The Influence of Academic Values and Belongingness

Home , Belongingness, School belonging

The Influence of Academic Values and Belongingness Concerns on Achievement Goals, Self-Efficacy, and Perceived Stress in First Quarter Freshmen: Relationships to Academic Performance and the Mediating Role of Procrastination

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Gary J. Kennedy, MA

Graduate Program in Educational Policy & Leadership

The Ohio State University

2009

Dissertation Committee:

Dr. Bruce W. Tuckman: Advisor

Dr. Lynley H. Anderman

Dr. Richard G. Lomax

Gary J. Kennedy

2009 Abstract

This study assesses the influence of student values on long-term self-regulatory decisions defined in terms of a tendency to procrastinate and how these values, indirectly through procrastination, but also directly, affect important motivational, affective, social and behavioral academic outcomes of first quarter freshmen. Results of a structural equation model showed that a concern over potential social exclusion may significantly increase the likelihood of procrastination, but that academic task and grade values may attenuate this influence. In addition, procrastination tendency assessed early in the first term had a significant direct negative influence on self-efficacy, and school belongingness, and was significantly positively related to perceived stress near the end of the term. There was also a significant negative total effect of procrastination on end of term grade point average. No significant relationship was found between procrastination and the achievement goal orientations (i.e., mastery-approach and performance-approach).

Academic and social values had the predicted differential relationships with achievement goal. Specifically, academic task values were positively related to a mastery-approach, but not a performance-approach goal orientation and a concern over social exclusion had a positive relationship with a performance-approach, but not a mastery-approach goal orientation. Finally, as expected, self-efficacy had a significant positive relationship and perceived stress a significant negative relationship with end of term grade point average.

An unexpected direct negative relationship and a non-significant total effect were found

ii between perceived school belongingness and grade point average. However, this may

have been due to a suppressor effect implying that school belongingness may actually have an indirect beneficial influence on academic performance through its positive

relationship with self-efficacy and negative relationship with perceived stress. Total

indirect effects substantiate this claim. Results are discussed in terms of student social

identity formation in the context of a goal hierarchy conceptualization of self-regulation

and its influence on students’ motivational resources and will.

iii Dedication

To Lisa (and maggie the mooch, too)

Eccl 12:1

Acknowledgments

This project and I have had the support of many very good and gracious people. First, I want to thank my advisor, Dr. Bruce Tuckman, from whom I have learned a great deal.

You can teach old dogs new tricks but you need a teacher the caliber of Dr. Tuckman.

His work, motivated by a desire to help students, was a major impetus for my decision to start and complete this project. His support and encouragement throughout this project is truly appreciated and I am honored to be able to claim that I am his student. I am also extremely grateful to Dr. Lynley Anderman and Dr. Richard Lomax. Their advice and support is immeasurable for me and I count myself as fortunate for having had the pleasure and the privilege of being the recipient of their expertise. Dr. Heather Davis and

Dr. John Gibbs were also instrumental in helping me get my bearings early in my program. Their instruction has influenced me considerably and is reflected in this current study. Completing this project while working full time would not have been possible without the support of Ms.Gail Stephenoff. There are no words to express my appreciation for her support throughout my tenure as a student. Many other people were also instrumental in various phases of this project. I want to express my thanks to Dr.

Martha Garland, Dr. Melinda McDonald, Ms. Beth Pittman, Dr. Tracy Tupman, and Dr.

John Wanzer and all of the advisors and instructors in the Exploration and Business units for allowing me access to the students in their charge. Of course, I would also like to

v thank these students – God-speed to all of you. Also, Dr. D’Arcy Oakes was my IT benefactor. His willingness to help get the project off the ground was a godsend. Thanks also to the folks who work with CARMEN for helping extract the data. I would also like to express my thanks and appreciation for Dr. Virginia Gordon, Dr. George Steele, and

Dr. Thomas Minnick. It has been my privilege to have worked with and for them over the years and I am truly thankful for their advice, support, and friendship. I have learned a great deal from them. I would also be remiss if I didn’t mention that learning and working at The Ohio State University has been a pleasure. This wonderful university has provided so much support over the years. I cannot sufficiently express my gratitude for having had the privilege of being here. Finally, I want to thank my wife Lisa, first for just putting up with me, but also for her love and support. She shared me with this program and project for four years and usually got the tired part. There is really nothing else to say except that I am blessed for having her in my life.

vi Vita

March, 1981…………………………B.S. Psychology, Youngstown State University

June, 1983……………………………M.A. Psychology, The Ohio State University

September, 1983…………………… Graduate Teaching Associate, Department of Psychology, The Ohio State University

September, 1987……………………..Graduate Administrative Associate, University College, The Ohio State University

September, 1990…………………… Academic Advisor, University College, The Ohio State University

September, 1992 to present………… Adjunct Lecturer, Psychology, Columbus State Community College

September, 1992……………………..Program Manager, Alternatives Program, University College, The Ohio State University

September, 1993…………………… Adjunct Lecturer, Psychology, Otterbein University

January, 1994………………………. Statistical Specialist, Exel Logistics, Westerville, Ohio

March, 1995………………………... Academic Advisor/Research Assistant, University College, The Ohio State University

August, 1997……………………….. Data Manager/Computer Administrator, University College, The Ohio State University

September, 2001 – June, 2003………Adjunct Lecturer, Psychology, Capitol University

September, 2000 to present………… Senior Statistical Specialist, Office of Enrollment Management, The Ohio State University

vii Publications

Steele, G., Kennedy, G. J., & Gordon, V. (1993). Retention of major-changers: a longitudinal study. Journal of College Student Development, 34, 58 – 62.

Kennedy, G. J., Gordon, R. L., & Gordon, V. (1995). Changes in social and academic integration in freshmen of high and average ability: Implications for retention. NACADA Journal, 15, 9-19.

Fields of Study

Major Field: Educational Policy & Leadership Specialization: Educational Psychology

viii Table of Contents

Abstract...... ii

Dedication...... iv

Acknowledgments...... v

Vita...... vii

List of Tables...... xii

List of Figures...... xiii

Chapter 1: Introduction and Review of the Literature...... 1 Introduction...... 1 Goals and Values...... 3 Goal Structure...... 6 Goal Content...... 7 Regulation of the Action Hierarchy and the Need for Belongingness...... 7 Regulation of the Action Hierarchy and the Role of Affect...... 15 Regulation of the Action Hierarchy and the Role of Personality...... 16 Goal Processes...... 18 Hierarchical Processes...... 19 Congruency of Academic and Social Goal Processes: The Relevance of School Belongingness, Perceived Social Exclusion, and Group Processes...... 22 Achievement Goal Specificity and Goal Setting...... 42 Goal Focus and Orientation...... 44 Values...... 48 Definition and Relevance...... 48 Values and Interest...... 50 Values and Goal Theory...... 51 Self-Regulation: Linking Values to Achievement Goals, School Belongingness, Self-Efficacy, and Perceived Stress...... 58 Procrastination, Value Conflict, and the Potentially Debilitating Effects of Perceived Social Exclusion ...... 61

ix The Consequences of Procrastination: Effects on Learning and Performance Goals...... 65 The Consequences of Procrastination: Effects on Self-Efficacy 68 The Consequences of Procrastination: Effects on Perceived Stress...... 71 Self-Efficacy as a Mediating Influence on Perceived Stress and Academic Performance...... 73 The Role of Self-Efficacy on Academic Outcomes...... 73 The Influence of Student’s Achievement Goals On Self-Efficacy...... 74 The Relationship of Perceived Stress to Self-Efficacy...... 77 Perceived Stress, the Depletion Resource Model, and the Benefit of Feelings of School Belonging...... 79 Summary and Overview of the Model and Hypotheses...... 82 Overview of the Proposal...... 82 Hypotheses...... 94

Chapter 2: Methods 99 Participants 99 Procedure...... 100 Data Collection...... 100 Analyses...... 101 Measurement of Constructs...... 102 Description of Instruments...... 103 The Construction of Parcel Indicators: Rationale...... 107 The Construction of Parcel Indicators: Procedure...... 111 Estimation of the Measurement Model...... 116 Estimation of the Structural Model and Testing of the Structural Hypotheses...... 117

Chapter 3: Results...... ……………………………………………...... 119 General Descriptive Statistics of Gender, Ethnicity, and Ability...... 119 Preliminary Exploratory Factor Analyses Determining Dimensionality of Constructs...... 121 Confirmatory Factor Analysis: Measurement Model...... 126 Structural Model: Direct Effects...... 130 Structural Model: Total and Indirect Effects………………………..136

Chapter 4: Discussion...... ……………………………………………...... 140 Summary and Interpretation of Results……………………………..140 Significance of the Study…………………………………………....154 Limitations of the Study...... 159 Suggestions for Future Research...... 161 Conclusion...... 163

References...... 166

Footnotes...... 190

Appendix A: Cover Letter to Students ……………………..………………………191

Appendix B: E-mail Notifications to Students……………………………………...193

Appendix C: Online Consent to Participate in Research…………………………....196

Appendix D: E-mail Notification to Students Verifying Submission of Questionnaire and Consent to Records Access………………………………….. 198

Appendix E: Descriptive Statistics and Multivariate Skewness and Kurtosis of Items By Construct...... 199

Appendix F: Descriptive Statistics and Multivariate Skewness and Kurtosis of Parcel Indicators by Construct and Assigned Item...... 205

Appendix G: Correlation Matrices of Items Used in Preliminary Factor Analysis... 210

Appendix H: Pattern Loading and Factor Structure coefficients Defining the Dimensionality of Six Latent Constructs………………………………………... 220

