in how some correctional officers spoke to prisoners. Several staff said that there needed to be some training incorporated into the training academy for both new and current correctional officers to ―teach CO‘s how to talk to inmates. One can still be forceful and follow rules and regulations… But it would cut down tremendously on the frustrations of inmates.‖ One CO at HSC said there needed to be ―some programs…Some kind of training for staff to address that male ego issue.‖ He attributed the ―pig-headedness on both sides to male ego shit‖ but did note that it was an issue with some female correctional officers as well. A few staff mentioned the golden rule—one correctional counselor saying that staff needed to learn better social skills and the rule of ―treat me as you want to be treated.‖ One lieutenant at Maximum said he had learned to keep his cool in most situations. He said,

―You‘ve got to have the ability to override the (urge) to act out. I do it all the time.

Inmates call me names, I‘m over that.‖ Another lieutenant explained that it helped so much to just listen to inmates, ―Just to talk to the guys and listen to them a little. You don‘t have to give them feedback, just let them vent. It might or might not be a legitimate gripe.

Sometimes I explain why things can‘t be the way they want.‖ He said that often helped calm inmates down. A few staff also noted that the system should take care to hire more people who ―really care,‖ and are willing to act more professionally.

277

Many of the prisoners brought up some of the same points mentioned by staff.

When asked how the system could help prisoners cope with problems, one said, ―Don‘t add to them (the problems). (COs) talk a lot of crap to you. If they don‘t like someone, they‘ll keep on talking to them until they flip out.‖ He called it verbal abuse but clarified that ―not every CO is like that but the majority of them are.‖ Other typical comments included,

―Stop treating us badly;‖ ―Be respectful to me, I‘ll be respectful to you;‖ and ―Talk more with inmates, not talk at them. The atmosphere here is ‗do what I say‘ and ‗I don‘t care what you have to say.‘‖ One prisoner recommended that ―Instead of confrontation and antagonism, there should be a bridge of communication where inmates and staff aren‘t hostile to one another.‖

There were a few prisoners who said that staff had the discretion to be more lenient in the enforcement of rules, but always chose instead to book inmates. One prisoner in

HSC said that officers there rarely use the two-nighter option and instead usually booked inmates right away. He recommended segregation be used only for those prisoners involved in serious infractions, not for petty infractions, which could be dealt with using the one- or two-nighter options. Another prisoner at Maximum said he thought officers could benefit from cultural awareness training. He explained, ―There are many young inmates here who are Hispanic and black from different cultures. Sometimes they have a different swagger and attitude because they come from a different environment. (But officers see this and) they‘ll say ―okay are you a tough guy?‖ and go after that guy. ―They don‘t understand the culture.‖

278

OTHER SUGGESTIONS FOR REDUCING MISCONDUCT There were a few other suggestions for reducing misconduct among the population including handing out rule books to prisoners (―people getting booked for charges that they don‘t know about‖); being more consistent in rule enforcement from shift to shift; adding more phones per block and yard (―highly stressful area that causes unnecessary arguments and fights‖) and in the HSC ensuring that prisoners are supplied with basic needs (soap, boxers, t-shirts, toilet paper). One other suggestion mentioned by a few prisoners was to take prisoners‘ concerns about their safety more seriously. One spoke about gangs being housed together. He said, ―Instead of putting all gangs in the same block—there are Bloods and Crips right next to each other in the same block and there have been brawls of people fighting—you need to separate gangs from each other.‖ He said that when a prisoner has lots of general enemies in a prison and asks to be moved out of state, he should be taken more seriously. Another prisoner agreed saying, ―Truthfully, I think most comes from the streets. The DOC doesn‘t take seriously the enemy situations. They will put someone back in population until something happens. Then to come out of protective custody status, you have to sign a paper but that piece of paper isn‘t going to stop you from doing anything.‖

Prisoners said that there needed to be a better system to protect prisoners from general enemies as well as specific ones.

*****

In summary, prisoners and staff were asked to articulate those practices and policies already in place that have contributed to keeping down levels of serious misconduct and violence and without which misconduct would increase. Staff, more than prisoners, were able to articulate more practices and policies that were already proving to be successful.

279

The top five categories of beneficial policies and practices included: vigilance of staff/direct supervision of prisoners; the disciplinary and good-time policies; programs; classification and the SRG process; and the attempt of staff, especially at Moran, to address prisoners‘ issues and communicate in a positive manner. About three-fifths of the prisoners had been involved in some kind of program, usually after having been booked, sanctioned, and going into segregation. Most of the prisoners gave positive feedback about their program involvement, some giving glowing accounts of programs that helped them cope and change their outlook on their incarceration.

When asked for solutions, most staff and prisoners were able to think of concrete solutions to help reduce serious prison misconduct and violence, although several prisoners said that the prison had many policies and practices in place and that it was now up to the prisoners themselves. Most of the prisoners and a little more than half of the staff thought that there was more the department could do to help prisoners cope better. The other half of the staff questioned the department‘s ability to help more and the cost, many concluding that enough was already being done. Interestingly, the ideas for reducing prison misconduct were basically the same as the ideas for increasing prisoners‘ positive coping skills. The top three were: increasing programming and constructive activities for prisoners; increasing staff to help prisoners solve their problems; and decreasing the disrespect that staff directed at prisoners.

Almost all of the staff and prisoners interviewed believed there needed to be more programming, more slots of current programming, more education, and more constructive activities for prisoners. They believed that cognitive behavioral programming would increase prisoners‘ coping abilities, and lessen idle time, both contributing to a decrease in

280 serious misconduct. They also believed that there needed to be more treatment and constructive activities for prisoners with mental health problems, and that more staff were needed to work directly with prisoners to help them solve their problems in an appropriate manner. Both staff and prisoners believed more staff would help prisoners cope better and staff believed it would reduce misconduct because it would help resolve issues, prod inmates into constructive activities, and provide positive interaction that would reduce conflict. Finally, almost all of the prisoners and about one-fourth of the staff believed that there needed to be a change in the way that many correctional staff addressed prisoners.

They said that the disrespect shown by some staff to prisoners contributed to prisoners‘ frustrations, often leading to misconduct. The staff members who brought up this issue believed there needed to be more training about the reasons why staff should not disrespect prisoners and the best ways to do it while still maintaining control and authority.

281

IX. DISCUSSION

This discussion chapter is divided into three subsections: a reconceptualization of serious misbehavior and violence; a detailed discussion of each of the areas that affect misbehavior and violence; and a discussion of the limitations of the study.

A RECONCEPTUALIZATION OF SERIOUS MISBEHAVIOR AND VIOLENCE

The focal research question of this study was whether prisoners‘ ways of coping affected their likelihood of being involved in serious prison misconduct and violence and if so, which aspects of coping were more likely to exert influence. Another underlying assumption of the research was that the negative trait emotions of anger, anxiety, and depression affect one‘s ultimate behavior in a stressful situation. It was assumed that prisoners who easily become angry, depressed, and anxious may be more likely to engage in behaviors that are negative such as serious prison misconduct and violence. The study also explored the relationship between coping skills and these negative trait emotions. In addition, the conceptual model acknowledged that there are other predictors that affect serious prison misconduct and violence. These predictors, all measured at the individual level, include those found in previous research to have a significant effect on prison misconduct and others that have not been previously studied. The current study controlled these predictors but also studied their relationship to ways of coping and the trait emotions.

Therefore the research tested whether one‘s coping skills and associated emotions interacted with or mediated any of those other individual predictors.

282

MODELING SERIOUS PRISON MISBEHAVIOR USING THE QUANTITATIVE RESULTS The negative binomial analysis mostly confirmed the first two hypotheses of the current research. Both Models 1 (annual misconduct score) and 2 (annual violent dreports), which included only the eight coping scales, were statistically significant, indicating that they were significantly different from the null model with the intercept alone. Seven of the eight coping scale parameters in Model 1 were statistically significant, though not all in the anticipated direction. Specifically, it was anticipated that active coping would be associated with lowering serious prison misconduct, but instead it was associated with a greater level of misconduct. In contrast, it was anticipated that disengagement would be associated with higher levels of serious disciplinary behavior, but instead it was associated with lower levels of misconduct. Only four of the coping scales‘ parameters (get support, humor, acceptance, and active coping) were statistically significant with the dependent variable of violent behavior, with active coping again not in the anticipated direction.

The negative binomial analyses in Models 3 (annual misconduct score) and 5

(violence alone) with only the three trait emotions revealed mixed results regarding the third hypothesis—that prisoners with stronger levels of negative trait emotions will be involved in greater levels of serious misconduct. Two of the trait emotions, anger and depression, were associated with greater levels of serious prison misconduct as measured by the annual misconduct score. However, the third trait emotion, anxiety, was associated with lesser levels of serious prison misconduct. Only one of the trait emotions, anger, was associated with greater prison violence as measured by the annual rate of violent dreports.

Next individual-level predictors were found to be associated with serious prison misconduct and violence (Models 7 and 10). First, the personal predictors that were found

283 to result in greater prison misbehavior and violence included: being young; not Hispanic; being single; prior drug use; prior psychiatric treatment; adjudication as a delinquent; street gang affiliation; prior incarceration; B&E/drug offense; not a sex offense; and having a shorter sentence. Again, the results varied by the measure of the misconduct and each of these individual predictors will be discussed in more detail in the next section.

Diagram 2 below shows the actual results of the final models of this research. That is, it includes results from Model 9 (predicting the annual misconduct score) and Model 12

(predicting violence only). These full models included the predictors, the ways of coping, and the trait emotions. The fourth hypothesis assumed that ways of coping would mediate the relationship between the personal predictors and serious misconduct and violence. As mentioned previously, for mediation to occur there would need to be a statistically significant relationship between the personal predictors and misconduct, between the personal predictors and ways of coping, between ways of coping and misconduct, and in the full model, the relationship between the personal predictors and misconduct would be decreased if there was partial mediation or disappear if there was full mediation (Baron &

Kenny, 1986). For the most part, this did not occur. Instead the statistically significant relationship between several ways of coping and misconduct disappeared with the introduction of the personal predictors. However, five ways of coping remained statistically significant in at least one of the two final models after the addition of the personal predictors and trait emotions. Specifically getting emotional and instrumental support, venting one‘s emotions, and reliance on religion were predictors of the annual misconduct score. Getting support, use of humor, and active coping were predictors of violence in the final model. Thus, prisoners who used both emotional and instrumental

284 support when faced with hardships were less likely to be involved in serious misconduct and violence. On the other hand, prisoners who regularly vented their emotions upon facing stressors were more likely to be involved in serious misconduct in general but not specifically in violent misconduct. Curiously, reliance on religion as a way of coping was also associated with higher levels of serious misconduct in the final model. There had been no relationship between religion as a way of coping in either the bivariate analysis or in the coping and trait models alone. However, the inclusion of the personal predictors appears to have had a suppressor effect on religion so that it became positively associated with misconduct once the model controlled for the personal predictors. Specifically the correlational analysis had shown that reliance on religion was associated with prisoners who were more likely to be older, black, and not single. Thus the introduction of these predictors unmasked a positive relationship between reliance on religion and the annual

Diagram 2 – Final Models Predicting Serious Prison Misconduct and Violence

COPING SKILLS Lack of support; venting emotions; use of humor; active coping, reliance on religion

TRAIT EMOTIONS Anxiety; lack of anger

SERIOUS PRISON MISCONDUCT AND VIOLENCE PERSONAL PREDICTORS Young; not Hispanic; single; prior drug use; prior psych treatment; prior incarceration; B&E/drug offense; not a sex offense

285 misconduct score; prisoners who relied on religion as a way of coping were more likely to have higher annual misconduct scores. Active coping, which surprisingly was found to be positively associated with serious misconduct and violence in earlier models, retained its relationship only with violent misconduct when the personal predictors and trait emotions were added into the final model. Lastly prisoners who used humor as a way of coping were more likely to be involved in violence according to the final model.

Three ways of coping, disengagement, denial, and acceptance lost their statistical significance in the full model with the addition of the personal predictors. In the coping only models, prisoners who disengaged were less likely to be involved in serious misconduct, while prisoners who used denial were more likely to be involved in serious misconduct. However, given their relatively weak correlations with misconduct, it is not too surprising that both lost their statistical significance with the addition of the personal predictors. However, it is more surprising that the strongest way of coping in both coping only models, acceptance, also lost its statistical significance with the introduction of the personal predictors. Needless to say, one can safely conclude from this research that at least five of the eight prisoners‘ ways of coping do matter to either serious misconduct in general and/or specifically violence.

The bivariate analyses found several of the prisoners‘ activities in prison to be predictive of serious misbehavior and violence. Specifically, prisoner activities that were found to differentiate between prisoners in the low-level vs. high-level disciplinary groupings included: idleness; not working; engaging in mostly prisoner-initiated activities; not engaging in prison-initiated activities; and prior prison misbehavior. Unfortunately due to missing data issues, none of these predictors were included in the multivariate analyses.

286

As mentioned in the findings chapter, there was a lot of missing data in the four ―activities‖ variables, not because prisoners did not answer the question but because they did not provide the actual number of hours associated with activities, instead providing qualitative answers like ―a lot.‖ Prior misbehavior could not be used in the multivariate analysis not because of missing data (there was actually no missing data) but because 85 of the prisoners had not been incarcerated during the two years for which prior misbehavior data was collected. Since prisoners who had shorter sentences were much more likely to be involved in misbehavior, excluding 85 of them in the multivariate analysis because they could not possibly have this data would have skewed the results.

Finally, multivariate analysis explored the relationships between prison hardships, fear of victimization, and victimization with serious prison misconduct and violence. A number of prison hardships were positively associated with serious prison misconduct including: missing family, freedom, and possessions; conflicts with staff; boredom; cleanliness and overcrowded conditions; and concerns about safety. Fear of victimization and victimization by staff, but not victimization by other prisoners was found to be correlated to serious misbehavior. However, one must view these as tentative findings because no personal characteristics or ways of coping were included in the analysis.

MODELING MISCONDUCT USING THE QUANTITATIVE AND QUALITATIVE RESULTS The qualitative data did confirm the results of the quantitative analyses, amazingly in almost every facet. This confirmation of the quantitative data using the qualitative data has been woven throughout previous findings sections and is reiterated in the subsequent section that summarizes and discusses each of the predictors of serious misconduct and violence. The qualitative results also helped to explain a number of the quantitative

287 findings. For example, though quantitative analyses of prison hardships included conflicts with prisoners, interviews revealed that several specific issues were especially likely to lead to prisoner conflict, including: gambling and owing gambling debts; being strong-armed; being disrespected; and general peer pressure among youthful offenders. Table 45 below lists the various groupings of findings from both the qualitative and quantitative results.

Several variable groups were reconfigured and new groups were added to delineate a better understanding of the predictors. Ways of coping and the trait emotions were consolidated into a grouping labeled ―personal traits‖ as was mental illness and lack of impulse control, factors brought up by many staff. Several predictors were added to the personal predictor category (personal upbringing; being designated to protective custody status due to one‘s offense or having been a witness; and being a weak prisoner) as prisoners and staff had discussed their roles in contributing to prison misconduct. One group, prisoner hardships was divided into two separate groups (prisoner hardships and staff skills discussed below). Similarly another group, personal predictors, was also divided into two groups (personal predictors and prisoner activities). While the original conceptual model treated the personal predictors as one group, they are divided into two distinct groups to delineate those predictors that predate their incarceration (personal predictors) and those that have to do with the actual activities of prisoners since they have been incarcerated

(prisoner activities).

In the planning of this research, the prison hardships grouping included how difficult prisoners found various hardships, their level of fear, and their actual physical and verbal victimization by staff. However it was clear from the interviews with prisoners and staff that there was probably a lot more going on with the conflicts with staff and the

288 perceived and reported victimization by staff so it was removed to its own grouping.

Indeed Bottoms incorporated a similar factor in his model of good behavior, labeling it as

―staff deployment, approaches, and skills‖ (1999, p. 258). While it has been labeled here simply as ―staff skills,‖ it does incorporate many of the same facets of Bottoms‘ concept including staff‘s way of communicating with prisoners, their attention to helping prisoners solve problems, their perceived fairness, their ability to perceive tension and address it, their level of consistency, their skill at direct supervision, and their vigilance, all having been linked by staff and prisoners to the lowering of disciplinary issues.

Table 45 - Predictors of Misconduct Found in Quantitative and/or Qualitative Results Reasons Predictors Found in Predictors Found Only in Quantitative Analysis*(-) Qualitative Data Prison Hardships Fear* Gambling (Models with just prison Missing family* Strong-arming hardships and misconduct) Missing freedom* Disrespect Missing possessions Peer pressure Boredom* Unclean & crowded conditions Personal Traits Getting support from others(-)* Mental health problems (Final Models 9 and 12) Venting emotions* Lack of impulse control Active coping* Acceptance Humor* Anger(-)* Anxiety* Personal Predictors Young* Prisoners‘ upbringing (Final Models 9 and 12) Not Hispanic Weak Single/not divorced Protective custody Prior drug use Adjudicated delinquent Prior psychiatric treatment* Prior incarceration B&E/drug offense Not a sex offense Prisoner Activities Idleness Program involvement (-) (Bivariate analysis) Not working* Gang involvement In prisoner-initiated activities Prior prison misconduct* Staff Skills Threatened by staff* Quality of interaction (-) (Models with just victimiza- Insulted by staff* Vigilance/direct supervision(-) tion and misconduct) Injured by staff* Problem-solving (-)

* Also found in qualitative results; (-) denotes a negative relationship

289

In both the bivariate and multivariate analyses, concern for one‘s safety and the fear of victimization scale were both directly correlated to serious misbehavior. Interviews with prisoners confirmed this relationship. Although fear is listed as one of several prison hardships, it is central to many of the other group factors and indeed, prisoners and staff discussed at great length their interrelationship with fear and misconduct. Specifically, personal traits, staff skills, personal predictors, and some of the prison hardships were all thought to influence fear, which in turn sometimes led to misconduct. For example, prisoners‘ personal characteristics such as being in a street gang or being known for one‘s sex offenses or other heinous crime could lead to fear. Prison hardships such as being disrespected, being strong-armed, owing a gambling debt, and general peer pressure were all itemized by staff and prisoners as leading to fighting and other types of misconduct.

This is especially interesting since the hardship, conflicts with prisoners, yielded mixed results in the bivariate and multivariate analyses. In the former, prisoners in the high-level disciplinary group were much more likely than those in the low-level group to rate conflicts with prisoners as being very difficult. However, in the multivariate analysis it was not a statistically significant predictor of misbehavior. Finally, in the interviews staff skills were seen to affect whether prisoners‘ fears resulted in misconduct. Staff can help prisoners solve problems related to other staff or prisoners; can take their concerns about personal safety seriously; can use their communications skills to listen; can calm prisoners who are afraid; and can minimize prisoner conflicts through direct supervision and vigilance.

TOWARD A PREVENTIVE MODEL ADDRESSING SERIOUS MISBEHAVIOR AND VIOLENCE Much of the research on prison violence and misconduct had been conducted amid the larger debate about prison adjustment with researchers adopting either the ―importation‖

290 or ―deprivation‖ model as a framework within which to understand prison misconduct. In many of the studies, prison misconduct was used as one indicator of prison adjustment, along with other variables such as trait emotions (anger, anxiety, and depression), and suicide attempts. The ―get-tough‖ era begun in the 1980s and leading to mass incarceration shifted the focus slightly to overcrowding as a predictor of misconduct and violence but still within the importation/deprivation debate. In the mid-1980s Porporino and Zamble (1984) and Gaes and McGuire (1985) both underscored the fact that serious prison misbehavior and violence could not possibly be caused by one of these sets of factors to the exclusion of the other. Thus research has become more integrative, incorporating individual-level predictors, as well as prison-level factors (e.g. overall make-up of staff or of the prison population) sometimes even incorporating situational or transactional factors. But one cannot help but question the underlying focus of the prison misconduct research in the last ten years. Many of the researchers have focused on one predictor or set of predictors

(similar to this study‘s focus on coping). However, in the analysis it has not been uncommon to give equal weight to whatever variable yielded statistical significance. It has also not been uncommon to see results that find prior criminal history to be one of the strongest predictors of misconduct. It is indeed a fine variable as a predictor, possibly to be included in a classification or risk-assessment instrument or as a control variable in other analyses. But other than that, it does not provide any relevant information to help a prison administrator reduce the level of prison misconduct.

Although the importation/deprivation debate is almost dead (though one can still find researchers arguing the merits of one vs. the other in their studies of prison violence) it did provide researchers with an enticing focus. It is proposed here that the focus of further

291 research on prison misconduct and violence be redirected toward its prevention in prisons.

There is no need here to rehash all of the reasons, as was discussed in the Introduction, that prison misconduct and especially violence is bad. Admittedly, there have been pockets of excellent research (outlined in the Literature Review chapter) on specific problematic areas like the proliferation of prison gangs and its effect on misconduct and the maladjustment of those mentally ill prisoners who are both ―mad and bad.‖ Many researchers‘ attention has rightly been turned to studying the problems of supermax prisons and more recently their populations. Several other researchers have suggested incorporating dynamic inmate variables into prison misconduct research (Harer & Langan, 2001; and Camp, et al., 2003).

Indeed this suggestion was one of the impetuses that led to the current research as one‘s coping skills could be improved through programming and counseling. However, in addition to including dynamic predictors Bottoms (1999) has exhorted researchers to examine the prisoners who misbehave, the prisoner environment, and the interaction of the two, especially paying attention to how prisoners interpret their environment and then in turn, shape and transform it. He has called for a more subtle approach that examines the daily minutiae of prison and how the various aspects may contribute to misconduct. One contribution of his research is his ―speculative and interactive model‖ of how prisons maintain good order. Thus research on prison misconduct and violence can yield richer results than being only an indicator of prison adjustment. It can provide prison administrators with detailed findings on various groupings of predictive factors so that they might use the information to lower serious prison misconduct and perhaps the need to resort to supermax prisons as a solution to misconduct.

