University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange
Doctoral Dissertations Graduate School
8-2004
The Role of Demographic Diversity in Predicting Worker Psychological Safety
Ann Marie Callahan University of Tennessee - Knoxville
Follow this and additional works at: https://trace.tennessee.edu/utk_graddiss
Part of the Sociology Commons
Recommended Citation Callahan, Ann Marie, "The Role of Demographic Diversity in Predicting Worker Psychological Safety. " PhD diss., University of Tennessee, 2004. https://trace.tennessee.edu/utk_graddiss/1961
This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:
I am submitting herewith a dissertation written by Ann Marie Callahan entitled "The Role of Demographic Diversity in Predicting Worker Psychological Safety." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the equirr ements for the degree of Doctor of Philosophy, with a major in Social Work.
Charles Glisson, Major Professor
We have read this dissertation and recommend its acceptance:
John S. Wodarski, David Dupper, Lawrence R. James
Accepted for the Council:
Carolyn R. Hodges
Vice Provost and Dean of the Graduate School
(Original signatures are on file with official studentecor r ds.)
To the Graduate Council:
I am submitting herewith a dissertation written by Ann Marie Callahan entitled “The Role of Demographic Diversity in Predicting Worker Psychological Safety.” I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Social Work.
Charles Glisson Major Professor
We have read this dissertation and recommend its acceptance:
John S. Wodarski
David Dupper
Lawrence R. James
Accepted for the Council:
Anne Mayhew
Vice Chancellor and Dean of Graduate Studies
(Original signatures are on file with official student records.) The Role of Demographic Diversity in Predicting Worker Psychological Safety
A Dissertation Presented for the Doctor of Philosophy Degree
The University of Tennessee, Knoxville
Ann Marie Callahan August 2004 ii
Dedication
I would like to dedicate this dissertation to Kathy Kilby and Toni McDaniel. Without your support, this work would not have been completed. I would also like to extend a special thanks to my family and committee members.
iii Abstract
This study examined the influence of demographic diversity on case manager perceptions
of psychological safety in child welfare and juvenile justice case management teams. The study was based on survey data from 82 case managers in 10 teams that were collected as part of a larger study. A series of regressions indicated that demographic diversity influenced a case manager’s psychological safety differently depending on the case manager’s characteristics. Diversity contributed to a decrease in psychological safety for non-Caucasians and an increase in psychological safety for Caucasians. Men reported an increase in psychological safety and women reported a decrease in psychological safety with greater team diversity. Finally, diversity contributed to a decrease in psychological safety for people without aggressive personalities and an increase in psychological safety for people with aggressive personalities.
iv Table of Contents
Chapter Page
1. Introduction…………………………………………………………….1
2. Demographic Diversity…………………………………………….…..5
3. Psychological Safety…………………………………………………..11
4. Linking Demographic Diversity and Psychological Safety…………...16
5. Methodology…………………………………………………………...21
6. Results………………………………………………………………….27
7. Discussion………………………………………………………………38
References………………………………………………………………49 Appendix………………………………………………………………..62 Vita……………………………………………………………………...85
v List of Tables
Table Page
1. Descriptive Statistics…………………………………………………64
2. Measurement Alphas…………………………………………………65
3. Correlation Matrix…………………………………………………....66
4. Hierarchical Regression Analysis for HI and CV Unstandardized/Standardized Beta Coefficients……………………..68
5. Unstandardized Beta Coefficients for HI and CV Significant Interaction Equations………………………………………………....70
6. F Statistic for Simple Main Effects…………………………………..71
7. Hierarchical Regression Analysis for RD Unstandardized/Standardized Beta Coefficients……………………..72
8. Unstandardized Beta Coefficients for RD Significant Interaction Equations…………………………………………………74
9. Significant Simple Main Effects for Race…………………………....75
1 Chapter 1
Introduction
Each year more than three million children are reported as abused or neglected in the United States (McDonald and Associates, 2004; Fromm, 2001). Many of these children rely on child welfare and juvenile justice services to help them mature into healthy and productive adults. Child welfare and juvenile justice case managers have the responsibility of coordinating the care of these children. Examples of case management responsibilities include developing client care plans, arranging residential placements, coordinating social and mental health services, and monitoring client progress. Case managers are often required to navigate through a complex system of service providers and confront systemic barriers. Case managers are further required to be emotionally available to engage in healthy relationships with clients, many with serious behavioral and emotional problems. Despite these significant challenges, case managers must provide the highest quality of care. One way to assist case managers in this process is to better understand how case managers perceive their work environment.
A worker in any organization attaches meaning to and attempts to make sense out of work environment characteristics (e.g., policies, practices, and procedures). A worker attends most closely to and places the greatest value on work environment characteristics that could potentially influence his or her own personal welfare (James and McIntyre,
1997). The psychological climate of a work environment is a worker’s perception of the psychological impact that the work environment has on his or her own personal welfare
(Glisson, 2000; Glisson and James, 2002; James and McIntyre, 1997; Doverspike and
Blumental, 2001; Hemmelgarn, Glisson, and Dukes, 2001). 2 Perceived congruence between personal values and organizational values can
have a positive impact on a worker’s psychological climate (James and McIntyre, 1997;
Parker, Baltes, Young, Huff, Altmann, Lacost, and Roberts, 2003). Some writers have described a positive psychological climate as “psychologically safe” (Brown and Leigh,
1996; Khan, 1990). More specifically, psychological safety is a worker’s perception of the freedom to fully engage in work responsibilities without fear of negative consequences to one’s status, career, or self-image (Khan, 1990).
Demographic diversity within child welfare and juvenile justice case management teams is hypothesized to be one factor that could influence the psychological safety experienced by case managers. Case management teams can be diverse in many ways but diversity is frequently expressed as variance in a team’s demographic characteristics such as gender, age, race, and ethnicity. Diversity can also be a function of the extent to which teams vary in characteristics such as education, tenure, personality, and values (Tsui an
Gutek, 1999; Harrison, Price, and Bell, 1998; Miliken and Martins, 1996). As a result, diversity would be expected to exist to some extent in all child welfare and juvenile justice case management teams.
Based on the social categorization theory (Tajfel, 1981; Turner, 1987), people
categorize others by referencing demographic characteristics, which are used to define
expectations for their behavior. People are assumed to engage in behaviors associated
with those who have similar demographic characteristics. Case management teams with a
greater degree of demographic diversity could be perceived as threatening and
“psychologically unsafe” given the team member’s limited ability to predict the behavior of others associated with different social categories (Kahn, 1990). Therefore, diversity 3 within a case management team can create the conditions that decrease a team member’s
psychological safety.
Organizational demography research has determined that demographic diversity
can influence outcomes related to psychological climate. Examples of these outcomes
include perceptions of conflict (Randel, 2002; Tsui and O’Reilly, 1989), role ambiguity
(Tsui and O’Reilly, 1989), psychological attachment (Tsui, Egan, and O’Reilly, 1992),
cooperation (Ely, 1994; Chatman, Polzer, Barsade, and Neale, 1997; Cox, Lobel, and
McLeod, 1991; Espinoza and Garza, 1985; Garza and Santos, 1991; Lau, 1998; Watson,
Kumar, and Michaelson, 1993), and advancement opportunities (Cox, Welch, and
Nkomo, 2001; Riordan and Shore, 1997; Greenhaus, Parasuraman, and Wormley, 1990;
Kirchmeyer, 1995).
Some organizational demography research suggests that demographic diversity
can influence outcomes related to psychological climate differently depending on the worker’s demographic characteristics. For example, Tsui, Egan, and O’Reilly (1992)
found that gender dissimilarity had a more negative effect on the psychological
attachment of men with an increased number of women in workgroups than on the
psychological attachment of women with an increased number of men in workgroups.
Chattopadhyat (1999) also found that racial dissimilarity negatively influenced altruism
reported by Caucasians in minority-dominated workgroups, but not for minorities in
Caucasian-dominated workgroups.
Research Focus
Additional research is needed to determine whether demographic diversity
influences the psychological safety of case managers in child welfare and juvenile justice 4 case management teams. This relationship is important because initial research suggests
that psychological climate can affect work attitudes, service quality, and client outcomes
in child welfare and juvenile justice case management teams (Glisson, 2002; Glisson and
James, 2002; Glisson and Durick, 1988; Glisson and Hemmelgarn, 1998). For example,
Glisson and James (2002) found that climate perceptions predicted organizational
commitment and job satisfaction over and above the influence of other predictors such as
organizational culture. Glisson and Hemmelgarn (1998) also found that the
organizational climate was the primary predictor of children’s improved psychosocial
functioning and a significant predictor of service quality based on several standards of
care (e.g., service continuity).
The present study will begin by discussing the meaning of demographic diversity
and psychological climate along with the theoretical background and measurement issues
associated with each construct. Organizational demography research is reviewed to
explain how demographic diversity might influence worker perceptions of psychological
safety. Study hypotheses are proposed based on the idea that demographic diversity can
have a different effect on a worker’s perception of psychological safety depending on the
worker’s own demographic characteristics. Finally, a series of regressions are conducted
to examine the relationship between demographic diversity and perceptions of
psychological safety.
5 Chapter 2
Demographic Diversity
Diversity is a work environment characteristic that represents teams of two or more people with different demographic characteristics (Williams and O’Reilly, 1998). A common distinction among diverse team members is based on observable or ascribed demographic characteristics relative to gender, race, ethnicity, and age. Another type of diversity is based on underlying or achieved demographic characteristics relative to education, technical abilities, functional background, and tenure. Additional underlying characteristics that capture diversity within a work group include attitudes, beliefs, values, and personality, but these characteristics have been studied less frequently in organizational research (Harrison, Price, and Bell, 1998; Miliken and Martins, 1996). The degree of demographic diversity within a work group is often a function of occupation, employee selection, and turnover rate (Tsui and Gutek, 1999).
Organizational demographers have focused on quantifying diversity to determine the implications for individual, group, and organizational functioning. The majority of research suggests that demographic diversity can inhibit the ability of workgroups to meet workgroup member expectations resulting in negative affective outcomes (e.g., lower job commitment) (Riordan and Shore, 1997; Tsui et al., 1992). In addition, research has consistently shown that those who are demographically different get lower performance ratings (Greenhaus, Parasuraman, and Wormley, 1990, Tsui and O’Reilly,
1989) and have higher levels of absenteeism and turnover (Tsui et al., 1992; Alexander,
Nuchols, Bloom, and Lee, 1995). Observable demographic characteristics, such as gender and race, usually have larger negative effects than underlying characteristics, such as 6 education and tenure (Williams and O’Reilly, 1998; Miliken and Martins, 1996).
Additional research is needed to understand how and why demographic diversity influences workers in a variety of work contexts.
Theoretical Foundation
Social Categorization Theory. One of the most commonly used theories to explain
the influence of demographic diversity is the social categorization theory (Williams and
O’Reilly, 1998). According to the social categorization theory (Tajfel, 1981; Turner,
1987), people organize information about themselves and others by placing people into
social categories. While any personal characteristic can be used for categorization, some
characteristics are considered more salient than others. Often people are categorized
based on characteristics that are easily observable and accessible such as gender, race,
and age. Education may also be used to define one person’s social category, while job
tenure may be considered salient for another. A “psychological group” is formed by a
collection of people who identify themselves as members of the same social category
(Tsui et al., 1992).
Social group membership and how groups are distinguished from one another
(e.g., minority versus majority team members) can be used to support a positive social
identity. A person’s social group is likely to be considered special and regarded positively
by that person and group members. On the other hand, people tend to view members of
other groups as less trustworthy, dishonest, or uncooperative compared to members of
their own group (Chattopadhyay, 1999; Tsui et al., 1992; Pelled, Cummings, and Kizilos,
2000; Edmondson, 1999). People also reference the social categories of others to help
them form expectations of their behavior. Therefore, the social categorization process 7 helps people establish a sense of social identity, anticipate the behaviors of others, and
interpret the social dynamics between people from different social categories.
Two Research Perspectives
Social work scholars have expanded our knowledge about employee diversity (see
Fong and Gibbs, 1995; Hyde, 1998; Mor Barak, 2000; Gummer, 1994; Ferguson, 1996),
but, in the process, have failed to address some important issues undertaken by
organizational demographers (Tsui and Gutek, 1999). The majority of social work
research focuses on the prescribed need for organizational change to achieve greater
cultural sensitivity, rather than exploring both the positive and negative effects of
diversity on individual, group, and organizational functioning. Social work research also
focuses on the experience of people from traditionally oppressed groups (e.g., based on
race), but does not address the experience of people with different demographic characteristics (e.g., based on personality). Finally, social work research often employs
qualitative methods or only addresses the main effects of demographic characteristics,
rather than quantifying the degree of diversity in such a way that would allow for the
testing of both main and interaction effects. Therefore, a new direction in social work
research is needed to further explore the effects of demographic diversity on case
managers in child welfare and juvenile justice case management teams.