Appendix I: Correlation Matrix of the Parcel Indicators…………………………... 226

Appendix J: Factor Loadings for Measurement Models…………………………… 228

Appendix K: Factor Loadings for Final Structural Model S2 Compared to Measurement Model M2………………………………………………………… 229

xi List of Tables

Table 1.1: Hypothesized Directions of the Structural Relationships...... 96

Table 2.1: Hypothesized Directions of the Structural Relationships (Reproduction)...... 118

Table 3.1: Gender and Ethnicity Distributions of All Students, Non-Responders, Partial Responders, and Complete Responders...... 119

Table 3.2: Descriptive Statistics of Ability Measures for All Students, Non-Responders, Partial Responders, and Complete Responders...... 120

Table 3.3: Parallel Analyses of Dimensional Structure: Eigenvalues for the Sample and Randomly Generated Correlation Matrices...... 121

Table 3.4: Maximum Likelihood Analyses of Dimensional Structure: Goodness of Fit Measures from the Extraction of the Number of Dimensions Determined by Parallel Analyses...... 123

Table 3.5: Factor Determinacy Values Resulting from the Extraction of p and p+1 Dimensions………………………………………………………….... 124

Table 3.6: Fit Indices for Measurement Models…………………………………… 127

Table 3.7: Correlation Matrix of Latent Variables from Model M1……………….. 128

Table 3.8: Correlation Matrix of Latent Variables from Model M2……………….. 130

Table 3.9: Fit Indices for Structural Models……………………………………….. 131

Table 3.10: Hypothesized Direction and Standardized Coefficients of Direct Effects…………………………………………………………………….. 132

Table 3.11: Correlations between Exogenous Variables in the Structural Model (S1)...... 134

Table 3.12: Standardized Total and Indirect Effects for Values, School Belongingness, and Procrastination...... 138

xii List of Figures

Figure 1: Path Diagram of the Final Structural Model (S2)……………………….. 137

xiii

Chapter 1: Introduction and Review of the Literature

Introduction

The first year in college is perhaps the most crucial, not only influencing

students’ decision to remain at an institution (Tinto, 1993), but also positioning

students academic achievement. In particular, for many students, the first academic

semester or quarter is a time of hectic, yet exciting transition that may have

potentially enduring consequences for achievement in college. Specifically,

unpublished institutional research data (Kennedy, 2008) suggests that the level of

academic achievement measured as first quarter grade point average is highly

predictive of four, five, and six year graduation. Thus, students’ first term in college

may be a pivotal time. One reason for the importance of the first term is that this

transition is necessarily a time of separation from family and past social and

educational associations (Tinto, 1993). In separating from past associations, students

must make this transition, in large part, through the formation of new associations.

Tinto (1993) has shown how difficulties in integrating in the academic and

social milieu of an institution can be a very important factor in students’ decision to

leave that institution voluntarily and perhaps never earn a degree. His analysis suggests that an important factor in decisions to leave college early is perceived

incongruence between the student and institution. Incongruence or the “mismatch

between the individual and the institution” (p. 51) is both academic and social in

nature. Academic incongruence involves a mismatch between one’s abilities and

interests and the demands of the academic system of the institution. Social

incongruence occurs when students’ values do not match those of the formal values

of the social system of the institution and/or the values of other individual students,

faculty, or staff informally assessed through day-to-day, interactions (Tinto, 1993).

However, social and academic incongruence occurs not only between the

student and the institution, but also within a student. Specifically, the perceptions

students have of their tenure at an institution are affected at least as much by how

they integrate their own social and academic values as by how these values are

integrated in the institution. Tinto’s sociological and institutionally-based analysis, while giving a nod of acknowledgement to the importance of student effort and motivation, has policy implications that appear to involve more of what institutions can do “to” students as opposed to opportunities to help students do “for” themselves.

While not denying the validity or usefulness of Tinto’s recommendations, it would seem that a psychological analysis emphasizing the congruence or incongruence of students’ academic and social values is a necessary component in informing policy decisions.

This study focuses on academic and social values students possess as they enter their first term at a major research university and how these values and the goals they charge influence their tendency to procrastinate generally. The tendency to procrastinate is a measure of a lack of self-regulation that can have debilitating

influences on motivational, affective, social, and behavioral outcomes over the course of a student’s first term and ultimately affect academic achievement. The hypothesized source of these outcomes is the congruence or incongruence of social and academic values that guide in the modulation of the links between one’s sense of social identity, a high-level goal of affiliation, and lower levels goals, particularly achievement goals. Values help guide decisions and procrastination and self- regulation in general are hypothesized to be substantial decisions that can play a rather significant role in students’ academic outcomes. Institutional intervention techniques of adding programs intended to help integrate students within the academic and social milieu, while clearly beneficial, may be deficient if at least some of these programs do not help students reflect (and possibly change) on their own motivational circumstances. This process begins with goals and values.

Goals and Values

Students pursue many goals (Harackiewicz, Barron, Pintrich, Elliot, & Thrash,

2002; Harackiewicz, Barron, Tauer, & Elliot, 2002; Pintrich, 2000a; Urdan & Maehr,

1995; Wentzel, 1989; Wentzel, 1993) and while education may have different meanings for students (Henderson-King & Smith, 2006), there is a considerable amount of research that suggests that students attempt to achieve two fundamental goals in academic contexts – to develop competence (i.e. mastery or learning goals) and/or to appear competent (or not to appear incompetent) (i.e. performance goals)

(e.g., Ames & Archer, 1988; Dweck & Legget, 1988; Elliot & Dweck, 1988;

Harackiewicz, Barron, Pintrich et al., 2002; Nicholls, 1984). In addition, the pursuit

of competence-related goals may depend to some extent on social goals (Urdan and

Maehr, 1995; Wentzel 2005) including seeking social approval, solidarity,

compliance, dependability, and belongingness (Wentzel, 2005; see also Ford, 1992)

that can affect academic achievement outcomes. For example, Wentzel (2005) has

shown how social competence goals can influence academic competence goals.

Goals, in general, are thought to be fundamental to the motivation of behavior.

To quote Bandura (1997), “behavior is motivated and directed by cognized goals

rather than being pulled by an unrealized future state” (p.128). Furthermore,

regardless of theoretical orientation, it is generally agreed that self-regulation and decision making occur necessarily as processes responsible for the initiation and maintenance of and striving toward goals (Bandura, 1997; Carver & Scheier, 1998;

Locke & Latham, 1990). Self-regulatory behavior, manifested as choices a person makes, implicates the goals a person is attempting to achieve and reflects what an individual deems as important and valued. Behavior that is self-regulated is behavior

attempting to achieve a valued goal. In the context of college level academic

achievement, the major thesis of this study is that the values students place on certain

goals will influence not only the importance of goals but will also help determine how

behavior is self-regulated, and through this, influence motivational, affective, and

behavioral processes relevant to achievement. Goals vary in importance and it this

degree of value associated with different goals that drives decision making and self-

regulation to the level of actual performance. In one sense this is similar to Bandura’s

(1997) notion of proactive forethought and the “setting [of] valued performance

standards that create a state of disequilibrium” (p. 131). However, in the current

conceptualization value is associated not only with performance standards, but also

with standards at a higher level in the goal hierarchy that the actual performance

standards serve and are ultimately related to a large extent to social identity and its

relationship to the goal of affiliation and the need for belongingness. Thus, given the

importance of goals, it is thus informative to review the concept of goals. This will

be done according to the outline provided by Austin and Vancouver (1996) by

reviewing the core goal properties of structure, content, process, and relevance defined in terms of their values or valences charged to them. In doing so special reference to achievement situations will be made.

According to Austin and Vancouver (1996) goals are “internal representations

of desired states, where states are broadly construed as outcomes, events, or

processes” (p. 338). With respect to goal structure, a commonly accepted conceptualization is of a goal hierarchy where higher level goals have an influential role in lower level subgoals and where goals at any sublevel can be means or ends

(Austin & Vancouver, 1996: Shah & Kruglanski, 2000; 2003). In addition, within

this hierarchy goals vary along dimensions of importance, complexity, difficulty,

specificity, and temporal range. Goal content exists within this dimensional,

hierarchical structure within particular classifications or taxonomies (e.g., see Ford,

1992 for a similar conceptualization) organized along two dimensions of valence

(approach-avoidance) and person (intrapersonal-interpersonal). Furthermore, given

the hierarchical structure, the goal content dimensions are organized along a conic

structure where explicit, conscious action goals are “etched” on the surface of the

cone and intrinsic goals are a “lattice work” inside the cone that etch the surface

(Austin & Vancouver, 1996, p. 358) variously represented as needs, emotions, or

traits and are often implicit or unconscious. Finally, goal processes refer to the setting and regulation of goals. Thus, goal processes involve activities intended to achieve goals and include goal establishment, planning, monitoring progress, persistence, and revision. Thus, the self-regulation and control of motivation and behavior are integral aspects of the processes involved in achieving goals.

Goal Structure

An important example of the action goal hierarchy is that of Carver and

Scheier (1998) where the goals that people have can be differentiated according to layers or levels of abstraction. Using their example, the top of the hierarchy includes one’s self-concept and principles (such as “be thoughtful”) that are directly tied to this concept. Below the principle goals level are programs linked to the principles.

Thus, “prepare dinner” (a program goal) is linked to the higher level principle of “be thoughtful”. Below the program goal level are the motor control goals that sequence the motor actions necessary (e.g. “slice broccoli”) to complete the higher level programs (Carver & Scheier, 1998, p. 72).