292

Thus an attempt will be made to further categorize the types of variables incorporated in this study of misconduct, with an eye toward grouping variables by their preventative worth. This would make the discussion of this and other studies more meaningful and help researchers think through the implications for further research, but more importantly, the implications for policy. Diagram 3 below reconfigures the findings of this study; it includes those groupings that were found to be predictive of serious misconduct and violence in both the quantitative and qualitative results. The predictors included in the model have been divided into two categories: static and dynamic predictors.

Predictors that are static are those that are set prior to a person‘s incarceration that cannot be changed. While age itself is certainly not static, the age of a prisoner at the time of a serious disciplinary incident cannot be changed. It is also static in that prison staff cannot cause changes in it though they can target programs to that age group. Another example is prior incarceration, which is a characteristic of somebody entering the prison system that cannot be changed. The static personal predictor category has itself been divided into two sub-categories: universal and local subculture predictors. Although ―universal‖ does not mean that every study has found these or similar factors to be predictors of serious misconduct or violence, it means that these factors have been found to be predictive despite the location of the prison and the local subculture from which its population is derived.

That is, age and prior incarceration have been found to be strong predictors of prison misbehavior regardless of the local subculture. Those variables that come under the ―local subculture‖ sub-category are those variables that in some respect reflect the subculture(s) from which offenders came. It is not surprising that there are completely mixed results in studies of race/ethnicity and misconduct if one makes the assumption that they have more

293 to do with the demographic make-up of offenders in any given state or area of a state, than they do with misconduct per se. For example, in this study Hispanics were less likely to be involved in serious misconduct and violence than either whites or blacks; however, in another study, the opposite has been found (Berg & DeLisi, 2006). Similarly, in this and another study (Cunningham & Sorensen, 2007 ) those with shorter sentences for offenses like B&E or drugs were more likely to have serious disciplinary problems while other researchers have found such problems with prisoners serving longer sentences for violent offenses (Kuanliang, et al., 2008).

Diagram 3 - Static vs. Dynamic Predictors of Serious Misconduct and Violence

DYNAMIC PERSONAL PREDICTORS DYNAMIC PREDICTORS

STAFF SKILLS

PRISONER ACTIVITIES SERIOUS PRISON MISCONDUCT AND VIOLENCE PRISON HARDSHIPS & FEAR

STATIC PREDICTORS STATIC PERSONAL PREDICTORS UNIVERSAL PREDICTORS & LOCAL SUBCULTURAL PREDICTORS

294

The static personal predictors, whether universal or associated with the local subculture, are useful mostly as predictors in risk assessment scores, or other types of classification tools.

The local subculture predictors can help prison staff identity those groups of offenders that might need extra attention whether it be increased monitoring, more direct supervision, or directed programming. For example, those offenders incarcerated for B&E or drug offenses might be assumed to be involved with drugs and targeted for further assessment to determine whether participation in substance abuse services is warranted. Finally, researchers may create a static predictor index that could both account for the variance in the misconduct associated with these static personal predictor variables, but also allow the researcher to use the additional statistical power saved by an index for exploration of the other more important dynamic variables in the study. In addition to the static personal predictors, another set of static predictors, not included in this study, would be the physical characteristics of a prison itself that cannot be changed (type of prison, type of living quarters, programmatic space).

Unlike static predictors, dynamic predictors can be affected by prison staff and prison policies. Some of them, like staff skills can be directly changed by the prison; while others, like the mental health of a prisoner can only be indirectly affected by prison staff and policies. For example, staff could be trained in the identification of prisoners with mental health problems and learn the best ways to de-escalate rather than escalate situations where mentally ill inmates are stressed. However, staff cannot change the actual mental health of prisoners but can change policies toward ones that help the mentally ill cope and away from ones that exacerbate their problems. Dynamic factor groupings include prison hardships, prisoner activities, and staff skills. The ―dynamic personal predictors‖ grouping includes the ways of coping that affected serious prison misconduct and violence, the trait emotions, and the predictors discussed by staff

295 and prisoners in the interviews: prisoners with mental illness and those with low impulse control.

These are all possibly personal traits that can be ignored thus resulting in greater levels of misconduct or they can be addressed to hopefully decrease misconduct.

Finally one can see all of the interconnections between these groupings of predictors. As noted previously, fear, is related to many of these group factors. For example, somebody who is young or who was in a street gang might be more fearful; those who are strong-armed might be more fearful; those who are mentally ill or have poor coping skills might be more fearful; prisons or areas of a prison with no direct supervision by staff might cause fear; and in turn, fear of victimization might cause misbehavior and violence. Similarly, several of the group factors

(prison hardships and fear, static personal predictors, and even other dynamic factors) affect prisoners‘ trait emotions such as anger and depression and their ways of coping, which are directly related to serious misconduct and violence. Having gang members in a prison system does not in and of itself cause violence. It can be mitigated by staff‘s skills at detection, proper assessment, direct supervision, classification as well as activities that redirect the energy of youth towards more pro-social activities like work and programs. In addition, staff could assess local subculture personal predictors of prisoners to determine their likelihood of serious misconduct— namely prior placement in a juvenile correctional facility, prior incarceration, and substantial drug abuse, as well as dynamic trait predictors such as evidence of mental health problems.

*****

All in all, not all predictors of serious prison misconduct and violence are created alike.

Some are static predictors, good for prediction and risk assessment and to use as control variables in research on other predictors. Local subculture predictors can help identify prisoners

296 that may be particularly likely to be involved in misconduct. Dynamic predictors are abundant.

The great news is that prisons can manipulate policies and practices and train staff to address the multitude of dynamic factors of which they have some control. The difficult part is for each prison to assess where they want to spend their valuable and scarce resources of time and money.

DISCUSSIONS AND CONCLUSIONS BY STATIC AND DYNAMIC GROUPINGS

This section summarizes the findings of each of the groupings of predictors and discusses their interconnectedness to other factors. It begins with the static predictors in this study, which are all personal predictors. This is followed by four sets of dynamic predictors: dynamic personal predictors; prison hardships; prisoner activities; and staff skills.

STATIC PERSONAL PREDICTORS STATIC PERSONAL PREDICTORS Static predictors are those predictors of serious misconduct or violence over which prison administrators have no control and which are unlikely to be changed. This category is divided into two sub-categories: universal and local subcultural predictors. Universal predictors have been found in many or even most studies to be predictive despite the location of the prison and the local subculture from which its population is derived. Whereas local subcultural predictors are those that might be associated with a location‘s subculture and thus might vary from state to state or even vary by locations within a state.

Universal Predictors

The current study included a number of variables that one might categorize as universal predictors including: age, marital status, education, previous employment, prior adjudication and

297 placement in a juvenile correctional facility, and prior adult incarceration. There was no relationship between serious misconduct and education or previous employment.

AGE

By far the strongest predictor in the bivariate and multivariate analysis of misconduct was age. As mentioned previously, prior research on age has found it to be one of the strongest predictors of serious disciplinary behavior in prison (Schnur, 1949; Zink, 1958; Wolfgang, 1961;

Ellis, et al., 1974; Flanagan, 1983; Ekland-Olson, et al., 1983; Toch, et al., 1989; Anson &

Hancock, 1992; Gendreau, et al., 1997; Lovell, et al., 2000; Harer & Langan, 2001; Wooldredge, et al., 2001; Gaes, et al., 2002; Camp, et al., 2003; Peck, Jr., 2004; Griffin & Hepburn, 2006;

Cunningham & Sorensen, 2007; Steiner & Wooldredge, 2008; Kuanliang, et al., 2008). It was the only personal characteristic that consistently remained in all of the models for both the annual misconduct score and for violence that included prisoner characteristics. The qualitative data also bore out the relationship of age to misbehavior.

Initially it was thought that young prisoners were more likely to be involved in serious disciplinary issues because they had poorer coping skills and were more likely to be involved in gangs. However, age remained a strong predictor even with the introduction of the coping and gang variables leading one to believe that other factors associated with their lack of socio- biological development (immaturity, lack of self-control, impulsivity, faulty decision-making) or their prison environment (peer pressure, campaign for respect) are responsible for misconduct.

Many staff and prisoners pointed to peer pressure and the need to be able to protect one‘s self as issues that were important to young prisoners. Concomitant with these issues was the campaign for respect among the young. Most of the young prisoners believed that one could not

298 walk away from situations where they had been disrespected, because the disrespect would continue and escalate. Cesaroni and Alvi (2010) conducted an exploratory analysis of how 350 young males negotiated their incarceration experiences through qualitative interviews and found two main themes. They found that young males needed to ―measure up to a standard of hegemonic masculinity. The importance of not showing weakness, standing up for oneself, and being tough and unemotional were all aspects of a dominant masculine performativity‖ (2010, p.

315). Hegemonic masculinity is the dominant form of masculinity, against which other categories of masculinities are measured (Connell & Messerschmidt, 2005; Abrams, Anderson-

Nathe, & Aguilar, 2008). Though few real men practice hegemonic masculinity on a daily basis, it is an ideal image that ―encourages competition and dominance and supports the subordination of women, gay men, and other, ―lesser‖ forms of masculinities‖ (Abrams, et al., 2008, pp. 24-

25). Abrams and her colleagues interviewed young males incarcerated in two juvenile correctional facilities and determined that while youths brought with them images of what it means to be a man, the institutional culture of both facilities supported hegemonic masculinities through rules, routines, interpersonal interactions, and the allowance of bullying and aggressive behavior between residents. In addition, the institutions ―supported qualities associated with hegemonic masculinities, such as power, competition, stoicism, sexism, and homophobia‖

(Abrams, et al., 2008, p. 38). The tough, unemotional exterior described in these two studies matches the use of humor and bravado used as a way of coping by the young, street-related prisoners in the current study. Cesaroni and Alvi also found that young prisoners resisted their own incarceration in two ways: through direct confrontation with correctional staff (their stand- in parental authoritative figures) and ―through symbolic and expressive acts of trivial

299 subversion.‖ Edgar and O‘Donnell‘s (1998) study of the reasons for assault included both the campaigns for status and respect among prisoners, especially those who were young.

Another completely different approach to understanding youthful misbehavior in prison is to understand those situations on the street that lead to violence among youth. Maimon and

Browning (2010) studied the prevalence of neighborhood violence hypothesizing that unstructured socializing among peers would result in a greater likelihood of violent offending and that living in a neighborhood with high collective efficacy would result in a lower likelihood of offending as a result of the existence of informal social control mechanisms. They found that unstructured socializing among peers varied by neighborhood and was reduced when collective efficacy was present. They found unstructured socializing among peers to be significantly associated with adolescents‘ violent behavior. Finally there was a ―significant cross-level interaction between neighborhood collective efficacy and individual-level unstructured socializing. This cross-level interaction suggests that as collective efficacy increases, the positive effect of unstructured socializing on violent behavior decreases‖ (Maimon, & Browning

2010, p. 468). This finding can be applied to prison settings where young prisoners have varied amounts of time to socialize with their peers in the yard, the gym, and housing units. Prisons that display collective efficacy would include the close monitoring of prisoners in these settings, along with the willingness to intervene and confront prisoners who are exploiting other prisoners or acting disturbingly in these public spaces. Indeed, the current study did find that prisoners who engaged in mostly prisoner-oriented activities were more likely to be involved in serious prison misconduct and violence compared to those who engaged in prison-oriented ones. Thus prisons and even locations within prisons might vary in their use of collective efficacy, which

300 affects the level of unstructured socializing and an inevitable spike in violence. This will be further discussed in the section on staff skills.

MARITAL STATUS

Closely associated with age, was marital status, with younger prisoners more likely than older prisoners to still be single. Marital status differentiated those prisoners in the high-level vs. low-level disciplinary group, with single prisoners more apt to be in the high-level group. While being single was a predictor of the annual misconduct score in the model of personal characteristics alone, it was not a predictor of the annual misconduct score in the final model with both coping and trait emotions. However, it was statistically significant in the final model predicting violence alone. Research findings on the relationship between marital status and serious disciplinary behavior have been mixed (Toch, et al., 1989; Wooldredge, et al., 2001).

Being single was not just a proxy for age but remained a predictor of serious misconduct on its own in the final model. While being married might make offenders more reluctant to be involved in misconduct for fear of losing their visitation with their wife and children, it might also be that prisoners who had remained single do so because they are too immature and possibly too selfish to commit to a serious relationship.

CRIMINAL HISTORY

Next to age, prior incarceration was the next strongest personal predictor, similar to findings in most other studies on the factors associated with misconduct and violence. Two- thirds of those experiencing serious disciplinary problems had been previously imprisoned as an adult, compared to a little more than half of those without disciplinary problems. Yet there were no substantial differences between the two groups when focusing on whether the prior incarcerations were for violent or non-violent offenses. Prior incarceration was found to be a

301 predictor of misconduct in the multivariate analysis of both the annual misconduct score and of violence. In the literature, next to age, prior criminal history has been the strongest predictor of prison misconduct over decades of research (Schnur, 1949; Johnson, 1966; Toch, et al., 1989;

Gendreau, et al., 1997; Harer & Langan, 2001; Wooldredge, et al., 2001; Gaes, et al., 2002;

Camp, et al., 2003; Huebner, 2003; Griffin & Hepburn, 2006; Steiner & Wooldredge, 2008).

Steiner and Wooldredge (2008) recently found prior incarceration to be one of the four predictors able to predict all three of their prison misconduct categories: assaults, drug and alcohol use; and other non-violent misconduct.

There was no relationship between gang affiliation and serious misconduct in the bivariate analysis or in the models with just the personal predictors. Nor was it a predictor in the final multivariate models predicting the annual misconduct score and violence. Perhaps the presence of other street cultural variables such as number of drugs, prior incarceration,

B&E/drug charges, and age are so highly correlated with gang affiliation that they masked its effect on misconduct. The qualitative research demonstrated that the relationship between gang affiliation and misconduct was not clear-cut as most would assume but instead was more complex—that often neighborhood turf issues or prior altercations on the street were branded as gang issues by staff. More research using better measurements of gang affiliation and distinguishing between street gang affiliation and prison gang affiliation would need to be undertaken in order to fully understand this relationship. As mentioned previously, gang affiliation has been consistently found to be a strong predictor of serious prison misconduct and violence with few exceptions (Gaes, et al., 2002; DeLisi, et al., 2004; Huebner, 2003; Griffin &

Hepburn, 2006; Trulson, 2007; Cunningham & Sorensen, 2007; Kuanliang, et al., 2008).

302

About half of the high-level group had been adjudicated delinquent and had been placed in a secure juvenile facility compared to a third of the low-level group. Nevertheless this variable was not a predictor in any of the multivariate analysis. However it surfaced in the qualitative interviews. Many staff discussed how some prisoners involved in serious prison misconduct had begun their criminal careers early. In addition many of the high-disciplinary prisoners interviewed had been adjudicated delinquent and placed in juvenile correctional facilities. As mentioned previously, Trulson‘s study (2007) of misconduct in juvenile correctional facilities found that the younger the age of a youth‘s first formal referral to the juvenile system, the more likely he was to be both dangerous and disruptive while confined. In essence those prisoners who become enmeshed in the justice system at such an early age are state-raised youth—those youths who grew up mostly in foster care, lived in juvenile residential programs, and served time in juvenile correctional facilities before graduating to adult prison.

Though state-raised youth make up a small minority of any prison, they are responsible for a larger percentage of disruption and dangerousness in the prison, according to all who have studied them (Irwin, 1970; Abbott, 1991; Wright, 1991; Lovell et al., 2000; Johnson, 2002).

Indeed, Abbott himself understood the rage and inadequacies that belie this unfortunate group.

―It is not so for the state-raised convict...A prisoner who is not state-raised tolerates the situation because of his social maturity prior to incarceration. He knows things are different outside prison. But the state-raised convict has no conception of any difference. He lacks experience and, hence, maturity. His judgment is untempered, rash; his emotions are impulsive, raw, unmellowed‖ (Abbott, 1991, p. 12).

Unfortunately, it is probable that this small group of prisoners is not only responsible for a greater proportion of the serious misconduct, but that their cavalier attitude and fearless demeanor might cajole or intimidate other young prisoners into following their lead. Thus, the

303 very group of prisoners who should be marginalized and monitored might be setting the tone for other young prisoners.

Finally, being involved in serious misconduct in the past is a good predictor of future misconduct, as has been found by others (Harer & Langan, 2001; Camp et al, 2003; Cunningham

& Sorensen, 2007). Overall it is probably not surprising that most prisoners who misbehave at one point in time continue with their misbehavior. However, the most interesting finding regarding prior misbehavior is that 31 percent of the prisoners who were in the low-level group with either no or few dreports in 2008 and 2009 had received serious predatory and non- predatory dreports prior to 2008. Thus one-third had not continued down that negative path.

Local Subcultural Predictors

The current study included a number of variables that one might categorize as local subcultural predictors including: race and ethnicity; sentence length; time incarcerated; crime type; chronic substance abuse; and prior gang involvement.

RACE AND ETHNICITY

There was no statistically significant relationship in the bivariate analysis between race and serious disciplinary behavior in the current sample. However, because it has been such a focal variable in other research studies of prison misconduct, it was broken down into dichotomous variables and included in the multivariate analysis with whites being the referent category. Being black or of another race did not correlate with either serious misconduct in general or with violence alone. However, not being Hispanic (thus being white, black, or other) was directly correlated with the annual misconduct score (but not violence) throughout the multivariate analysis. Much of the previous research on race and prison misconduct had also

304 found mixed results, with some researchers finding more prison misbehavior among blacks and/or minorities in general, (Carroll, 1974; Toch, et al., 1989; Gaes, et al., 2002; Berg & DeLisi,

2006; Trulson, 2007), others finding no such relationship between prison misconduct and race

(Ellis, et al., 1974; Wright, 1989), and still others finding mixed results (Petersilia, 1983; Harer

& Steffensmeier, 1996; Camp, et al., 2003). How can one presume that race or ethnicity by itself is a predictor of prison misconduct without considering the local subculture from which prisoners come to prison? Even studies that use race as a means to study whether importation is more important than deprivation cannot possibly be meaningful unless one has conducted an in- depth study of the local criminal subculture from which prisoners were imported. Jacobs exhorted researchers to consider the geographic area when studying race relations in prison

(1979). He said, ―It should make a difference whether prisons are located in the southwest, with its large Mexican-American population, in the north central states, where native American movements are becoming significant, or in the south, with its distinctive history of paternalistic race relations‖ (1979, p. 22). The geographic areas from which prisoners are drawn are similarly important to the development of the prisoner subculture that affects prison misbehavior. That blacks and whites are more likely in Rhode Island to be involved in serious misbehavior compared to Hispanics is simply emblematic of the local subculture of young offenders, discussed in more detail below.

OFFENSE TYPE AND SENTENCE LENGTH

The bivariate analysis revealed little difference in the percentages of low-level and high- level disciplinary prisoners who were in for a violent offense (slightly more than half of both disciplinary groupings), or for the less frequent weapons or non-violent charges. However, a much greater percentage of the low-level misbehavior group was made up sex offenders, while a

305 much greater percentage of the high-level misbehavior group was made up of prisoners involved in B & Es and drug charges. Given the prevalent crime types involved in each group, it makes sense that high-level misbehavior prisoners have, on the average, shorter sentences than low- level misbehavior prisoners and that those in the former group are less likely to be serving life in prison. Multivariate analyses consistently found being incarcerated for a B&E/drug offense to be a significant predictor of the annual misconduct score. Not being a sex offender was also directly related to a higher annual misconduct score in the final model with both ways of coping and trait emotions. However, neither of these offense types were related to prisoners‘ level of violent misbehavior. Also, neither total sentence length nor time incarcerated were related to either the annual misconduct score or violence in the final models.

The few studies in the literature that examined seriousness of current offense yielded mixed results. Two studies found offense type to be predictive of prison misconduct (Harer &

Langan, 2001; and Griffin, & Hepburn, 2006), while another did not (Wooldredge et al., 2001).

Steiner and Wooldredge‘s (2008) recent bi-level analysis of prison misconduct also found mixed results by study year, by type of sentenced offense, and by type of prison misconduct. There were similar mixed results regarding the relationship between sentence length and prison misconduct with some studies finding a correlation (Zink, 1958; Flanagan, 1980; Berg & DeLisi,

2006; Griffin & Hepburn, 2006; Trulson, 2007; Kuanliang, et al., 2008) and others finding no correlation or mixed results (Wolfgang, 1961; and Wooldredge et al.; 2001). However a very recent study of prison misconduct in Florida by Sorensen and Cunningham (2010) had similar findings as the current study—that short sentences (presumably associated with less serious offenses) correlated with prison misbehavior while both violent offenses and longer sentences did not. In this study, while those with serious disciplinary issues were more likely to have

306 started their criminal careers early and to have been previously incarcerated as an adult, they were no more likely to be violent offenders, but more have committed less serious property offenses and drug charges, both associated with street-life crime. Again like race, studies in the literature that examined seriousness of current offense and sentence length yielded mixed results, a similar indicator that they might be more a reflection of the local subculture of youth criminals on the street than indicative of an actual link between crime type and prison misbehavior.

SUBSTANCE ABUSE

The bivariate analysis uncovered little differentiation between the low- vs. high-level disciplinary groups for most of the substance abuse variables. However, chronic substance abuse and the specific use of marijuana did correlate with serious prison disciplinary behavior at the bivariate level of analysis. The number of drugs used in the six months prior to incarceration was positively correlated to the annual misconduct score in two of the multivariate models. In the model predicting the annual misconduct score, number of drugs was statistically significant with just the personal predictors in the model. However, when the ways of coping were added, the relationship between drug use and serious misconduct disappeared entirely. Yet in the final model with both coping and trait emotions, number of drugs used was positively correlated with serious prison misconduct and violence. Interestingly, in the model predicting violence alone, the opposite occurred—drug use was not a statistically significant predictor of violence until the addition of the ways of coping. With the introduction of coping, number of drugs used had a negative relationship to violence—that is prisoners who used more drugs were less likely to be involved in violence. However, once trait emotions were added to the final model, the number of drugs lost its statistical significance as a predictor. More detailed analysis would need to be

307 undertaken in order to understand the complex interaction between drug use, ways of coping, trait emotions, and serious misconduct and violence.