Current Study
The current study will employ methods that are consistent with organizational
demography research to identify the positive and negative effects of gender, race, and
personality diversity on case managers in child welfare and juvenile justice case
management teams. It is necessary to consider the influence of gender and racial diversity 8 (or lack of diversity), because female and Caucasian workers dominate many case
management teams such as those included in the present study. Independent of the
demographic distribution of characteristics, gender and race are among the most readily
observable characteristics that workers can use to distinguish themselves from one
another. We examine both gender and racial diversity here.
The study also includes diversity in one personality characteristic, aggression,
because personality has been an understudied demographic characteristic in organizational research (Miliken and Martins, 1996; Williams and O’Reilly, 1998).
Given that personality was not a readily observable characteristic, like gender or race, team members would have to make a judgment about the aggressiveness of each team member (Williams and O’Reilly, 1998). Team member perspectives and style of interaction would likely reflect one’s tendency to engage in aggression. Hence, team members would assess the behavior of others to determine the level of risk aggression would pose to one’s personal welfare.
Measurement Issues. Gender, race, and personality diversity will be measured
based on the compositional approach and the relational approach. These methods have
been central to the quantification of demographic diversity in organizational research.
Compositional Approach. The compositional approach involves measuring a
team’s distribution of individuals with a particular demographic characteristic (Tsui and
Gutek, 1999). Pfeffer (1983) originally developed the concept of “compositional demography” to refer to the distributional characteristics of an organization according to variation of a particular demographic characteristic in a social unit (e.g., team). This measurement approach is important because it quantifies diversity as a property of a team 9 that can be compared across teams. For example, a team member’s perception of fairness
in a team with a majority of men might be compared to a team member’s perception of
fairness in teams with a majority of women to determine if perceptions of fairness vary as
function of the proportion of male and females in each team. Therefore, the use of this
approach is based on the assumption that the dynamics between individuals in one case
management team would be different from the dynamics between individuals in another
case management team depending on the collective demographic profile of the team
members (Tsui and Gutek, 1999).
Relational Approach. The relational approach focuses on each team member’s
experience. The relational approach involves measuring an individual’s degree of
difference from other team members in a particular demographic characteristic (Tsui and
Gutek, 1999). This measurement approach quantifies the degree of “relational
demography,” a concept that was first introduced by Wagner, Pfeffer and O’Reilly (1984)
and was later formalized by Tsui and O’Reilly (1989). The relational approach is
important because it quantifies diversity as a property of an individual that can be
compared to other individuals in the same team. For instance, a team member’s perception of fairness might be a function of how different he or she is from other members of the team. Therefore, the use of this approach is based on the assumption that
a case manager’s demographic characteristics compared to the characteristics of other team members have a unique impact on that individual (Tsui and Gutek, 1999).
10 Advantages. Measuring gender, race, and personality diversity by using the
compositional approach and relational approach has several advantages. It enables the
collection of specific information about how workers vary from one another, consistent
quantification of demographic diversity, measurement of diversity on both the individual
and team level, and testing of interaction effects. The way in which workers vary from
one another can be measured to identify when demographic diversity can have either positive or negative effects. The measurement of diversity on the individual and team levels can be used to determine whether the influence of diversity is consistent across levels. Finally, interaction effects can be tested between a worker’s demographic characteristics and the degree of demographic diversity among team members.
11 Chapter 3
Psychological Safety
Psychological climate refers to a worker’s perception of the psychological impact of a work environment on his or her own personal welfare (Reichers and Schneider,
1999; Glisson, 2000; Glisson and James, 2002; James and McIntyre, 1997; Doverspike and Blumental, 2001; Hemmelgarn, Glisson, and Dukes, 2001; James and James, 1989).
Psychological climate represents a worker’s perception of a work environment, while organizational climate represents agreement among co-workers or shared perceptions of a work environment (Glisson, 2000; James ad McIntyre, 1997; James and Jones, 1974).
The focus of the present study is on psychological climate. A positive psychological climate has been described as “psychologically safe” (Glisson and James, 2002; Glisson,
2000; Brown and Leigh, 1996). More specifically, psychological safety is as a worker’s perception of freedom to fully engage in work responsibilities without fear of negative consequences to one’s status, career, or self-image (Kahn, 1990).
Work environments that promote psychological safety may increase a worker’s perception of fairness, growth and advancement opportunities, personal accomplishment, role clarity, and cooperation as well as decrease a worker’s perception of emotional exhaustion, depersonalization, role conflict, and role overload (Glisson and James, 2002).
Authors have theorized how each work environment can affect a worker’s psychological safety differently depending on the characteristics that distinguish it (Kopelman, Brief, and Guzzo, 1990). The type of workers attracted, selected, and retained by an organization distinguish a work environment (Lindell and Brandt, 2000); however, beyond theoretical conjecture, there is little evidence to show that these work 12 environment characteristics influence a worker’s sense of psychological safety (James
and McIntyre, 1997; Kahn, 1990).
Theoretical Foundation
Valuation. James and McIntyre (1997) suggested that psychological climate depends on the extent to which a worker’s values are reflected in a work environment. A worker’s values are used as standards for assessing one’s personal and organizational welfare (James and James, 1989). When a worker perceives that organizational values are consistent with his or her own values, this congruence supports the perception of psychological safety. For example, if a worker perceives that a supervisor provides a fair performance review, then the worker who values fairness would be more likely to experience psychological safety. A worker’s psychological safety would be supported by the congruence between one’s value of fairness and the organizational value of fairness conveyed by how a supervisor treated the worker. If, on the other hand, a worker perceives one’s supervisor provides an unfair performance review, then the worker who values fairness would be more likely to experience a decrease in psychological safety.
That worker would not perceive the freedom to fully engage in work responsibilities for fear of not being treated fairly in the future.
Kahn (1990) provided several other examples of how a work environment might decrease a worker’s psychological safety. These examples reflect some of the dynamics that occur when a worker perceives the authority of his or her preferred status is threatened by a co-worker. For example, a female worker felt threatened by a male co- worker when she attempted to exercise her authority at a summer camp:
13 There are times when I’m trying to get a girl camper to go to bed, and some male counselor starts flirting with the girl. It makes me look bad and undermines me incredibly. So I have to be ‘the bitch.’ If I didn’t, and just dealt with the kids as I’d like to, they’d just hassle me and not listen to me (p. 710).
Workers also used intimidation to suppress new workers from questioning their authority as part of an informal hierarchy. Therefore, based on Kahn’s research, a worker in a preferred status team could experience a decrease in psychological safety upon perceiving a threat to one’s authority. This worker might further engage in behavior that perpetuates one’s preferred status, even if that meant threatening the psychological safety of co-workers outside one’s social group.
Evolution of Climate Research
Although psychological safety is a relatively new construct, it is the product of
many years of climate research. Research on climate began in the 1950s as an outgrowth
of the science of applied psychology. The focus of early research was on the role of
climate relative to team and organizational functioning. Several key publications in the
1960s provided the most notable introduction. Litwin and Stringer (1968) conceptualized
and operationalized climate similar to how the construct is studied today (i.e., based on
worker perceptions of various climate indicators). In the 1970s, there were numerous
literature reviews and critiques of the climate construct with new methods of
measurement introduced (Reichers and Schneider, 1990). James and Jones (1974) were
the first to distinguish between psychological climate and organizational climate. Among
other advancements, efforts to develop a unified measure of psychological climate took
place in the 1980s and 1990s (James and James, 1989). Based on ethnographic research, 14 Kahn (1990) introduced the concept of psychological safety to explain why workers personally engaged or disengaged at work. Psychological safety was defined as the freedom to fully engage in work responsibilities without fear of negative consequences to one’s status, career, or self-image. Brown and Leigh (1996) further operationalized psychological safety as a positive psychological climate.
Current Study
There has been little consistency in the measurement of psychological climate.
Perceptions of almost every aspect of a work environment have been measured including perceptions of co-workers, supervision, and top management (Parker, Baltes, Young,
Huff, Altman, Lacost, and Roberts, 2003). To ensure accurate measurement of psychological climate, the current study will be based on an empirically validated model.
Measurement Issues. Worker perceptions of the work environment generally cluster into four dimensions or types of perceptions that are used as indicators of psychological climate (James and McIntrye, 1997; James and James, 1989; Jones and
James, 1979; Lock, 1976). These dimensions include (1) role stress and lack of harmony,
(2) leadership facilitation and support, (3) job challenge and autonomy, and (4) workgroup cooperation, friendliness, and warmth. Based on the work of James and James
(1989), these dimensions are manifest indicators of a latent, underlying factor called psychological climate-general (PCg). The presence of an underlying factor means that individuals employ a more integrated cognitive structure to appraise work environment characteristics. Each of the four climate dimensions indicate an overall assessment of the degree to which the work environment is perceived to be personally beneficial versus personally detrimental to the organizational welfare of a worker. The finding of a general 15 factor has gained substantial empirical support (e.g., Brown and Leigh, 1996; Glisson and
Hemmelgarn, 1998; Glisson and James, 2002). Therefore, the present study will be based
on a hierarchical cognitive model that assumes a single, general factor (PCg) underlies
different types of climate perceptions (James and McIntyre, 1997).
The present study will also determine whether demographic diversity predicts a
positive psychological climate or psychological safety among case managers in child
welfare and juvenile justice case management teams. Glisson and James (2002) found
that case managers on child welfare and juvenile justice case management teams
described positive climates or perceptions of psychological safety as a function of
depersonalization (e.g., “I worry that this job is hardening me”), role conflict (e.g., “I do
things that are against my better judgment”), and emotional exhaustion (e.g., “I feel used
up.”). These three variables were indicators of a negative psychological climate (PCg).
Climate perceptions further predicted work attitudes relative to job satisfaction and
organizational commitment, beyond the influence of other work environment
characteristics (e.g., organizational culture).
As cited earlier, Glisson and Hemmelgarn (1998) concluded that the climate
perceptions of case managers were the primary predictor of positive service outcomes as
indicated by improved psychosocial functioning of clients and were a significant
predictor of service quality as indicated by several standards of care (e.g., service
comprehensiveness). Therefore, additional research is needed to determine how
individual and work environment characteristics influence perceptions of psychological climate. Research in this area is important because the results may be useful in improving
the effectiveness of child welfare and juvenile justice organizations. 16 Chapter 4
Linking Demographic Diversity and Psychological Safety
The present study is based on the belief that demographic diversity is a work environment characteristic that can influence a worker’s perception of psychological safety. Organizational research has determined that demographic diversity can influence outcomes related to psychological climate. Outcomes related to psychological climate that have been predicted include perceptions of conflict (Randel, 2002; Tsui and
O’Reilly, 1989), role ambiguity (Tsui and O’Reilly, 1989), psychological attachment
(Tsui et al., 1992), cooperation (Ely, 1994; Chatman et al., 1997; Cox et al., 1991;
Espinoza and Garza, 1985; Garza and Santos, 1991; Lau, 1998; Watson et al., 1993), and advancement opportunities (Cox, Welch, and Nkomo, 2001; Riordan and Shore, 1997;
Greenhaus et al., 1990; Kirchmeyer, 1995). The most relevant studies are those that were conducted with workgroups rather than with dyads (e.g., supervisors-subordinates) or organizations; measured gender, race, or personality diversity; quantified diversity based on compositional or relational approaches; tested for interaction effects; and predicted psychological climate. These studies will be reviewed to inform study hypotheses related to gender, racial, and personality diversity.
Gender Diversity
Chattopadhyat (1999) compared workgroups to determine whether men in groups dominated by women were more likely to have lower levels of altruism toward workgroup members compared to women in groups dominated by men. The author found no significant gender effects. In contrast, Tsui, Egan, and O’Reilly (1992) found significant gender effects in the level of psychological attachment to workgroup 17 members. Men reported a significant decrease in levels of psychological attachment with
greater gender diversity, while women reported a significant increase in levels of psychological attachment with greater gender diversity. Increased gender diversity was
indicated by the degree to which an individual did not have the same gender as team members. However, these gender effects were no longer significant after controlling for
company variables (i.e., job satisfaction, hierarchical level of the individual, and size of
the organizational unit).
In this study, the effects of diversity in teams dominated by women are likely to
be opposite of that initially reported by Tsui et al. Women may view men as a threat to
their social status in a female-dominated team because men are perceived to have more
social power. At the same time, women likely adhere to traditional gender roles by
attending to the needs of men and treating them as informal leaders despite personal
aspirations to be respected as leaders themselves (Martin, 1985). Men often experience
little hostility and are socially well integrated in groups dominated by women (Fairhurst
and Snavely, 1983; Konrad, Winter, and Gutek, 1992). Therefore, men may enjoy
working with women because they feel supported by them and do not perceive women as
a threat to them professionally. These feelings are likely stronger than men’s potential
negative feelings about being in a female-dominated team (Macke, 1981; Wharton and
Baron, 1987). This rationale leads to the first hypothesis:
Hypothesis 1: The effect of gender diversity on perceptions of psychological
safety will be positive for males and negative for females.