Carver and Scheier (1998) use Vallacher and Wegner’s (1987) action identification theory as a basis from which to discuss the processes involved in the functioning of the goal hierarchy. In this scheme the “way people think about their actions is informative about the goals they’re using to guide their actions” (Carver &

Scheier, 1998, p. 74). Furthermore, when people think about or identify their actions at a lower-level in the hierarchy (e.g. “study”) they tend to naturally drift toward the

higher, more abstract levels in the hierarchy. In academic contexts, what this means

is that if a student becomes comfortable at the lower-level, the more abstract levels of

identification tend to emerge. Alternatively, if a student is experiencing difficulty at a

higher level of identification, the tendency is to drift downward toward the lower, more concrete level to accommodate and handle the difficulty. Thus, there is a continual cycling, naturally, up toward the higher, more abstract layers and down toward the more concrete, lower levels provoked by difficulty. Finally, since higher

level identification is an emergent property of the lower levels, it is possible for

people at a lower level identification to be distracted and adopt different higher

identification levels. For example, a student at a program level involving the choice

of studying or going out with friends, might be distracted from studying such that her

action identification is absorbed into the principles of “it is not important to do well

on an upcoming exam” or “it is more important to have fun”. However, people at the

higher levels of identification tend to maintain that identification in the event they

need to drift to the more concrete levels to accommodate difficulties.

Goal Content

The intrinsic goals function in the regulation of the action hierarchy and are

conceived in terms of needs, emotions, and personality traits (Austin & Vancouver,

1996).

The Regulation of the Action Hierarchy and the Need for Belongingness

Needs have been variously conceptualized in terms of number, content, and importance (e.g., Baumeister & Leary, 1995; DeShon & Gillespie, 2005; Kuhl, 1994;

McClelland, 1985) and are thought to function as modulators of the linkages in the action hierarchy implicitly (Kuhl, 1994; McClelland, Koestner, and Weinberger,

1989). Implicit modulation is beyond conscious accessibility and correlates more with behavioral trends rather than with choices. This is similar to Kuhl’s (1994)

“hovering homunculi” that modulate the linkages among goals in the hierarchy and are responsible for its organization. It should be noted, however, that linkages in the goal hierarchy may also be modulated explicitly in a manner that can be independent of needs through the more conscious aspects of the two structural dimensions of goal complexity or connectedness and/or goal importance or value (Carver & Scheier,

1998). To the extent that values relate to conscious choices, this would be similar to the more explicit or “self-attributed” motives discussed by McClelland, et al., (1989).

This aspect will be discussed more fully below.

With respect to content, of particular relevance to the current study is what

Baumeister & Leary (1995) refer to as the need for belongingness defined as “a need to form and maintain at least a minimum quantity of interpersonal relationships” (p.

499). In line with some of the metatheoretical requirements they state as criteria for a construct to be considered a human motive, Baumeister and Leary cite evidence that shows that people form relationships easily and resist dissolution of relationships. In addition, they show that human relationships are an important basis for human thought and emotion and deprivation of relationships can be a potent source of behavioral and emotional problems. Gardner, Pickett, & Brewer (2000) showed how belongingness needs are associated with greater processing of socially relevant information. Specifically, when participants were made to feel excluded in a

computerized chat room, they recalled significantly more explicitly social events in a

diary they read prior to participating in the chat room than participants made to feel included in the chat room. There was thus a satiation effect that is expected for any need construct. That is, according to Gardner et al. (2000) “those who should have

felt social satiety from a recent social acceptance experience showed less selective

memory for social information than their rejected and thus socially hungry counterparts” (pp. 490 – 491).

Recently, DeWall, Baumeister, and Vohs (2008) have also shown satiation effects. Specifically, DeWall et al. (2008) showed that participants who were led to

believe that they should expect a future of good social relationships subsequently performed poorly on a variety of tasks (e.g. hand-eye coordination, dichotic listening, cold pressor, persistence on a unsolvable anagram task) only when these tasks were framed as diagnostic of traits desirable for good social relationships but not when framed in non-diagnostic ways. However, participants who were led to believe that they should expect a future of poor social relationships and loneliness or were rejected for what they believed were personal reasons performed better when tasks were framed as diagnostic of traits desirable for good social relationships. These results support the contention that once a need is satiated, the desire to perform behaviors to satisfy the need is decreased and should preclude optimal performance.

Thus, participants who believed that their futures were socially bright were in effect satiated at the time they performed the socially diagnostic tasks and did not have a need to perform well on those tasks. However, for participates with an active need for belongingness, as would be the case for participants who were rejected or believed

that their social futures were relatively dismal, performance on tasks that were diagnostic of social skills should be better since they offer the chance of being accepted.

Baumeister and Leary (1995) also cite evidence that interpersonal conflict is more upsetting and stressful than many other potential, non-personal stressors such as work overload, transportation or financial problems (Bolger, DeLongis, Kessler, &

Schilling, 1989). Indeed, it is the contention of Baumeister and Tice (1990) that one major source of anxiety (in general) is the possibility of social exclusion. According to these authors, the potential for experiencing anxiety resides in any event, immediate or inferred, actual or possible, direct or indirect, that implicates some deficiency in self that would sever relationships with others. Furthermore, there is recent evidence that the negative affect people experience due to social exclusion – the “pain” or “hurt” – may be related to the same neural structures involved in physical pain (Eisenberger, Lieberman, & Williams, 2003; MacDonald, & Leary,

2005). In addition, it appears that the body responds to social pain in a way analogous to physical pain, specifically by experiencing an emotional numbness thus creating a type of analgesic effect as a protection while the injury persists (DeWall &

Baumeister, 2006). Thus, for example, DeWall and Baumeister (2006) have shown that participants given false negative feedback about the implications of their extraversion/introversion score from the Eysenck Personality Questionnaire as indicating that their score predicted a future of being alone or rejected by others, showed reduced sensitivity to physical pain as well as reduced emotional reactivity

(joy or sadness) at the potential win or loss of a future football game and less empathy over another’s suffering from a romantic breakup or broken leg.

The effects of social exclusion are not limited to physical pain or its analogs.

In line with the notion that a thwarted need to belong should affect a wide variety of situations (Baumeister and Leary, 1995), experimental evidence exists that feelings of insecure social relationships can implicitly affect persistence, cognition, and self- regulation (Baumeister, Twenge, & Nuss, 2002; Baumeister, DeWall, Ciarocco, &

Twenge, 2005). Baumeister et al. (2005) reported experimental results showing that participants who expected to be alone in life showed significantly less persistence on an unsolvable tracing task relative to those given false positive feedback (e.g., their personality results indicated a high likelihood of successful relationships in the future). Baumeister, et al. (2002) also showed significant declines in complex cognitive performance involving thinking and reasoning in participants who believed that they would end up alone in life. Because a control group exposed to a manipulation involving making participants feel bad by suggesting that their life would be filled with mishaps unrelated to social rejection did not show reduced cognitive processing or self-regulatory ability, the conclusion is that emotions do not mediate the effects social exclusion has on cognitive and self-regulatory processes.

Thus, it is possible that social rejection interferes with executive processing and/or decreases the desire or willingness for self-regulation. (Baumeister & DeWall, 2005).

Perceived Social Exclusion in Relation to Social Identity. One hypothesis of this study is that students achieve or underachieve academically, in part, through fundamental processes engendered by a need for belongingness (Baumeister and

Leary, 1995) that, while affecting classroom and school behavior, has as it source

factors that transcend the situational contingencies of being “in school” (Steinberg,

Brown, & Dornbusch, 1996). The need for belongingness is conceptualized here as a high level need modulating the linkages from the goal of affiliation (DeShon &

Gillespie, 2005) that contributes in influencing decisions (academic or otherwise)

students make. When affiliation goals are thwarted, perceptions and concerns over

the potential of social exclusion are made manifest.

But why would the need for belongingness and its potential lack of

fulfillment, instantiated as a concern over social exclusion, be important in academic

achievement situations? In addition to socialization processes in general,

achievement goals arise in processes that occur in social groups (L. H. & E. M.

Anderman, 1999; L. H. Anderman & Jensen, 2004; Freeman, L. H. Anderman, &

Jensen, 2007; Roeser, Midgley, & Urdan, 1996; A. M. Ryan, 2001; Juvonen, 2006),

processes that involve a continual dialectic of optimal distinctiveness (Brewer, 1991)

created through the desire for individuation, personal integrity, and continuity on one

hand and collectiveness, connectedness, and security on the other (Brewer, 2003).

These processes are the core of students’ social identities and help contribute to defining who they are and what they do (Spears, Scheepers, Jetten, Doosje, Ellemers,

& Postmes, 2004). As such it has been found that significant others play a prominent role in the goals students pursue (Fitzsimons & Bargh, 2003; Shah, 2003) and forces

(perceived or actual) that contribute to exclusion or rejection from desired others may instigate an internal monitoring system which defines self-esteem (Leary & Downs,

1995; Leary, Tambor, Terdal, & Downs, 1995). This is relevant in educational

achievement contexts because, according to one theory, social goals such as

affiliation are high level goals served by lower levels goals such as achievement

(DeShon & Gillespie, 2005). Furthermore, affiliation may be a high level goal

because social interaction and relationship form the very hypostasis of self: much of

the glue that holds the self together is relationship with others. Brewer and Gardner

(1996) for example point out that individuals define themselves not only as unique or

differentiated from others but also interpersonally (e.g., intimate dyadic relationships)

and collectively (i.e., emergent from a larger group often represented by a social

category). They showed that priming the ‘we’ concept lowered the threshold of

similarity as measured by the increase in response latency of participants to make

judgments of ambiguous statements dissimilar to their own relative to judgments of

statements that were similar to their own. Likewise, priming the ‘they’ concept

facilitated dissimilar relative to similar judgments (Experiments 1 and 2). This

implies that different levels of perceived inclusiveness can influence how the self is

construed.