Several prior research studies did find a significant relationship between prior substance abuse and prison misconduct (Gendreau, et al., 1997; Dhami, et al., 2007; Steiner &

Wooldredge, 2008) though not all (Berg & DeLisi, 2006). Steiner and Wooldredge (2008) found that prior drug abuse was one of only four variables that successfully predicted the three types of prison misbehavior that they studied (assaults, drug/alcohol offenses, nonviolent offenses).

Chronic substance abuse prior to incarceration is indicative of both personal dysfunction, but possibly also is another factor that might describe the local criminal subculture from which prisoners came. Fortunately, while chronic prior substance abuse is considered a static predictor, it is used usually as a proxy variable for prisoners‘ current substance abuse problem or propensity for addiction—a dynamic factor that prisoners can work on while incarcerated.

STREET CULTURE

Up to this point, the local subcultural predictors begin to paint a picture of young, mostly urban offenders involved in street crime (B&Es and drug crimes) who might also be chronic drug abusers. It appears that these young offenders attempt to replicate their street culture within the prison system, although further research would be necessary to validate this conclusion.

However many of the issues found in this study such as the campaign for respect, a tough reputation, and the need to exact retribution if either of these are threatened are encompassed in

Anderson‘s (1999) description of street culture. The street culture is an oppositional culture developed in mostly urban, often minority neighborhoods where access to jobs and a living wage are largely non-existent and despair is pervasive. Though mostly inhabited by ―decent‖ families, these neighborhoods are dominated by those residents who embrace the ―street code‖ that

308 promotes the campaign for respect and values aggressive behavior and the use of violence if deemed necessary. Anderson explains this campaign for respect:

―In the street culture, especially among young people, respect is viewed as almost an external entity that is hard-won but easily lost, and so must constantly be guarded. The rules of the code in fact provide a framework for negotiating respect. The person whose very appearance—including his clothing, demeanor, and way of moving—deters transgressions feels that he possesses, and may be considered by others to possess, a measure of respect. With the right amount of respect, for instance, he can avoid ―being bothered‖ in public. If he is bothered, not only may he be in physical danger but he has been disgraced or ―dissed‖ (disrespected). Many of the forms that dissing can take might seem petty to middle-class people (maintaining eye contact for too long, for example), but to those invested in the street code, these actions become serious indications of the other person’s intentions. Consequently, such people become very sensitive to advances and slights, which could well serve as warnings of imminent physical confrontation‖ (Anderson, 1994, p. 82).

Such a street culture would support the use of violence to exact respect and build one‘s reputation. If young prisoners bring their street culture into prison with them, it is no wonder that bullying, threats, spontaneous assaults, and planned violence would predominate.

Stewart and Simons (2010) recently tested Anderson‘s theories through a longitudinal study on whether neighborhood street culture and/or individual commitment to the code of the street affected adolescent violence using the Family and Community Health Study in multiple neighborhoods in two states. Their multilevel analyses revealed that the presence of neighborhood street culture predicted violence. Although adolescent violence was tempered by the addition of an individual‘s commitment to the street code, they also found that individuals committed to that code were more likely to engage in violent delinquency when they resided in a neighborhood where the street code predominated (Stewart, & Simons, 2010, p. 591-591). This is particularly relevant to prisons. That prisoners bring their street code values into prison with them is worrisome. However, prisons must be careful to not allow the prison itself to become a

309 milieu where street culture can flourish. Anderson discussed the difficulty with which youths from ―decent‖ families have in navigating the dangerous path between doing what is right and keeping true to the street culture to maintain one‘s safety. Imagine the difficulty for new young prisoners! A final point to make is that many who have grown up in the street culture have a negative view of law enforcement (Anderson, 1999) thus prisoners coming into prison with a street culture attitude start their incarceration with a dislike and mistrust for correctional staff.

It would be a mistake to assume that all prisoners who entered the prison abiding by the street culture were also gang members or that prisoners who were not gang members do not abide by the street culture. The extent of the intersection between prisoners who abide by street culture values and those who are street gang members is unclear. Nevertheless gang issues exacerbate the prevalence of serious prison misconduct and violence beyond street culture values because of the additional turf issues between gangs and the likelihood that gang members will be involved in lucrative illegal activity. Twice as many prisoners in the high-level disciplinary group reported being in a street gang compared to those in the low-level group, though the differences were not statistically significant. Though most likely underreported, about 12 percent of those in the high-level disciplinary group admitted to street gang membership.

Admitting to gang involvement in prison was even lower so a decision was made to analyze the

Security Risk Group designation instead. Though prisoners in the high-level disciplinary group were more likely to be SRG-designated than those in the low-level group, SRG status can be given for other negative behaviors besides gang activity. Still only 25 percent of the high-level group was SRG-designated. The picture becomes even blurrier with the multivariate analysis and the consideration of qualitative data. There was no correlation between street gang membership and violent misbehavior in any of the negative binomial analyses. Similarly, when

310 only the personal predictors were in the model, there was no correlation between street gang membership and the annual misconduct score. However, as mentioned previously, with the introduction of the ways of coping into the model, a negative association occurred. However it disappeared again with the introduction of the trait emotions. Either way the results demonstrate that knowledge of street gang membership itself does not necessarily equate with serious prison misbehavior and violence.

As mentioned previously, most of the research studies on gangs in prison have discovered that gang affiliation is a strong predictor of serious prison misconduct and violence. Other research has found a link between the likelihood of bullying, gang-related activity, and the tendency to morally disengage (Wood, Moir, & James, 2009) That is, prisoners involved in gangs choose to morally disengage in order to abide by the mandates of the gang including bullying those who are weaker than themselves. Trulson (2007) defined gang affiliation as known gang association at time of commitment. Most of the other studies either adopted criteria for gang affiliation used by the Federal Bureau of Prisons (self-proclaimed membership, gang- identifying tattoos, appearance on gang membership lists, possession of gang paraphernalia, association with certified gang members, and other indicators of membership) or a combination of known street affiliation and/or prison participation in gangs, sometimes referred to as security threat group designation (Gaes, et al., 2002; Huebner, 2003; Griffin & Hepburn, 2006).

However, DeLisi, Berg, and Hochstetler (2004) differentiated between street gang affiliation and security threat group designation. Their study revealed that prisoners with just prior street gang involvement or just prison gang involvement were no more likely than other prisoners to engage in prison violence. However, they also created a multiplicative variable of the two and discovered that prisoners with street gang history and prison gang involvement were significantly

311 more involved in prison violence compared to non-gang prisoners. As mentioned previously the gang variables were significantly predictive of prison violence only in the full model when various types of gang membership were considered and their overall effects were smaller than a dozen other risk factors.

Indeed as noted before, RIDOC staff disagreed about the prevalence of real gangs vs. neighborhood turf issues, and about the prevalence of gang involvement in violence. While ten of the 25 prisoners interviewed did admit to some type of gang affiliation, either on the street or in prison, they mostly discussed the peer pressure they found and the need to adhere to such street culture codes as the campaign for respect and a tough reputation. Yet one cannot help but think that prisoners who were involved in gangs on the street and who seek out gang membership in prison are probably more likely to be involved in serious prison misbehavior and violence.

However, a more nuanced assessment of gangs and their contribution to prison violence and disruption might be warranted in the future.

*****

In summary, several static predictors were found to be correlated to serious prison misconduct and violence. Static predictors included those that were more universal (age, being single, and prior adult incarceration) as well as those that were local subcultural predictors (not being Hispanic, being incarcerated for B&Es/drug offenses, and prior chronic substance use).

These predictors yield three or possibly four groups of prisoners that might be more at risk for involvement in serious disciplinary issues. The first are young prisoners. Just as any parent, high school teacher, or police officer knows, teens and young adults are more likely to make poor decisions, be negatively influenced by their peers, and misbehave. It is no surprise that

312 teens and young people in prison behave similarly but unfortunately with more serious misbehavior yielding higher stakes. However, some of these young offenders who misbehave might be likened to those who Moffitt (1993) designated as ―adolescent-limited‖—that is, as they get older they will ―age out‖ of involvement in general prison misbehavior. The trick for prison staff is to determine which prisoners are indeed ―life-course persistent‖ offenders who will be repeatedly involved in serious misbehavior and violence throughout their incarceration. Is it those prisoners who have been involved in the local street criminal subculture (B&E/drug offenses, short sentences, chronic substance abuse)? Or is it perhaps an even smaller group of prisoners involved not only in the street criminal subculture but also who were and are still involved in gangs? Or is it that even smaller subset of prisoners who are state-raised youth who identify with the local street subculture and with gangs but who are so dysfunctional that misbehavior is predictable? While the lines between each of these subsets are blurry, care must be taken to not equate the disrespectful, rebellious street-sassiness of a nineteen-year-old with the disturbed and potentially ominous threats of a state-raised youth. The consequences range from over punishment and over classification of the former to the potentially fatal underestimation of the dangerousness of the latter.

DYNAMIC PREDICTORS DYNAMIC PERSONAL PREDICTORS

Coping and Trait Emotions

The current study had several hypotheses that centered on whether ways of coping and trait emotions affected the level of prisoners‘ serious misconduct and violence. The study utilized the COPE instrument (Carver, et al., 1989) to measure various ways of coping, which resulted in eight separate subscales. The negative binomial analysis models (for both the annual

313 misconduct score and the annual rate of violent dreports using only the eight coping scales) were statistically significant, indicating that they were significantly different from the null model with the intercept alone.

In the model predicting the annual misconduct score from just the ways of coping, seven of the eight coping scales were statistically significant (but not religion), though not all in the anticipated direction. It had been anticipated that active coping would be associated with a lower level of serious prison misconduct, but instead it was associated with high levels of misconduct.

In contrast, it was anticipated that disengagement would be associated with more serious disciplinary behavior, but instead it was associated with less misconduct. Prisoners who got emotional and instrumental support, used acceptance and disengagement and who did not use humor, active coping, denial, or venting to cope with stresses had lower misconduct scores.

When the personal predictors and trait emotions were added into the final model predicting the annual misconduct score (Model 9), getting support and not venting one‘s emotions were still related to lower annual misconduct scores. It also resulted in the use of religion becoming statistically significant even though it had not been statistically significant in the coping only model. However, the addition of the personal predictors and trait emotions resulted in the disappearance of five of the ways of coping: humor, active coping, denial, disengagement, and acceptance.

In the model predicting violence from just the ways of coping, four of the coping scales‘ parameters (use of humor, getting support, active coping, and acceptance) were statistically significant, with active coping again not in the anticipated direction. However, again the addition of the personal predictors resulted in acceptance no longer being statistically significant.

314

Humor lost its statistical significance when the personal predictors were added into the model

(Model 11) but it became statistically significant again in the final model with the trait emotions

(Model 12). Getting support and active coping also remained strong predictors of violence even with the addition of the personal predictors and trait emotions in the final model.

In addition to the ways of coping, the study incorporated three trait emotions: anxiety; anger; and depression. In the negative binomial analysis with trait emotions only two of the trait emotions, anger and depression, were associated with greater annual misconduct scores.

However, anxiety was associated with a lower level of serious prison misconduct. Only anger was associated with more prison violence as measured by the annual rate of violent dreports.

However, similar to the ways of coping, the relationship between the trait emotions and misconduct changed with the addition of the personal predictors in the final two models.

Specifically in the final model predicting the annual misconduct score, the relationship between the trait emotions flipped; anxiety had a positive relationship with the misconduct score and anger had a negative one. In the model predicting violence, only anger was correlated with violence, again with a negative relationship. Depression was not statistically significant in either of the final models.

WAYS OF COPING

The COPE differentiates between getting emotional support, which happens in the primary appraisal and getting instrumental support, which happens in the secondary appraisal. In this study, the questions associated with both loaded onto one factor and were thus combined.

Prisoners who reported getting support from others as a coping mechanism were less likely to be involved in serious misbehavior and violence in both the bivariate and multivariate analyses.

Getting support remained a strong predictor throughout the analyses for the annual misconduct

315 score and for violence; it was the sole way of coping that remained statistically significant for both measures of misconduct once the personal predictors were added into the models.

The study collected data on some proxies of social support including visits, letters, and number of friends in prison. Though the number of letters received was greater for those with disciplinary issues, this is probably due to the fact that these prisoners spend more time in segregation and lose their visiting privileges, resorting instead to letters. Neither of the other two social support predictors—number of visits received or number of friends—turned out to be strong predictors of serious misconduct and violence indicating that it is more likely to be the quality of one‘s relationship to others rather than the quantity of contacts that gauge actual social support in prison. Interestingly, one-third of the prisoners interviewed said that they had nobody to talk to for either emotional support or for help in solving their problems. The other two-thirds either went to family, friends in prison, or staff for emotional and instrumental support.

Prisoners with family members elicited emotional support, but fewer of them asked family members to help them solve problems in prison. More than half of the prisoners interviewed said they had friends in prison; others said one could not have ―true friends‖ in prison.

Nevertheless, two-thirds said they approached other prisoners for help with problems, but few did so for emotional support. Getting support was one of the four ways of coping that interviewed prisoners said they used to address their most difficult prison hardships. Eliciting emotional support from one‘s family or friends in prison was often used to deal with everyday interaction and conflicts with staff.

Focusing on and venting one‘s emotions happen during the primary appraisal—it is how people react initially to a new stress or hardship. Prisoners in the high-level disciplinary group were more likely to report venting their emotions as a way of coping. In addition, prisoners who

316 were more likely to vent were also more likely to have higher annual misconduct scores.

Venting emotions was only one of two ways of coping that retained its statistical relationship with the annual misconduct score once the personal predictors and trait emotions were introduced into the final model. There was no similar relationship between venting and the use of violence. Venting was closely correlated to the trait emotion of anger. Thus it appears that prisoners who vented their emotions upon confronting a stressful situation were more likely to vent their anger and frustration, thus acting out and ending up with bookings, though not bookings that were necessarily violent in nature. For example, these prisoners might be more likely to swear and threaten other prisoners and staff, but are not necessarily acting on those threats with actual violence.

Reliance on religion did not differentiate between the low-level and the high-level disciplinary groups in the bivariate analysis. In the coping only models predicting both the annual misconduct score and violence, reliance on religion was not statistically significant. It was also not a statistically significant predictor of violence in the final model with both the personal predictors and the trait emotions. However, reliance on religion as a way of coping was associated with a higher annual misconduct score in the final model. It appears that the inclusion of the personal predictors resulted in a suppressor effect on religion so that it became positively associated with serious misconduct once the model controlled for variables correlated with religion—namely, being older, black, and not single.

The active coping scale in this study was made up of all but one item in five of Carver‘s subscales (1989), including: positive reinterpretation and growth; suppression of competing activities, restraint coping; planning; and active coping. While positive reinterpretation and growth takes part during the primary or emotion-focused phase of coping, the remaining four are

317 all part of the secondary or problem-solving phase. It is worth repeating Carver‘s definition of planning and active coping. According to Carver, planning is thinking about how to cope with a stressor, ―what steps to take and how to handle a problem.‖ Active coping refers to the ―process of taking active steps to try to remove or circumvent the stressor or to ameliorate its effects...(and) includes initiating direct action, increasing one‘s efforts, and trying to execute a coping attempt in stepwise fashion‖ (Carver, et al., 1989, p. 268). Analysis of the individual questions that make up this scale revealed that half of the active coping and planning questions might as easily result in behavior that is negative as behavior that is positive because they describe taking direct action. Three of the questions involve more thinking, reflection, and the taking of sequential steps. As previously concluded, when the dependent variable is taking a negative action vs. a positive action, as it is in this study, Carver‘s scales do not capture the subtleties of the differences between coping strategies that might result in positive behavior and those that might result in either positive or negative behavior. Instead, there seems to be a difference between action that is impulsive and direct (and possibly negative) and action that is reflective (and therefore more likely to be positive). Thus the 19-item active coping scale is a complex variable that encompasses several facets of active coping.

In the bivariate analysis active coping could not differentiate between the high- vs. low- level disciplinary groups. However, it was statistically significant in several of the negative binomial models of serious misconduct. In the model predicting the annual misconduct score with just the ways of coping, prisoners who actively coped were more likely to have higher annual misconduct scores. However, this relationship disappeared when the personal predictors were included in the model. However, in the model predicting violence, active coping had a positive relationship with violence even after the personal predictors were included. It is one of

318 just two ways of coping that retained its statistically significant relationship with violence even after the personal predictors were added. In all of these results, prisoners who resorted to active coping were more likely to be involved in serious prison misconduct and violence. Thus it appears that these prisoners take direct action to address prison stressors even when the solutions they plan involve negative behavior. Several staff noted that prisoners who were involved in gangs had better coping skills compared to other prisoners involved in serious prison misconduct and violence. Perhaps it is not just true for gang members, but might also be true for those prisoners who come from a street culture where conflicts with other persons are not an unusual occurrence. Indeed adhering to a street code and/or the ideal of hegemonic masculinity as explained earlier would demand action that is direct and impulsive, rather than reflective and thoughtful. In in-depth interviews prisoners discussed using negative- or positive-problem solving when facing their most difficult prison hardships. However, they were most likely to actively try to solve problems with other prisoners. While several of the prisoners said they had come to realize that there were more positive ways to resolve problems with prisoners (better communication or sometimes ignoring them) there were two issues that almost all prisoners agreed could not be ignored or handled in a positive manner: being disrespected in front of others or being threatened and/or attacked by others. A few prisoners said that they had learned to try to verbally diffuse situations when they were disrespected; however nobody thought one could walk away from being threatened or attacked because of the inevitability for its escalation. Thus active coping seems to be the preferred way of coping when dealing with conflicts with other prisoners. Unfortunately, in prison, active coping most often results in negative behavior. Thus in this case, it is not a deficit in coping that is leading to serious misconduct and violence, but the conclusion that violence is the preferred solution. Again this conclusion leads back to the street

319 culture from which many of these prisoners came; one where hyper-masculinity, a campaign for respect, and a lack of empathy reign.

Using humor as way of coping was a strong predictor of serious misconduct and violence in the bivariate analysis. In the bivariate analysis using humor or joking differentiated between the low- vs. high-level misbehavior groups. In the negative binomial analysis, using humor was a predictor of both the annual misconduct score and the rate of violence when only the ways of coping and trait emotions were in the model. However, in the final model predicting the annual misconduct score (Model 9) with the personal predictors and trait emotions, humor was no longer statistically significant. In contrast, in the final model predicting violence (Model 12) humor was positively related to violence. Humor was the way of coping that was the most correlated with the street culture variables of being single, not having a prior incarceration, having a B&E/drug charge, and more prior drug use. In the qualitative results, prisoners who used humor, joking, or bravado when facing problems or stressors were more likely to be involved in the street culture where showing weakness or hesitancy was not an option. Instead they were more likely to joke, be sarcastic, and act impulsively in the face of danger or other types of problems.

Behavioral disengagement involves freeing oneself from doing anything about the problem at hand, while mental disengagement involves actually becoming involved in other types of activities or behavior to also take one‘s mind off of the problem. Though the questions from both of the subscales did not load on the same factor, reliability analysis showed greater strength as a scale when added together. In the bivariate analysis, prisoners involved in more misconduct were more likely to mentally and behaviorally disengage than prisoners with little or

320 no misconduct problems. However, disengagement had no effect on serious misconduct and violence once personal predictors and trait emotions were included in the final models.

Nevertheless, interviews with prisoners revealed that they sometimes practiced mental and behavioral disengagement when dealing with their most difficult prison hardships. This was especially true for those who were unable to maintain communication with family. Prisoners disengaged physically from their problems when they determined they had no chance in solving them. It was one of several coping mechanisms used by prisoners to handle conflicts and everyday interaction with staff. However, prisoners who were faced with problems with other prisoners did not usually resort to disengagement, especially prisoners who were young and who were associated with the street culture. Thus is may not be surprising that the street culture characteristics might have conditioned the relationship between disengagement and serious misconduct. Prisoners used myriad ways of disengaging; some were more positive (work, school, programs, sports) than others (hanging out and socializing with friends).

Denial also takes place during the primary appraisal of a problem and involves either refusing to believe the problem exists or trying to act as if there is no problem. Prisoners in the high discipline group were more likely to use denial than were those in the low group. Similarly, prisoners who reported using denial as a coping mechanism were more likely to have a higher annual misconduct score in the coping only model. However, like disengagement, the relationship between denial and the annual misconduct score disappeared when the personal predictors and trait emotions were added into the final model. Denial was not a predictor of violence in any of the models. Moreover, prisoners never mentioned being in denial of their problems during the in-depth interviews.

321

Acceptance is the opposite of denial in that one is acknowledging that a problem or stressor exists. It is also a coping response in that one can accept not only that a problem exists, but also that this is the way it is and one cannot do anything about it. In the bivariate analysis, acceptance did not distinguish between prisoners in the low-level vs. high-level disciplinary groups. Though it was one of the strongest predictors of both the annual misconduct score and the annual rate of violence in the models with just coping and trait emotions; once other personal predictors and trait emotions were introduced in the final models, acceptance lost its statistical significance.

Interestingly, in the in-depth interviews prisoners were asked whether they tried to actively solve their problems or ignored them. In the probes of how they ignored them, prisoners alternatively discussed behavior that could be categorized as mental disengagement but also behavior that could be categorized as acceptance. In the former prisoners did not acknowledge their problems or engaged in behavior that would redirect their thoughts away from the stressor.

In the latter, prisoners acknowledged the stressor and also acknowledged that it was a problem they had to ―deal with‖ and could not do anything about. Several of the prison hardships they faced (missing family, freedom, and possessions) were stressors that they had to just accept.