18 Racial Diversity
Chattopadhyat (1999) also compared groups to determine whether racial diversity
negatively influenced altruism toward workgroup members. Non-Caucasians did not
report any change in response as a function of increased proportions of Caucasians, but
Caucasians reported a significant decrease in altruism in groups with more non-
Caucasians. Riordan and Shore (1997) determined that it was not until Caucasians
became the numerical minority that they reported a significant increase in negative
attitudes and a decrease in perceptions of advancement opportunities. In contrast, Tsui et
al. (1992) found that Caucasians reported a decrease in psychological attachment with
increased racial diversity even when non-Caucasians represented a small fraction of the
group. Taken together these studies suggest that Caucasians are likely to have a negative reaction in teams dominated by non-Caucasians while non-Caucasians are not likely to have a significant reaction in teams dominated by Caucasians.
Despite the threat of a threshold effect, the present study extends the work of Tsui et al. (1992) by conjecturing that Caucasians will report a significant decrease in
perceptions of psychological safety with an increased number of non-Caucasians. This conjecture is based on the assumption that Caucasians may experience discomfort with an
increased number of non-Caucasians in teams due to the lack of experience in being a
numerical minority. Non-Caucasians may interpret an increased number of Caucasians
with negative consequences e.g., less opportunity for advancement, but one’s
psychological safety would not be significantly effected due to familiarity with being a
numerical minority and having related coping skills. Unlike the mutual attraction that
exists between gender groups, there is no mutual attraction between racial groups that 19 could enhance psychological safety (Chattopadhyat, 1999). Hence, the second hypothesis is as follows:
Hypothesis 2: The effect of racial diversity on perceptions of psychological
safety will be non-significant for non-Caucasians and negative for Caucasians.
Personality Diversity
After an extensive literature review supplemented by the work of other authors
(Tsui and Gutek, 1999; Miliken and Martins, 1996; Williams and O’Reilly, 1998), the
only studies found that tested personality diversity predicted individual or group
performance (Hoffman, 1959; Hoffman and Mair, 1961; Harrison, Price, Gavin, and
Florey, 2002; Altman and Haythorn, 1967; Mohammed and Angell, 2003; van Vianen
and De Dreu, 2001; Neuman, Wagner, and Christiansen, 1999). Hence, these studies did
not address the outcome variables relevant to the present study. They also operationalized
personality relative to the “Big Five” indicators of personality (i.e., extraversion,
emotional stability, openness to experience, agreeableness, and conscientiousness) (see
Neuman et al., 1999), rather than relative to a measure of aggression. Therefore, previous
research provided little guidance for hypothesis formation used to predict the influence of
personality diversity on psychological safety.
In this study, personality is operationalized relative to the degree to which an
individual is considered to be “aggressive.” An individual is considered to be
“aggressive” when he or she reports a consistent pattern of reasoning that justifies
aggression (James, 1998). Aggressive personality traits, rather than more benign 20 personality traits, are likely to evoke a significant change in psychological safety. Hence, it is conjectured that a non-aggressive team member might perceive that a team dominated by aggressive team members is psychologically unsafe given the likelihood that aggressive team members may act aggressively toward the other team members. The aggressive team member might not perceive a threat surrounded by non-aggressive team members, especially if the aggressive team member is treated differently due to being viewed by others as a dominant team member. This rationale supports the third hypothesis:
Hypothesis 3: The effect of personality diversity on perceptions of psychological
safety will be negative for people with non-aggressive personalities and positive
for people with aggressive personalities.
21 Chapter 5
Methodology
Sample
The sample includes 82 case managers from 10 child welfare and juvenile justice
case management teams. The sample participated in a larger study of case management
teams serving 30 counties in Tennessee. That study used a true experimental design with random assignment of case management teams to experimental and control conditions.
The data for the present study were provided at follow-up by the teams in the control
condition. The respondents completed Likert-type scaled instruments that included the
Children’s Services Organizational Climate Survey (Glisson and Hemmelgarn, 1998;
Glisson and James, 2002) and the Conditional Reasoning Test of Aggression (CRTA)
(James, 1998) at a regularly scheduled case management team meeting. Characteristics of
the sample are presented in Table 1 of the Appendix and summarized below.
As shown in Table 1, 69 percent of the respondents were Caucasian; 80 percent
were female. For their level of education, 51 percent of the respondents had a bachelor’s
degree, 30 percent completed some graduate work, 10 percent had a graduate degree, 5
percent had some college, and 5 percent had a high school diploma. Seven percent of the
respondents were new team members. The average age was 40. Respondents also had an
average of 9 years of experience and 6 years of organizational tenure.
On the team level, 45 percent of the teams were in an urban area with 55 percent
of the teams in a rural area. The size of each team ranged from 6 to 12 members with an average of 8 members per team. Each team had an average of 1 new team member and 5 team members that had left the organization the previous year. 22 Predictor Variables
Race and gender. Respondents were assigned either no (0) or yes (1) on “Female” for gender and no (0) or yes (1) on “Caucasian” for race. Non-Caucasian respondents were largely African-American (96 percent).
Aggression. Aggression was measured as a continuous variable based on the number of aggressive responses (0 - 9) selected by the respondent on the Conditional
Reasoning Test of Aggression (CRTA) (James, 1998). The average score on the CRTA was 4, ranging from 0 to 8 for each respondent. The CRTA was used to quantify personality for each respondent relative to patterns in reasoning that would enhance or detract from one’s tendency to engage in aggression. Respondents with a higher CRTA score are more likely to justify the use of aggression than respondents with a lower
CRTA score (James, 1998).
Heterogeneity Index and Coefficient of Variation. The heterogeneity index (HI)
(or diversity index) (Teachman, 1980; Shannon and Weaver, 1949) quantified the degree
of gender or racial dissimilarity among case managers on that team. As found in the
Appendix, the diversity formula for team HI is 1 minus the sum of i squared (i being the
proportion of the team with the ith characteristic) (Smith, Smith, Olian, Sims, O’Bannon,
and Scully, 1994). For example, case management teams with 50 percent non-Caucasian
members and 50 percent Caucasian members would have a racial HI of .50. The HI for
gender and race ranged from 0 to .50 for each case management team. A higher HI
indicated greater diversity in gender or race among case managers.
The coefficient of variation (CV) measured the distribution of people with
particular CRTA scores, being a continuous variable that denoted a respondent’s level of 23 aggressiveness (Tsui and Gutek, 1999). The personality CV is calculated by dividing the standard deviation of the CRTA score by its mean, which represents the second diversity formula used in the study (see Appendix) (Alexander, Lichtenstein, Jinnet, D Aunno, and
Ullman, 1996). The personality CV ranged from .29 to .71 for each case management team. Similar to the heterogeneity index, a higher CV indicated greater team member diversity in personality.
Relational Demography. Relational Demography (RD) represented the degree to which a case manager was different from his or her team members on gender, race, and personality. The Euclidean Distance Measure is the standard way of quantifying RD
(Tsui and Gutek, 1999). The diversity formula for RD is the square root of the summed squared differences between an individual’s (Si) value on the characteristic of interest and the value of the characteristic for every other team member (Sj) divided by the total number of team members (n) (see Appendix) (Tsui et al, 1992). For example, one
Caucasian in a team with nine non-Caucasians would have a racial relational demography score of .95 (square root of 9/10). The RD score for gender and race ranged from 0 to .95 for each case manager. The RD score for personality was calculated the same way except with the continuous variable from the CRTA. The RD score for personality ranged from
3 to 56 for each case manager. Overall, a higher RD score indicated a greater degree of one’s difference from team members on that demographic characteristic.
Control Variables
Employee Turnover. The prevalence of turnover in this sample combined with research suggesting turnover has a negative influence on employee relations supports the need to control for this variable (Alexander et al., 1996; Watson et al., 1993; Alexander et 24 al., 1995). Three variables were calculated to control the influence of employee turnover.
A dichotomous variable called “New Team Member” was calculated for each respondent
to indicate who changed teams the previous year (no = 0, yes = 1). This data was
gathered by comparing one’s team membership at baseline to team membership at
follow-up.
A continuous variable called “Number of New Team Members” was calculated to control the effect of team level team turnover. This variable was calculated by comparing
team membership at baseline to team membership at follow-up. This variable accounts
for the total number of respondents who either entered or left each team. Another
continuous variable called “Organizational Turnover” was calculated as the number of
team members who resigned from the organization in the previous year. According to
baseline and follow-up data, six respondents changed teams. Six teams experienced
change in membership due to team transfer the previous year. Ten teams had team
members that resigned the previous year.
Final Selection of Control Variables. A common factor analysis and preliminary
regression analyses were conducted to guide the selection of control variables and
decrease the risk of multicollinearity in the final regression model. Predictor and control
variables were subjected to a common factor analysis. A preliminary regression analysis
was conducted for each control variable with a squared multiple correlation greater than
.50. The regressions also included predictor variables and psychological safety as the
outcome variable. This process was employed until there was little decrease in
standardized coefficients, while including the necessary variables to maintain the
integrity of the model. As a result, the following control variables were selected to be 25 included in the final series of regressions used for hypothesis testing: age, education,
new team member, organizational turnover, and number of new team members.
Outcome Variables
Psychological Safety. The Children’s Services Organizational Climate Survey
was used to determine a respondent’s perception of psychological safety (Glisson and
Hemmelgarn, 1998; Glisson and James, 2002). There were scales of 14 indicators often
used to measure climate included in the survey. Scores for all climate subscales were computed. The average score on the climate survey was 68, ranging from 33 to 97.
Depersonalization, role conflict, and emotional exhaustion represented psychological climate-general (PCg) based on a study that used the same data set by Glisson and James
(2002). Subscale scores for these indicators were reverse coded, computed, and combined to provide the final climate score for each person. The alpha reliability for the psychological climate scale was .78 (Table 2). Other climate subscales included fairness, role clarity, opportunities for growth and advancement, personal accomplishment, role overload, and cooperation. Alpha reliabilities ranged from .65 to .92 (Table 2).
Missing Data
Missing data were managed in several ways prior to data analysis. Questions with
missing data were compared to the same questions on the original survey to determine if
the answer was overlooked in the data entry process. When follow-up data were not available for a particular demographic variable, baseline data were used to replace
follow-up data or the data remained missing. The final percentage of missing data in the
data set ranged from 1 percent to 5 percent, depending on the variable. There were a total
of six respondents with missing data on race, gender, and personality across five teams. 26 One respondent was missing race and gender data so racial RD or gender RD could not be calculated for this individual. Missing data on race and gender also lowered the total number of team members counted when calculating the racial RD and HI and gender RD and HI for other team members on the respondent’s team. Five respondents were missing personality data so personality RD could not be calculated for these individuals. Missing
data lowered the total number of team members included when calculating the
personality RD and CV for other team members across four teams.
27 Chapter 6
Results
Correlation Matrix
As indicated by the correlation matrix (Table 3), race was the only variable among race, gender and personality that was correlated with psychological safety.
Caucasians reported less psychological safety, overall. Caucasians further reported a decrease in fairness and less opportunities for growth and advancement, as well as more role conflict and role overload. In addition, several diversity measures were correlated with perceptions related to psychological safety. Gender diversity was negatively correlated with psychological safety, overall. Racial diversity was positively correlated with fairness. Finally, personality diversity was negatively correlated with opportunities for growth and advancement and positively correlated with depersonalization and emotional exhaustion.
Procedure
Hypotheses predicted that women, Caucasians, and people with non-aggressive
personalities would report less psychological safety with greater gender, racial, and
personality diversity. In a series of regressions, all predictor variables which included
relational demography variables (e.g., racial relational demography) and control variables
were entered in Step 1. In Step 2, each interaction factor (e.g., Caucasian × racial
relational demography) was entered and tested independently from all other interaction
factors. Variables in a second series of regressions were entered likewise, with predictor
and control variables including the heterogeneity index (e.g., racial heterogeneity index) 28 or coefficient of variation variable entered in Step 1 and interaction factor (e.g.,
Caucasian × racial heterogeneity index) entered in Step 2.
Main Effects in Step 1. Regressions indicated that there were no significant main effects for women compared to Caucasians and people with aggressive personalities
(Tables 5 and 7). Caucasians reported a decrease in perceptions of fairness, opportunities for growth and advancement, and role clarity. Caucasians further reported an increase in role conflict and role overload. People with aggressive personalities reported fewer
opportunities for growth and advancement. There were also almost no measures of
diversity that significantly predicted psychological safety (Tables 5 and 7). Respondents
in teams with more variation in the number of team members with aggressive
personalities reported an increase in depersonalization. Beyond there being significant
main effects, there were many more significant interaction effects.
Hypothesis 1
The effect of gender diversity on perceptions of psychological safety will be
positive for males and negative for females.
A multiple regression indicated a statistically significant interaction between
gender and gender heterogeneity in predicting perceptions of psychological safety, B =
-55.52, t(56) = -2.59, p = .01 (Table 4). Based on the interaction equation (Table 5), the
relationship between gender heterogeneity and perceptions of psychological safety was
not significant for males, B = 13.57, t(56) = .76, p = .45, but this relationship was
significantly more negative for females as seen in Figure 1a of the Appendix. After 29 splitting the sample based on gender (i.e., males versus females), an analysis of variance
(ANOVA) was used to test for simple main effects. Using a split sample likely reduced
the statistical power of the interaction, but it was useful in providing additional
information. Results indicated that the simple main effects for males, F(1,16) = 1.74, p =
.21 and females, F(1,60) = 2.66, p = .11 were not significant (Table 6). H1 was partially
supported given a significant interaction effect.