In this context then understanding academic achievement begins with the

realization that it serves larger processes of identity formation and is in large part a

social process. Social interactions are a space within which students find their place in the world. This space contains the ever changing cognitive scripts and schemas students write not only about generic cognitive structures, but also about themselves and social situations (Crick & Dodge, 1994), in collaboration with others and within

the push-and-pull of accommodative and assimilative processes (Piaget, 1977) that

contribute to continual translation and transformation of self. This collaborative give

and take, this transforming and translating, implies that self-schemas are to some

extent shared with others and not completely of one’s own making. The

collaborations are the source of an emergent dialectic of what Brewer (1991) refers to

as optimal distinctiveness, that self-identity is a rapprochement between individuality

and connectedness (Brewer, 2003). Thus, relating this to the goal hierarchy

conceptualization, linking an academic achievement goal with a higher level self-

concept goal might involve one’s self-concept tied to a common-identity group

(Prentice, Miller, & Lightdale, 1994) or “prototypicality that locates one within the fabric of the group” (Hogg, 2005, p. 246) such as “good student” or “not a good student” that could be linked to a principle such as “it is important/not important to do well on an upcoming exam”. The program then might be something like “study versus go out with friends” for example. The motor sequence would then involve the necessary actions for studying or going out with friends.

It is assumed that goals related to social identity, like affiliation, will be highly

valued. Goals, such as achievement, linked to affiliation (DeShon & Gillespie, 2005)

and modulated by the need for belongingness will thus also be highly valued. The

differentiation of academic achievers and underachievers may be based to some

extent on these values influencing interest, desire, and effort that derive ultimately

from the formation of identity rooted in the processes students negotiate in group life

and manifested as self-regulatory processes determining cognitive, motivational, and

behavioral processes. Thus, in order to begin identifying achievers from

underachievers, it is important to focus on what students attend to, what goals they

value, what they expect to achieve, and whether or not they believe they can achieve

it. What one will find, ultimately, is a choice rooted in social identity and motivated to some extent by social exclusion.

Regulation of the Action Hierarchy and the Role of Affect

Baumeister and Leary (1995) contend that “many of the strongest emotions people experience, both positive and negative, are linked to belongingness” (p. 508).

Whereas needs function as drivers, emotions signal changes from a desired or accustomed state (Ben-Ze’ev, 2000). Specifically, emotions may play a regulatory role as signals to the rate of progress toward goal attainment (Carver & Scheier,

1998). When progress is slowed or thwarted, negative affect such as depression or anxiety can result depending on whether one’s goal is to decrease (approach) or increase (avoidance) “distance” from a goal. Particularly relevant here is the relationship of this mechanism to the experience of stress (Carver & Scheier, 1998).

To the extent that goal establishment is a self-regulatory process relating to one’s conception of self (Markus & Wurf, 1987) any disruption in behaviors should be related to increase levels of anxiety and stress (Carver & Scheier, 1998). This may be particularly true for the disruption of social or affiliation goals because of their intimate relationship to one’s self-concept in that the self is not only partly dependent on interaction with others (Markus & Wurf, 1987), but also because other non-social goals are often shared with others (Millar, Tesser, & Millar, 1988). For example

Millar et al., (1988) asked first quarter freshmen females to indicate the frequency of various activities performed at home with a close significant other prior to starting their freshmen year. They found that disruptions of activities at home were positively

related to experiences of loneliness and depression especially when a companion at

school could not be found.

Regulation of the Action Hierarchy and the Role of Personality

Finally, personality traits are included as intrinsic goals by Austin and

Vancouver presumably to provide an aspect of consistency to the goal content

taxonomy. Personality traits might be viewed as moving averages or smoothing functions of self-regulatory activity across time that produce goal directed behavior

(DeShon & Gillespie, 2005). Ford (1992) captured this notion with the concept of the

behavior episode schemata (BES) as “a complex repertoire of enduring behavior

patterns that . . . represents an integrated ‘package’ of thoughts, feelings, perceptions,

actions, biological processes, and relevant contexts” (p. 26). Furthermore, BES are anchored in the changing goals and situations people encounter thus providing attentional, cognitive, and affective guidance over time. Thus, personality is seen as a set of stable behavioral schemata or scripts anchored in goals and contexts resulting in temporal stability but also flexibility in behavior.

However, it should not be expected that all behavioral schemata in the sense used by Ford (1992) are adaptive. Perhaps the poster child for this notion of maladaptive behavioral schemata is procrastination. Indeed, Steel (2007) referred to procrastination as “quintessential self-regulatory failure”. Tuckman (1991) defines

procrastination as the “absence of self-regulated performance” or “the tendency to put

off or completely avoid an activity under one’s control” (p. 474). There is evidence that procrastination may be a pattern of behavior characterizing a deficit of self-

regulatory behavior across time and situations. In a meta-analysis of studies on procrastination van Eerde (2003) reported a significant negative relationship between it and the trait of Conscientiousness (McCrae & Costa, 1997). Schouwenburg (2004) discussed research that showed that in addition to Neuroticism, low

Conscientiousness in combination with “trait procrastination, weak impulse control, lack of persistence, lack of work discipline, lack of time management skill, and an inability to work methodically . . . [provides] little justification for viewing

procrastination as a separate trait.” Schouwenburg concluded that the entire cluster

might be best viewed as a lack of self-control in the self-regulatory process. To the

extent that procrastination is a trait (or perhaps, more accurately, a BES) related not

only to low Conscientiousness, but also to a lack of self-control, weak impulse

control, and an unwillingness to defer or delay gratification of short-term pleasures

(Ferrari & Emmons, 1995), it should be expected that this cluster should show long term effects. Support for this comes from studies using the delay of gratification paradigm (e.g., Mischel & Baker, 1975; Mischel & Ebbesen, 1970) to investigate self-regulatory activity across time. In this paradigm, children face a dilemma: wait for an extended time (e.g., 15 or 20 minutes) for a large reward, for example two cookies, or take an immediately available smaller reward. What is amazing about this paradigm is that the amount of time a 4- or 5-year old child waits for a reward is predictive of social, emotional, and cognitive competencies later in life. For example,

Shoda, Mischel, & Peake (1990) showed that if children are offered no alternative

strategies to think about during the waiting period, the delay time actually waited by

4- and 5-year old children was significantly correlated with parents’ ratings of self-

control in frustrating situation, coping skills, concentration ability, intelligence,

perseverance, and resistance to temptation 10 years later. Perhaps even more

remarkable was the significant correlations between delay time as young children and

performance on the SAT exam. It is important to note that the significant correlations

were found only for children who were offered no suggestions at effortful control

(Metcalfe & Mischel, 1999) during the waiting time and whose rewards were exposed during this time. Nevertheless, to the extent that procrastination involves an unwillingness to defer gratification of short-term gains, these results corroborate the notion that procrastination can be assessed as tendency for low self-regulation with potential long-term consequences.

Goal Processes

The final piece on the nature of goals involves goal processes. Austin and

Vancouver (1996) refer to goal processes as “behaviors and cognitions related to striving toward multiple goals” (p. 347). The discussion above on intrinsic goals and the action goals derived from them implies that goals are functional self-regulatory operators on these behaviors and cognitions which according to Austin and

Vancouver (1996) include the actual establishment (goal-setting), planning, monitoring, and revising of goals that can occur over time. A good example of a goal process model in academic settings is Zimmerman’s (2000a). According to this model, self-regulation is a cyclical process incorporating a feedback process in which comparisons of goals or standards with the effects of one’s behavior on the environment and on the self evoke future behavior that will either change or maintain

the status quo. For Zimmerman, forethought grounds performance or volitional control that produces self-reflections which in turn feeds future forethought. It is

comprised of goal setting and planning (strategies) as well as self-efficacy beliefs,

outcome expectations and goal value and orientation. Performance is defined by

attention strategies and self-observation, monitoring, and self-instruction. Self-

reflection involves judging one’s performance and attributions of one’s performance

to effort and ability.

Goal setting and planning are thus crucial aspects to self-regulatory learning

and behavior. Recently Zimmerman (2008) laid out a specific conceptualization,

extending that of Locke & Latham (1990), of the self-regulatory functional operations

of goals on goal setting as a mapping of advantageous goal properties onto goal

setting behavior. Zimmerman (2008) lists eight goal properties (hierarchical

organization in the form of short-term versus long-term goals, congruence or lack of

conflict, specificity, proximity, difficulty or challenge, origin (self-set versus

assigned), conscious quality, and focus).

Hierarchical Processes

First, the hierarchical organization of goals seems to imply that people cycle

between distal and proximal goals as a natural consequence of the drifting between

higher and lower levels of the goal hierarchy. In fact, it might be conceived that

distal goals are to some extent emergent properties of proximal goals and that the

benefits to students of setting proximal goals are seen only to the extent that they are

integrated with more distal goals (Zimmerman, 2008). Proximal goals provide

beneficial feedback for students who maintain higher levels of identification (more

distal goals) because this feedback can be used to accommodate difficulties without

outside sources of motivation. In addition, the desire and motivation required in arduous or stressful tasks may be increased with a focus on proximal goals that are

within one’s capabilities or situation (Ford, 1992).