However, mostly all prisoners interviewed had also come to the conclusion that one must use acceptance to deal with staff and the rules of the prison. They had realized that they could not solve problems that derived from how officers spoke to them or from practices that they perceived to be unfair. Thus prisoners who can acknowledge their lack of efficacy and their lack of autonomy when dealing with prison staff and facing the prison structure might be less likely to be involved in serious misconduct and violence.

322

TRAIT EMOTIONS

In addition to the ways of coping, this study incorporated three trait emotions: anxiety, depression, and anger. One of the most important findings regarding trait emotions was that prisoners in the sample were much more likely to be in a constant state of depression, anxiety, and anger compared to samples of non-incarcerated males (Spielberger, 1972, 1980, 1983). It was assumed that prisoners who were perennially angry, anxious, and depressed would be more likely to be involved in serious misconduct and violence. The bivariate analysis revealed that prisoners in the high-level group were much more likely to be in a state of anger compared to those in the low-level group. Similarly, in their study of self-control, prison infractions, and victimization, Kerley and his colleagues found that anger (temper) was the only dimension of self-control associated with prison infractions (Kerley, Hochstetler, & Copes, 2009).

However, the multivariate analysis in this study revealed some surprising findings. In the models with trait emotions only or trait emotions and ways of coping, anger and depression indeed had a positive relationship with the annual misconduct score. Contrary to assumptions made at the start of the study, anxiety had a negative relationship with the annual misconduct score. Anger was found to have a positive relationship with violent misbehavior in the model with trait emotions only; prisoners who were angrier were more likely to be violent. However, with the introduction of the ways of coping, anger‘s statistical significance was lost. However, with the introduction of the personal predictors in the final violence model, being angry was again statistically significant, however, in the opposite direction. Interestingly a similar phenomenon happened in the final model of the annual misconduct score. Prisoners who were angrier were less likely to have high scores and prisoners who were more anxious were more likely to have higher annual misconduct scores.

323

Anger was highly correlated with several of the personal predictors including being young, in a gang, previously adjudicated as a delinquent, having a B&E/drug offense, having been treated for psychiatric problems, and prior drug use. One can only surmise that the introduction of these personal predictors would all serve to control for those prisoners involved in the street culture prior to incarceration. When those predictors were not in the model, being angry became a proxy for those variables. However, once the model controlled for them, the relationship between anger and misbehavior changed. This is likely to be a case of reciprocal suppression—that is when two independent variables or sets of variables mutually suppress irrelevant variance in the other (Lancaster, 1999). In the negative binomial analysis, the relationship between each independent variable and the dependent variable is based on controlling all other independent variables in the model. Thus in the final model, once the anger of the prisoners involved in the street culture was controlled, the remaining prisoners involved in serious misbehavior and violence were less likely to be angry.

Interestingly, there had also been a high correlation between prisoners who had been involved in psychiatric treatment and anxiety. Perhaps once the variance in the misbehavior of street-involved inmates was controlled, the remaining variance could be explained by other types of prisoners involved in serious misbehavior, like those who were mentally ill or who might need protective custody—thus prisoners who are less likely to be angry, but more likely to be anxious.

Although prisoners were not asked about their anger, anxiety, or depression specifically in the in- depth interviews, many of the prisoners who had been involved in the street culture did discuss how they had so often felt angry, and how that anger was often the determinant of whether they used acceptance or positive problem-solving rather than engaging in negative problem-solving.

324

Alternatively, prisoners who had been treated for psychiatric problems and those with protective custody issues discussed their anxiety about interaction with both prisoners and staff.

Thus prisoners are less likely to be involved in serious prison misconduct and violence when they get support from family members and friends. In contrast, prisoners who are involved in serious prison misconduct and violence are more likely to vent their emotions, especially anger, when facing prison hardships. In addition, those involved in the street culture are more likely to use bravado and denial when faced with stressors and to react directly without reflection, choosing responses that are negative, especially in situations where they are being disrespected, threatened, and of course, physically attacked.

Prisoners with Mental Health Issues

Since this researcher did not have access to mental health data, prisoners were asked to report psychiatric treatment prior to and during incarceration as a proxy measure for mental illness. Those prisoners in the high disciplinary group were more likely to have reported some type of treatment for psychiatric or emotional problems both before and during their incarceration compared to prisoners in the low group. Prisoners exhibiting misbehavior were also more likely to have been on medication for psychiatric and emotional problems prior to prison and to have been transferred to a psychiatric unit or facility while in prison.

Over half of state prisoners and almost half of federal prisoners had mental health problems, based on personal interviews with prisoners in 2004 that used recent history or symptoms of mental health to determine the existence of mental health problems (James &

Glaze, 2006). Their Bureau of Justice Statistics report found that 43 percent of state prisoners suffered from mania, 23 percent suffered from major depression, and 15 percent suffered

325 symptoms of psychotic disorders (2006, p. 1). They also found that prisoners with mental health problems were more likely to have had more prior incarcerations, to have been chronic drug abusers, and to have been in a fight since entering prison. They were also more likely to have prior or current offenses for violence, to be homeless, to have had parents with substance abuse problems, and to have been physically or sexually abused.

Many researchers have discovered the predominance of mentally ill prisoners among those involved in serious misbehavior and violence and among those residing in supermax-like units (Toch & Adams, 1986; Toch, et al., 1989; Toch & Adams, 1994; Lovell, et al., 2000; Peck,

Jr., 2004; Toch & Kupers, 2007; Lovell, 2008). In the current study, prisoners who were in the high disciplinary group were more likely to have reported some type of treatment for psychiatric or emotional problems and specifically to have been on medications for those problems prior to their incarceration. However, there were no clear associations between mental health treatment in prison and level of disciplinary infractions. The majority of prisoners who had received treatment on the street reported receiving treatment in prison; though this was true of a greater percentage of the high-level disciplinary group compared to the low-level group. Still one-fifth of the high group and one-third of the low group who had gotten treatment for mental health issues prior to incarceration were not getting mental health treatment in prison. Conversely, about one-fifth of both groups who had not been in any mental health treatment on the street were currently receiving some kind of mental health treatment in prison.

In the multivariate analysis, the variable ―psychiatric treatment‖ included those prisoners who had either received psychiatric treatment prior to incarceration or who had been transferred to a psychiatric unit while incarcerated. Having received psychiatric treatment did not correlate with serious misconduct and violence in general (annual misconduct score). Interestingly, it also

326 did not correlate with level of violence when the model included just the personal characteristics or the personal characteristics and the ways of coping. However, once the trait emotions were added into the final model, psychiatric treatment was directly related to violent misbehavior.

Since it had been highly correlated with all three trait emotions, perhaps once the variance in the rate of violence that was associated with trait emotions was controlled it unmasked the relationship between psychiatric treatment and violence. Psychiatric treatment had also been highly correlated with a number of other variables including chronic drug use, not being black or

Hispanic, prior incarceration, residing in the HSC, and receiving fewer letters. Thus there emerged a positive relationship between having received psychiatric treatment and violence—the likely result of another suppressor effect.

In the in-depth interviews, most of the administrators and lieutenants identified mental illness as contributing to serious prison misconduct and sometimes violence. Most staff believed that those with mental illness made up a substantial proportion of those with serious disciplinary problems, especially those housed in the HSC. Ten of the 25 prisoners interviewed revealed that they had mental health issues. Every staff person acknowledged the existence of prisoners with mental health problems, which were exacerbated by the deinstitutionalization of the mentally ill in the 1980s. Administrators discussed steps they or their staff took to accommodate the needs of this population but acknowledged that more needed to be done.

PRISONER HARDSHIPS

Delineating the Pains of Imprisonment

In The Society of Captives Gresham Sykes (1958) outlined the ―pains of imprisonment‖ in his study of the New Jersey State Maximum Security Prison in Trenton. In a prison where the maintenance of social order and the suppression of misbehavior were paramount, Sykes found

327 the most difficult deprivations for prisoners were freedom, autonomy, security, goods and services, and lack of heterosexual relationships. Others have since studied the deprivations of prison or prison stresses and have expanded the list (Johnson & Toch, 1988; Johnson, 2002;

Zamble & Porporino, 1988; Maitland & Sluder, 1998). Although each study varied in their research methodology (some asking open-ended questions about what is difficult and others asking prisoners to rate the hardships) two of Sykes‘ deprivations have come up as very difficult in this and most other studies: missing freedom and missing family, the latter being an expansion of Sykes‘ lack of heterosexual relationships. These hardships are the crux of being punished by incarceration and though prisons have changed since Sykes completed his study, they are still the most difficult consequences for prisoners to endure. In the current study missing family placed second to missing freedom, but was rated as ―very difficult‖ by three-fourths of prisoners.

Although there are still many prisons in the nation that only allow visits behind glass using phones, in Rhode Island prisoners have ample opportunity for contact visits in an open visiting area as well as access to telephones in their units and in the yard. This level of contact has shifted the burden from one where contact was yearned for to one where prisoners are more acutely aware of what they are missing and unable to help resolve family problems. With the opportunity for almost daily contact, prisoners hear about the minutiae of their families‘ daily lives such as their son‘s first steps or the inability of their wife to pay household bills. Thus the pain of missing family has evolved from the pain of physical detachment to the pain of enforced helplessness, leading to feelings of inadequacy and shame. One might wonder whether the increased stress that prisoners might experience from such frequent contact would be worth the increased contact. However, Johnson lists the need to be appreciated and to be cared for as an essential dimension of the prison environment. Though prisoners may come across staff who are

328 caring and empathic, there is no substitution for the love of one‘s family. Indeed prisoners touted their contact with families as helping them to cope emotionally with the pain of incarceration and no prisoner interviewed in this study attributed bad behavior to missing family or freedom. The exploratory multivariate analysis did find that prisoners who experienced these deprivations as particularly difficult were more apt to be involved in serious misbehavior and violence. Many of the staff interviewed discussed how bad news about one‘s family often affected prisoners‘ moods and their ability to cope with daily prison issues. Prisoners also discussed how the competition for telephone access to families was a frequent source of conflict among prisoners. But since most prisoners miss their freedom and family, one must ask whether the difficulty experienced by those involved in misbehavior is derived from their inability to cope with the deprivations, the circumstances of their family relationships (not seeing family because in segregation; families are particularly dysfunctional; families less likely to maintain contact), or a combination of both.

The current study found over half of the prisoners also rated lack of privacy as being particularly difficult though there was no association between privacy and misbehavior. Lack of privacy was only brought up by a fifth of prisoners in Zamble and Porporino‘s study (1988) and

Johnson (2002) included the need for privacy and peace as one of seven ecological dimensions that prisons needed to address.

After missing freedom and family and the lack of privacy, there was a group of prison hardships that about one-third to one-half of prisoners experienced as particularly difficult.

These included Sykes‘ (1958) lack of autonomy and lack of goods and services, along with those added by others including regrets about the past, concerns about the future, boredom, and those difficulties that one may categorize as conditions of confinement (quality of food and medical

329 care, excessive noise, eating environment, and overcrowded conditions). However, in this study only two were found to be associated with greater levels of misbehavior: lack of goods and services and boredom. Specifically in this study, prisoners who found missing possessions and boredom to be particularly difficult to deal with were also more likely to be involved in disciplinary issues. It is unclear again whether prisoners who are likely to rate missing their possessions as particularly difficult have poorer resources with which to cope generally with the deprivations of prison or if this is reflective of the characteristics of those who misbehave in

Rhode Island—those who are young. That is, do prisoners who misbehave have a deficit that makes living without their ―things‖ more difficult than others or is it that those who misbehave are young and thus more likely to value material goods like their cell phone and MP3 player?

As mentioned, a high degree of difficulty with boredom was associated with a higher annual misconduct score, but also was one of only two hardships associated with high levels of violence alone. Both staff and prisoners discussed at length the consequences of boredom in prison, including: too much time to dwell on one‘s current and potential problems; rumination about negative past events; too much time to think about and carry out acts of misbehavior and violence; lack of opportunity to engage in constructive activities that might result in increasing prisoners‘ self-esteem; and lack of participation in activities that might otherwise improve the lives of prisoners (earning canteen money; becoming educated; addressing personal problems).

Closely related to boredom were the mental and physical activities in which prisoners participated, another ecological dimension mentioned by Johnson (2002) and brought up by a small percentage of prisoners in Zamble & Porporino‘s study (1988). In the current study, prisoners who spent a greater percentage of time idle in their cell and who did not work were more likely to be involved in misconduct.

330

The least difficult prison hardships for prisoners in the entire sample to endure were concerns about their safety, conflicts with prisoners and staff, and following prison rules.

Zamble and Porporino (1988) had similar findings in regards to concern about safety, though slightly more prisoners in their study cited conflicts with other prisoners as a common problem.

As mentioned previously, two-thirds of the current study‘s sample either felt safe or very safe and over three-fourths were not worried about being attacked and felt the chance of attack was not high. This was so despite the fact that almost half of the sample had listed enemies in the

RIDOC database and almost half had been physically hurt in some way by other prisoners.

Nevertheless, the lack of safety and fear of victimization have been the subjects of many studies and have been linked to psychophysiological disturbances for inmates (McCorkle, 1993); a decrease in general well-being (Maitland & Sluder, 1996); and the incidence of post-traumatic stress and depression (Hochstetler, et al., 2004). Though not rated as very difficult by most prisoners, concerns about safety, conflicts with prisoners and staff, and following rules differentiated the low-level from the high-level misbehavior groups in this study. Those in the high disciplinary group were much more likely to have rated conflicts with prisoners and staff and following rules as much more difficult to deal with. They were also likely to have reported being concerned about their safety and were much more likely to fear victimization and believed their chances of being attacked were much higher than did those in the lower disciplinary group.

While both groups reported the same number of enemies and similar levels of actual victimization by other prisoners, high misbehaving prisoners were significantly more likely than low misbehaving ones to report victimization by staff. In the multivariate exploratory analyses those prisoners who rated conflicts with staff and concerns about their safety as being very difficult were also more likely to have higher annual misconduct scores. Also an increased fear

331 of victimization and specifically having been hurt by staff were associated with a higher annual misconduct score. The only other hardship (besides boredom) that was associated with a greater level of violence was an increased concern for one‘s safety.

Focus on Fear and Respect

Indeed, fear was at the epicenter of many prisoners‘ problems with misbehavior. Safety concerns arose out of prisoners‘ involvement with gambling and gambling debts, being strong- armed, and peer pressure to demonstrate toughness. It was also tied to prior activity on the street, either due to participation in criminal activity involving other prisoners or their families as victims or as a result of neighborhood turf or gang activities. Fear was also exacerbated by dynamic personal predictors such as anxiety, being mentally ill, and being weak and unable to cope with potential danger. Fear also resulted from negative experiences with staff, whether they included threats or actual injury. Those who had a specific reason to be concerned about their safety (e.g. prisoners with heinous charges, notorious cases, snitches, or those who had given state‘s evidence against others) had a higher likelihood of either preemptive strikes to avoid being attacked, defending oneself in attacks, or becoming involved in serious and/or bizarre behavior to ensure removal from the general population. Their fears were fueled by constant threats of attack as well as actual attacks. As a matter of fact research on bullying in prison showed that over half of prisoners agreed that informers ―deserve to be bullied;‖ while three- fourths felt sex offenders deserve bullying also (O‘Donnell & Edgar, 2010). Thus it is not surprising that many prisoners experience fear given, as Sykes has pointed out that ―the individual prisoner is thrown into prolonged intimacy with other men who in many cases have a long history of violent, aggressive behavior‖ (1958, p. 77). However, more recent research on victimization and fear of victimization has found that the less frequent but more spectacular acts

332 of prison violence (murder, serious assault, and sexual assault) have overshadowed the perniciousness of the more mundane but routine victimization in prison and its effects

(O‘Donnell & Edgar, 1998; Wolff & Shi, 2009). O‘Donnell and Edgar found that the routine victimization ―shapes the social ethos of prisons and young offender institutions. The potential for assault, theft and verbal abuse grinds down prisoners and shifts their attitudes about the boundaries of acceptable behavior‖ (1998, p. 277). This is also not unlike what happens in those neighborhoods where street culture reigns. While Anderson (1999) does discuss how some violent altercations on the street do lead to death, it is more the constant threat of verbal abuse, theft, and simple assault that shapes the attitudes and ethics of those who live by the street culture and weigh down on those ―decent‖ folks who must constantly navigate an unpredictable course. Thus the pervasiveness of a street culture or worse, a culture of violence in prison affects all prisoners—even those who ―just want to do their time‖ but who are sometimes ensnared into

―doing business‖ in reaction to assaults, threats, or repetitive verbal abuse.

Closely associated with fear was the prisoners‘ campaign for respect. If fear is at the epicenter of prisoners‘ problems with misbehavior, disrespect is the cloud that hangs over all prisoner and staff interaction. Thus while many prisoners have fears for their safety, they cannot show any weakness by going to staff for help, backing down in front of others, or using precautionary behavior that would label them as fearful and earn them disrespect. This is similar to what happens in a street culture, where youths might be fearful, but dare not show it, choosing instead to act tough to maintain their reputation and ensure the respect of their peers (Anderson,

1999). However, the campaign for respect is not only associated with street culture, but also with age. As mentioned previously, McCorkle (1992) found that age was the best predictor of aggressive precautionary behavior. Interview data bore this out with prisoners more likely to

333 lash out at the smallest insult by staff or other prisoners rather than chance losing disrespect among their peers and being labeled as weak. One can therefore see how fear, the campaign for maintaining respect, and the ever present peer pressure among young prisoners result in misbehavior that is often erratic, over the top, and sometimes preemptive.

*****

In the end one must ask whether prison hardships contribute to prison misbehavior and violence. One set of variables found to be associated with more misbehavior both in the bivariate and multivariate analysis was tautological—it is not surprising that prisoners highly concerned about conflicts with staff and other prisoners and about following rules would be the ones that become involved in serious misbehavior and violence. However, two other sets of variables were more puzzling. One set was found to be correlated with a high annual misconduct score in the multivariate analysis (missing family, possessions, and freedom) and another to be correlated with being in a low or high-level misbehavior group in the bivariate analysis (concerns about the future, again missing possessions, autonomy, excessive noise, and cleanliness). Even when those living in segregation were eliminated from the sample, prisoners with serious misbehavior and violence issues were more likely to be concerned about their future, to miss their possessions and autonomy, and to experience greater difficulty with noise and cleanliness.

Interviews with prisoners did not reveal any reasons why prisoners with disciplinary issues should miss their families or even their freedom more than other prisoners. Nor did those prisoners have less autonomy and more opportunities for exposure to more noise and unclean conditions. Prisoner interviews did not for the most part reveal any connection between these types of hardships and disciplinary problems. Perhaps those with other predictors of misbehavior, those who are angry, those with mental illness, those with poor coping skills, those

334 who are young, those who have been through the prison experience previously are more likely to have difficulty with the hardships experienced by all prisoners. However, two hardships, fear and boredom, were found to be associated with serious prison misbehavior and violence in both the bivariate analysis, the multivariate analysis, and in qualitative interviews. Peer pressure and young prisoners‘ drive for respect complicate prisoners‘ attempts to stay safe. Both fear and boredom are pernicious in their ability to prevent prisoners from following a pro-social path that might lead to better lives but they are also dynamic predictors over which prison administrators have some control.

PRISONER ACTIVITIES Prisoner activities refer to anything that prisoners do while incarcerated. There were a number of prisoner activities that were correlated to serious misconduct and violence, including what prisoners do with their free time, and work. Prisoners in both the high- and low-level disciplinary groups spent the greatest percentage of their time watching television, listening to the radio, and socializing with friends and acquaintances. Both groups spent much less time in constructive activities such as reading, going to school, working out or involved in sporting events, and participating in programs. The activities that clearly differentiated the two disciplinary groups was working and being idle in one‘s cells. Overall prisoners involved in prison misconduct were more likely to spend time in their cell or in prisoner-initiated activities and less likely to spend their time working or in constructive prison-led activities. Huebner

(2003) discovered that ―remunerative controls‖ (paying prisoners to work) mediated the effects of traditional determinants of staff but not inmate assaults. Most recently Steiner and

Wooldredge (2008) found that prisoners who participated in programming and who worked more hours were less likely to be involved in prison assaults, substance abuse, and other types of

335 prison misconduct. These findings are relevant for all prisoners but especially for young prisoners who choose to mimic their street-life existence while in prison, spending their time hanging out in the yard, at most shooting hoops, but focusing more on the behavior of others

(prisoners, staffs, and visits) with little regard for their own self-improvement.

Wright (1991) conducted a study of prison adjustment that measured the effects of individual characteristics along with environmental differences and the congruence of the two.

One of his major findings was that increased activity and more support for self-improvement programs resulted in reductions in disruptive behavior. It was also one of six recommendations by the Commission on Safety and Abuse in America‘s Prisons to prevent prison violence

(Gibbons, & Katzenbach, 2006). Specifically the Commission recommended the promotion of productivity and rehabilitation and that prisons should ―invest in programs that are proven to reduce violence and to change behavior over the long term‖ (2006, p. 12). Studies that examined the effects of specific prison programming on misconduct are somewhat sparse. A number of studies conducted prior to 2001 that examined the effects of educational, vocational, or drug treatment programs on misconduct or the prevalence of programming in general yielded mixed results (Lahm, 2009) with some finding that programming reduced misconduct and others that it did not. However, more recent research has yielded somewhat more promising results. Lahm

(2009) found that prisoners involved in college education programs were less likely to receive dreports compared to prisoners taking classes in other levels of education. Camp, Daggett,

Kwon, and Klein-Saffran (2008) studied the Life Connections Program (LCP), which is a formal faith-based program operating within the Bureau of Prisons that uses cognitive behavioral therapy, promotes pro-social values and morals, and develops social support. They found evidence that program participation in LCP did lower prisoners‘ likelihood of being involved in

336 more serious misconduct, but not in less serious misconduct. Similarly, Welsh and his colleagues (2007) found evidence that participation in therapeutic community drug treatment programs in five Pennsylvania prisons lowered prisoners‘ involvement in all three levels of misconduct compared to those in a comparison group. However, their repeated measures effect

(time) was not itself significant indicating that the misconduct of those in treatment was not reduced at a greater rate over time than the comparison group. In addition prisoners who served more time in prison after treatment were more likely to be involved in misconduct. Both findings point to the influence of time yet to be served on treatment efficacy and its lasting effects on misconduct. Nevertheless, French and Gendreau (2006) conducted a meta-analysis on the effects of 105 prison-based treatment programs on misconduct. They found that participation in behavioral treatment programs had produced reductions in misconduct by about 26 percent and more effectively influenced misconduct than non-behavioral programs and education.