Perceptions Related to Psychological Safety. There were also significant
interactions between gender and gender diversity in predicting perceptions related to
psychological safety including role conflict and role overload (Table 4).
Role Conflict. There was a significant interaction between gender and gender
heterogeneity in predicting perceptions of role conflict, B = 27.94, t(60) = 3.15 , p = .003
(Table 4). Based on the interaction equation (Table 5), the relationship between gender
heterogeneity and perceptions of role conflict was not significant for males, B = -13.30,
t(60) = -1.75, p = .09, but this relationship was significantly more positive (i.e., more role
conflict) for females (Figure 1b). ANOVA results indicated that the simple main effects
for males, F(1,16) = 2.83, p = .12 and females, F(1,64) = .23, p = .63 were not significant
(Table 6). H1 was partially supported given a significant interaction effect.
Role Overload. In addition, there was a statistically significant interaction between gender and gender heterogeneity in predicting perceptions of role overload, B =
21.56, t(60) = 2.43 , p = .02 (Table 4). Based on the interaction equation (Table 5), the
relationship between gender heterogeneity and perceptions of role overload was not
significant for males, B = -8.43, t(60) = -1.11 , p = .27, but this relationship was
significantly more positive for females (Figure 1c). ANOVA results indicated that the 30 simple main effects for males, F(1,16) = 4.24, p = .06 and females, F(1,64) = .86, p = .36 were not significant (Table 6). H1 was partially supported given a significant interaction effect.
Hypothesis 2
The effect of racial diversity on perceptions of psychological safety will be non-
significant for non-Caucasians and negative for Caucasians.
A multiple regression indicated a statistically significant interaction between race and racial relational demography in predicting psychological safety, B = 55.15, t(56) =
3.33, p = .00 (Table 7). Based on the interaction equation (Table 8), the relationship
between racial relational demography and perceptions of psychological safety was
significant for non-Caucasians, B = -42.38, t(56) = -2.96 , p = .01. Non-Caucasians
reported a significant decrease in perceptions of psychological safety with a greater
number of Caucasian team members. The relationship between racial relational
demography and psychological safety was also significantly more positive (i.e., greater
psychological safety) for Caucasians (Figure 2a). An analysis of variance (ANOVA) was
used to test the simple main effects after dividing the sample by race (i.e., no-Caucasians
versus Caucasians). ANOVA results indicated that the simple main effects for non-
Caucasians were significant, F(1,23) = 17.83, p = .00, but they were not significant for
Caucasians, F(1,54) = 2.49, p = .12 (Table 6). H2 was partially supported given a significant interaction effect. 31 Perceptions Related to Psychological Safety. There were also statistically
significant interactions between race and racial diversity in predicting perceptions related
to psychological safety including perceptions of depersonalization, emotional exhaustion,
fairness, role clarity, role conflict, and role overload (Tables 4 and 7).
Depersonalization. There was a statistically significant interaction between race
and racial relational demography in predicting perceptions of depersonalization, B =
-10.15, t(60) = -2.55, p = .01 (Table 7). Based on the interaction equation (Table 8), non-
Caucasians did not report a significant change in perceptions of depersonalization in
teams with a greater number of Caucasian members, B = 5.99, t(60) = 1.73, p = .09, but
the relationship between racial relational demography and depersonalization was
significantly more negative for Caucasians (Figure 2b). ANOVA results indicated that the
simple main effects for non-Caucasians, F(1,25) = 8.75, p = .01 and Caucasians, F(1,56)
= 4.19, p = .05 were significant (Table 6). Non-Caucasians reported a significant increase
in perceptions of depersonalization with a greater number of Caucasian team members
and Caucasians reported a significant decrease in perceptions of depersonalization with a
greater number of non-Caucasian team members. H2 was partially supported given a
significant interaction effect.
Emotional Exhaustion. There was a statistically significant interaction between
race and racial relational demography in predicting perceptions of emotional exhaustion,
B = -19.22, t(60) = -2.63, p = .01 (Table 7). Based on the interaction equation (Table 8),
the relationship between racial heterogeneity and perceptions of emotional exhaustion
was significant for non-Caucasians, B = 17.25, t(60) = 2.72 , p = .01. Non-Caucasians
reported a significant increase in perceptions of emotional exhaustion with a greater 32 number of Caucasian team members. The relationship between racial relational
demography and emotional exhaustion was also significantly more negative (i.e.,
decreased emotional exhaustion) for Caucasians (Figure 2c). ANOVA results indicated
that the simple main effects for non-Caucasians were significant, F(1,25) = 13.05, p =
.001, but they were not significant for Caucasians, F(1,56) = .99, p = .33 (Table 6). H2
was partially supported given a significant interaction effect.
Fairness. There was a statistically significant interaction between race and racial
heterogeneity in predicting perceptions of fairness, B = 15.87, t(60) = 1.98, p = .050
(Table 4). Based on the interaction equation (Table 5), non-Caucasians did not report a significant change in perceptions of fairness in teams with a greater number of Caucasian members, B = -9.87, t(60) = -1.30, p = .20, but the relationship between racial
heterogeneity and fairness was significantly more positive for Caucasians (Figure 2e).
ANOVA results indicated that the simple main effects for non-Caucasians were not
significant, F(1,25) = .50, p = .49, but they were significant for Caucasians, F(1,56) =
4.39, p = .04 (Table 6). Caucasians reported a significant increase in perceptions of
fairness with a greater number of non-Caucasian team members. H2 was partially
supported given a significant interaction effect.
In addition, there was a statistically significant interaction between race and racial
relational demography in predicting perceptions of fairness, B = 11.39, t(60) = 2.70, p =
.01 (Table 7). Based on the interaction equation (Table 8), non-Caucasians did not report a significant change in perceptions of fairness in teams with a greater number of
Caucasian members, B = -6.28, t(60) = -1.71, p = .09, but the relationship between racial
relational demography and perceptions of fairness was significantly more positive for 33 Caucasians (Figure 2f). ANOVA results indicated that the simple main effect of race for
non-Caucasians were not significant, F(1,25) = 2.66, p = .12, but they were significant for
Caucasians, F(1,56) = 7.77, p = .01 (Table 6). Caucasians reported a significant increase in perceptions of fairness with a greater number of non-Caucasian team members. H2 was partially supported given a significant interaction effect.
Role Clarity. There was a statistically significant interaction between race and
racial heterogeneity in predicting perceptions of role clarity, B = 18.92, t(60) = 2.00, p =
.05 (Table 4). Based on the interaction equation (Table 5), the relationship between racial heterogeneity and perceptions of role clarity was significant for non-Caucasians, B =
-18.91, t(60) = -2.10, p = .04. Non-Caucasians reported a significant decrease in perceptions of role clarity with a greater number of Caucasian team members. The relationship between racial heterogeneity and role clarity was also significantly more positive for Caucasians (Figure 2d). ANOVA results indicated that the simple main effects for non-Caucasians were significant, F(1,25) = 4.90, p = .04, but they were not
significant for Caucasians, F(1,56) = .20, p = .66 (Table 6). H2 was partially supported
given a significant interaction effect.
Role Conflict. There was a statistically significant interaction between race and
racial heterogeneity in predicting perceptions of role conflict, B = -32.10, t(60) = -2.28,
p = .03 (Table 4). Based on the interaction equation (Table 5), non-Caucasians did not
report a significant change in perceptions of role conflict in teams with a greater number
of Caucasian members, B = 24.33, t(60) = 1.82, p = .07, but the relationship between
racial heterogeneity and perceptions of role conflict was significantly more negative for
Caucasians (Figure 2g). ANOVA results indicated that the simple main effects for non- 34 Caucasians, F(1,25) = 2.72, p = .11 and Caucasians, F(1,56) = .55, p = .46 were not
significant (Table 6). H2 was partially supported given a significant interaction effect.
In addition, there was a statistically significant interaction between race and racial
relational demography in predicting perceptions of role conflict, B = -20.76, t(60) =
-2.78, p = .01 (Table 7). Based on the interaction equation (Table 8), the relationship
between racial relational demography and perceptions of role conflict was significant for
non-Caucasians, B = 12.76, t(60) = 1.97, p = .05. Non-Caucasians reported a significant
increase in perceptions of role conflict in teams with a greater number of Caucasian
members. The relationship between racial relational demography and perceptions of role
conflict was also significantly more negative for Caucasians (Figure 2h). ANOVA results
indicated that the simple main effects for non-Caucasians were significant, F(1,25) =
5.59, p = .03, but they were not significant for Caucasians, F(1,56) = 2.56, p = .12 (Table
6). H2 was partially supported given a significant interaction effect.
Role Overload. There was a statistically significant interaction between race and
racial heterogeneity in predicting perceptions of role overload, B = -27.22, t(60) = -1.98,
p = .05 (Table 4). Based on the interaction equation (Table 5), non-Caucasians did not
report a significant change in perceptions of role overload in teams with a greater number
of Caucasian members, B = 23.53, t(60) = 1.80, p = .08, but the relationship between
racial heterogeneity and perceptions of role overload was significantly more negative for
Caucasians (Figure 2i). ANOVA results indicated that the simple main effects for non-
Caucasians, F(1,25) = 3.99, p = .06 and Caucasians, F(1,56) = .39, p = .54 were not
significant (Table 6). H2 was partially supported given a significant interaction effect. 35 In addition, there was a statistically significant interaction between race and racial
relational demography in predicting perceptions of role overload, B = -20.18, t(60) =
-2.79, p = .01 (Table 7). Based on the interaction equation (Table 8), the relationship
between racial relational demography and perceptions of role overload was significant for
non-Caucasians, B = 14.72, t(60) = 2.35, p = .02. Non-Caucasians reported a significant
increase in perceptions of role overload in teams with a greater number of Caucasian
members. The relationship between racial relational demography and perceptions of role
overload was also significantly more negative for Caucasians (Figure 2j). ANOVA
results indicated that the simple main effects for non-Caucasians were significant, F(1,25)
= 9.09, p = .01, but were not significant for Caucasians, F(1,56) = 3.05, p = .09 (Table 6).
H2 was partially supported given a significant interaction effect.
Hypothesis 3
The effect of personality diversity on perceptions of psychological safety will be
negative for people with non-aggressive personalities and positive for people with
aggressive personalities.
A multiple regression did not indicate a statistically significant interaction
between personality and personality coefficient of variation, B = 11.56, t(56) = 1.08, p =
.28 (Table 4) or personality and personality relational demography, B = .07, t(56) = .89,
p = .38 (Table 7) in predicting perceptions of psychological safety. Therefore, H3 was not
supported given that the effect of personality diversity on psychological safety was non- 36 significant for people without aggressive personalities and people with aggressive personalities.
Perceptions Related to Psychological Safety. There were significant interactions
between personality and personality diversity in predicting perceptions related to
psychological safety including fairness, cooperation, and role conflict (Tables 4 and 7).
Fairness. There was a significant interaction between personality and personality
coefficient of variation in predicting perceptions of fairness, B = 5.46, t(60) = 2.20, p =
.03 (Table 4). Based on the interaction equation (Table 5), the relationship between personality coefficient of variation and perceptions of fairness was not significant for people without aggressive personalities, B = -18.92, t(60) = -1.75, p = .09, but this
relationship was significantly more positive (i.e., increased fairness) for people with
aggressive personalities (Figure 3a). After splitting the sample based on personality (i.e.,
people without aggressive personalities versus people with aggressive personalities), an
Analysis of Variance (ANOVA) was used to test the simple main effects. ANOVA
results indicated that the simple main effects for people without aggressive personalities,
F(1,56) = 1.16, p = .29 and people with aggressive personalities, F(1,21) = 2.20, p = .15 were not significant (Table 6). H1 was partially supported given a significant interaction effect.
Cooperation. There was also a significant interaction between personality and personality coefficient of variation in predicting perceptions of cooperation, B = 4.64,
t(60) = 2.30, p = .03 (Table 4). Based on the interaction equation (Table 5), the relationship between personality coefficient of variation and perceptions of cooperation was significant for people without aggressive personalities, B = -18.03, t(60) = -2.05, p = 37 .05. People without aggressive personalities reported a significant decrease in perceptions of cooperation in teams with a greater number of people with aggressive personalities.
The relationship between personality coefficient of variation and perceptions of cooperation was also significantly more positive for people with aggressive personalities
(Figure 3b). ANOVA results indicated that the simple main effects for people without aggressive personalities, F(1,56) = .59, p = .45 and people with aggressive personalities were not significant, F(1,21) = .53, p = .48 (Table 6). H3 was partially supported given a significant interaction effect.
Role Conflict. There was a significant interaction between personality and personality relational demography in predicting perceptions of role conflict, B = -.07, t(60) = -2.07, p = .04 (Table 7). Based on the interaction equation (Table 8), the relationship between personality coefficient of variation and perceptions of role conflict was not significant for people without aggressive personalities, B = .35, t(60) = 1.79, p =
.08, but this relationship was significantly more negative for people with aggressive personalities (Figure 3c). ANOVA results indicated that the simple main effects for people without aggressive personalities, F(1,56) = .08, p = .78 and people with aggressive personalities, F(1,21) = .99, p = .33 were not significant (Table 6). H3 was partially supported given a significant interaction effect.