However, the hierarchical nature of goals suggests that setting only proximal

goals increases the likelihood of distraction when a student is faced with making a

decision between two conflicting goals. That is, due to the nature of the goal

hierarchy and the natural cycling that takes place (Vallacher & Wegner, 1987),

students who are studying simply to complete today’s assignment without a longer-

term goal such as fully understanding a difficult concept or even merely getting a

good grade in a course, run the risk of being distracted and pursuing a conflicting

higher level goal. According to Carver and Scheier (1998) “goal qualities at higher

levels are intrinsically more important than those at lower levels.” (p. 90) Thus, to the

extent that proximal goals are means to distal goals, they are at a lower level in the

goal hierarchy and thus are not expected to be as valued, per se, as higher level, more

distal goals. Proximal goals can accrue value only in reference to the pursuit of more distal, high-level goals (Carver & Scheier, 1998). Thus, when two or more proximal goals conflict the goal pursued will be the one linked to a more highly valued distal goal. In educational contexts the conflict may involve the choice of studying on one hand versus socializing or leisure activities on the other. For example, Fries & Dietz

(2007) showed in an experimental setting that high school students performing a

learning task showed less motivation for the task and more motivational interference

in performing the task when they knew an attractive video clip was available relative

to a group that had watched the video clip prior to performing the learning task. In the latter case, the manipulation removed the conflict by showing the video clip first.

Hofer, Schmid, Fries, Dietz, Clausen, & Reinders (2007), showed in a self-report study that when motivational conflicts are present, value orientations influence the choice students make in pursuing different goals. Specifically, Hofer et al. (2007) compared two value orientations: achievement and well-being/leisure values. The prototype student favoring the achievement value is one “who has clear goals, struggles through uncomfortable tasks and wants to achieve something in life” whereas the well-being/leisure prototype is characterized as “a student who spends a

lot of time with friends, loves diversion and spontaneous activities, and wants to have

fun in life . . .” (p. 21). Their structural equation model showed that low self-

regulation, identified with items measuring distractibility, shallow processing,

persistence, frequently switching from a learning task to other tasks, and mood (e.g. a

“bad mood” when working and knowing others are “having fun”) was significantly

positively related to the well-being/leisure orientation but negatively related to the

achievement orientation. In addition, achievement orientation was significantly

related to the amount of time students reported devoting to learning tasks. The well-

being/leisure orientation had a negative weight with time investment, but was not

significant. The “prototypes” Hofer et al. (2007) refer to reflect conflicting values

and imply differential emphasis in the higher level goals being pursued.

Congruency of Academic and Social Goal Processes: The Relevance of School

Belongingness, Perceived Social Exclusion, and Group Processes.

In addition to the importance of the integration of proximal and distal goals, to

the extent that students purse multiple goals, a second important goal process

necessary for academic achievement is goal congruency (Zimmerman, 2008).

Indeed, motivation is likely increased when multiple goals converge synergistically

(Ford, 1992). This may be particularly true with respect to academic achievement

goals and social goals (Urdan & Maehr, 1995; Wentzel, 2005).

The Relevance of Social Goals and Processes to Academic Pursuits. Urdan

and Maehr (1995) maintain that social goals are distinct from mastery and

performance goals and that there are different types of social goals that can have

different effects between and within mastery and performance goals. For example,

seeking approval can have advantageous academic effects if those from whom

approval is being sought have academic interests, but can have deleterious effects

otherwise. Seeking approval may also have different effects from the goal of being dependable which, in turn, can have differential effects depending on whether or not

meeting the social goal of dependability takes time away from studying. Urdan and

Maehr also point out that the consequences of pursuing social goals depend on four factors. First, the type of social goal orientation can produce different outcomes. For example, a student oriented toward seeking social approval may view academic achievement negatively if that student thinks that pursuing academic interests will have a negative response from peers. Alternatively, a student oriented toward complying with a teacher or parent who views academic pursuits positively will also

likely have a positive view toward academics. Second, the type of social goal can

interact with the values of the targets of social goals. Seeking approval can have two

different effects depending on whether the targets value or devalue academic work.

Third, the meaning of the goal orientation (mastery or performance) and achievement

situation will vary as a function of culture, ethnicity, and gender. Finally, the way in

which social goals are coordinated with other goals can produce differential academic

achievement outcomes. For example, if mastery, performance, and social goals all

converge to the same place (i.e., become synchronous) such as becoming competent

in learning new mathematics procedures and demonstrating this new ability to friends

who value this type of learning, then a positive achievement outcome should result.

The value of this approach is that social goals summarize a set of possible directions

students can take with respect choosing particular goal orientations. As such, it is

possible that social goals may be a fundamental source of academic achievement and

may be determinative in the more proximate and accessible achievement goals

(Pintrich, 2000a).

Thus, for adolescents, the pursuit of social goals can have profound effects,

both beneficial and detrimental, on academic achievement. As Zimmerman (2008)

has stated “the act of goal setting and pursuit is nested within a social group context

that can enhance or undermine students’ effectiveness in school” (p. 271). Bandura

(1986), for example, has discussed evidence that shows that self-regulatory functions are supported by and through social factors including social reward, modeling support, and negative social sanctions. Furthermore, Shah (2003) has shown how significant others can affect the willingness to persist at a task and the perceived value

of goals (experiments 1 and 2). For example, in Experiment 1 Shaw (2003) first

asked participants to name a significant other who they thought would believe that an

upcoming anagram task would be difficult. Participants who were then subliminally

primed with the name of this significant other showed significantly higher task persistence and performance relative to a no prime condition. Likewise, in a second experiment, when participants were subliminally primed with the name of a significant other who believed that an upcoming task was important, participants primed with this name also showed higher persistence and performance relative to a no prime condition. In addition, participants’ rating of how their significant other would value the task was positively related to their own value rating of the task. This significant interaction implied that participant’s own value rating was better predicted in the priming relative to the no priming condition.

While the influence of social goals mediated by peer groups appears to begin to take shape at the transition between elementary and middle school (e.g. Berndt,

1999, A. M. Ryan, 2001), peer groups continue to play an important function throughout adolescence (Heaven, 2001) suggesting that social goals will be valued throughout this period. Berndt and Keefe (1995) showed that in early adolescents

(seventh and eighth grade), not only the behavior of peers but also the quality of relationship can affect academic engagement. The valuing of social goals and the desire to have social goals satisfied is clearly a potent factor moderating the affect of peers on academic engagement. For example, in an experimental study, White,

Sanbonmatsu, Croyle, & Smittipatana (2001) found that participants solved fewer anagrams (underachieved) when a likable confederate failed to solve most of the

anagrams (Likeable/fail condition) relative to conditions where an unlikable

confederate failed to solve anagrams (Unlikeable/fail condition) or where a likeable

confederate solved an average number (Likeable/average condition). In addition, in

the “Likeable/fail” condition participants had a higher empathy index, greater concern

over the relationship with the confederate, and greater concern for the confederate’s feelings relative to the “Unlikable/fail” or “Likeable/average” conditions. The authors proposed that failure at tasks that are of high-relevance in the presence of close others is predicted to decrease self-esteem. However, for low-relevance tasks, failure at such tasks when close others also fail may be a way to maintain peer relationships. This implies that succeeding at these tasks when close others fail risks rejection. Similarly, Dishion, Spracklen, Andrews, and Patterson (1996) showed that

the effect peers have on one another depends to a large extent on the reinforcing

properties of the values in the group. For example, non-delinquent dyads tended to

react positively to prosocial conversation whereas delinquent dyads were more likely to react positively to rule-breaking topics of conversation. However, this effect may reflect a more fundamental need for social reinforcement or perhaps a need for belongingness (Baumeister & Leary, 1995) because Dishion et al. (1996) also found that even among the non-delinquent adolescents in their study, conversations involving socially reinforced discussions of rule-breaking predicted escalations in delinquency two years later.

Further evidence supporting Urdan and Maehr’s (1995) contention that students’ negative orientation toward pursuing academic interests may be influenced by perceived negative responses from peers is seen in the seminal study conducted by

Fordham and Ogbu (1986). In this study of what they describe as a “predominantly black high school” (p. 185) in a low-income area of Washington DC, Fordham and

Ogbu report on the academic underachievement of black students who “appear to have the ability to do well in school [and yet] choose to avoid adopting attitudes and putting in enough time and effort in their schoolwork because their peers (and they themselves) would interpret their behaviors as ‘white’” (p. 187). This “burden of acting white” is described as a counterresistance to an identity formed in a community of resistance to discrimination and racism. Acting white is a burden to black students because it represents an existential anxiety of exclusion from this community. It is this fear of losing the “black identity” through exclusion from the community that manifests itself as resistance one form of which is academic underachievement.

That this phenomenon is not limited to western culture is seen in Demerath’s

(2000) ethnographic work which presents another case study of students’ academic underachievement in the Manus province of Papua New Guinea. Elsewhere

Demerath (2003) observes that underachievement in these students is due in large part to a privileging and individualizing of their perceptions of the economic situation and their future that idealizes village life. That is, in interviews one common theme among students was that a grade 10 certificate was “worthless” because they could not get a job and that “their villages were there for them to return to after they finished school” (Demerath, 2000, p. 209). In Demerath’s (2000) study he noted that this individualization of perception of village life “circulated” throughout the peer groups such that the village was often times a mechanism used to dissuade other

students from studying. Thus, Demerath claims that this “village-based identity” with its “egalitarian ethic” was “at the core of the antiacademic student culture”

(Demerath, 2000, p. 216). Students who adopted particular behaviors or ways of speaking associated with modern lives in the cash sector were often disparaged as betraying the village identity. These students were characterized as “acting extra” or

“fancy” or “expensive”.