Furthermore they discovered an important correlation between the ―degree to which prison programs reduced misconducts and lower recidivism rates reported in the community‖ indicating that prison misconduct might be a proxy for future anti-social behavior in the community.

Although lowering serious misconduct is not the initial impetus for offering treatment and education to prisoners, it might prove to be a beneficial by-product.

STAFF SKILLS Another set of dynamic predictors that affect serious misconduct and violence is staff skills. In Bottoms‘ (1999) theoretical and interactive model for maintaining good order and good behavior in prisons, his category of ―staff deployment, approaches, and skills‖ act as a mediating force between many his other predictor categories and the outcome of good behavior and order.

In the current model of static and dynamic predictors on serious misconduct and violence, staff

337 skills similarly either attenuate or exacerbate the effects of three predictor categories: static personal predictors; dynamic personal predictors; and prison hardships, especially prisoners‘ sense of fear.

Interviews with prisoners and staff highlighted a number of staff skills that contributed to a reduction in misconduct and violence. One of these was vigilance—the staff‘s ability to continuously observe and assess prisoner behavior and to carry out routine correctional practices that reduce contraband and uncover other prisoner misbehavior. Another staff skill was direct supervision. Few RIDOC staff are tucked away behind glass in their monitoring of prisoners.

Instead they regularly interact with prisoners in the housing units, common areas, and in the yard, which results in many more pairs of eyes monitoring the everyday interactions of prisoners.

Also staff often hold prisoners accountable for adhering to minor rules like tucking in one‘s shirt and making one‘s bed with the hope that prisoners‘ attention to these small issues will deflect their more serious misbehavior. Essentially this is the Broken Windows theory of policing

(Wilson & Kelling, 1982) being applied to prisons—addressing the minor violations; demonstrating that one cares about everything prisoners do; and hoping that this attention to the minutiae of prisoners‘ everyday lives will mitigate their involvement in serious misconduct.

Prisoners also discussed the importance of being able to talk to staff, especially mental health staff to share their concerns and help them solve problems. Prisoners spoke of instances when staff‘s positive intervention helped to redirect their reactions to stressors from negative ones to ones that were more socially acceptable. Also staff‘s willingness to help prisoners‘ address their everyday problems helped to reduce prisoners‘ stress, anger, and obsession over unsolvable problems lessening the likelihood of their resorting to misbehavior and violence.

338

However, the study also found deficits in staff skills that contributed to prisoner misbehavior and violence. The multivariate analyses found those prisoners who rated conflicts with staff as being very difficult to deal with were also more likely to have higher annual misconduct scores. In addition having been previously hurt by staff was associated with a higher annual misconduct score. Prisoners and staff complained that some staff, especially some correction officers, engaged in interaction with prisoners that was disrespectful, confrontative, and inappropriate. While these negative behaviors by staff would be disliked by all prisoners or even most humans, it is particularly abrasive to those who view the need for respect as essential in their lives—those who live by the street culture and those who are young. However, even though every prisoner brought up confrontative, disrespectful treatment by officers as a problem sometime during their interviews, only one had given it as a reason for misbehavior. Yet it clearly raises the tension of the prison environment. Some staff who observed poor staff interaction believed some staff to be directly responsible for some prisoner misbehavior, but more often they found indirect links to misbehavior. Poor staff interaction causes prisoners to label all staff as unhelpful, unresponsive, and unfair and does not provide prisoners with good role models to emulate. When staff members denigrate prisoners, ignore their needs, and treat them as less than human, it contributes to a negative, tense, and dangerous prison climate.

Ultimately the ability of staff to communicate effectively, respond positively, and respectfully address prisoners is about public safety, both in the prison and eventually, on the street.

LIMITATIONS OF THE RESEARCH

Limitations of the current research study fall into two categories: operationalization of the variables and research methodologies.

339

LIMITATIONS ASSOCIATED WITH THE OPERATIONALIZATION OF THE VARIABLES Using dreports as the measure of serious prison misconduct and violence likely underestimated the level of misconduct and only measured that misconduct where the prisoner was caught and the staff member assesses the behavior as warranting a disciplinary report. Thus misconduct not caught or caught but addressed in a more informal manner might have been excluded from the current analysis. Although this is a definite limitation, the researcher is particularly interested in focusing on the formal process by which prisoners are written up, brought to a disciplinary hearing, and sanctioned. In addition, since this study is limited to mostly misbehavior that is considered very serious or violent, it is less likely that staff will choose to address this type of misbehavior in an informal manner and also less likely that it will go undetected.

The psychological literature does not address any possible limitations in either the instruments to measure coping skills or the instruments to measure emotions. However, both instruments rely on words that might have different meanings to different people. Although this is true of any self-report instrument, one‘s definition of sadness might be much more or less severe than another‘s definition of sadness. Also the operationalization of both coping and negative emotions involves the use of state or dispositional instruments. This means that both measure how people generally feel or generally act.

Prisoners who are faced with life-or-death decisions, extremely stressful situations, or events that restrict one‘s options might not react in the same way as they might in other situations. However, these instruments have been used in myriad studies to assess dispositional coping and emotions and given the circumstances, are the best that are available.

Access to actual mental health diagnoses of prisoners was not given so instead a proxy for mental health problems was used. This included having previously received psychiatric treatment for

340 mental health problems or receiving such treatment while incarcerated. Thus the study cannot distinguish between different forms of mental illness nor on the intensity of the problems.

LIMITATIONS ASSOCIATED WITH THE RESEARCH DESIGN There are research limitations associated with the general design of the research. One limitation is that the study did not include other variables such as differences in the physical structures of each of the four prisons; differences between the staffing of each prison (e.g. how closely the staff resembles the prisoner population); or any variables that have to do with the management philosophy or management style of the RIDOC or the individual prisons. It also did not account for other possible causes of serious prison misconduct and violence such as lack of self-control, impulsivity, head injury, or prior abuse as a child.

Other limitations have to do with the use of self-reports and in-depth interviewing of prisoners and staff. All depended on the truthfulness of those being interviewed, thus putting the validity of the study at stake. In the instructions regarding the survey and their participation, it was reiterated that truthfulness was very important and that respondents should not answer any question that they did not want to answer truthfully nor try to determine what answer would best serve the study‘s outcome.

Another limitation is that the sampling methodology for the in-depth interviewing was purposive and thus not random. Those prisoners who were willing to participate in an in-depth interview reported so at the end of the survey and thus made up the list of potential interview participants. Care was taken to attain a somewhat diverse sample of prisoners who had been involved in serious prison misconduct and violence. The prisoners were diverse in terms of age, race, facility, and the categories most associated with misbehavior: gang involvement; chronic substance abuse; prior adjudication as a juvenile delinquent; mental illness; and age. The sample might be somewhat biased because it might not have included some types of prisoners who did not volunteer for the interviews and overall might

341 not be generalizable to the entire population of prisoners who are involved in serious prison misconduct and violence in the RIDOC.

Another limitation might have resulted from the process of administering the surveys. The surveys were administered individually to a few prisoners, but mostly in groups of from three to thirty prisoners, depending on the level of security, the number of prisoners to be interviewed from a unit or housing area, and the recommendations of staff facilitating the process. There is some question as to whether prisoners answered differently if they had been in a group of five vs. a group of fifteen. There is also the possibility of within-group contamination—that is did any of the participants make comments that might have affected the participation or answers of the others in the group? The researcher was careful to explain the project, the confidentiality, and the parameters of participation and in the same exact manner to each group of participants.

Finally, further analysis might have been conducted that would have allowed for more in- depth interpretation of the study‘s results. For example, further analysis could help explain the relationship between ways of coping and the trait emotions. In addition, the study only included dispositional coping and trait emotions. How prisoners react to specific stressful situations (situational coping and state emotion) might be different than how they generally cope and feel.

342

X. FUTURE RESEARCH AND POLICY RECOMMENDATIONS

The focus on dynamic and static predictors of serious misconduct and violence will hopefully begin to address the disjunction between prison practice that ―locates trouble primarily in the dispositions of individual prisoners‖ and penologists who also find trouble rooted in the prison system (Bottoms, 1999, p. 205). Indeed Bottoms has proposed that ―One reason for such a disjunction is, perhaps, that prison scholars have failed adequately to develop socially contextualized accounts of prison violence that make real connections to the lived daily experience of prison administrators. Researchers speak of ―interactionist‖ approaches, but they have rarely addressed the minutiae of the average prison day, or considered in detail how violence can arise within this social order‖ (Bottoms, 1999, p. 205). Hopefully the recommendations for both future research and for policy recommendations listed below offer ideas that are practical and realistic.

FUTURE RESEARCH

PRISON HARDSHIPS Prisoners have to cope with many prison hardships during their incarceration. More research needs to be conducted on the relationship between prison hardships and prison misbehavior, particularly those that have been found to be correlated with misbehavior such as missing freedom, family, and possessions. Do troublesome prisoners have more difficult relationships with their families or is it their inability to cope? Has increased contact exacerbated or mitigated missing one‘s family? Why do prisoners who are likely to be involved in misbehavior find it more difficult to miss their freedom or their possessions?

343

Though the connection between fear and misbehavior and also boredom and misbehavior are more understandable, more research needs to be undertaken to understand the circumstances under which fear and boredom lead to serious misconduct and violence. Though quantitative research methodologies would help in untangling the paths from these variables to misconduct, qualitative methods will be essential to unraveling the underlying dynamics of fear, which is an especially complex emotion.

COPING AND TRAIT EMOTIONS This study found that prisoners‘ ways of coping do affect their level of involvement in serious prison misconduct and violence. However, more research needs to be carried out on this topic. In particular, research could focus on the congruence between prisoner characteristics and coping, especially in helping to delineate those relationships where the introduction of the personal predictors caused the relationship between ways of coping and serious misconduct to disappear. What are the deficits in coping for different types of prisoners (vulnerable, mentally ill, street culture, substance abusers, long-termers) and how do these coping deficits specifically contribute to serious misconduct and violence? In addition, more research needs to be conducted on the relationship between ways of coping and prisoners‘ trait emotions. Finally, research could also examine the relationship between institutional characteristics (security level, level of structure, activity level) and how ways of coping affect disruptive and violent behavior.

INFLUENCE OF LOCAL STREET CULTURE AND GANGS The current research led to the tentative conclusion that prisoners might be bringing their street culture values into the prison based on feedback from prisoners, observations by staff, and the quantitative analysis that showed that several local street subcultural variables were strong predictors of serious prison misconduct and violence. However, certainly more qualitative

344 research would have to be carried out simultaneously in the prison and in the local neighborhoods from which street subculture is derived. This would be a huge undertaking but one that might be valuable in understanding the transference of values and behavior between the streets and prison. Research also needs to be conducted on the relationship between prisoners committed to the street culture values and that of gang members. Are most prisoners who adhere to street culture values also gang members? Certainly not every state prison system‘s gang problem is similar; Rhode Island is not California. In areas where gangs have not taken over prison systems, what are the differences between gang members and those involved in lesser turf issues and what are indicators that might differentiate these two groups, if in fact, they are different? Finally, what do states need to do to prevent gangs from proliferating and taking more and more control of the prison environment?

PROGRAMS AND PRISONER ACTIVITIES More research needs to be conducted on the differential effects of treatment programming on the prevalence of misconduct. In particular, what kinds of programming reduce the incidence of serious misconduct and violence? What types of prisoners are more apt to benefit from these kinds of programs? Finally what skills do prisoners acquire in these programs and what else might happen in these programs that eventually leads to decreased misconduct? Is it, as has been suggested by Johnson (2002), that prisoners involved in programming increase their individual coping skills and in turn, their self-esteem, which leads eventually to mature coping and lower levels of serious misconduct? For example, does participation in a college program increase prisoners‘ abilities to critically think and communicate and lead to an increase in self-esteem and the decision to be more responsible for one‘s actions? Even though the reduction of serious

345 misconduct and violence is not the primary reason for providing treatment programming to prisoners, it might be a valuable by-product.

In addition to treatment programs that provide prisoners with skills and knowledge, there are other activities that might benefit them, such as work, involvement in sports, and other types of prisoner activities (e.g. Toastmasters, veterans‘ groups, book clubs). Research could be undertaken on whether prisoners with misbehavior are less likely to try to obtain prison employment or less likely to be hired and more likely to be fired than other prisoners. Since prison jobs can range from providing prisoners with from two to sixty hours of work per week, researchers could try to determine the optimal number of hours for prisoners to work that would keep them busy enough, give prisoners some money, and leave enough time for other programming that might lead to other types of self-improvement. Finally, what other types of activities are beneficial to prisoners and what benefits (team work, communication skills, chances to practice coping skills, involvement in a group, development of a niche) might they provide beyond keeping prisoners constructively busy?

POLICY IMPLICATIONS

This research has generated a great many ideas for improving coping skills and addressing serious misconduct and violence. Fascinatingly, the ideas from staff and prisoners for both increasing coping skills and reducing serious prison misconduct were nearly identical. The top three were: increasing programming and constructive activities for prisoners; increasing staff to help prisoners solve their problems; and decreasing the disrespect that staff directed at prisoners. Still other ideas for changes in policy emerged from the data itself and were also influenced by others‘ research on improving the conditions of confinement, and especially the incidence of violence in prisons. The policy recommendations below are those that were

346 supported by both the qualitative and quantitative data. Thus, specific recommendations about some staff issues, which were derived mostly from the qualitative data, were not included.

INCREASING TREATMENT PROGRAMMING AND ACTIVITIES Most interviewed staff and prisoners suggested more types of programming, more slots of existing programming, more education, and more constructive activities for prisoners. They believed that cognitive behavioral programming would increase prisoners‘ coping abilities, and lessen idle time, both contributing to a decrease in serious misconduct. Indeed other researchers and outsiders who have been commissioned to study serious prison misconduct and violence have come to similar conclusions (Wright, 1991; French & Gendreau, 2006; Gibbons &

Katzenbach, 2006; Camp et al., 2008; Lahm, 2009). Recommendations are divided into those regarding treatment programming as well as for other types of constructive activities.

INCREASING TREATMENT PROGRAMMING Treatment programs are defined here as any program and initiative that purports to help prisoners improve themselves in addition to the benefits that might be derived from being involved in a constructive activity that keeps prisoners busy. Most staff and prisoners believed that the RIDOC needed to increase the level of treatment programming for prisoners. Staff thought that treatment programming, especially that which used cognitive behavioral therapy, would help prisoners change their criminogenic thinking patterns and teach them better ways to cope. Many of the interviewed prisoners who had participated in programs had divulged that the programs had taught them how to stop and reflect on problems, techniques for reducing stress, and alternatives for addressing prison problems. Most of the prisoners praised the programs in which they participated, explaining how they had helped them cope and change their outlook on

347 their incarceration. Prisoners wanted more programs that addressed substance abuse, gave them vocational skills, and taught them how to improve their communication skills.

The RIDOC does offer a number of vocational and academic educational programs, and offers a substantial variety of rehabilitative programs in its facilities geared to various individual problems (substance abuse, suicide prevention, sexual abuse), offender types (murderers‘ counseling, domestic violence, and sex offender therapy), and other treatment/rehabilitative needs (cognitive behavioral groups, AIDS education and counseling, anger management, meditation, and various religious services). However, there are not enough existing slots to address the demand for programming by prisoners or their need. With the exception of the new programming for youthful offenders who have encountered disciplinary problems at Maximum, program participants are generally selected by potential release date—that is prisoners who are closer to release have participation priority over longer-term prisoners. While it might make sense to apportion access to educational and vocational programs (and possibly substance abuse programs) on this basis, cognitive behavioral therapy-based programs serve to target criminogenic thinking and improve prisoners‘ coping skills—key issues to reducing violence in prison and aiding prisoners in their adjustment. Therefore, it would probably be important for the RIDOC to not only increase the number of program slots, but to also rethink selection criteria in order to address prisoners who are at high risk to seriously misbehave in prison.

Andrews and Bonta (1998) found that programs that focused on higher risk offenders were five times more effective in reducing recidivism than were programs that focused on lower risk offenders. Thus program selection that involves waiting for the prisoner to be close to release or alternatively to display ―good behavior‖ might be counterproductive. Prisoners should be selected for these types of programming based on need. In addition, programs should adhere

348 as much as possible to the Principles of Effective Intervention developed by Gendreau and

Goggin (2000), including treatment that is designed to match key offender characteristics and learning styles, what they refer to as the responsivity principle. It should also be multi-modal, in that it offers a variety of interventions to address a range of offender needs, particularly criminogenic needs. A wide variety of programs helps prisoners build autonomy by letting them find programs in which they feel comfortable and where they can develop a niche. Such innovative programs as the aforementioned Life Connections Program in the Federal Bureau of

Prison (Camp, et al., 2008) or the drama-based program in Canada called ―Insult to Injury‖ that has been found to reduce anger aggression (Blacker, Watson, & Beech, 2008) would increase the variety of program modalities and possibly the efficacy of overall programming. In addition the intensity of treatment should involve at least 100 hours of direct service over a three to four month period, but also include an on-going aftercare component that is less intensive but that helps reinforce what was learned during the initial treatment phase. Needless to say, all of these principles must occur within programming where there is a high quality of program implementation, staff, training as well as knowledge dissemination to line staff and participation of staff in program decisions. Admittedly this is a tall order for prisons at any time, but is especially difficult when there are fiscal challenges. Nevertheless if treatment results in the reduction of serious misconduct and recidivism, money invested up front would eventually pay off in that prisoners would serve less time if not involved in serious misconduct and might be less likely to return to prison.

INCREASING OTHER PRISONER ACTIVITIES In addition to increasing treatment programming, prisons could increase work opportunities and other types of activities for prisoners. Work opportunities provide prisoners

349 with more than money. They keep prisoners constructively occupied and provide them opportunities for practicing responsibility, being prompt, taking orders, proper communication, and sometimes being part of a team. For prisoners who have hardly worked and for those who have not been successful at it, these are important opportunities to learn how to be part of the work force. Currently, work is considered a privilege that prisoners must earn. However, like treatment programming, it is likely that giving work opportunities to those who are at the highest-risk might produce the greatest benefits. Perhaps there could be a formal hierarchy of jobs that prisoners could work toward, culminating in the prison industries. Either way, it would be optimal to create more jobs at all levels, perhaps splitting into two those jobs in the kitchen or other areas that currently involve the longest number of hours. In addition, prisons could become creative, providing prisoners with jobs that would aide in the provision of other prison activities. For example, a prisoner could be hired by the library to do most of the work to operate several book clubs. While the prisoner would not be responsible for choosing the prisoners to be involved, he could be responsible for writing drafts of letters for the librarian to bookstores soliciting books, creating and posting flyers announcing book club meetings, maintaining waiting lists for book clubs, selecting books to read, and ensuring that attendance sheets are turned into the librarian. This would provide one prisoner with one job, but might also provide a pro-social activity that would be prohibitive for staff to undertake on their own.

Finally, each department in the prison could be asked to develop at least one new inmate job that might ease the burden of staff and hopefully engage prisoners in constructive activities.

Many prisoners discussed the crushing boredom they experienced, especially those serving long sentences. Not all of these prisoners require education or expensive treatment programming, yet helping them fill the void with pro-social and constructive activities would

350 help set the tone for whole prisons, providing young prisoners with a role model for how to do one‘s time. While some of these types of activities do currently exist, especially at Moran, more need to be developed at Maximum, at Price, and at the HSC. Though programs would need to be tailored to the level of security and though staff monitoring would need to be shifted, it is preferable to have prisoners involved enthusiastically in constructive activities that are controlled by the prison, than in unsupervised social engagement that mimics street culture activities.

PRISONERS WITH MENTAL HEALTH PROBLEMS This section provides recommendations for treating prisoners with mental health problems who seriously misbehave differently than other prisoners who similarly misbehave.

Indeed, many staff, especially administrators believed that prisoners with both mental health and behavioral problems needed extra care, stricter monitoring, and a less punitive approach. Thus there was a feeling that current practice in dealing with these ―special care inmates‖ was not only too harsh but also ineffective. Another reason for changing approaches is costs. Lovell and

Jemelka conducted a cost analysis of infractions at a medium security prison in Washington and found each infraction roughly costs $970 to cover the cost of writing up the offense, investigation, holding disciplinary boards, hearing appeals, carrying out sanctions, maintaining electronic records of all these processes, loss of good time, and transfers to higher security facilities (1996, pp.175-176). They extrapolated the costs to the entire prison system (11,000 prisoners) and estimated that major and minor infractions cost the state over $9 million annually.

Since they had found that mental illness accounted for 41 percent of the infractions in that system, they concluded that the misbehavior by the mentally ill costs $3.5 million annually.

They acknowledge that many of the costs associated with the disciplinary and sanctioning process are fixed and thus in the short run could not be reduced. However, they argue that

351 providing in-house treatment programs for psychologically disturbed prisoners would, in the long run, reduce prison costs by reducing their time incarcerated, by removing them from expensive isolation units, and by possibly reducing their recidivism. Recommendations for addressing the misbehavior of mentally ill prisoners have generally been divided into three categories: proper identification of this population, appropriate treatment options, and helping equip correction officers for the care of this population (Toch & Adams, 1986; Toch, et al., 1989;

Toch & Adams, 1994; Lovell, et al., 2000; Peck, Jr., 2004; Toch & Kupers, 2007; Lovell, 2008).