38 Chapter 7
Discussion
The present study expanded the common understanding of diversity in social
work research. First, results clearly suggested the importance of interaction effects in
predicting worker psychological safety. Significant interactions revealed that the effect of
diversity was relative to worker demographic characteristics. The lack of significant main
effects further reinforced the importance of interaction effects. The only significant main
effect was for personality diversity relative to personality coefficient of variation.
Respondents in teams with an increased number of people with aggressive personalities
reported more depersonalization, which represented less psychological safety. When
interaction effects were taken into account, however, only people without aggressive personalities reported less psychological safety while people with aggressive personalities reported more psychological safety.
There were also multiple interaction effects related to gender diversity as well as racial diversity in predicting perceptions of psychological safety. The most substantial findings involved the relationship between race and racial diversity (Table 9). Overall, non-Caucasians reported a decrease while Caucasians reported an increase in psychological safety with greater racial diversity. It was no surprise that racial diversity
appeared to be the most influential factor compared to gender and personality diversity.
The issue of racial diversity in the workplace has taken center stage since the passage of
civil rights legislation in the 1960s. As a result, organizations have provided a forum for the manifestation of racial tension that has defined American culture for centuries. Social 39 service organizations, like most organizations, have been confronted with the challenge of promoting and managing racial diversity without adequate resources to do so.
It is important to note that significant interactions were associated with two different levels or measures of diversity. Diversity on the individual-level (relational demography) quantified the degree of one’s difference from team members on a particular attribute. Diversity on the team-level (team heterogeneity) quantified the degree of variation among team members on a particular attribute. Racial diversity across levels had a similar effect on perceptions of psychological safety. For non-Caucasians, an increase in racial heterogeneity contributed to a decrease in perceptions of role clarity,
while an increase in racial relational demography contributed to an increase in
perceptions of depersonalization, emotional exhaustion, role conflict, and role overload.
Hence, such continuity in the type and direction of perceptions seemed to be related to a
single, underlying construct defined as psychological climate-general (PCg).
However, each level of diversity was not always equally important in predicting
perceptions of psychological safety. Gender heterogeneity had a significant effect on
psychological safety, but the influence of gender relational demography was not
significant. Gender relational demography may not have been significant due to the role
of more influential variables. Tsui et al. (1992) found that gender relational demography
significantly predicted psychological attachment, but it was no longer significant after
controlling for organizational structure and culture. Nevertheless, gender heterogeneity
had a similar effect on perceptions of psychological safety. Continuity in the type and
direction of perceptions provided support for the overall effect of gender heterogeneity. 40 Gender heterogeneity had a more positive effect on the psychological safety of men and more negative effect on the psychological safety of women.
Study results also suggested that the effect of diversity was relative to obscure characteristics like personality. Organizational demography researchers had rarely tested the impact of personality and when it was tested only considered the influence of more benign personality traits (e.g., extroversion) (see Mohammed and Angell, 2003). This study tested a worker’s pattern of reasoning to identify which workers were more likely to justify the use of aggression. As expected, the interaction between a worker’s degree of aggressiveness and personality diversity significantly predicted worker psychological safety. These results were even more significant considering that there were very few
“aggressive” team members based on the restricted range in respondent aggressiveness scores. Any results would have been important, however, since the role of personality had never been tested like this before. Therefore, significant results in this study validated the inclusion of personality and personality diversity in predicting worker perceptions of psychological safety.
Findings on the role of personality and personality diversity were consistent with previous research on workplace aggression. Previous research suggested that personality traits (Baron, Neuman, and Geddes, 1999; Skarlicki, Folger, and Tesluk, 1999; Skarlicki and Folger, 1997; Douglas and Martinko, 2001) and perceptions of provocation (James,
1998; Skarlicki and Folger, 1997; Greenberg and Barling, 1999; Greenberg, 1993, 1990;
Leck, Saunders, and Charbonneau, 1996; Baron et al., 1999; Cole, Grubb, Sauter,
Swanson, and Lawless, 1997; Anderson and Martin, 1999) were reliable predictors of aggression. Aggression could have been expected in teams with more team members with 41 aggressive personalities. Hence, worker psychological safety may have been the product of aggressive behavior perpetrated by team members with more aggressive personalities or perceived threat of aggression experienced by team members with less aggressive personalities when exposed to team members with more aggressive personalities.
In addition, unlike traditional research on diversity, this study focused on the experience of both minority and non-minority workers. Minority workers represented individuals often ascribed less social power such as women and non-Caucasians. Study results indicated that minority workers reported diversity had a negative effect on their perceptions of psychological safety. This finding was ironic given greater potential for the acceptance of diversity among workers in a helping profession. Nevertheless, minority workers perceived some degree of threat to warrant feeling unsafe with an increased number of non-minority workers. Results also indicated that non-minority workers, like men and Caucasians, reported diversity had a positive effect on perceptions of psychological safety. Caucasians reported a positive response in teams where they remained the numerical majority, but a small sample size limited research on the experience of Caucasians in teams with a different racial profile.
It was interesting to note how the degree of psychological safety reported by minority and non-minority workers fluctuated depending on the amount of team-level diversity. For example, the degree of psychological safety reported by women compared to men was not very different in teams with little gender heterogeneity, but was very different in teams with more gender heterogeneity. Interestingly, this relationship was just the opposite for racial and personality diversity. For instance, the degree of psychological safety reported by non-Caucasians compared to Caucasians was very 42 different in teams with little racial heterogeneity, but was not very different in teams with more racial heterogeneity. Therefore, it was only more gender diversity that had a dramatically different effect on the degree of psychological safety, while more racial and personality diversity seemed to minimize differences reported by minority compared to non-minority workers. The social categorization theory helps to explain these results.
Based on the social categorization theory, perceptions of psychological safety may have been the partial product of expectations of co-workers in a particular category.
For instance, female workers may have expected male workers to be dominant team members, while male workers may have expected female workers to be passive team members. The more these contradictory expectations were uniformly shared and/or had a positive or negative effect on workers within a particular gender category, the more likely the degree of psychological safety would have been different between gender categories.
In contrast, there may not have been contradictory expectations that were uniformly shared by workers within a particular racial or personality category. Hence, a lack of uniformity in expectations and/or positive or negative effect on workers in a particular category could have resulted s similar degree of psychological safety between categories.
Study Limitations
As suggested earlier, other factors could have influenced the effect of diversity on
worker psychological safety beyond demographic characteristics. Some of these variables
will be addressed as study limitations. Two limitations included the presence of
uncontrolled variables and measurement problems.
43 Uncontrolled Variables. Despite steps to avoid this study limitation, uncontrolled variables could have influenced the variables of interest. For example, this study treated race as categorical variable without accounting for the racial identity strength for each respondent (Randel, 2002; Cox et al., 2001). There is some evidence to suggest that racial identity strength could have important implications for an individual’s values and perceptions of a work environment (Cox et al., 2001; Watts and Carter, 1991). Watts and
Carter (1991) found that African-American civil service workers who reported weak identification with one’s racial group also reported more favorable perceptions of working in a predominantly Caucasian company compared to workers that reported a stronger identification with one’s racial group. The influence of a worker’s racial identity could have also been dependant upon the salience of race for that particular worker or work environment (Tsui and Gutek, 1999). Race may not have been salient for a
Caucasian in a Caucasian-dominated team, but it may have been salient for a non-
Caucasian in a Caucasian-dominated team.
The workgroup value of diversity, called a diversity perspective, is another potentially influential variable (Chen and Eastman, 1997; Larkey, 1996; Kossek and
Zonia, 1993; Ely and Thomas, 2001; Cox, 1991). Ely and Thomas (2001) tested the conditions under which a workgroup’s diversity perspective enhanced or detracted from work group functioning. The results fell under three distinct categories: discrimination- and-fairness, access-and-legitimacy, and integration-and-learning. Relative to the value of cultural identity, groups with the discrimination-and-fairness perspective said cultural differences should be assimilated into the dominant culture to ensure justice and equality.
Groups with the access-and legitimacy perspective said cultural differences should be 44 used only as a resource to gain access to and legitimacy with diverse clients. Groups with the integration-and-learning perspective reported cultural differences should be integrated
into core work group processes. It is reasonable to assume that a workgroup’s diversity perspective would also have implications for workgroup member’s perception of psychological safety relative to one’s cultural differences.
Measurement Problems. Measurement problems may have further influenced the study results due to missing data and choice of study measures. Respondents without
information on the demographic variables of interest could not be counted in the total
number of team members when calculating the corresponding diversity scores (Appendix
A). As a result, error variance was factored into the relational demography score, heterogeneity index, and coefficient of variation for respondents in five out of ten case management teams (see “Missing Data” section). The lack of inclusiveness in the study measures created another limitation. Even though there did not appear to be problems, the climate and personality measures had not been tested on people with different demographic characteristics. The cross-cultural sensitivity of these instruments needed to be tested prior to utilization in the study to minimize this additional source of variance
(Rubbin and Babbie, 2001; Sedlacek, 1994). Furthermore, the current study only measured the main and interaction effects of single demographic characteristics. The development of a multi-characteristic measure would have been required to test the role of cumulative or higher-order interaction effects (Tsui and Gutek, 1999).
There were other problems associated with the methods of measurement used in this study. The common denominator of team size was required for the calculation of each diversity score (Appendix A) (Cohen and Cohen, 1983). This could explain why 45 some of the variables were correlated like gender HI and racial HI (Table 3) which
influenced subsequent regressions to some extent. The HI and CV provided a measure of
team-level diversity, but did not provide information about where the variance was
occurring. Since the coefficient of variation is the mean divided by the standard
deviation, variations could have occurred from either term (Williams and O’Reilly,
1998). Finally, none of the measures in this study adequately described the proportional differences of case managers with particular characteristics in each team (Williams and
O’Reilly, 1998). For example, O’Reilly, Williams, and Barsade (1997) initially determined that team heterogeneity on ethnicity positively affected team innovation.
Subsequent analyses revealed that the proportion of Asians in each team had influenced the results. Other studies determined that the proportion of men and women could explain the influence of gender diversity (Ely, 1994; Konrad et al., 1992).
Study Innovations
The majority of social work research on diversity has largely focused on ways to
combat institutional oppression within social service organizations or increase cross-
cultural sensitivity of workers in the provision of care to minority clients (see Fong and
Gibbs, 1995; Hyde, 1998; Mor Barak, 2000; Gummer, 1994; Ferguson, 1996). This study
attempted to deconstruct the actual affects of diversity by using methodology found in
organizational demography research. The first method involved the quantification of
diversity on the individual- and team-level. These two measures provided a standardized
method of quantifying diversity rather than reliance on a more subjective approach. In addition, the definition of diversity was extended to include more obscure characteristics like personality. Aggressive personality was incorporated because a worker’s tendency to 46 engage in aggression would have likely influenced a worker’s psychological safety.
Research on psychological climate was also used to identify outcome variables based on the assumption of a single, underlying variable called psychological climate-general
(PCg). The most important innovation was the study population. These results revealed the implications of diversity specific to case managers in child welfare and juvenile justice case management teams.
Summary
In summary, these results clearly identified the importance of interaction effects
in predicting worker psychological safety. Such interactions required a closer look at
worker and team member demographics to better understand the effect of diversity.
Beyond the influence of more traditional characteristics like race and gender, the effect of
diversity was also relative to more obscure characteristics like personality. There were
consequences of diversity for both minority and non-minority workers. Workers most
socially vulnerable often reported diversity had a negative effect one’s psychological
safety. It was ironic that diversity had a negative effect even within a helping profession
where psychological safety among co-workers would have been expected. It was also
interesting to find a different degree of psychological safety reported by minority
compared to non-minority workers with more team-level diversity. This study provided
additional support for the need to identify how to create a safe work environment for all
workers in social service organizations. These results implied that the effects of diversity
may not be addressed with policy alone, but may require creative interventions to manage
the undercurrents of emotion related to diversity. Therefore, placing different people 47 together may not be a reasonable end in itself for it could bring with it unexpected
consequences that could inhibit worker psychological safety.
Implications for Social Work. Study innovations provided the foundation for new
directions in addressing the effect of diversity on worker psychological safety. Combined with findings of previous research, social service administrators must deal with diversity
in a way that enhances worker psychological safety. If workers do not experience psychological safety, they may not be willing to solicit the emotional support of co- workers in order to be more emotionally available to their clients or solicit the expertise of co-workers to be more creative in resolving client problems. Addressing diversity in a way that enhances worker psychological safety will require administrators to move beyond policy planning and diversity training. One of the first methods of intervention could involve identifying what the organizational culture communicates about the role of diversity by exploring patterns in promotion and demographic make-up of organizational leaders. Also, it would be necessary to deal directly with worker discomfort. Thus more team building, staff support/monitoring, and other hands-on intervention would be essential to help workers manage their personal discomfort within professional bounds.