Thus, when the value of social goals antithetical to academic goals takes priority over the value of academic goals, poor academic achievement and behavior problems may result (Juvonen, 2006). On one hand, this may happen because students have a tendency of choosing friends similar to themselves (Osterman, 2000) and so a common pattern observed is a process in which deviant students tend to affiliate with other deviant students resulting in a cycle of lack of participation in school leading to disengagement from school (Finn, 1989). This process may also involve students’ goals and values. For example, Urdan (1997) has shown that a student’s affiliation with peers whose orientation toward school is positive or negative is related to whether or not that student is a high or low achiever respectively. High achieving students tend to associate with other high achieving students who value academic pursuits whereas the opposite pattern is observed for low achieving students. Furthermore, student affiliation with positively or negatively oriented students is related to the type of goal orientation assumed by a student.

Students with a mastery goal orientation tended to affiliate with peers who value academic achievement. However, students with a performance goal orientation tended to affiliate with peers who value academic achievement regardless of

academic ability and with students who tended not to value academic achievement but only if they were also of high ability – students who tended to affiliate with low ability students and did not value academic achievement tended to not have a high performance goal orientation. While this may indicate that performance goals are not

necessarily maladaptive (Urdan, 1997), the important point here is that a student’s

peers can have an important influence on her/his decisions regarding academic achievement.

The Importance of Feelings of School Belonging. Perhaps one mechanism of this decision is a student’s perception of the degree of inclusion or exclusion with peers in and out of school. The contention is that values of social goals antithetical to academic goals may take priority over the value of academic achievement when students fear social exclusion from academically oriented peers and/or from school in the main. Perceived exclusion may then motivate some students to bond with others whose social goals conflict with academic goals (Hymel, Comfort, Schonert-Reichl,

& McDougall, 1996; Juvonen, 2006; Newman, Lohman, & Newman, 2007).

It seems then that a crucial factor for understanding engagement with or disengagement from school, as well as academic achievement in general, is that of a student’s sense of belonging in school in general as well as a sense of belonging at a particular school (L. H. Anderman & Freeman, 2004). There is evidence that a sense of school belonging may be positively related to a students’ academic motivation and achievement as well as their overall subjective well being (Anderman, 2003; L. H.

Anderman & Freeman, 2004; Freeman, et al., 2007). Importantly, the review of the research provided by L. H. Anderman and Freeman (2004) corroborates the notion of

the importance of congruency between social and academic goals in understanding academic achievement, motivation, and well-being and that this congruency depends in part on a student’s sense of belongingness “in school” as well as “at a school”. To the extent that a student feels that s/he belongs to an academy, in the general sense of that term (i.e., “in school”) and/or at a particular school, the values held and goals attempted will more likely be those associated with those endorsed by the academy and/or the school. The mechanism of this sense of belongingness can work through friends (e.g. Berndt, 1999; Berndt & Keefe, 1995; Goodenow & Grady, 1993; Hymel, et al., 1996), peer networks (e.g. Cairns, Cairns, & Neckerman, 1989; Kindermann,

1993; Kindermann, et al., 1996), family and parents (e.g. Gutman & Midgley, 2000), teachers and classroom climate (e.g. L. H. Anderman, 2003; Roeser, et al., 1996;

Roeser, Eccles, & Sameroff, 2000), and/or a sense of community with a particular local school a student attends (e.g. Battistich & Hom, 1997; Battistich, Solomon,

Kim, Watson, & Schaps, 1995; Goodenow, 1993a; 1993b; Osterman, 2000).

School belongingness may not only be affected by relationships with teachers and peers, but may also affect students’ behaviors such as academic engagement and aggression (Juvonen, 2006) as well as motivation, cognition, self-regulation, and affect (L. H. Anderman, 1999; L. H. Anderman & Freeman, 2004). For example, L.

H. Anderman and E. Anderman, (1999) showed that even with a decline in achievement motivation from fifth to sixth grades, students who perceived a sense of belonging with their school were more likely to pursue mastery goals. However, the sense of belonging had no effect on the pursuit of performance goals in the sixth

grade. L. H. Anderman (1999) showed that perceived school belonging, in addition to

responsibility and social goals, were positively related to affect in school.

Both the E. M. Anderman & L. H. Anderman, (1999) and L. H. Anderman

(1999) studies showed how perceived school belonging can have beneficial effects

during a period of transition. In these studies, the transition involved children moving

from elementary to middle school, a period that can have detrimental effects on students’ motivation and self-concept (e.g. Midgley, E. M. Anderman, & Hicks,

1995). Another important academic transition in many students’ lives is that between high school and college. Interestingly, many of the same buffering effects on motivation and achievement behavior seen in younger students are also seen in late adolescents’ transition to college (Freeman, L. H. Anderman, & Jensen, 2007;

Pittman & Richmond, 2007; 2008). For example, in a longitudinal study Pittman and

Richmond (2008) showed that perceived university belonging was positively related to academic and social competence and negatively related to feelings of depression and anxiety as students transitioned from high school to college. Freeman, L. H.

Anderman, and Jensen (2007) also showed the importance of perceived class belonging in college. That is, students sense of class belonging was positively related to self-efficacy, intrinsic motivation, and academic value during students first semester at a university. In addition, they showed that while a sense of class belonging was not significantly related to an overall sense of belonging at the university after perceived social acceptance of peers and professor caring were controlled, the latter two factors were positively related to an overall sense of university belonging. What is particularly important about this study is that the

finding that students’ perception of social acceptance by peers was so strongly related

to their sense of university belonging underscores the importance of the role (direct or

indirect) of peer relationships in affect, motivation, and achievement in college.

However, even though most of the research on school belongingness points to

its overall beneficial academic effects, there are some indications that this may need

to be qualified. L. H. Anderman and Hughes (2003, cited in L. H. Anderman &

Freeman, 2004) reported that seventh grade students’ likelihood of adopting a personal performance-avoidance goal orientation in the spring semester in response to their perception of their classroom performance-avoidance orientation in the fall was higher for students reporting higher levels of school belongingness. Thus, according to L. H. Anderman and Freeman (2004), “students’ sense of belonging appeared to promote an undesirable orientation, rather than a more adaptive one”. (p. 32) E. M.

Anderman (2002) in a hierarchical linear modeling study reported that while students’ sense of belonging in school was related to many positive outcomes (e.g. higher GPA and sense of optimism, lower reported depression, rejection, and behavioral problems) at the individual level, students in schools with higher aggregate (school level) belongingness reported higher levels of perceived rejection and behavioral problems. This apparent paradox is at least partially resolved when students’ perceived sense of individual social exclusion in a context of perceived aggregate inclusion is taken into account (E. M. Anderman, 2002). Schools with high aggregate belongingness provide a context in which some students may believe that, while the school as a whole is a supportive environment for other students, the school and the other students that belong there are impenetrable for them. Thus, as E. M. Anderman

states, “the potential positive buffering effects of an individual student’s perception of

belonging matter less in schools in which many students report a high sense of belonging”. (p. 807) It would seem that the most reasonable explanation for this is

that a given student’s perceived level of belongingness is less than the perceived level

of belongingness in the school. The result is a higher likelihood of perceived social

exclusion.

Perceived school belongingness, while linked to many beneficial behavioral,

motivational, affective, and cognitive outcomes as one potential antecedent, may

itself be an outcome in its own right (L. H. Anderman & Freeman, 2004). The antecedents of school belongingness might, in turn, account for its beneficial and detrimental effects. One particularly important antecedent for the present work is students’ concern about being socially excluded. Specifically, concerns about potential social exclusion or, equivalently, a desire for inclusion may motivate students to seek out other students and faculty and thereby increase the likelihood of becoming involved in activities at the institution as well as feeling a sense of

belonging in the academy. To this extent a desire for social inclusion can have beneficial consequences working through increases in feelings of school belonging.

However, concerns over exclusion can be a double-edged sword in that it can have deleterious effects on self-regulation (Baumeister et al., 2002; Baumeister &

DeWall, 2005; Baumeister, et al, 2005). And to the extent that a lack of self- regulation for academic pursuits reflects an increased propensity for non-academic pursuits, it might be expected that concerns over social exclusion, working through a decrease in self-regulation, might be indirectly related to perceived decreases in

feelings of school belongingness. Specifically, self-regulation is proposed as a partial

mediating factor between social values and school belongingness. Furthermore since

self-regulated performance partially reflects these values through the choices students

make (Tuckman, 1990), it incorporates not only social values such as a desire for

inclusion but academic task values as well. There is evidence, for example, that

perceived academic task value is positively related to perceptions of school belonging in middle school students (Anderman, 2003). It is therefore reasonable to assume that in addition to the direct influence of academic and social values, self-regulation should also influence feelings of school belonging by virtue of its role in decision making based directly on these values. Thus, it is hypothesized that self-regulation

partially mediates the link between social and academic values and school

belongingness that, in turn, mediates the link between self-regulation and

motivational, affective, and behavioral outcomes.

Social Identity and Social Exclusion Processes Affecting Goal Congruency.

Having already discussed the relation of social exclusion and social identity as

a structural consequence of the need to belong, it is now necessary to examine in

more detail how this relationship is important as an aspect of the goal process of

congruency.