IDENTIFICATION OF THE MENTALLY ILL Obviously before one can propose to treat prisoners involved in serious misbehavior who are mentally ill, they need to be identified. Much has been written about the juxtaposition of mental illness and disruptiveness in prison and the myriad assessments and tools that correctional systems can use to identify this group of prisoners (Young, Justice, & Erdberg, 2004; Lovell,

2008; Scott, 2008; Adams & Ferrandino, 2008). In fact the RIDOC has within the last three years put into place procedures that will help identify prisoners whose mental health problems might cause them to harm themselves or others. This includes a policy that mandates that staff notify the shift commander or mental health services staff if they observe a prisoner whose emotional or psychiatric status would result in harm to themselves or others (RIDOC, 2007).

Mental health staff can then put prisoners on one of three observational levels, depending on their evaluation of the situation. In addition, individuals who enter the Intake Service Center

(ISC) are screened for mental health issues and suicidal ideations, which if identified trigger a more in-depth mental health assessment. The results from all of these procedures are now entered into the Electronic Medical Record (EMR)—an electronic database that provides access to the information by medical and mental health staff. As mentioned previously, there is already

352 in place a policy that allows staff who preside over the disciplinary process to notify the shift commander or mental health services staff that a prisoner who was charged with a booking might have a mental health status that should be examined (RIDOC, 2009).

Yet despite these laudable screening processes, it appears that the mental health problems of many prisoners caught up in a repetitive cycle of disruption and segregation remain undetected. Indeed, while the screening processes at the ISC might pick up on mental health problems of prisoners committed more recently, there are still many prisoners in population who could benefit from such detection. Yet identification at this stage depends on correctional staff having both the will and ability to do so. Unfortunately, non-mental health staff members do not have sufficient training about mental health behavior to even be able to judge whether or not a prisoner‘s behavior might be due to mental health issues so they can even make a referral. Thus all staff, especially correction officers, should be mandated over the course of a period of time to receive in-depth training on various diagnoses and associated symptoms of mental illness; training on how one might preliminarily identify prisoners with such problems in order to be able to make a referral; and tips on how to better communicate with prisoners with mental health issues. Staff who have regular direct contact with prisoners who have mental health issues (in units where they might be concentrated) should have additional training on how to communicate and best handle these special population prisoners.

In addition to training non-mental health staff, there needs to be increased collaboration between mental health staff and non-mental health staff. Though mental health staff might already be stretched thin, increased collaboration, though time-consuming in itself, might pay off in the long-term. Mental health staff cannot visit every prisoner with mental health problems every day and thus there needs to be more communication between mental health workers and

353 correctional line staff so that line staff can know how to deal with these prisoners in a way that supports, rather than ignores or at worst usurps mental health workers‘ decisions. After all it is correction officers who spend the bulk of time with all prisoners. Dvoskin and Spiers suggest that correction officers can be instrumental in the identification and treatment of mentally ill prisoners through four collaborative roles (2004, p. 41). The first they refer to as ―counseling and psychotherapy—talking with inmates‖ though they do not intend COs to be amateur psychotherapists, they do acknowledge that it is usually COs that are present when these prisoners face crises and how CO‘s react and communicate might determine the outcome of most crises. Thus COs are regarded as ―functional professionals‖ who need to be trained to ―calm down‖ these special care prisoner in times of crisis rather than exacerbate the situation. A second role for COs is ―consultation—talking about inmates‖ which means that COs can refer prisoners for mental health assessment, but can also serve as the eyes and ears of mental health workers since they spend so much more time with these prisoners. A third role for correction officers is to participate in the treatment regimes for prisoners assigned to special housing or activities or to behavioral programs. For example, a treatment regime might involve rewarding certain positive behaviors and implementing other measures to address negative behaviors.

Mental health staff cannot always be present to implement these types of regimes—instead correctional staff must keep track of the behaviors and possibly implement the agreed-upon responses. Finally, correction officers can be essential in helping mental health staff monitor reactions to changes in medication, noting behaviors that are unusual and serving as the eyes and ears of staff. Other researchers have summarized the benefits of collaboration between mental health staff and non-mental health staff, especially correction officers including: identifying violent and other disruptive patterns among prisoners and devising effective and workable

354 interventions (Dvoskin & Spiers, 2004; Toch & Kupers, 2007; Adams & Ferrandino, 2008). For example, if correction officers, especially lieutenants, had more training on the identification of prisoners with mental health problems, they would be able to appropriately refer prisoners they encounter in the disciplinary process to mental health staff for assessment, possibly resulting in treatment that could temper future disruptive behavior. As Toch and Kupers so aptly concluded,

―The challenge is one of helping staff to understand motives and dynamics in the violent behavior of (mentally ill) offenders so that they see that one can attempt to interrupt violent careers, as opposed to locking the offenders up, leaving their predispositions inviolate.

Clinicians can assist custody in this enterprise, assuming they are themselves willing to take risks by understanding, engaging, and reforming violent prisoners‖ (2007, p. 25).

TREATMENT OF DISRUPTIVE PRISONERS WITH MENTAL HEALTH ISSUES This study did not examine the treatment needs of prisoners with mental health issues.

However, in studying those prisoners who are involved in serious misconduct and violence, it became apparent that prisoners who are mentally ill make up a sub-group of this disruptive population. Many of the staff interviewed reiterated the need to provide more treatment to prisoners with mental health issues. Some staff suggested there was a need for more mental health workers or social workers and separate units in each facility where they could have their issues addressed. Needless to say it is not within the parameters of this study to detail recommendations for treatment. Adams and Ferrandino (2008) have written an excellent piece on the management of mentally ill prisoners, citing the tension between treatment and control, environment as therapy, the distinction between risks and stakes, and the use of medication and isolation to manage these special care prisoners. They and others also cite the increasing successful use of specialized units or therapeutic communities for the mentally ill (Toch &

355

Kupers, 2007; Kupers, et al., 2009). One example of a successful specialized unit was outlined by Kupers and his colleagues (2009). As a result of litigation in Mississippi, the Department of

Corrections under two federal consent decrees was mandated to reduce the population in administrative segregation in Unit 32, review the entire classification system, and remove prisoners from administrative segregation with serious mental illness and provide them with treatment. As a result, they created a step-down unit within Unit 32, separate from the administrative segregation housing, for prisoners with serious mental illness who still needed segregation and general population prisoners with serious mental illness. Prisoners on the upper tier remain segregated until they are ready to step down into the open unit. Treatment is modeled on the ―assertive community treatment approach,‖ which focuses less on one‘s mental illness and more on how participants can use their intact strengths, faculties, and ambitions to overcome their illness. Through group therapy and psychoeducation, they learn how to better cope with stresses, their anger, and their impulses and are given incentives, including more freedom as rewards for good behavior. Prisoners requiring inpatient psychiatric services were transferred to an inpatient psychiatric unit at another facility. Kupers reports that prisoners, custody staff, and mental health staff have all reported favorably on the step-down unit. Review of the disciplinary records of 43 prisoners who had completed and been successfully discharged from the step-down unit found that these prisoners‘ averaged 4.7 dreports six months before participation, 1.2 dreports during participation, and 0.6 in the six months after leaving the step-down unit (2009, p.

1046). Included among these graduates were a number of Security Threat Group leaders who had been involved regularly in violent and disruptive behavior. Although costs and other resources are certainly a consideration, a step-down unit with two phases could improve on the

RIDOC‘s current Observation Security Unit where prisoners with mental health problems are

356 observed and stabilized before transfer back to general population. Such a unit might also stop the cycle of disruption/isolation that other prisoners with both mental health and behavioral problems repeatedly experience.

ADDRESSING FEAR Interviews with staff and prisoners alike along with observations revealed that the practice of direct supervision of prisoners has certainly contributed to low levels of serious prison misconduct and violence in Rhode Island. In addition, staff‘s vigilance and the department‘s policy of SRG classification has added to the increased security, safety, and order of the prison. Yet prisoners involved in serious misbehavior and especially violence have expressed concerns of victimization both in the self-reported surveys and in interviews. Policies to address violence can be divided into three categories: policies that reduce an atmosphere of violence; those that address the individual concerns of prisoners; and those that address fears associated with staff use of physical force. The recommendations associated with these first two categories of policies are outlined below.

REDUCING AN ATMOSPHERE OF VIOLENCE While most prisoners were not too worried about being attacked, it was a tangible concern for young, mostly urban, offenders incarcerated for B & E/drug offenses and serving short sentences. Many of these prisoners had begun their criminal careers very early, often serving time in secure juvenile correctional facilities. Many had either been in gangs or involved in street criminal activity that had sometimes led to neighborhood turf issues. While some of these young prisoners could have had specific issues that increased their fears, there was a cycle of violence that was exacerbated by the campaign for ―respect‖ and the concomitant peer pressure to demonstrate toughness. Indeed staff at Maximum aptly picked up on this trend of

357 violence among newly-incarcerated young offenders and developed an initiative to address it.

They developed a closed-ended program carried out by the facility‘s staff and outside vendors to teach prisoners to think first and use alternative non-violent means to address problems and fears. Anecdotal evidence from both staff and prisoners of this new initiative has shown it to be helpful in addressing misbehavior among this population. Similar cognitive behavioral therapy initiatives have been shown to increase prisoners‘ mature coping skills (Zamble & Porporino,

1988; MacKenzie, et al., 1989; and McDonald, 2006). Unfortunately, many of the prisoners involved in serious misbehavior and violence had been exposed to violence and/or experienced victimization on the street, in juvenile correctional facilities, or while awaiting trial.

Interestingly, a number of these were youths who were incarcerated before the age of 18 and thus were placed in the HSC for protective custody reasons until they reached the age of 18 years old.

A few of them recounted how they had been placed in segregation within the HSC, thus being exposed to the most violent and dangerous prisoners classified to the HSC. Thus the anticipation of violence and victimization experienced by these young offenders in the street, in former juvenile correctional facilities, and in the HSC affects the social atmosphere of both the ISC and

Maximum. Indeed prisoners did recount instances, especially in the ISC, early on in their incarceration where they felt they had to either respond to threats and actual violence or preemptively ward off violence by practicing it themselves.

Thus the RIDOC should take steps that might reduce the atmosphere of violence created by this convergence of young, street-involved, repeat offenders. Prison administrators could take steps to set the tone at the beginning of young prisoners‘ incarceration. This could include paying particular attention (increased direct supervision and vigilance) to young offenders who are convicted and residing in the ISC or who are newly arrived at Maximum. Perhaps instead of

358 being mixed in with the oldest, most violent offenders in the P units when they enter Maximum, particularly young at-risk offenders could be placed in a housing unit for young offenders, though one might argue that putting them all together could backfire. Another option would be to designate one unit as an orientation unit for all new prisoners transferred to Maximum.

The incentive to remain dreport free by providing good-time for program participation and good behavior is particularly strong, though it seems that young prisoners, for some reason, have not understood this incentive in its entirety at the beginning of their sentences. Given that they are mostly serving short sentences, this could be a powerful incentive for good behavior.

Thus when prisoners go through the orientation to each prison, they should be given both written and verbal explanations of how the good-time policy works as well as visual examples of how good time can be accumulated through program participation and good behavior or decimated by misbehavior. At this orientation, special emphasis should be given to the importance of programs, not only for good time, but how it might help prisoners upon release. Though somewhat controversial, the prison could also use older prisoners to explain the importance of doing one‘s time, of taking advantage of the good time policy, and participating in programs.

Prisoners are more likely to listen to other prisoners who have experienced similar pressures rather than listening to staff. A small group of prisoners could be selected by staff, trained, and monitored to provide concrete examples of the consequences they experienced for negative behavior and the benefits of positive behavior. Just as Anderson (1999) observed that older black males who served as role models could lower adolescent violence in disadvantaged neighborhoods, so might older prisoners who grew up in the streets but who eventually renounced their commitment to street culture values do so in prison.

359

In addition, staff could assess the risk of young prisoners for serious misbehavior by predictors other than SRG status and prior incarceration. These could include prior multiple drug usage, prior treatment for mental health problems and/or actual screening for mental health issues, prior placement in a juvenile correctional facility, and possibly prior physical abuse. It appears that these predictors are cumulative in that prisoners who exhibit more of these predictors are also more likely to be involved in serious misconduct. This risk assessment would allow young prisoners to be identified for cognitive behavioral therapy close to entry into

Maximum instead of waiting for these prisoners to actually act out. Finally, staff should make a concerted effort to engage these prisoners in prison-initiated activities such as academic or vocational education (possibly mandating education until the attainment of a G.E.D.) that would steer them away from prisoner-initiated ones.

Finally prisons could try to redefine what it means to ―be a man.‖ As mentioned previously, young prisoners in general and those who grew up in neighborhoods where street culture dominated have skewed images of what it means to be a man. The identification with either hegemonic masculinities and/or street culture makes it all about respect, toughness, intimidation, preemption, and maintaining one‘s reputation and status. Perhaps that is what they learned and all they know. But prisons are given the opportunity to take and shape prisoners‘ thoughts over periods of months and even years. Through role modeling by staff, education, and possibly other initiatives that have been created for this very purpose, prisons could attempt to redefine masculinity as more than toughness; that it involves taking responsibility for one‘s actions and helping others, especially one‘s family members, succeed in life. Together all of these steps could set a more constructive tone for young prisoners as they begin their incarceration, hopefully reducing their likelihood of misbehavior and violence.

360

ADDRESSING PRISONERS’ INDIVIDUAL CONCERNS FOR SAFETY Although the majority of RIDOC prisoners reported that they felt relatively safe, there was a small number of prisoners who did not. In addition to reducing the atmosphere of violence, prison officials also need to address prisoners‘ individual concerns. Indeed, the

RIDOC does maintain the names of enemies in the database, taking into consideration prisoners‘ enemies before transfer of a prisoner from one facility to the next. There are also a small number of prisoners who are in protective custody status in the HSC. However, several prisoners interviewed complained that movements to protective custody or out-of-state were often not carried out until the prisoner had been shown to be at risk via multiple dreports or a serious attack. This was particularly true of prisoners who had given states‘ evidence against another prisoner in the courtroom or whose crime was particularly heinous. These prisoners recounted how they had needed to protect themselves from other prisoners in several instances, often resulting in their receipt of multiple dreports and reclassification to higher security before any staff took their requests for protection seriously. Perhaps a more proactive mechanism needs to be created for identifying those few prisoners who for whatever reason might not be able to live in general population at least at the beginning of their sentence or who need more careful monitoring to keep them safe. In addition, prisons can adopt policies that mandate that staff not publicize prisoners‘ crimes or other protective custody issues to other prisoners and train staff on the merits of abiding by that policy. Finally there were a few prisoners who felt very unsafe in population despite not having specific enemies that they could name. Instead they were generally fearful of prisoners that they might have come up against on the street, from rival gangs, other neighborhoods, or had been on the other side of their crimes. Perhaps prison administrators could address this more nebulous form of potential predatory behavior.

361

*****

This study focused on the predictors of serious prison misconduct and violence in the

Rhode Island Department of Correction. One must first acknowledge the fact that most prisoners involved in this study felt relatively safe and unlikely to be attacked. The RIDOC administration should take pride in the policies and practices that have led to this perception of safety by most prisoners, especially in contrast to other state prison systems where violence is endemic.

Nevertheless much was learned about the predictors of serious misconduct and violence in this prison system, including the context within which it is exacerbated or mitigated. First, some of prisoners‘ ways of coping did matter. Prisoners who learned to elicit both emotional and instrumental support from loved ones, fellow pro-social prisoners, and staff were less likely to be disruptive or violent. Those who relied on religion or coped through venting their emotions, bravado, and who charged into reacting to stressors were more likely to be disruptive and to misbehave. Second, trait emotions did affect misconduct. That is, when the personal predictors and coping were taken into consideration, prisoners who were angry were less likely to be involved in misconduct and violence, while those who were anxious were more likely to be involved in serious misconduct. Third, the quantitative and qualitative data yielded at least five categories of prisoners that are more apt to be involved in serious misconduct and violence: prisoners with mental health problems, young prisoners, weak prisoners, gang members, and those prisoners, usually more well-behaved, who fought back when victimized by others. After trying to categorize the 25 prisoners who were interviewed, it was determined that they all fit into at least one of these categories; however, two-thirds of them fit into multiple categories. In addition, many of them were repeat offenders, chronic substance abusers, and state-raised youths who had been incarcerated as juveniles indicating the complexities of addressing serious

362 misconduct and violence and the need to individualize responses to this very small percentage of prisoners who are responsible for the largest percentage of violence. To make matters worse, aggressiveness and violence among young prisoners may be exacerbated by the urban street culture from which many of these prisoners came and by adherence to values promoted by hegemonic masculinities in the prisons.

It is also recommended that a reconceptualization of predictors of serious prison misconduct and violence is in order. The importation/deprivation model is impractical and emulates the ―us vs. them‖ mentality of prisoners vs. staff/administration. Instead, predictors should be categorized by their usefulness; by whether they are static predictors to be used for classification and identification purposes or dynamic predictors about which prison administrators can do something to lessen disruption and violence in prison. Some of the dynamic predictors are personal predictors that prisons can address, such as prisoners‘ trait emotions, their ways of coping, and mental illness. However others are related to prison policies and practices and include reducing fear, increasing staff skills, deterring inappropriate staff behavior, and expanding treatment and non-treatment activities, as well as work opportunities.

The goal of prisons is ultimately public safety. Courts and correctional agencies should expect that prisons will do all they can to change the trajectories of offenders so that when they are released, they will not reoffend. Much of the effort should be focused on preventative measures: averting a culture of violence, addressing prisoners‘ needs and problems, and increasing their ability to cope in a positive manner. However, prison officials should make a particularly concerted effort to view the serious disruption and violence of a small percentage of prisoners as an opportunity to make a large impact on violence in prison and potentially on future crime and violence on the street. Prisoners who are perennially disruptive in prison and who

363 commit violence are indeed a captive audience, spending periods of time in and out of segregation and special housing. Prison administrators, correctional line staff, and treatment- oriented staff would find their jobs both more interesting and more rewarding if allowed to collaborate, analyze, and address the needs of these particularly difficult prisoners.

364

APPENDIX A – RIDOC DISCIPLINARY SEVERITY SCALE

365

366

367

368

369

370

371

APPENDIX B – PRISONER SELF-REPORT SURVEY

372

Self-Report Survey How Prisoners Cope with the Strains of Prison Please respond to each of the following items by circling the number for the answer that is best for you, using the response choices listed below. Please try to answer each item separately in your mind from each other item. Choose your answers thoughtfully, and make your answers as TRUE FOR YOU as you can. Please answer every item. There are no “right” or “wrong” answers, so choose the most accurate answer for YOU—not what you think “most people” would say or do or what you think is best for this study. Prison Stresses Listed below are some problems prisoners often face in prison. How hard has each of the following been for you? The responses range from “not hard” to “very hard.” Please circle the response that best matches with how hard each problem has been for you.