Social work researchers could contribute by expanding social work knowledge on the relationship between demographic diversity and psychological safety. An important first step would involve collecting data in a way that avoided some of the limitations associated with the present study. At a minimum, that would require a larger number of teams for more variation within and between teams in the study. It would also be important to replicate the current findings and test the influence of additional demographic characteristics such as age, tenure, and educational level. More complex 48 interactions between worker characteristics could be tested as well. For instance, the experience of Caucasian men in teams with an increased number of Caucasian team members might be explored. One of the most important directions for future research would be to determine whether psychological safety mediated the effect of diversity on individual, group, and organizational functioning. Therefore, efforts to address the relationship between diversity and psychological safety may be used to benefit social workers and, ultimately, the people they serve.
49 References
50 Alexander, J. A., Lichtenstein, R., Jinnett, K., D Aunno, T. A., and Ullman, E.
(1996). The effects of treatment team diversity and size on assessments of team
functioning. Hospital and Health Service Administration, 41(1), 37-54.
Alexander, J., Nuchols, B., Bloom, J., and Lee, S. (1995). Organizational demography and turnover: An examination of multiform and nonlinear heterogeneity.
Human Relations, 48(12), 1455-1481.
Altmann, I. And Haythorn, W. (1967). The effects of social isolation and group
composition on performance. Human Relations, 20, 313-339.
Anderson, C. M. and Martin, M. M. (1999). The relationship of
argumentativeness and verbal aggressiveness to cohesion, consensus, and satisfaction in
small groups. Communication Reports, 12(1), 21-31.
Baron, R. A., Neuman, J. H., and Geddes, D. (1999). Social and personal determinants of workplace aggression: Evidence for the impact of perceived injustice and the type a behavior pattern. Aggressive Behavior, 25, 281-296.
Brown, S. P. and Leigh, T. W. (1996). A new look at psychological climate and its relationship to job involvement, effort, and performance. Journal of Applied
Psychology, 81(4), 358-368.
Calabrese, K. R. (2000). Interpersonal conflict and sarcasm in the workplace.
Genetic, Social, and General Psychology Monographs, 126(4), 459-494.
Chatman, J. A., Polzer, J. T., Barsade, S. G., and Neale, M. A. (1998). Being different yet feeling similar: The influence of demographic composition and organizational culture on work processes and outcomes. Administrative Science
Quarterly, 433, 749-780. 51 Chattopadhyat, P. (1999). Beyond direct and symmetrical effects: The influence
of demographic dissimilarity on organizational citizenship behavior. Academy of
Management Journal, 42(3), 273-287.
Chen, C. C. and Eastman, W. (1997). Toward a civic culture for multicultural
organizations. Journal of Applied Behavioral Science, 33(4), 454-470.
Cohen, J. and Cohen, P. (1983). Applied multiple regression/correlation analysis
for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.
Cole, L. L., Grubb, P. L., Sauter, S. L., Swanson, N. G., and Lawless, P. (1997).
Psychological correlates of harassment, threats, and fear of violence in the workplace.
Scandinavian Journal of Work, Environment, and Health, 23(6), 450-457.
Cox, T., Jr., (1991). The multicultural organization. Academy of Management
Executive, 5(2), 34-47.
Cox, T., Jr., Lobel, S., and McLeod, P. (1991). Effects of ethnic group cultural differences on cooperative and competitive behavior on a group task. Academy of
Management Journal, 34, 827-847.
Cox, T., Jr., Welch, J., and Nkomo, S. M. (2001). Research on race and ethnicity:
An update and analysis. In R. T. Golembiewski (Ed.), Handbook of Organizational
Behavior (2nd ed., 255-286). New York: Marcel Dekker, Inc.
Douglas, S. C. and Martinko, M. J. (2001). Exploring the role of individual
differences in the prediction of workplace aggression. Journal of Applied Psychology,
86(4), 547-559. 52 Doverspike, D. and Blumental, A. (2001). Gender issues in the measurement of
physical and psychological safety. The Journal of Prevention and Intervention in the
Community, 22(1), 21-34.
Edmondson, A. (1999). Psychological safety and learning behavior in work
teams. Administrative Science Quarterly, 44(2), 350-383.
Ely, R. J. (1994). The effects of organizational demographics and social identity
on relationships among professional women. Administrative Science Quarterly, 39(2),
203-239.
Ely, R. J. and Thomas, D. A. (2001). Cultural diversity at work: The effects of diversity perspectives on work group processes and outcomes. Administrative Science
Quarterly, 46, 229-273.
Espinoza, J., and Garza, R. (1985). Social team salience and inter-ethnic
cooperation. Journal of Experimental Social Psychology, 21, 380-392.
Fairhurst, G. T. and Snavely, B. K. (1983). A test of the social isolation of male
tokens. Academy of Management Journal, 26(2), 353-361.
Ferguson, S. A. (1996). Towards an anti-racist social service organization. Journal
of Multicultural Social Work, 4(1), 35-48.
Fong, L. G. W. and Gibbs, J. T. (1995). Facilitating services to multicultural communities in a dominant culture setting: An organizational perspective. Administration
in Social Work, 19(2), 1-24.
Fromm, S. (2001). Total estimated cost of child abuse and neglect in the united
states Text [On-line]. Available: http://nccanch.acf.hhs.gov/general/stats/index.cfm 53 Garza, R. and Santos, S. (1991). Inteam/outteam balance and interdependent
inter-ethnic behavior. Journal of Experimental Social Psychology, 27, 124-137.
Geddes, D. and Baron, R. A. (1997). Workplace aggression as a consequence of negative performance feedback. Management Communication Quarterly, 10(4), 433-454.
Glisson, C. (2002). The organizational context of children’s mental health
services. Clinical Child and Family Psychological Review, 5(4), 233-253.
Glisson, C. (2000). Organizational climate and culture. In R. J. Patti (Ed.), The
handbook of social welfare management (pp. 159-218). Thousand Oaks, CA: Sage
Publications.
Glisson, C. and Durick, M. (1988). Predictors of job satisfaction and
organizational commitment in human services organizations. Administrative Science
Quarterly, 33, 61-81.
Glisson, C. and James, L. R. (2002). The cross-level effects of culture and climate in human service teams. Journal of Organizational Behavior, 23, 767-794.
Glisson, C. and Hemmelgarn, A. L. (1998). The effects of organizational climate
and interorganizational coordination on the quality and outcomes of children’s service
systems. Child Abuse and Neglect, 22, 401-421.
Greenberg, J. (1993). The social side of fairness: Interpersonal and informational
classes of organizational justice. In R. Cropanzano (Ed.), Justice in the Workplace:
Approaching Fairness in Human Resource Management (pp. 79-103). Hillsdale, NJ:
Lawrence Erlbaum Associates, Publishers.
Greenberg, J. (1990). Employee theft as a reaction to underpayment inequity:
The hidden cost of pay cuts. Journal of Applied Psychology, 75(5), 561-568. 54 Greenberg, J. and Alge, B. J. (1998). Aggressive reactions to workplace injustice.
In R. W. Griffin, A. O’Leary-Kelly, and M. Collins (Eds.), Dysfunctional behavior in
organizations: Violent and deviant behavior (pp. 83-117). Stamford, CT: Jai Press, Inc.
Greenberg, L. and Barling, J. (1999). Predicting employee aggression against
coworkers, subordinates, and supervisors: The roles of person behaviors and perceived
workplace factors. Journal of Organizational Behavior, 20, 897-913.
Greenhaus, J. H., Parasuraman, S., and Wormley, W. (1990). Effects of race on organizational experiences, job performance evaluations, and career outcomes. Academy
of Management Journal, 33(1), 64-86.
Gummer, B. (1994). Managing diversity in the work force. Administration in
Social Work, 18(3), 123-140.
Harrison, D. A., Price, K. H. and Bell, M. P. (1998). Beyond relational
demography: Time and the effects of surface- and deep-level diversity on work group
cohesion. Academy of Management Journal, 41(1), 96-107.
Harrison, D. A., Price, K. H., Gavin, J. H., and Florey, A. T. (2002). Time, teams, and task performance: Changing effects of surface- and deep-level diversity on group
functioning. Academy of Management Journal, 45(5), 1029-1045.
Hemmelgarn, A. L., Glisson, C. and Dukes, D. (2001). Emergency room culture
and the emotional support component of family-centered care. Children’s Health Care,
30(2), 93-110.
Hoffman, L. (1959). Homogeneity and member personality and its effects on
group problem solving. Journal of Abnormal and Social Psychology, 58, 27-32. 55 Hoffman, L. and Maier, N. (1961). Quality and acceptance of problem solutions
by members of homogeneous groups. Journal of Abnormal and Social Psychology, 62,
401-407.
Hyde, C. H. (1998). A model for diversity training in human service agencies.
Administration in Social Work, 22(4), 19-33.
James, L. R. (1998). Measurement of personality via conditional reasoning.
Organizational Research Methods, 1(2), 131-163.
James, L. A. and James, L. R. (1989). Integrating work environment perceptions:
Explorations into the measurement of meaning. Journal of Applied Psychology, 74, 739-
752.
James, L. R. and Jones, A. P. (1974). Organizational climate: A review of theory and research. Psychological Bulletin, 81(12), 1096-1112.
James, L. R. and McIntyre, M. D. (1997). Perceptions of organizational climate.
In K. R. Murphy (Ed.), Individual differences and behavior in organizations (pp. 416-
450). San Francisco, CA: Jossey-Bass Publishers.
Jones, A. P. and James, L. R. (1979). Psychological climate: Indicators and relationships of individual and aggregated work environment. Organizational Behavior
and Human Performance, 23(2), 201-251.
Kahn, W. H. (1990). Psychological conditions of personal engagement and
disengagement at work. The Academy of Management Journal, 33(4), 692-724.
Kirchmeyer, C. (1995). Demographic similarity to the work group: A
longitudinal study of managers at the early career stage. Journal of Organizational
Behavior, 16(1), 67-83. 56 Kopelman, R. E., Brief, A. P., and Guzzo, R. A. (1990). The role of climate and
culture in productivity. In B. Schneider (Ed.), Organizational climate and culture (pp.
282-318). San Francisco, CA: Jossey-Bass Publishers.
Kossek, E. E. and Zonia, S. C. (1993). Assessing diversity climate: A field study of reactions to employer efforts to promote diversity. Journal of Organizational Behavior,
14, 61-81.
Larkey, L. K. (1996). Toward a theory of communicative interactions in culturally
diverse workgroups. Academy of Management Review, 21(2), 463-491.
Lau, D. C. (1998). Demographic diversity and faultlines: The compositional
dynamic of organizational groups. Academy of Management Review, 23(2), 325-340.
Leck, J. D., Saunders, D. M., and Charbonneau, M. (1996). Affirmative action
programs: An organizational justice perspective. Journal of Organizational Behavior, 17,
79-89.
Lichenstein, R. and Alexander, J. A. (2000). Perceived promotional opportunities
in veteran’s affairs hospitals: A reexamination of relational demography theory. Journal
of Applied Behavioral Science, 36(3), 269-296.
Lindell, M. K. and Brandt, C. (2000). Climate quality and climate consensus as
mediators of the relationship between organizational antecedents and outcomes. Journal
of Applied Psychology, 85(3), 331-348.
Litwin, G. H. and Stringer, R. A. (1968). Motivation and organizational climate.
Cambridge, MA: Harvard Business School, Division of Research. 57 Locke, E. A. (1976). The nature and causes of job satisfaction. In M. D. Dunnette
(Ed.), Handbook of industrial and organizational psychology (pp. 1297-1349). Chicago,
IL: Rand McNally.
Macke, A. S. (1981). Token men and women: A note on the salience of sex and occupation among professionals and semiprofessionals. Sociology of Work and
Occupations, 8(1), 25-38.
Martin, P. Y. (1985). Team sex composition in work organizations: A structural- normative model. Research in the Sociology of Organizations, 4, 311-349.
McDanold, W. R. and Associates (2004). Child maltreatment 2002. Text [On-
Line] Available http://nccanch.acf.hhs.gov/frames/frameset_c.cfm?url=http%3A%2F%
2Fwww%2Eacf%2Ehhs%2Egov%2Fprograms%2Fcb%2Fpublications%2Fcm02%2Find ex%2Ehtm
Miliken, F. J. and Martins, L. L. (1996). Searching for common threads:
Understanding the multiple effects of diversity in organizational groups. Academy of
Management Review, 21(2), 402-433.
Mohammed, S. and Angell, L. C. (2003). Personality heterogeneity in teams:
Which differences make a difference for team performance? Small Group Research,
34(6), 651-677.
Mor Barak, M. E. (2000). Beyond affirmative action: Toward a model of diversity and organizational inclusion. Social Services in the Workplace, 23(3/4), 47-68.
Neuman, G. A., Wagner, S. H., and Christiansen, N. D. (1999). The relationship between work team personality composition and the job performance of teams. Group and Organization Management, 24(1), 28-45. 58 O’Reilly, C. A., III, Williams, K. and Barsade, S. (1997). Group demography and
innovation: Does diversity help? In E. Mannis and M. Neale (Eds), Research in the
management of groups and teams (Vol. 1). Greenwich, CT: JAI Press.