The work of Baumeister and his colleagues (e.g. Baumeister et al., 2002;

Baumeister & DeWall, 2005; Baumeister, et al, 2005) have shown that perceived

social exclusion impairs cognitive and self-regulatory processes. In addition there are

neurological substrate similarities between perceived social exclusion and physical

pain (e.g. Eisenberger, et al. 2003: MacDonald & Leary, 2005) and the emotional

numbing reminiscent of the decreased sensitivity to physical pain that is common in physical injury (DeWall & Baumeister, 2006). This is important because it implies that perceived social exclusion to the extent that it is a significant threat to one’s psychological safety and security, is similar to perceived physical danger, to the extent that it is a threat to one’s physical safety and security. The difference, however, is that in most modern societies physical danger may be sporadically potent whereas social exclusion is an abiding potential. Thus, people must be constantly vigilant regarding potential social exclusion. This may account for the finding that forces

(perceived or actual) that contribute to exclusion or rejection from desired others instigate an internal monitoring system which defines self-esteem (Leary & Downs,

1995; Leary, et al, 1995).

Social exclusion affects one’s desire and interest, and thus choice, for a task

(Tice, Bratslavsky, & Baumeister, 2001) and in this sense will influence cognitive, motivational, and affective processes. However, social exclusion motivates effort, cognition, and affect differently relative to the values and goals of those with whom one has important and meaningful relationships. These characteristics of social exclusion and its relationship to self-regulatory processes likely have their roots, in part, in social constructive processes that begin in childhood. For example, in

Piagetian conservation tasks, children with mutually wrong ideas about conservation

(e.g. length) come to a mutually agreed upon conclusion through a sharing of perspectives that each child initially ignored (Ames & Murray, 1982). Importantly, these conclusions were not those internalized by the direction of adults (the experimenters), but were constructed in social interaction. Thus, it is through this

process of social reciprocity that regulatory processes evolve more generally. Piaget

(1977) maintained that all regulatory activity results in either the repetition or correction of that activity. Self-regulation or “optimizing equilibration” involves construction and compensation, assimilation and accommodation. To the extent that these processes occur in a social context with peers provoking, cooperating, negotiating, and opposing one another, social exclusion can do nothing but disrupt these interactions and thereby disrupt a core aspect of self-regulatory activity.

The disruption of self-regulation by means of social exclusion occurs because regulatory activity in general develops and emerges from the core of cognitive adaptation in social contexts. And while the antecedents of self-regulation exist early in life (prior to 3 years of age) they are nevertheless socially situated (Kopp, 1982).

Regulatory activity through the cycle of disequilibration-equilibration guides the formation and transformation of schemas and scripts (Piaget, 1977) regarding one’s knowledge of the world and self. To the extent that educational achievement occurs in a social context, the deleterious effects of social exclusion should be expected. In fact, Patrick (1997) shows that many of the same processes required for academic self-regulation are necessary for social self-regulation. Importantly, she argues that

“being accepted by one’s peers, being able to make and keep friends, and being socially responsible are all positive correlates of school success. Most strikingly, being actively rejected by one’s peers is a strong correlate of school-related difficulties.” (p. 211) Using models of academic self-regulation (e.g., Garcia &

Pintrich, 1994; Zimmerman, 1989) she suggests that students who have difficulty with self-regulation in one domain are likely to show difficulties in the other.

As mentioned above, given the aversiveness of perceived social exclusion

(Baumeister et al., 2005; Baumeister & DeWall, 2005; Eisenberger, et al., 2003;

MacDonald, & Leary, 2005), the need for belongingness (Baumeister & Leary, 1995) may play an important and fundamental role in this process. Furthermore, relying on evidence suggesting the important role of group processes in the self-concepts students create in the push and pull of individuation and connectedness (Brewer,

1991; 2003), the need for belongingness and the social goals directed toward its satisfaction may be group based (Spears, Ellemers, Doosje, & Branscombe, 2006).

Specifically, the fulfillment of social goals may be based in two processes reciprocally influencing one another and reflecting two types of identities and groups.

The distinction between these processes is important in understanding how some of the social influences in academic context relate to academic achievement.

According to Spears et al. (2004), group distinctiveness occurs through two mechanisms. One mechanism occurs when there is a lack of objective social structure but a perceived subjective essential quality is shared among people. Group distinctiveness is created based on these essentialized qualities and provides meaning to the social context thereby giving it the structure it lacks. A second mechanism is reflective where, given a social structure exists, the form and content of the structure provide for defining the inferred essential qualities of a group. Thus, the creative function is an in-group-motivated process whereby groups create and differentiate themselves from others, from the bottom-up. The reflective function is based on a top-down meaning that exists for groups to reflect upon for instrumental reasons such as achieving common goals. As such, group distinctiveness defines the self and

motivates the self’s reactions to particular social structures. The perception of the group as distinctive and whole “provides a means to invest the social self with ‘a’ social structure and provides a base from which to resist and contest aspects of ‘the’ social structure” (Spears et al. 2004, p. 315). The emergent qualities of group identity are seen in Postmes, Spears, Lee, & Novak, R. J. (2005) discussion of the influence groups have on individuals. One route or continuum of group identity is characterized as a top-down quality of groups. The authors refer to this as a deductive route to social identity. Group identity forms around a common attribute or attributes that attain meaning at a supraindividual level. They take shape from sociohistorical factors and occur in an intergroup context. It is possible to be in a group formed deductively and not know (or even like) the members of the group. The main criteria for belonging is sharing or agreeing to a set of superordinate attributes defining the group. For example, the academic underachievement and resistance of some African

American students to Standard English (Fordham and Ogbu (1986) might be viewed as a manifestation of a deductive route to the identity of being Black or conversely and equally, not ‘acting White’. Individuality is created and expressed by reference to these deductive norms. The second route to group identity is an emergent, bottom-up quality of groups and is referred to as an inductive route to social identity. This occurs through communication and, paradoxically, through expressions of individuality.

According to Postmes et al. (2005) “it is through this process of induction that the individual actions of group members . . . can shape group identity. . .” (p. 749).

Postmes et al. (2005) presented experimental results that showed that individuals in groups formed experimentally based on the belief that the group members could have

been close, personal friends (the inductive route) were more likely to be influenced by other group members when all members of the group were “individuated” (members were identified by a photograph) and were less likely to be influenced when members were “depersonalized” (members were not identified by a photograph and communicated via computer). The results were reversed for groups formed through the deductive route where group membership was based solely on the belief that group members simply shared a common world-view. In this case, group members were more likely to be influenced when members were depersonalized and less likely to be influenced when members were individuated.

Prentice, et al. (1995) argued for a distinction between two types of groups that reflect the creative/inductive versus the reflective/deductive processes just discussed. They suggested that people in common-identity groups cohere “because of members’ attachment to the group itself” but that common-bond groups operate at an interpersonal level such that the group coheres “because of the members’ attachment to one another . . . [such that the group is] . . . an aggregate of individual bonds”

(p.491). Prentice et al. (1995) dichotomized these groups, but in the current conceptualization these different kinds of groups seem to reflect the continual cyclical process described above such that common-identity groups represent the goal of one’s self-concept and that common-bond groups represent goals similar to the program level in Carver and Scheier’s (1998) goal hierarchy. In other words, people aren’t “in” a common-identity versus a common-bond group. Based in part on action identification theory (Vallacher and Wegner, 1987) and motivated action theory

(DeShon & Gillespie, 2005, described in more detail below), this conceptualization

views people as guided by a high level identity goal whose focus tends to float to the top of the hierarchy in order to achieve or maintain this goal (e.g. “I am a good student”, “I am a Chess player”) but it will always drive other goals like affiliation and agency to the extent that they are tied up with the identity goal. And in cases where that identity is in doubt or when there is some perceived difficulty with the identity, lower levels goals like achievement may be invoked. Thus, for example, the common-identity of “student” may call a program with personal and interpersonal implications (e.g., “study alone” vs. “study with friends” vs. “go to a party with friends”). If a person’s common-identity of “student” is in doubt due, for example, to a recently received grade lower than what was expected or desired, the program of

“study alone” may be implemented. The nature of the programs called and implemented is part of what defines the common-bond group because the bonds are formed and maintained in the process of performing behaviors implemented to achieve lower level goals (e.g. “study for an upcoming exam”, “take a break and go to a movie with friends”). Thus, even a program such as “study alone” which is a physically solitary activity can nevertheless be implicitly affected by social bonds

(Shah, 2003). In effect, “study alone” may be a common-bond activity to the extent that a student’s social bonds are with other students who also sometimes prefer to study alone. To paraphrase Spears, et al., (2006) the individual is in the group.

Notice, however, that once said, the student also has a common-identity of “student”

(the group is in the individual) to the extent that studying alone is one aspect of this identity. However, people at the lower, interpersonal, common-bond level may be distracted and absorbed into a higher level goal different from the one they started

with. Because there are so many behavioral options at the lower levels of the goal

hierarchy, a student may not choose to study alone, but rather study with friends. If

studying with friends is chosen, behaviors required to fulfill lower level goals provide

even more options and potential for distraction (e.g. “ask about this Calculus

problem” vs. “talk about the most recent piece of gossip”, etc). These choices, over

the long term, can have implications for subsequent formation and/or transformation

of higher-level identity. For example, a student who desires to have a common-

identity of “Mechanical Engineer” and consistently chooses to put off studying the

Calculus in favor of going out with friends early in his tenure as a student will at some point need to reform his common-identity goal.

Thus, in academic contexts the reciprocal push and pull of the “individual in

the group” versus the “group in the individual” (Spears, et al., 2006) may help

account for some of the findings in the literature that suggest that school belongingness can have both beneficial and detrimental effects on students’ academic motivation, achievement, and well-being (L. H. Anderman & Freeman, 2004) and addresses the issue of academic and social goal congruency. For example, D. S.