How hard has each of the following been for you? Not hard at all Very Hard a. missing family or friends 1 2 3 4 5 b. missing certain activities 1 2 3 4 5 c. conflicts with prisoners 1 2 3 4 5 d. regrets about the past 1 2 3 4 5

e. concerns about the future 1 2 3 4 5 f. missing personal possessions 1 2 3 4 5 g. boredom 1 2 3 4 5 h. lack of privacy 1 2 3 4 5

i. excessive noise 1 2 3 4 5 j. quality of medical care 1 2 3 4 5 k. missing freedom 1 2 3 4 5 l. conflicts with staff 1 2 3 4 5

373

How hard has each of the following been for you? Not hard at all Very Hard m. not being able to make my own decisions 1 2 3 4 5 n. quality of food 1 2 3 4 5 o. environment where we eat 1 2 3 4 5 p. cleanliness of the facility 1 2 3 4 5

q. following prison rules 1 2 3 4 5 r. overcrowded conditions 1 2 3 4 5 s. concerns about my safety 1 2 3 4 5

++++++++++++++++++++

Coping

There are lots of ways people respond when they confront difficult or stressful events during their prison sentence. This section asks you to answer what you generally do and feel when you experience stressful events. Of course different events bring out somewhat different responses, but think about what you usually do when you are under a lot of stress. Indicate what YOU usually do when YOU experience a stressful event in prison. I usually: 1= don’t do this at all 2=do this a little bit 3=do this a medium amount 4= do this a lot

1. I try to grow as a person as a result of the experience. 1 2 3 4

2. I turn to work on other activities to take my mind off things. 1 2 3 4

3. I get upset and let my emotions out. 1 2 3 4

4. I try to get advice from someone about what to do. 1 2 3 4

374

Indicate what YOU usually do when YOU experience a stressful event in prison. I usually: 1= don’t do this at all 2=do this a little bit 3=do this a medium amount 4= do this a lot

5. I concentrate my efforts on doing something about it. 1 2 3 4

6. I say to myself “this isn’t real.” 1 2 3 4

7. I put my trust in God. 1 2 3 4

8. I laugh about the situation. 1 2 3 4

9. I admit to myself that I can’t deal with it, and quit trying. 1 2 3 4

10. I restrain myself from doing anything too quickly. 1 2 3 4

11. I discuss my feelings with someone. 1 2 3 4

12. I get used to the idea that it happened. 1 2 3 4

13. I talk to someone to find out more about the situation. 1 2 3 4

14. I keep myself from getting distracted by other thoughts or activities. 1 2 3 4

15. I day dream about things other than this. 1 2 3 4

16. I get upset, and am really aware of it. 1 2 3 4

17. I seek God’s help. 1 2 3 4

18. I make a plan of action. 1 2 3 4

19. I make jokes about it. 1 2 3 4

20. I accept that this has happened and that it can’t be changed. 1 2 3 4

21. I hold off doing anything about it until the situation permits. 1 2 3 4

375

Indicate what YOU usually do when YOU experience a stressful event in prison. I usually: 1= don’t do this at all 2=do this a little bit 3=do this a medium amount 4= do this a lot

22. I try to get emotional support from friends or relatives. 1 2 3 4

23. I just give up trying to reach my goal. 1 2 3 4

24. I take additional action to try to get rid of the problem. 1 2 3 4

25. I refuse to believe that it has happened. 1 2 3 4

26. I let my feelings out. 1 2 3 4

27. I try to see it in a different light, to make it seem more positive. 1 2 3 4

28. I talk to someone who could do something concrete about the problem. 1 2 3 4 29. I sleep more than usual. 1 2 3 4

30. I try to come up with a strategy about what to do. 1 2 3 4

31. I focus on dealing with this problem, and if necessary, let other things slide a little. 1 2 3 4 32. I get sympathy and understanding from someone. 1 2 3 4

33. I kid around about it. 1 2 3 4

34. I give up the attempt to get what I want. 1 2 3 4

35. I look for something good in what is happening. 1 2 3 4

36. I think about how I might best handle the problem. 1 2 3 4

37. I pretend that it hasn’t really happened. 1 2 3 4

38. I make sure not to make matters worse by acting too soon. 1 2 3 4

376

Indicate what YOU usually do when YOU experience a stressful event in prison. I usually: 1= don’t do this at all 2=do this a little bit 3=do this a medium amount 4= do this a lot

39. I try hard to prevent other things from interfering with my efforts at dealing with this. 1 2 3 4 40. I watch TV or movies, to think about it less. 1 2 3 4

41. I accept the reality of the fact that it happened. 1 2 3 4

42. I ask people who have had similar experience what they did. 1 2 3 4

43. I feel a lot of emotional distress and I find myself expressing those feelings a lot. 1 2 3 4 44. I take direct action to get around the problem. 1 2 3 4

45. I try to find comfort in religion. 1 2 3 4

46. I force myself to wait for the right time to do something. 1 2 3 4

47. I make fun of the situation. 1 2 3 4

48. I reduce the amount of effort I’m putting into solving the problem. 1 2 3 4

49. I talk to someone about how I feel. 1 2 3 4

50. I learn to live with it. 1 2 3 4

51. I put aside other activities in order to concentrate on this. 1 2 3 4

52. I think hard about what steps to take. 1 2 3 4

53. I act as though it hasn’t even happened. 1 2 3 4

54. I do what has to be done, one step at a time. 1 2 3 4

55. I learn something from the experience. 1 2 3 4

56. I pray more than usual. 1 2 3 4

377

++++++++++++++++++++ Living in Prison Experiences

1. Aside from sleeping, how much time in the average week do you spend on the following activities? About how many hours do you spend each week on: a. work ______b. attending school or a training program ______c. visits and letter writing ______d. hanging around with friends, socializing ______e. group meetings (e.g. AA or NA or anger management groups) ______f. watching TV, listening to radio or music ______g. sports and games (e.g., cards), hobbies ______h. reading ______i. sitting in my cell doing nothing ______j. other things ______(Please specify______)

2. How many close friends do you have here? ______

3. About how many letters do you get in a typical month? ______

4. About how many visits do you get in a typical month? ______

5. Circle which of the following would best describe the way you live here?

a. Avoid trouble, keep busy and do what’s necessary to get out as soon as possible. b. Do something to improve or better myself while I’m in here. c. Forget the outside, learn how to operate in the institution and have as good a time as possible here. 6. During this sentence, have you been treated for psychiatric or emotional problems? (Circle all that apply.)

a. No b. Met individually with psychiatric staff. c. Group meetings with psychiatric staff d. Transferred to hospital/psychiatric unit

378

e. Received or am taking medication for psychiatric or emotional problems. f. Other ______7. Do you currently belong in a prison gang? YES NO

For the questions below, please circle the number that best corresponds to how you feel. a. How safe do you feel in this prison? Very Safe 1 2 3 4 Not Safe at All b. During this sentence, how much have you worried that you will be attacked? Very Little 1 2 3 A Great Deal c. Do you feel the chance of being attacked in this prison is…. Low 1 2 3 High ++++++++++++++++++++

Victimization Experiences During this sentence, please answer if other prisoners have tried to or done any of the following things to you. Then answer if staff have tried to or done any of the following things to you. Prisoners Staff a. used any sort of weapon on you YES NO YES NO b. hit, kicked, punched, or slapped you YES NO YES NO c. scratched you YES NO YES NO d. pushed or shoved you YES NO YES NO

e. bit you YES NO YES NO f. physically hurt you in any way YES NO YES NO g. made a threat that scared or worried you YES NO YES NO h. said that they would hurt you if you didn’t give them some of your property YES NO YES NO i. called you names or said mean things to you YES NO YES NO

379

Prisoners Staff j. played games with your mind YES NO YES NO k. forced sexual activity on you YES NO YES NO l. made sexual comments to you that made you uncomfortable YES NO YES NO

m. made you give them money/valuables YES NO YES NO n. stolen property that belonged to you YES NO YES NO o. disrespected you in front of others YES NO YES NO ++++++++++++++++++++ Background Information 1. Did you use drugs prior to this incarceration? YES NO 2. Please think about the 6 months before you entered the RIDOC? Indicate whether you used these substances during those 6 months. Please answer yes or no.

a. Alcohol (beer, wine, or hard liquor)? YES NO b. Tobacco (cigarettes, chewing tobacco, etc.)? YES NO c. Marijuana/hashish (pot, grass, reefer, or blunts)? YES NO d. Crack/powder cocaine? YES NO e. Heroin? YES NO f. Inhalants (paint thinner, glue, poppers)? YES NO g. Other______? YES NO

3. Did you finish high school or get your G.E.D. prior to this incarceration? YES NO

4. Were you employed during the three months prior to this incarceration? YES NO

5. Were you involved in an intimate relationship during the three months prior to this incarceration? YES NO

380

6. Had you ever been adjudicated delinquent and committed to a secure juvenile correctional facility? YES NO

7. Had you ever been sentenced to a prison or jail prior to this incarceration? YES NO

8. Before this incarceration, did you belong to a gang? YES NO

9. Before being sentenced to this incarceration, had you been treated for psychiatric or emotional problems? (Circle all that apply.) a. No b. Had one or more individual treatment sessions. c. Attended one or more group treatment sessions. d. Admitted to a hospital/psychiatric unit e. On medication for psychiatric or emotional problems. f. Other______

10. Have any of your family members (mother, father, brother, sister, other family who lived with you) been incarcerated for 30 days or longer? YES NO

11. Have any of your immediate family members ever been treated for a problem with drugs or alcohol? YES NO

12. Have any of your family members ever abused drugs or alcohol? YES NO

13. Have any of your family members been involved in a gang? YES NO

++++++++++++++++++++

381

++++++++++++++++++++ General Feelings (only one example of anxiety, anger, and depression are given) Listed below are a number of statements that people have used to describe themselves. Read each statement and then circle the appropriate number to the right of the statement to indicate how you generally feel. There are no right or wrong answers. Do not spend too much time on any one statement but circle the answer that seems to describe how you generally feel.

1= Almost never 2= Sometimes 3= Often 4= Almost always

1. I feel nervous and restless. 1 2 3 4

2. I am quick tempered. 1 2 3 4

3. I feel gloomy. 1 2 3 4

Thank You!

Would you be interested in taking part in a more in-depth individual interview about coping, the strains of prison and other prison experiences? ______Yes ______No

382

APPENDIX C – PRISONER INTERVIEW SCHEDULE

383

RIDOC Prisoner Interview

1. What is the most difficult part about being imprisoned? (Why, how, how often?)

2. Is that difficulty also the most stressful part of being imprisoned? (Why, why not?)

3. Is there anything that you do that helps you cope with [name the problem reported in the first question]?

4. What are the next two or three most difficult or stressful parts of being in prison? (difficult and/or stressful?)

5. What do you do to cope with each of [name them] those difficulties?

6. What events or problems have you had during this incarceration that have been particularly difficult for you to deal with? (When, where, why, how?)

7. How did you cope with each of those events or problems [one by one]?

8a, When you are confronted with a problem or stressful situation while in prison, do you talk to anyone for emotional support? (who? How likely? How often?)

b. For help or advice in how to solve the problem or deal with the situation? (who? How likely? How often?)

9. Are you more likely to try to solve problems or stressful situations or try to put them out of your mind and ignore them? What types of problems can be solved? What types can be ignored? What types have you just accepted as unsolvable and something you must accept?

10. Do you feel like you can solve problems and address stressful situations that you confront in prison? What type have you solved? What type have you found cannot be solved? What do you do with those?

11. Are there other prisoners who you would consider your friends? How many?

384

12. Have you ever relied on other prisoners to help ensure your safety? Have you ever helped to ensure the safety of other prisoners?

13. Have other prisoners ever helped you solve your problems or deal with stressful events? If yes, how much do you rely on other prisoners to help you through your prison experience?

14. Are you currently in a prison gang? If yes, has being in a gang helped you address problems and stressful situations? How?

15. How much do you rely on friends or relatives outside of prison to help you through your prison experience? (How help? Who rely on? How often?)

16. How much do you rely on RIDOC staff in prison to help you through your prison experience? (How help? What type of staff member? How often?)

17a. Do you have problems with mental health issues? How much has this been a factor in how you deal with problems or stressful situations? (Get details of how mental health has affected life in prison.) b. -Are you receiving any type of help for your mental health issues, including medication? (What kind of help? Started here or from before? Medication?)

18. Are there programs or procedures in place now that have helped you address problems or stressful situations? What are they? What kinds of problems do they help address? How often have you been helped by this?

19. What could the DOC do to help prisoners address their problems and stressful situations?

20. Have you ever been hurt by staff or other prisoners during this incarceration? Explain circumstances, how you dealt with it, prevalence of this victimization, facility where took place, end result(s), etc…

21. Have you ever been disrespected by staff or other prisoners during this incarceration? (How did they disrespect you? How did you deal with this?

22. Do you have any issues or problems from your past that you think have made your incarceration more difficult? Explain.

385

23. Do you have any issues or problems from the outside that you think have made your incarceration more difficult? Explain

24. In general, what do you think are the main reasons prisoners are involved in serious prison misconduct and violence?

25. What institutional correctional practices, if any, do you see that help reduce serious prison misconduct?

26. What would you recommend that this prison or the system do to decrease the likelihood of prison violence? Of other types of serious disciplinary misconduct?

27. Have you been booked in the last year? a. In general do you think the bookings issued are fair? b. Did you go through a disciplinary hearing? What do you think of the disciplinary process? Is it fair? c. Are the sanctions fair? What changes, if any, would you like to see in the booking process, the disciplinary process, or the sanctions?

28. Are there things that the DOC could do to help you not get further bookings?

29. Is there anything else about stresses in prison, and coping in prison that you think I should know?

386

APPENDIX D – STAFF INTERVIEW SCHEDULE

387

RIDOC Staff Interview

In general, what do you think are the main reasons prisoners are involved in serious prison misconduct and violence?

In general, do you think that prisoners’ involvement in serious prison misconduct is volitional or circumstantial or a combination of both?

How would you categorize prisoners involved in serious prison misconduct and violence?

What institutional correctional practices, if any, have you seen that have helped reduce serious prison misconduct? Are there individual correctional practices that you or other officers (administrators) have implemented to reduce the likelihood of violence?

What types of correctional practices or policies would you recommend to decrease the likelihood of prison violence? Of other types of serious disciplinary misconduct?

What do you think of the current disciplinary process? Of the current sanctioning process? Are there other sanctions/solutions that you would like to see implemented?

What is your best guess as to the percentage of inmates in the RIDOC that are involved in serious prison misconduct who also have mental health problems? What evidence or experiences are you relying on to come up with that figure?

Are there any prison strains, stresses, or circumstances that contribute to inmates’ involvement in serious prison misconduct? If so, what do you think could help address that problem?

Do you think that inmates who are involved in serious prison misconduct and violence have poorer, better, or similar abilities to cope with those stresses? Please explain. (Explain coping if necessary).

Do you think that the prison system could help inmates cope better? How?

388

REFERENCES

Abbott, J. H. (1991). In the belly of the beast: Letters from prison. New York: Random House.

Abrams, L. S., Anderson-Nathe, B., & Aguilar, J. (2008). Constructing masculinities in juvenile corrections. Men and Masculinities, 11, 22-41.

Adams, K., & Ferrandino, J. (2008). Managing mentally ill inmates in prisons. Criminal Justice and Behavior, 35, 913-927.

Akers, R. (1985). Deviant behavior: A social learning approach. Belmont, CA: Wadsworth.

Amirkhan, J. H. (1990). A factor analytically derived measure of coping: The coping strategy indicator. Journal of Personality and Social Psychology, 59, 1066-1074.

Anderson, E. (1994). Code of the street. Atlantic Monthly, 273, 80-94.

Anderson, E. (1999). Code of the street: Decency, violence, and the moral life of the inner city. New York: W. W. Norton.

Andrews, D. A., & Bonta, J. (1998). The psychology of criminal conduct. Cincinnati, OH: Anderson.

Anson, R., & Hancock, B. (1992). Crowding, proximity, prisoner violence and the Eighth Amendment. Journal of Offender Rehabilitation, 17, 123-132.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavior change. Psychological Review, 84, 191-216.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Belsey, D. A., Kuh, E., & Welsh, R. E. (1980). Regression diagnostics: Identifying influential data and sources of multicollinearity. New York: Wiley.

Berg, M. T., & DeLisi, M. (2006). The correctional melting pot: Race, ethnicity, citizenship, and prison. Journal of Criminal Justice, 34, 631-642.

Publication Date: 2006Berk, R., & MacDonald, J. M. (2008). Overdispersion and Poisson regression. Journal of Quantitative Criminology, 24, 269-284.

Biggam, F. H., & Power, K. G. (1999). A comparison of the problem-solving abilities and psychological distress of suicidal, bullied, and protected prisoners. Criminal Justice and Behavior, 26, 196-216.

389

Blacker, J., Watson, A., & Beech, A. R. (2008). A combined drama-based and CBT approach to working with self-reported anger aggression. Criminal Behaviour and Mental Health, 18, 129-137.

Bottcher, J. B., & Ezell, M. E. (2005). Examining the effectiveness of boot camps: A randomized experiment with long-term follow up. Journal of Research in Crime and Delinquency, 42, 309-332.

Bottoms, A. (1999). Interpersonal violence and social order in prisons. In M. Tonry & J. Petersilia (Eds.), Crime and justice: A review of research (Volume 26, pp. 205-281). Chicago: University of Chicago Press.

Bowman, B. J. (1997). The impact of mandatory relocation on female maximum security prisoners. International Journal of Offender Therapy and Comparative Criminology, 41, 375-377.

Braga, A. (2003). Serious youth gun offenders and the epidemic of youth violence in Boston. Journal of Quantitative Criminology, 19, 33-54.

Bureau of Justice Statistics. (2005). Census of state and federal adult correctional facilities, 2005. Washington, D. C.: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Camp, S. D., Daggett, D. M., Kwon, O., & Klein-Saffran, J. (2008). The effect of faith program participation on prison misconduct: The Life Connections Program. Journal of Criminal Justice, 36, 389-395.

Camp, S.D., Gaes, G.G., Langan, N.P., & Saylor, W.G. (2003). The influence of prisons on prisoner misconduct: A multilevel investigation. Justice Quarterly, 20, 501-533.

Carroll, L. (1974). Hacks, blacks, and cons. Lexington, MA: Lexington Books.

Carroll, L. (1988). Race, ethnicity, and the social order of the prison. In R. Johnson and H. Toch (Eds.) The pains of imprisonment (pp. 181-208). Prospect Heights, Illinois: Waveland Press.

Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A theoretically based approach. Journal of Personality and Social Psychology, 56, 267- 283.

Carver, C. S. (n.d.). The COPE Inventory. Retrieved on 3/30/08 from http://www.psy.miami.edu/faculty/ccarver/sclCOPEf.html

Cesaroni, C., & Alvi, S. (2010). Masculinity and resistance in adolescent carceral settings. Canadian Journal of Criminology and Criminal Justice, 52, 303-321.

Cesaroni, C., & Peterson-Badali, M. (2005). Young offenders in custody: Risk and adjustment. Criminal Justice and Behavior, 32, 251-277.

390

Clemmer, D. (1940). The prison community. New York, NY: Holt, Rinehart and Winston.

Cohen, A. (1955). Delinquent boys: The culture of the gang. New York: Free Press.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences, 2nd Edition. Hillsdale: Lawrence Erlbaum Associates.

Colvin, M. (1992). The penitentiary in crisis: From accommodation to riot in New Mexico. Albany: State University of New York Press.

Connell, R., & Messerschmidt, J. W. (2005). Hegemonic masculinity: Rethinking the concept. Gender and Society, 19, 829-859.

Creative Research Systems. (2011). Sample size calculator. Retrieved at: http://www.surveysystem.com/sscalc.htm.

Cunningham, M. D., & Sorensen, J. R. (2007). Predictive factors for violent misconduct in close custody. The Prison Journal, 87, 241-253.

CustomInsight.com. (2011). Survey random sample calculator. Retrieved at: http://www.custominsight.com/articles/random-sample-calculator.asp.

DanielSopers.com. (2011). Statistics calculators. Retrieved at: http://www. danielsoper.com/statcalc/calc01.aspx.

DeLisi, M., Berg, M. T., Hochstetler, A. (2004). Gang members, career criminals and prison violence: Further specification of the importation model of prisoner behavior. Criminal Justice Studies, 17, 369-383.

Dhami, M. K., Ayton, P., & Loewenstein, G. (2007). Adaptation to imprisonment: Indigenous or imported? Criminal Justice and Behavior, 34, 1085-1100.

Douglas, C. I. (1983). Crisis of imprisonment: Coping with stress adjustment to forced separation. Ann Arbor, MI: University Microfilms International.

Dvoskin, J. A., & Spiers, E. M. (2004). On the role of correctional officers in prison mental health. Psychiatric Quarterly, 75, 41-59.

D‘Zurilla, T. J., & Chang, E. C. (1995). The relations between social problem solving and coping. Cognitive Therapy and Research, 19, 547-562.

Edgar, K., & O‘Donnell, I. (1998). Assault in prison. British Journal of Criminology, 38, 635-643.

Edgar, K., O‘Donnell, I., & Martin, C. (2003). Prison violence: The dynamics of conflict, fear and power. Portland, OR: Willan Publishing.

391

Ekland-Olson, S., Barrick, D.M., & Cohen, L.E. (1983). Prison overcrowding and disciplinary problems: An analysis of the Texas prison system. Journal of Applied Behavioral Science, 19, 163-176.

Ellis, D., Grasmick, H.G., & Gilman, B. (1974). Violence in prisons: A sociological analysis. American Journal of Sociology, 80, 16-43.

Epstein, S. (1990). Cognitive-experiential self-theory. In L. A. Pervin (Ed.), Handbook of personality: Theory and research (pp. 165-192). New York: Guilford Press.

Feld, B. (1981). A comparative analysis of organizational structure and inmate subcultures in institutions for juvenile offenders. Crime & Delinquency, 27, 336-363.

Flanagan, T.J. (1980). Time served and institutional misconduct patterns of involvement in disciplinary infractions among long-term and short-term prisoners. Journal of Criminal Justice, 8, 357-367.

Flanagan, T.J. (1983). Correlates of institutional misconduct among state prisoners: A research note. Criminology, 21, 29-39.

Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-239.

Folkman, S., & Lazarus, R. S. (1985). If it changes it must be a process: Study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology, 48, 150-170.

Forst, M.J., & Brady, J.M. (1983). The effects of determinate sentencing on prisoners misconduct in prison. Prison Journal, 63, 100-113.

Fox, J. G. (1988). Women in prison: A case study in the social reality of stress. In R. Johnson & H. Toch (Eds.), The pains of imprisonment. (pp. 205-220). Prospect Heights, Illinois: Waveland Press.

French, S. A., & Gendreau, P. (2006). Reducing prison misconducts: What works!. Criminal Justice and Behavior, 33, 185-218.

Fritch, A. M. (2006). Trauma, coping, and health in incarcerated women. Dissertation submitted to Idaho State University.

Gaes, G.G. (1985). The effects of overcrowding in prison. In M. Tonry & N. Morris (Eds.) Crime and justice: An annual review (Volume 6, pp. 95-146). Chicago: University of Chicago Press.

Gaes, G.G., & McGuire, W.J. (1985). Prison violence: The contribution of crowding versus other determinants of prison assault rates. Journal of Research in Crime & Delinquency, 22, 41-65.

392

Gaes, G.G., Wallace, S., Gilman, E., Klein-Saffran, J. & Suppa, S. (2002). The influence of prison affiliation on violence and other prison misconduct. Prison Journal, 82, 359- 385.

Garson, G. D. (2009). Generalized linear models and general estimating equations. Raleigh, NC: North Carolina State University. From http://faculty.chass.ncsu.edu/garson/PA765/gzlm_gee.htm (accessed June 12, 2010).

Gendreau, P., & Goggin, C. (2000). Correctional treatment: Accomplishments and realities. In P. Van Voorhis, M. Braswell, & D. Lester (Eds.) Correctional counseling and rehabilitation, 4th edition. Cincinnati, OH: Anderson Publishing Co.

Gendreau, P. (1996). The principles of effective intervention with offenders. In A. Harland (Ed.), Choosing correctional options that work: Defining the demand and evaluating the supply (pp. 117-182). Thousand Oaks, CA: Sage.

Gendreau, P., Goggin, C.E., & Law, M.A. (1997). Predicting prison misconduct. Criminal Justice and Behavior, 24, 414-431.

Giallombardo, R. (1966). Society of women: A study of a women’s prison. New York: John Wiley & Sons.

Gibbons, J. J., & Katzenbach, N. de B. (2006). Confronting confinement: A report of The Commission on Safety and Abuse in America’s Prisons. New York: Vera Institute of Justice.