Pelled, L. H., Cummings, T. G., and Kizilos, M. A. (2000). The influence of
organizational demography on customer-oriented prosocial behavior. Journal of Business
Research, 47(3), 209-216.
Pfeffer, J. (1983). Organizational demography. Research in Organizational
Behavior, 3, pp. 1-52.
Randel, A. E. (2002). Identity salience: A moderator of the relationship between
group gender composition and work group conflict. Journal of Organizational Behavior,
23, 749-766.
Reichers, A. E. and Schneider, B. (1990). Climate and culture: An evolution of constructs. In B. Schneider (Ed.), Organizational climate and culture (pp. 5-39). San
Francisco, CA: Jossey-Bass Publishers.
Riordan, C. M. and Shore, L. M. (1997). Demographic diversity and employee attitudes: An empirical examination of relational demography within work units. Journal
of Applied Psychology, 82(3), 342-358.
Rubin, A. and Babbie, E. (2001). Research methods in social work (4th ad.).
Belmont, CA: Wadsworth/Thomson Learning.
Sadlacek, W. E. (1994). Issues in advancing diversity through assessment. Journal
of Counseling and Development, 72, 549-553.
Schneider, B. and Rentsch, J. (1988). Managing climates and cultures: A futures
perspective. In J. Hage (Ed.), Futures of Organizations: Innovating to adapt strategy and 59 human resources to rapid technological change. Issues in organization and management
series (pp. 181-203). Lexington, MA, England: Lexington Books.
Shannon, C. E. and Weaver, W. (1949). The mathematical theory of
communication. Urbana: University of Illinois Press.
Skarlicki, D. P. and Folger, R. (1997). Retaliation in the workplace: The roles of distributive, procedural, and interactional justice. Journal of Applied Psychology, 82(3),
434-443.
Skarlicki, D. P., Folger, R., Tesluk, P. (1999). Personality as a moderator in the
relationship between fairness and retaliation. Academy of Management Journal, 42(1),
100-108.
Smith, K. G., Smith, K. A., Olian, J. D., Sims, Jr., H. P. O'Bannon, D. P. and
Scully, J. A. (1994). Top management team demography and process: The role of social integration and communication. Administrative Science Quarterly, 39, 412-438.
Tajfel, H. (1981). Human groups and social categories: Studies in social
psychology. Cambridge, England: Cambridge University Press.
Teachman, J.D. (1980). Analysis of population diversity. Sociological Methods and Research, 8, 341-362.
Thomas, D. A. (1993). Racial dynamics in cross-race developmental
relationships. Administrative Science Quarterly, 38, 169-194.
Turner, J. (1987). Rediscovering the social group: A social categorization theory.
Oxford, UK: B. Blackwell. 60 Tsui, A. S., Egan, T. D., and O'Reilly, C. A., III (1992). Being different:
Relational demography and organizational attachment. Administrative Science Quarterly,
37, 539-579.
Tsui, A. S. and Gutek, B. A. (1999). Demographic differences in organizations:
Current research and future directions. New York, NY: Lexington Books.
Tsui, A. S. and O’Reilly, C. A., III (1989). Beyond simple demographic effects:
The importance of relational demography in superior-subordinate dyads. Academy of
Management Journal, 32(2), 402-423.
Van Vianen, A. E. M. and de Dreu, C. K. W. (2001). Personality in teams: Its
relationship to social cohesion, task cohesion, and team performance. European Journal
of Work and Organizational Psychology, 10(2), 97-120.
Wagner, W. G., Pfeffer, J., and O’Reilly, C. A., III. (1984). Organizational
demography and turnover in top management groups. Administrative Science Quarterly,
29, 74-92.
Watts, R. J. and Carter, R. T. (1991). Psychological aspects of racism in organizations. Team and Organization Studies, 16(3), 328-344.
Watson, W. E., Kumar, K., and Michaelsen, L. K. (1993). Cultural diversity's
impact on interaction process and performance: Comparing homogeneous and diverse
task groups. Academy of Management Journal, 36(3), 590-602.
Wharton, A. S. and Baron, J. N. (1987). So happy together? The impact of gender
segregation on men at work. American Sociological Review, 52, 574-587. 61 Williams, K. Y. and O’Reilly, C. A. (1998). Demography and diversity in organizations: A review of 40 years of research. Research in Organizational Behavior,
20, 77-140.
62
Appendix
63
Diversity Formulas
Heterogeneity Index (HI) and Coefficient of Variation (CV): The degree to which the distribution of a particular characteristic varies among team members.
For race and gender: (1 − ∑ i² )
One minus the sum of i squared (i being the proportion of the team with the ith characteristic) for the team racial HI and gender HI. A higher score indicates greater variability on the team for that characteristic (Smith et al., 1994).
For personality: (SD ÷ 0)
Divide the standard deviation of the aggressive score by the mean aggressive score for team personality CV. A higher personality CV indicates greater variability in aggressive scores in the team (Alexander et al., 1996).
Relational Demography (RD): The degree to which one team member is different from the rest of the team members in a particular category.
For race, gender and personality: n [ 1/n ∑ (Si − Sj)² ] ½ j = 1 The formula represents the square root of the summed squared differences between an individual Si’s value on a specific demographic variable and the value on the same variable for every other individual Sj in the sample for the work unit, divided by the total number of respondents in the unit (n). N, being the total number of individuals in the unit, includes the person i who is being calculated to derive a metric (ranging from 0 to .95) that specifically indicated the degree of one’s difference from others in the work unit.
64
Table 1. Descriptive Statistics
Variable Percentage Mean SD Individual Level (n = 82): Gender Male 20 Female 80 Race Non-Caucasian 30 Caucasian 70 Education High School Diploma 5 Some College 5 Bachelor’s Degree 50 Some Graduate Work 30 Graduate Degree 10 New Team Member 10 Aggressive Personality (range 0 – 8) 3.69 1.72 Age (range 21 – 62) 39.74 11.47 Years of Experience 9.00 8.31 (range 0 – 29) Organizational Tenure 5.85 8.03 (range 0 – 29) Gender RD .48 .27 (range .00 – .95) .49 .26 Racial RD (range .00 – .95) Personality RD 15.16 12.32 (range 3 – 56) Psychological Climate 68.22 15.39 (range 33 – 97)
Team Level (n = 10): Urban Locale 45 .45 Team Size (range 6 – 12) 8 2.50 Number of New Team Members 1.00 .66 (range 0 – 2) Organizational Turnover 5 3 (range 2 – 11) Gender HI .30 .16 (range .00 – .50) Racial HI .30 .16 (range .00 – .48) Personality CV .47 .11 (range .29 – .71)
65
Table 2. Measurement Alphas (n = 82)
Number of Items Alpha Reliability Psychological Climate 20 .78
Climate Subscales: Depersonalization 5 .68 Emotional Exhaustion 6 .92 Fairness 6 .65 Growth & Advancement 5 .84 Personal Accomplishment 6 .65 Role Clarity 6 .84 Role Conflict 9 .87 Role Overload 8 .86 Cooperation 4 .77
66 Table 3. Correlation Matrix (n = 82)
Female Caucasian Personality Racial Gender Pers. Racial Gender Pers. RD RD RD HI HI CV Female 1 Caucasian .07 1 Personality -.03 -.03 1 Racial RD -.08 -.48** .03 1 Gender RD -.69** .04 .00 -.12 1 Pers. RD .15 .10 .14 .07 -.21 1 Racial HI .03 -.28* .08 .81** -.19 .19 1 Gender HI -.29* .03 -.03 -.23* .83** -.09 -.25* 1 Pers. CV .05 .15 .09 .08 -.21 .66** .13 -.14 1 Psych. Safety .10 -.26* -.13 .11 -.23* .02 .08 -.16 -.21 Depers. -.06 .14 .17 -.11 .14 .01 -.10 .13 .23* Emotional -.08 .09 .09 .03 .19 -.01 .02 .16 .22* Exhaustion Fairness .05 -.26* -.06 .24* -.18 -.03 .23* -.10 -.08 Growth .09 -.30** -.16 .22 -.13 -.10 .15 -.08 -.22* and Adv. Personal -.12 -.10 .18 .06 -.05 .05 -.00 -.09 .03 Acc. Role Clarity .01 -.12 .04 .06 -.02 -.13 -.01 .07 .02 Role Conflict -.10 .34** .11 -.19 .13 -.01 -.10 .02 .13 Role Over. -.02 .30** .21 -.16 .11 -.02 -.07 .04 .14 Cooperation -.16 .14 -.03 .02 .20 .04 .09 .17 .01 *p<.05; **p<.01 (two-tailed).
67
Table 3. Continued.
Psych. Depers. Emo. Fairness Growth Personal Role Role Role Co Safety Exhaust. and Adv. Acc. Clarity Conflict Over. op. Female Caucasian Personality Racial RD Gender RD Pers. RD Racial HI Gender HI Pers. CV Psych. Safety 1 Depers. -.72** 1 Emotional -.91** .60** 1 Exhaustion Fairness .58** -.30** -.41** 1 Growth .51** -.32** -.44** .50** 1 and Adv. Personal .07 -.07 -.09 -.12 .22* 1 Acc. Role Clarity .42** -.17 -.31** .30** .43** .13 1 Role Conflict -.91** .49** .71** -.64** -.51** -.08 -.50** 1 Role Over. -.86** .48** .76** -.53** -.43** -.06 -.44** .86** 1 Cooperation .12 -.18 -.12 .04 .18 -.02 .42** -.09 -.05 1 *p<.05; **p<.01 (two-tailed).
68
Table 4. Hierarchical Regression Analysis for HI and CV Unstandardized/Standardized Beta Coefficients
Variable Psych. Safety Depers. Emo. Exhaust. Fairness Growth and Adv. Step 1: Constant 104.51 6.16 2.78 21.64 22.23 Female 1.19 .03-.732 -.09 .26 .02 .03 .00 -.31 -.03 Caucasian -9.19* -.28* .511 .07 1.52 .11 -2.32* -.27* -3.47*** -.40*** Aggressive -1.13 -.13 .191 .010 .48 .12 -.36 -.15 -.78** -.32** Education -5.86** -.35** .36 .10 2.09* .28* -1.78*** -.39*** -.50 -.11 New Team 2.32 .03 -1.22 -.07 -4.15 -.13 -2.86 -.14 -2.13 -.11 Member Age .23 .17-.07 -.25 -.08 -.13 .07 .18 -.08 -.23 Org. Tenure -.84*** -.46*** .10 .24 .35** .44** -.23*** -.47*** -.15* -.30* No. of New .92 .04 -.63 -.12 .38 .04 -.07 -.01 .42 .07 Team Members Racial HI 2.63 .03 -2.60 -.12 1.79 .04 4.03 .15 2.97 .11 Gender HI -18.76 -.19 3.47 .17 8.30 .21 -3.06 -.12 -2.58 -.10 Personality CV -13.17 -.09 8.35* .28* 3.47 .06 3.04 .08 -1.15 -.03 R Square .34 NA NA .36 NA
Step 2: Female × -55.52** -.67** 7.23 .40 15.75 .45 -7.83 -.35 -4.50 -.21 Gender HI Caucasian × 56.57 .68 -8.96 -.50 -17.71 -.50 15.87* .72* 6.68 .31 Racial HI Aggressive × 11.56 .82 -1.44 -.45 -3.29 -.53 5.46* 1.39* 2.83 .73 Personality CV *p<.05; **p<.01; ***p<.001. NA, not applicable-- R Square for equations only with significant interactions in Step 2.
69
Table 4. Continued.
Variable Personal Acc. Role Clarity Role Conflict Role Overload Cooperation Step 1: Constant 22.53 29.87 7.44 12.13 14.47 Female -1.46 -.15 -.45 -.04 -.47 -.03 .94 .06 -.94 -.13 Caucasian -.97 -.12 -2.89* -.31* 6.20*** .39*** 4.48** .32** .94 .15 Aggressive .41 .19 -.09 -.04 .75 .17 .80 .20 -.09 -.05 Education .76 .18 -1.36* -.27* 3.30*** .39*** 2.70** .36** -.43 -.13 New Team -2.55 -.14 -6.03* -.27* 3.16 .09 -2.78 -.08 -3.24 -.22 Member Age -.08 -.25 -.07 -.18 -.03 -.04 -.03 -.05 .00 .00 Org. Tenure .12 .26 -.09 -.17 .41*** .46*** .24* .29* -.01 -.02 No. of New -.07 -.01 .66 .10 -1.14 -.10 -.47 -.05 -.53 -.12 Team Members Racial HI -1.43 -.06 -2.34 -.08 -3.79 -.08 -.31 -.01 3.41 .18 Gender HI -1.68 -.07 .01 .00 2.97 .07 4.13 .10 4.19 .23 Personality CV -1.60 -.05 3.45 .09 .65 .01 .75 .01 .63 .02 R Square NA .23 .39 .28 .17
Step 2: Female × -6.53 -.32 -5.40 -.23 27.94** .70** 21.56* .60* 2.87 .18 Gender HI Caucasian × 11.24 .56 18.92* .80* -32.10* -.80* -27.22* -.76* -10.90 -.69 Racial HI Aggressive × -1.11 -.31 1.50 .36 -7.19 -1.01 -5.17 -.81 4.64* 1.65* Personality CV *p<.05; **p<.01; ***p<.001. NA, not applicable-- R Square for equations only with significant interactions in Step 2.
70
Table 5. Unstandardized Beta Coefficients for HI and CV Significant Interaction Equations
Variable Psych. Fairness Role Role Role Overload Coop. Safety Clarity Conflict Step 2 Constant 92.22 26.86 30.57 36.10 12.65 -3.1316.15 3.17 22.06 Female 2.12 .04 .71 -.43 -.80 -.50.69 .92 -.36 Caucasian 6.73 -7.90** -2.39* -9.53** -1.81 17.48***-1.71 14.04** .88 Aggressive -1.62 -.39 -3.14* -.13 .91* .81 .92* .85 -2.45* Education -5.51** -1.80*** -1.52** -1.38* 3.23*** 3.33*** 2.64** 2.72** -.21 New Team -.65 -2.72 -3.65 -5.87* 4.77 2.88 -1.54 -3.02 -3.91* Member Age .28 .06 .06 -.08 -.04 -.01.04 -.02 -.003 Org. Tenure -.90*** -.24*** -.20** -.10 .42*** .42*** .24* .24* .02 No. of New -.19 .98 -.30 .85 -.56 -1.48 -.16 -.75 -.73 Team Members Racial HI -1.44 -9.87 4.74 -18.91* -2.77 24.33 .48 23.53 4.01 Gender HI 13.57 -4.32 -1.02 -1.50 -13.30 5.52 -8.43 6.29 5.91* Personality CV -5.24 3.97 -18.92 4.56 -1.75 -1.24 -1.10 -.85 -18.03* Female x -55.52** NA NA NA 27.94** NA 21.56* NA NA Gender HI Caucasian x NA 15.87* NA 18.92* NA -32.10* NA -27.22* NA Racial HI Aggressive x NA NA 5.46* NA NA NA NA NA 4.64* Personality CV R Square ∆ .07 .04 .05 .05 .05 .09 .04 .07 .07 *p<.05; **p<.01; ***p<.001. NA, not applicable-- interaction factor not included in the model. 71
Table 6. F statistic for Simple Main Effects
Variable Psych. Depers. Emo. Fairness Growth and Personal Role Role Role Coop. Safety Exhaust. Advance. Accomp. Clarity Conflict Over. Gender x GHI: Male 1.74 NA NA NA NA NA NA 2.83 4.24 NA Female 2.66 NA NA NA NA NA NA .23 .86 NA Gender x GRD: Male NA NA NA NA NA NA NA NA NA NA Female NA NA NA NA NA NA NA NA NA NA
Race x RHI: Non-Caucasians NA NA NA .50 NA NA 4.90* 2.72 3.99 NA Caucasians NA NA NA 4.39* NA NA .20 .55 .39 NA Race x RRD: Non-Caucasians 17.83*** 8.75** 13.05** 2.66 NA NA NA 5.59* 9.09** NA Caucasians 2.49 4.19* .99 7.77** NA NA NA 2.56 3.05 NA
Aggressive x PCV: Non-Aggressive 1.16 .59 Aggressive NA NA NA 2.20 NA NA NA NA NA .53 Aggressive x PRD: Non-Aggressive .08 Aggressive NA NA NA NA NA NA NA .99 NA NA
*p<.05; **p<.01; ***p<.001. NA, not applicable-- no significant interaction effects.
72
Table 7. Hierarchical Regression Analysis for RD Unstandardized/Standardized Beta Coefficients
Variable Psych. Safety Depers. Emo. Exhaust. Fairness Growth and Adv. Step 1: Constant 108.04 10.20 -1.43 24.67 21.83 Female -5.36 -.14-.59 -.07 3.28 .20 -1.45-.14 -.86 -.09 Caucasian -9.31* -.28* .54 .08 2.20 .16 -1.85 -.21 -3.42** -.39** Aggressive -1.51 -.17 .23 .12 .71 .18 -.35 -.15 -.85** -.36** Education -5.68** -.34** .29 .08 2.03* .28* -1.76*** -.38*** -.41 -.09 New Team 3.55 .05 -1.92 -.12 -4.14 -.13 -2.82 -.14 -2.04 -.10 Member Age .23 .17 -.08 -.29 -.06 -.10 .07 .18 -.08 -.22 Org. Tenure -.89*** -.49*** .13* .32* .36*** .45*** -.23*** -.46*** -.17** -.34** Racial RD -.43 -.01 -1.68 -.13 2.72 .11 2.34 .15 1.71 .11 Gender RD -17.42 -.30 1.17 .09 7.10 .29 -4.41 -.29 -1.70 -.11 Personality RD .07 .06 .02 .05 -.08 -.13 .01 .04 .04 .10 No. of New -.06 -.00 -.14 -.03 .96 .10 .15 .02 .17 .03 Team Members R Square .35 .16 .27 .38 NA
Step 2: Female × 36.79 .59 -.28 -.02 -1.17 -.04 33.72 2.04 13.65 .84 Gender RD Caucasian × 55.15** .98** -10.15** -.83** -19.22** -.80** 11.39** .76** 5.72 .38 Racial RD Personality × .07 .40 -.01 -.15 -.01 -.11 .00 .09 .01 .11 Personality RD *p<.05; **p<.01; ***p<.001. NA, not applicable-- R Square for equations only with significant interactions in Step 2.
73
Table 7. Continued.
Variable Personal Acc. Role Clarity Role Conflict Role Overload Cooperation Step 1: Constant 24.58 33.04 6.18 8.78 14.87 Female -2.72 -.29-1.38 -.12 1.00 .05 3.31 .20 -.26 -.04 Caucasian -1.10 -.14-2.39 -.25 5.76** .36** 4.30* .30* .51 .08 Aggressive .36 .16 .01 .00 .81 .18 .91 .23 -.19 -.11 Education .67 .16 -1.53** -.31** 3.31*** .40*** 2.77** .37** -.40 -.12 New Team -2.42 -.13 -6.01* -.27* 2.88 .08 -2.95 -.09 -3.51 -.24 Member Age -.09 -.27-.08 -.20 -.03 -.04 -.02 -.03 .00 .01 Org. Tenure .12 .26 -.06 -.12 .43*** .48*** .25** .31** -.03 -.08 Racial RD -1.21 -.08 -.43 -.03 -2.94 -.10 -.54 -.02 .52 .05 Gender RD -2.76 -.20 -2.39 -.15 4.00 .14 5.62 .23 2.64 .24 Personality RD .00 .00 -.04 -.12 -.03 -.04 -.03 -.06 .05 .20 No. of New -.02 -.00 1.21 .18 -1.186 -.11 -.51 -.05 -.53 -.12 Team Members R Square NA NA .40 .30 NA
Step 2: Female × 16.73 1.11 24.82 1.39 -37.53 -1.25 -44.89 -1.68 10.74 .91 Gender RD Caucasian × 6.94 .50 3.02 .19 -20.76** -.76** -20.18** -.83** -3.21 -.30 Racial RD Personality × .02 .42 .00 .06 -.07* -.82* -.06 -.73 -.00 -.06 Personality RD *p<.05; **p<.01; ***p<.001. NA, not applicable-- R Square for equations only with significant interactions in Step 2.
74
Table 8. Unstandardized Beta Coefficients for RD Significant Interaction Equations
Variable Psych. Depersonalization Emotional Fairness Role Role Safety Exhaustion Conflict Overload Step 2 Constant 138 4.73 -11.79 30.81 -5.01 1.88 -2.10 Female -4.14 -.64 3.18 -1.40 .90 -.17 3.21 Caucasian -43.66*** 6.69** 13.86** -8.76** 18.35*** 5.90** 16.54*** Aggressive -1.73 .29 .81 -.41 .92 2.62** 1.02** Education -5.29** .19 1.85** -1.65*** 3.11** 3.17*** 2.58** New Team -.43 -1.33 -3.02 -3.48 4.09 2.36 -1.77 Member Age -.06 -.06 -.003 .03 3.43 -.06 .04 Org. Tenure -.69** .09 .29** -.19** .36** .39*** .17 Racial RD -42.38** 5.99 17.25** -6.28 12.76* -3.08 14.72** Gender RD -13.37 .66 6.14 -3.84 2.95 3.47 4.61 Personality RD .12 .004 -.10 .03 .05 .35 -.06 No. of New .24 -.16 .93 .17 -1.22 -1.66 -.54 Team Members Female x NA NA NA NA NA NA NA Gender RD Caucasian x 55.15** -10.15** -19.22** 11.39** -20.76** NA -20.18** Racial RD Aggressive x NA NA NA NA NA -.07* NA Personality RD R Square ∆ .11 .08 .08 .07 .07 .04 .08 *p<.05; **p<.01; ***p<.001. NA, not applicable-- interaction factor not included in the model.
75
Table 9. Significant Simple Main Effects for Race
High Racial Heterogeneity Index (RHI)
Non-Caucasians Caucasians
- role clarity + fairness
High Racial Relational Demography (RRD)
Non-Caucasians Caucasians
- psychological safety + fairness
+ depersonalization - depersonalization
+ emotional exhaustion
+ role conflict
+ role overload 76
Figure 1a. Perceptions of Psychological Safety:
90 81.94 80 77.27 70 75.15 60 56.29 50 Male 40 Female 30
Psychological Safety Psychological 20 10 0 Gender Heterogeneity Index
Figure 1b. Perceptions of Role Conflict:
35 31.09 30
25 24.57 23.77 20 Male 17.92 Female 15 Role Conflict Role 10
5
0 Gender Heterogeneity Index
Figure 1. Gender Interaction Graphs.
77
Figure 1c. Perceptions of Role Overload:
40 36.81 35 30.24 30 29.55 25 25.33 Male 20 Female
15 Role Overload 10
5
0 Gender Heterogeneity Index
Figure 1. Continued.
78
Figure 2a. Perceptions of Psychological Safety:
120
100 99.78
80 Non-Caucasian 68.25 Caucasian 60 56.12 59.52 40
Psychological Safety 20
0 Racial Relational Demography
Figure 2b. Perceptions of Depersonalization:
12 10.9 9.9 10
8 Non-Caucasian 6 Caucasian 6.95 4 4.21 Depersonalitzation 2
0 Racial Relational Demography
Figure 2. Race Interaction Graphs. 79
Figure 2c. Perceptions of Emotional Exhaustion:
25 20.06 20 17.53 15.66 15 Non-Caucasian 10 Caucasian
5 Emotional Exhaustion Emotional 3.67 0 Racial Relational Demography
Figure 2d. Perceptions of Role Clarity:
30 29.02 25 19.94 20 19.49 19.5 15 Non-Caucasian Caucasian
Role Clarity 10
5
0 Racial Heterogeneity Index
Figure 2. Continued. 80
Figure 2e. Perceptions of Fairness:
25 21.76 20 17.02 16.74 15 13.86 Non-Caucasian 10 Caucasian Fairness
5
0 Racial Heterogeneity Index
Figure 2f. Perceptions of Fairness:
25
21.21 20 17.38 15 Non-Caucasian 15.24 Caucasian 12.52
Fairness 10
5
0 Racial Relational Demography
Figure 2. Continued. 81
Figure 2g. Perceptions of Role Conflict:
35 29.96 30 26.23 25 24.16 20 Non-Caucasian 15 12.48 Caucasian
Role Conflict Role 10
5
0 Racial Heterogeneity Index
Figure 2h. Perceptions of Role Conflict:
180 160.75 154.52 160 153.15 140 142.4 120 100 Non-Caucasian 80 Caucasian 60 Role Conflict Role 40 20 0 Racial Relational Demography
Figure 2. Continued. 82
Figure 2i. Perceptions of Role Overload:
35 31.33 29.56 30 28.58 25 20 Non-Caucasian 17.29 15 Caucasian
Role Overload 10
5 0 Racial Heterogeneity Index
Figure 2j. Perceptions of Role Overload:
35 32.64 30.08 30 27.45 25
20 16.1 Non-Caucasian 15 Caucasian
Role Overload 10
5
0 Racial Relational Demography
Figure 2. Continued. 83
Figure 3a. Perceptions of Fairness:
20 17.6 18 16.71 16 14.18 14 12 Non-AG 12.25 10 AG
Fairness 8 6 4 2 0 Personality Coefficient of Variation
Figure 3b. Perceptions of Cooperation:
18 16.24 16 15.41 14 12 12.44 12.51 Non-AG 10 AG 8
Cooperation 6 4 2 0 Personality Coefficient of Variation
Figure 3. Personality Interaction Graphs. 84
Figure 3c. Perceptions of Role Conflict:
35 32.04 29.08 30 27.56 25
20 Non-AG 20.80 AG 15
Role Conflict 10
5
0 Personality Relational Demography
Figure 3. Continued.
85
VITA
Ann Callahan received her Ph.D. in Social Work from The University of Tennessee. Her research interests include organizational demography, social work service quality, and gerontology. Dr. Callahan is also interested in teaching evidenced-based practice and engaging in student mentorship. She has extensive social work practice experience as well as a license in clinical social work.