Gillespie, W. (2003). Prisonization: Individual and institutional factors affecting inmate conduct. New York: LFB Scholarly Publishing.

Goffman, E. (1961). Asylums. Garden City, NY: Doubleday

Goodstein, L., MacKenzie, D. L., & Shotland, R. L. (1984). Personal control and prisoner adjustment to prison. Criminology, 22, 343-369.

Goodstein, L., & Wright, K. (1991). Inmate adjustment to prison. In L. Goodstein & D. L. MacKenzie (Eds.), The American prison (pp. 229-251). New York: Plenum.

Gottfredson, M.R., & Hirschi, T. (1990). A general theory of crime. Palo Alto, CA: Stanford University Press.

Gover, A. R., MacKenzie, D. L., & Armstrong, G. S. (2000). Importation and deprivation explanations of juveniles‘ adjustment to correctional facilities. International Journal of Offender Therapy and Comparative Criminology, 44, 450-467.

Greer, K. (2002). Walking an emotional tightrope: Managing emotion in a women‘s prison. Symbolic Interaction, 25, 117-139.

393

Griffin, M., & Hepburn, J. (2006). The effect of gang affiliation on violent misconduct among inmates during the early years of confinement. Criminal Justice and Behavior, 33, 419-448.

Harer, M.D., & Langan, N.P. (2001). Gender differences in predictors of prison violence: Assessing the predictive validity of a risk classification system. Crime & delinquency, 47, 513-536.

Harer, M.D., & Steffensmeier, D.J. (1996). Race and prison violence. Criminology, 34, 323-351.

Harreveld, F. V., Van Der Pligt, J., Classsen, L., & Van Dijk, W. W. (2007). Prisoner emotion coping and psychological and physical well-being: the use of crying over spilled milk. Criminal Justice and Behavior, 34, 697-708.

Hensley, C., Koscheski, M., & Tewksbury, R. (2002). Does participation in conjugal visitations reduce prison violence in Mississippi? An exploratory study. Criminal Justice Review, 27, 52-65.

Hirschi, T. (1969). Causes of delinquency. Berkeley: University of California Press.

Hochstetler, A., Murphy, D. S., & Simons, R. L. (2004). Damaged goods: Exploring predictors of distress in prison inmates. Crime & Delinquency, 50, 436-457.

Howard, A., & Scott, R. A. (1965). A proposed framework for the analysis of stress in the human organism. Behavioral Science, 10, 141-160.

Huebner, B. M. (2003). Administrative determinants of inmate violence: A multilevel analysis. Journal of Criminal Justice, 31, 2, 107-117.

Hyman, S. M., Paliwal, P., & Sinha, R. (2007). Childhood maltreatment, perceived stress, and stress-related coping in recently abstinent cocaine dependent adults. Psychology of Addictive Behaviors, 21, 233-238.

Ingledew, D. K., Hardy, L., Cooper, C. L., & Jamal, H. (1996). Health behaviours reported as coping strategies: A factor analytical study. British Journal of Health Psychology, 1, 263-281.

Irwin, J. (1970). The felon. Englewood Cliffs, NJ: Prentice-Hall.

Jackson, P. B. (1992). Specifying the buffering hypothesis: Support, strain, and depression. Social Psychology Quarterly, 55, 363-378.

James, D. J., & Glaze, L. E. (2006). Mental health problems of prison and jail inmates. Washington, D. C.: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Janus, I. L. (1969) Stress and frustration. New York: Harcourt Brace Jovanovich (p. 198).

394

Johnson, E. (1966). Pilot study: Age, race and recidivism as factors in prisoner infractions. Canadian Journal of Corrections, 8, 268-283.

Johnson, R. (1979) Juvenile delinquency and its origins. Cambridge, MA: Cambridge University Press.

Johnson, R. (2002). Hard time: Understanding and reforming the prison (3rd edition). Monterey, California: Brooks/Cole Publishing Company.

Johnson, R. & Toch, H. (1988). The pains of imprisonment. Prospect Heights, Illinois: Waveland Press.

Kerley, K. R., Hochstetler, A., & Copes, H. (2009). Self-control, prison victimization, and prison infractions. Criminal Justice Review, 34, 553-568.

Klein, R. J. (1987). The role of anger and other demographic data in the prediction of disruptive behaviors for prisoners while incarcerated in protective custody. Ann Arbor, MI: University Microfilms International.

Kremelberg, D. (2011). Practical statistics: A quick and easy guide to IBM SPSS statistics, STATA, and other statistical software. Thousand Oakes, CA: Sage.

Kuanliang, A., Sorensen, J. R., & Cunningham, M. D. (2008). Juvenile inmates in an adult prison system: Rates of disciplinary misconduct and violence. Criminal Justice and Behavior, 35, 1186-1201.

Kupers, T. A., Dronet, T., Winter, M., Austin, J., Kelly, L., Cartier, W., Morris, T. J., Hanlon, S. F., Sparkman, E. L., Kumar, P., Vincent, L. C., Norris, J., Nagel, K., & McBride, J. (2009). Beyond supermax administrative segregation: Mississippi‘s experience rethinking prison classification and creating alternative mental health programs. Criminal Justice and Behavior, 36, 1037-1050.

Lahm, K. F. (2009). Educational participation and inmate misconduct. Journal of Offender Rehabilitation, 48, 37-52.

Lancaster, B.P. (1999). Defining and interpreting suppressor effects: Advantages and limitations. Paper presented at the Annual Meeting of the Southwest Educational Research Association (San Antonio, TX, January 21-23, 1999). Retrieved 4/14/11 @ http://www.eric.ed.gov/PDFS/ED426097.pdf.

Lattimore, P. K., MacDonald, J. M., Piquero, A. R., Linster, R. L., & Visher, C. A. (2004). Studying frequency of arrest among paroled youthful offenders. Journal of Research in Crime and Delinquency, 41, 37-57.

Laub, J. & Sampson, R. J. (2004). Shared beginnings, divergent lives. Cambridge, MA: Harvard University Press.

395

Lazarus, R. S. (1966). Psychological stress and the coping process. New York: McGraw- Hill.

Lazarus, R. S. (1981). The stress and coping paradigm. In C. Eisdorfer, D. Cohen, A. Kleinman, & P. Maxim (Eds.), Models for clinical psychopathology (pp. 174-214). New York: Spectrum.

Lazarus, R. S., Averill, J. R., & Opton, E. M., Jr. (1974). The psychology of coping: Issues of research and assessment. In G. V. Coelho, D. A. Hamburg, & J. E. Adams (Eds.), Coping and adaptation (pp. 249-315). New York: Basic Books.

Lazarus, R. S., Coyne, J. C., & Folkman, S. (1982). Cognition, emotion, and motivation: The doctoring of Humpty-Dumpty. In R. W. J. Neufeld (Ed.), Psychological stress and psychopathology (pp. 218-259). New York: McGraw-Hill.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer Publishing Company.

Liebling, A. (1999). Prison suicide and prison coping. In M. Tonry & J. Petersilia (Eds.) Prisons: Crime and justice a review of research (pp. 283-359) Chicago: University of Chicago Press.

Long, J. S. (1997). Regression models for categorical and limited-dependent variables. Thousand Oaks, CA: Sage.

Louscher, P.K., Hosford, R.E., & Moss, C.S. (1983). Predicting dangerous behavior in a penitentiary using the Megargee Typology. Criminal Justice and Behavior, 10, 269- 284.

Lovell, D. (2008). Patterns of disturbed behavior in a supermax population. Criminal Justice and Behavior, 35, 985-1004.

Lovell, D., Cloyes, K., Allen, D., & Rhodes, L. (2000). Who lives in super-maximum custody? A Washington State study. Federal Probation, 64, 33-38.

Lovell, D., & Jemelka, R. (1996). When prisoners misbehave: The costs of discipline. The Prison Journal, 76, 165-179.

Lyne, K., & Roger, D. (2000). A psychometric re-assessment of the COPE questionnaire. Personality and Individual Differences, 29, 321-335.

MacKenzie, D. L., & Goodstein, L. (1986). Stress and the control beliefs of prisoners: A test of three models of control-limited environments. Journal of Applied Social Psychology, 16, 209-228.

MacKenzie, D. L., Robinson, J. W., & Campbell, C. S. (1989). Long-term incarceration of female offenders: Prison adjustment and coping. Criminal Justice and Behavior, 16, 223-238.

396

MacKenzie, D. L., & Souryal, C. (1994). Multisite evaluation of shock incarceration. Washington, DC: National Institute of Justice, Department of Justice.

Maimon, D., & Browning, C. R. (2010). Unstructured socializing, collective efficacy, and violent behavior among urban youth. Criminology, 48¸ 443-474.

Maitland, A. S., & Sluder, R. D. (1996). Victimization in prisons: A study of factors related to the general well-being of youthful inmates. Federal Probation, 60, 24-31.

Maitland, A. S., & Sluder, R. D. (1998). Victimization and youthful prison prisoners: An empirical analysis. The Prison Journal, 78, 55-73.

Matza, D. (1964). Delinquency and drift. New York, NY: John Wiley.

McCorkle, R. (1992). Personal precautions to violence in prison. Criminal Justice and Behavior, 19, 160-173.

McCorkle, R. C. (1993). Fear of victimization and symptoms of psychopathology among prison prisoners. Journal of Offender Rehabilitation, 9, 27-41.

McDonald, D. (2006). Empowering female prisoners: An exploratory study of a prison therapeutic community and its impact on the coping skills of substance abusing women. Dissertation submitted to the Indiana University of Pennsylvania.

McGrath, J. E. (1970). A conceptual formulation for research on stress. In J. E. McGrath (Ed.), Social and psychological factors in stress. New York: Holt, Rinehart & Winston.

Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674-701.

Nacci, P., Teitlebaum, H., & Prather, J. (1977). Population density and prisoner misconduct rates in the federal prison system. Federal Probation, 41, 26-31.

Nachshon, I. & Rotenberg, M. (1977). Perception of violence by institutionalized offenders. Journal of Criminal Law and Criminology, 68, 454-457.

Nagy, C., Woods, D. J., & Carlson, R. (1997). The relationship between female prisoners‘ coping and adjustment in a minimum-security prison. Criminal Justice and Behavior, 24, 224-233.

Norušis, M. (2003). SPSS12.0: Statistical procedures companion. Upper Saddle River, NJ: Prentice Hall.

Nunnally, J. C., & Bernstein, I. H. (1994). Pychometric theory, 3rd Edition. New York: McGraw Hill.

397

O‘Donnell, I., & Edgar, K. (1998). Routine victimization in prisons. The Howard Journal, 37, 266-279.

O‘Donnell, I., & Edgar, K. (2010). Fear in prison. The Prison Journal, 79, 90-99.

Osgood, D. W. (2000). Poisson-based regression analysis of aggregate crime rates. Journal of Quantitative Criminology, 16, 21-43.

Owen, B. (1998). In the mix: Struggle and survival in a women’s prison. Albany: State University of New York Press.

Parisi, N. (1982). Prisoner‘s pressures and responses. In N. Parisi (Ed.), Coping with imprisonment: Perspectives in criminal justice (pp. 9-26). Beverly Hills, CA: Sage.

Paterline, B. A. & Petersen, D. M. (1999). Structural and social psychological determinants of prisonization. Journal of criminal justice, 27, 427-441.

Paternoster, R., Brame, R., Bachman, R., & Sherman, L. S. (1997). Do fair procedures matter? The effect of procedural justice on spouse assault. Law & Society Review, 31, 163-204.

Peck, Jr., L. W. (2004). Hoeing a long and hard row: Long term administrative segregation of a cohort of Texas prison prisoners. Dissertation submitted to Sam Houston State University. Huntsville: Texas.

Petersilia, J. (1983). Racial disparities in the criminal justice system. Santa Monica, CA: Rand.

Poole, E., & Regoli, R. (1980). Race, institutional rule breaking and institutional response: A study of discretionary decision making in prison. Law and Society Review, 14, 931- 946.

Porporino, F. J., & Zamble, E. (1984). Coping with imprisonment. Canadian Journal of Criminology, 26, 403-421.

Reisig, M. D., & Mesko, G. (2009). Procedural justice, legitimacy, and prisoner misconduct. Psychology, Crime & Law, 15, 41-59.

Rhode Island Department of Corrections. (2007). Management of offenders whose emotional/psychiatric status presents a danger to self or others: RIDOC policy and procedure 18.51-1 DOC, effective date 2/12/07. Cranston, RI: RIDOC.

Rhode Island Department of Corrections. (2009). Code of inmate discipline: RIDOC policy and procedure 11.01-5 DOC, effective date 3//9/09. Cranston, RI: RIDOC.

Rhode Island Department of Corrections. (2010a). Donald Price Medium Security Facility. Cranston, RI: RIDOC. Downloaded on 7/22/10 from http://www.doc.ri.gov/institutions/ facilities/price.php.

398

Rhode Island Department of Corrections. (2010b). High Security Center. Cranston, RI: RIDOC. Downloaded on 7/22/10 from http://www.doc.ri.gov/institutions/ facilities/hsc.php.

Rhode Island Department of Corrections. (2010c). John J. Moran Medium Security Facility. Cranston, RI: RIDOC. Downloaded on 7/22/10 from http://www.doc.ri.gov/institutions/ facilities/moran.php.

Rhode Island Department of Corrections. (2010d). Maximum Security. Cranston, RI: RIDOC. Downloaded on 7/22/10 from http://www.doc.ri.gov/institutions/ facilities/max_sec.php.

Rhode Island Department of Corrections. (2010e). Rhode Island Department of Corrections Planning and Research Unit summary of survey data and other interesting facts. Cranston, RI: RIDOC. Downloaded on 7/22/2010 from http://www.doc.ri.gov/administration/planning/docs/Pop%20Brochure%202009.pdf.

Rhode Island Department of Corrections. (2010f). RIDOC at a glance. Cranston, RI: RIDOC. Downloaded on 7/22/10 from http://www.doc.ri.gov/documents/ media/RIDOC%20Brochures/RIDOC%20At%20a%20Glance%2005-10.pdf.

Riveland, C. (1999). Supermax prisons: Overview and general considerations. Washington, DC: National Institute of Corrections.

Rocheleau, A. M., Forcier, M., & Jackson, J. (1998). Evaluation of the Massachusetts Department of Correction Department Disciplinary Unit. Unpublished report prepared for the Massachusetts Department of Correction.

Sappington, A. A. (1996). Relationships among prison adjustment, beliefs, and cognitive coping style. International Journal of Offender Therapy and comparative Criminology, 40, 54-62.

Sarason, B. R., Shearing, E. N., Pierce, G. R., & Sarason, I. G. (1987). Interrelationships between social support measures: Theoretical and practical implications. Journal of Personality and Social Psychology, 52, 813-832.

Schnur, A. C. (1949). Prison conduct and recidivism. Journal of Criminal Law and Criminology, 40, 36-42.

Scott, R. (2008). Interventions to address violence associated with mental illness. Australian Psychiatry, 16, 405-411.

Sherman, L. (1993). Defiance, deterrence, and irrelevance: A theory of the criminal sanction. Journal of Research in Crime and Delinquency, 30, 445-473.

Sieverdes, C. & Bartollas, C. (1986). Security level and adjustment patterns in juvenile institutions. Journal of Criminal Justice, 14, 135-145.

399

Silverman, M., & Vega, M. (1990). Reactions of prisoners to stress as a function of personality and demographic variables. International Journal of Offender Therapy and Criminology, 34, 187-196.

Smith, D. E. (1988). Crowding and confinement. In R. Johnson and H. Toch (Eds.) The pains of imprisonment (pp. 45-62). Prospect Heights, Illinois: Waveland Press.

Soderstrom, I. R., Castellano, T. C., & Figaro, H. R. (2001). Measuring ―mature coping‖ skills among adult and juvenile offenders: A psychometric assessment of relevant instruments. Criminal Justice and Behavior, 28, 300-328.

Sorensen, J., & Cunningham, M. D. (2010). Conviction offense and prison violence: A comparative study of murderers and other offenders. Crime & Delinquency, 56, 103-125.

Sorlie, T., & Sexton, H. (2001). The factor structure of ―The Ways of Coping Questionnaire‖ and the process of coping in surgical patients. Personality and Individual Differences, 30, 961-975.

Spielberger, C. D. (1972). Anxiety: Current trends in theory and research: I. New York, N.Y.: Academic Press.

Spielberger, C. D. (1980). Test Anxiety Inventory. Preliminary professional manual. Palo Alto, CA: Consulting Psychologists Press.

Spielberger, C. D. (1983). Manual for the State-Trait Anxiety Inventory (STAI). Palo Alto, CA: Consulting Psychologists Press.

Spielberger, C. D. (1995). State-Trait Personality Inventory (STPI): Sampler set, preliminary manual, response sheet, scoring directions. Menlo Park, CA: Mind Garden.

Spielberger, C. D., Ritterband, L. M., Sydeman, S. J., Reheiser, E. C., & Unger, K. K. (1995). Assessment of emotional states and personality traits: Measuring psychological vital signs (pp. 52-53). In J. N. Butcher (Ed.) Clinical personality assessment: Practical approaches. NY: Oxford University Press.

Steiner, B., & Wooldredge, J. (2008). Inmate versus environmental effects on prison rule violations. Criminal Justice and Behavior, 35, 438-456.

Stewart, E. A., & Simons, R. L. (2010). Race, code of the street, and violent delinquency: A multilevel investigation of neighborhood street culture and individual norms of violence. Criminology, 48, 569-603.

Stewart, J. (1999). John D. and Catherine T. MacArthur Research Network on Socioeconomic Status and Health: Anxiety. Retrieved 9/11/08 at: http://www.macses.ucsf.edu/Research/Psychosocial/notebook/anxiety.html.

400

Sykes, G.M. (1958). The society of captives. Princeton, NJ: Princeton University Press.

Thomas, C. W., & Foster, S. (1972). Prisonization in the prisoner contraculture. Social Problems, 20, 229-239.

Tittle, C. R., & Tittle, D. P. (1964). Social organization of prisoners: An empirical test. Social Forces, 43, 216-221.

Toch, H. (1975). Men in crisis: Human breakdowns in prison. Chicago: Aldine.

Toch, H. (1977). Living in prison: The ecology of survival. New York: Free Press.

Toch, H., & Adams, K. (1986). Pathology and disruptiveness among prison prisoners. Journal of Research in Crime and Delinquency, 23, 7-21.

Toch, H., Adams, K., & Grant, D. (1989). Coping: Maladaptation in prisons. New Brunswick, NJ: Transaction Books.

Toch, H., & Adams, K. (1994). The disturbed violent offender. Hyattsville, MD: American Psychological Association.

Toch, H., & Kupers, T. A. (2007). Violence in prisons, revisited. Journal of Offender Rehabilitation, 45, 1-28.

Trulson, C. R. (2007). Determinants of disruption: Institutional misconduct among state- committed delinquents. Youth Violence and Juvenile Justice, 5, 7-34.

Tyler, R. T. (1990). Why people obey the law. New Haven, CT: Yale University Press.

Walkey, F.H., & Gilmour, D.R. (1981). The relationship between interpersonal distance and violence in imprisoned offenders. Criminal Justice and Behavior, 11, 331-340.

Warr, M. (2002). Companions in crime. New York: Cambridge.

Wellford, C. (1967). Factors associated with adoption of the inmate code: A study of normative socialization. Journal of Criminal Law, Criminology, and Police Science, 58, 197-203.

Welsh, W. N., McGrain, P., Salamatin, N., & Zajaz, G. (2007). Effects of prison drug treatment on inmate misconduct: A repeated measures analysis. Criminal Justice and Behavior, 34, 600-615.

Wheeler, S. (1961). Socialization in correctional communities. American Sociological Review, 26, 697-712.

Wilson, J. Q., & Kelling, G. L. (1982). Broken windows. Atlantic Monthly, March. Retrieved at: http://www.theatlantic.com/magazine/archive/1982/03/ broken-windows/4465/

401

Wolff, N., & Shi, J. (2009). Feelings of safety inside prison among male inmates with different victimization experiences. Violence and Victims, 24, 800-816.

Wolfgang, M. (1961). Quantitative analysis of adjustment to the prison community. Journal of Criminal Law, Criminology and Police Science, 51, 607-618.

Wood, J., Moir, A., & James, M. (2009). Prisoners‘ gang-related activity: The importance of bullying and moral disengagement. Psychology, Crime & Law, 15, 569-581.

Wooldredge, J., Griffin, T., & Pratt, T. (2001). Considering hierarchical models for research on prisoner behavior: Predicting misconduct with multilevel data. Justice Quarterly, 18, 203-231.

Wortley, R. (2002). Situational prison control: Crime prevention in correctional institutions. Cambridge, UK: Cambridge University Press.

Wright, K. N. (1987). Improving correctional classification through a study of the placement of prisoners in environmental settings. Binghamton, NY: Center for Social Analysis.

Wright, K. N. (1989). Race and economic marginality in explaining prison adjustment. Journal of Research in Crime and Delinquency, 26, 67-89.

Wright, K. N. (1991). A study of individual, environmental, and interactive effects in explaining adjustment to prison. Justice Quarterly, 8, 217-242.

Young, M. H., Justice, J. V., & Erdberg, P. (2004). Assault in prison and assault in prison psychiatric treatment. Journal of Forensic Science, 49, 141-149.

Zamble, E. (1992). Behavior and adaptation in long-term prisoners. Criminal Justice and Behavior, 19, 409-425.

Zamble, E., & Porporino, F. J. (1988). Coping, behavior, and adaptation in prison prisoners. New York: Springer-Verlag.

Zamble, E., Y Porporino, F. J. (1990). Coping, imprisonment and rehabilitation. Criminal Justice and Behavior, 17, 53-70.

Zink, T. (1958). Are prison troublemakers different? Journal of Criminal Law, Criminology and Police Science, 48, 433-434.

402