Kaplan, Peck, and H. B. Kaplan (1997) have shown how years of perceived alienation and rejection from teachers and school-based peers is associated with affiliation with deviant peers and the increased risk of dropping out for middle-school students. They

observed the cyclical relationships that would be expected with an on-going top-

down/bottom-up process of identity formation and group belongingness. For

example, negative academic experiences can lead to a desire to quit school which

increases the likelihood for future failure and negative academic experiences.

Perceived teacher rejection and a devaluing of doing well academically can lead to

affiliation with deviant peers which increases the likelihood of future teacher rejection and devaluing academic grades. What is being observed in this study is a cycle of social identity whereby a student does not feel a part of the “school” – the student is not “in the school” psychologically – such that the “school” is not “in the student”. Finn’s (1989) participation-identification model outlines a very similar process. However, given the aversiveness of the social alienation and exclusion, many students may thus identify with other potentially deviant peers and to the extent that deviant behaviors are reinforced (Dishion, et al., 1996) students may then receive the respect of the group (Spears, et al, 2006) which strengthens a perception of being an “individual in the group”.

Thus, individuals tend to affiliate with other similar individuals which can lead to an amplification of core characteristics common to both. For example, Mounts and Steinberg (1995) showed that adolescents who had a predilection for drug use and were close friends with peers who also used drugs evidenced an increase in drug use over a one-year period. However, adolescents who were successful in school and had friends that were successful in school showed academic improvement over the same period. These bottom-up, inductive, individual-in-the-group effects extend beyond a small peer group into peer group networks where the top-down, deductive, group-in-the-individual effects are seen to the extent that individual group members begin to take on the characteristics of the group as sort of a regression to the mean

(Kinderman, McCollam, & Gibson, 1996). That is, students in networks associated with high academic engagement increased their engagement over time whereas

students associated with low academic engagement decreased their engagement over time. The top-down nature of this effect was seen in the stability of student profiles

regardless of turnover.

Achievement Goal Specificity and Goal Setting

The Concept of Goal versus Goal Orientation. A third important goal process

outlined by Zimmerman (2008) is goal specificity. Goal specificity is the setting of

explicit standards for performance (Schunk, Pintrich, & Meece, 2008). A common

finding is that setting specific, challenging goals leads to better performance than

setting general goals such as “do your best” (Locke & Latham, 1990; 2002) because

the latter type of goal allows for more variability and ambiguity in what is considered

to be acceptable performance (Locke & Latham, 2002). Setting specific goals thus

allows for more precision regarding the focus and standards of performance.

As mentioned above, in educational and organizational contexts two

achievement goals of particular interest are learning and performance goals. From a

goal-setting perspective, learning goals are aims organized around acquiring

knowledge and skills whereas performance goals are aims organized around the

implementation of knowledge (Seijts & G. P. Latham, 2005). This conceptualization

is similar to that of goal orientation where a mastery (or learning) goal orientation

focuses on developing new skills and a performance goal orientation focuses on

demonstrating those skills (Ames, 1992; Dweck & Legget, 1988; Midgley et al.,

1998; Pintrich, 2000a). However, achievement goal orientations, while related to

achievement goals, are different constructs (Elliot, 2005; Pintrich, 2000b; Schunk et

al., 2008), and there is some disagreement regarding the nature of goal orientations in

relation to goals. For Elliot (e.g. 2005; Elliot & Thrash, 2001) achievement goals are

not overall orientations (e.g. Ames & Archer, 1988), but rather aims. For others (e.g.

E. M. Anderman, Austin, & Johnson, 2002 Pintrich, 2000b; Zimmerman, 2008) goal

orientations are distinct reasons for approaching or avoiding tasks.

The term ‘achievement goal’ in this study refers to either an end or a means

and is a sufficient condition for its orientation. Further, the term ‘goal orientation’ is

used here not only to refer to a reason for goal pursuit, but also as an aspect of the establishment of a goal – that is, as a focus in the sense of orienting toward or focusing on one’s aim or goal. As such, ‘goal orientation’, while not synonymous with ‘achievement goal’ is conceptualized here as a particular way of thinking about pursuing a goal (DeShone & Gillespie, 2005) and is a necessary condition for the establishment of a goal. This conceptualization requires that one specify precisely the actual goal(s) one is referring to and to understand that goals can function as both means and ends (Shah & Kruglanski, 2000). In a goal hierarchy perspective, means often refer to motor level subgoals (Shah & Kruglanski, 2000), but goals higher in the hierarchy may be instrumental in achieving even higher level goals. Thus, for example, a student may set her sights on doing well in an upcoming exam in order to get a good grade in a course, in order to achieve a degree, in order to become a lawyer, in order to belong to a group called ‘Lawyer’. From the perspective of a goal hierarchy, it would seem that the aims-reasons distinction is moot. For example,

Elliot (2005) states:

In any given achievement context, an aim (e.g. to do well relative to others) is always undergirded by a more general reason (e.g., to show others I have ability, to feel the satisfaction of success, to avoid the shame of failure, to get the reward my mother promised me), so clearly both aim and reason are important in accounting for achievement behavior. However, as illustrated by the preceding examples, a single aim may be undergirded by many different reasons, and I think it is optimal to keep the aim and reason constructs conceptually separate . ... (p. 65)

However, this particular example simply points to the principle of equifinality (Austin

& Vancouver, 1996; Shah & Kruglanski, 2000) – “to show others I have ability”, “to feel the satisfaction of success”, “to avoid the shame of failure”, “to get the reward my mother promised me” are goals in service to at least one other goal “to do well

relative to others”. It is equally conceivable that the principle of multifinality could

apply here too if, for example, “to show others I have ability” simultaneously aims

toward not only a performance achievement goal such as “to do well relative to

others” but also toward some other goal such as “to get in the Chess Club”. Thus in the discussion that follows, performance and learning (mastery) goals are means and

ends whose implicit (if not explicit) establishment necessarily occur by virtue of their focus or orientation.

Goal Focus and Orientation

Focus refers to the enhancement of “learning processes versus performance outcomes or products” (Zimmerman, 2008, p. 275). As such, this suggests that as a goal is being established, the property of focus refers to how the goal is going to be achieved and implies differential patterns of thought and action as a result. This conceptualization is similar to the definitions of goal orientation provided by DeShon

and Gillespie (2005). Specifically, goal orientations “describe the pattern of

cognition and action that results from pursuing a mastery-approach, performance- approach, or performance-avoid goal at a particular point in time in a specific

achievement situation.” (p. 1114) Zimmerman and Kitsantas (1997) showed that for

the goal of dart-throwing, the best process for achieving the goal was to focus first on

the learning process of learning dart-throwing strategies and then, once the strategies

were learned to automaticity, to concentrate on the performance outcome of achieving

the highest score possible. Extending this to an academic context, Zimmerman and

Kitsantas (1999) showed that in a writing task involving learning how to make

sentence revisions, girls focusing first on mastering a learning strategy for making

correct revision and then switching to a performance goal of reducing the number of

redundant words, performed better than girls given only a performance or learning

goal. Thus, it seems that the point of the Zimmerman and Kitsantas studies is to

suggest that given a new and unfamiliar goal (e.g. dart-throwing), once the decision to

pursue the goal is made, the best pattern of thought and action is to first orient to

learning process goals and then switch to performance outcome goals once the

necessary strategies are learned. The choice of a goal implies its orientation.

The motivated action theory perspective of DeShon and Gillespie (2005)

utilizes the hierarchical goal structure discussed above. With respect to goal content,

these authors place achievement goals (i.e. mastery-approach, performance-approach, and performance-avoidance) at an intermediate layer in the goal hierarchy. As such, achievement goals are mechanisms in the pursuit of higher level principle-goals (e.g. growth, fairness, structure, social values) which in turn satisfy the highest level self-

goals such as affiliation, agency, and self-esteem. Actions devoted to the pursuit of

any of the achievement goals are done in an attempt to accomplish some higher level

goal and result in the cognitive and behavioral patterns referred to as goal

orientations. In this model, the higher level goals are principle-goals defined as

“general heuristics or behavioral principles that serve as guides for clusters of

behavior” (DeShon & Gillespie, 2005, p. 1106; see also Powers, 1973). Some

examples of principle goals listed by DeShon and Gillespie include growth, fairness,

integrity, being friendly etc. However, for new college freshmen, one general

heuristic or behavioral principle that may guide much of their behavior oriented toward achievement goals may be to simply succeed their first college term. Thus, it is postulated here that for first term (quarter) freshmen, one important principle level goal to accomplish is to “do well in college”, one measure of which, for better or worse, is a grade point average. A grade point average is postulated to represent a summary or reflection of how well one is achieving one’s higher level academic goal related to succeeding in college. However, because grade point average represents a demonstration (as opposed to the development) of competence, the higher level grade goal representing academic success is more likely served by performance achievement goals than it is by mastery achievement goals. Other higher level academic goals such as completing challenging courses for their own sake, regardless of grade, are more likely served by mastery achievement goals relative to performance goals.

Thus, in the process of setting or establishing a goal, one must focus on the processes required in achieving the goal. In a hierarchical system, these processes are

lower level goals. In the case of achievement goals, this clearly situates goal orientations as mechanisms of goal achievement because these differential orientations, as patterns of thought and action, focus one’s attention on the manner in which the higher level goal, once established, will be accomplished. Shaw and

Kruglanski (2003) have shown the importance of bottom-up priming of higher level goals. Specifically, when patterns of thought and activity related to specific higher level goals were primed, the accessibility of those goals was enhanced. Furthermore,

these authors were able to show that the priming of these differential patterns

increased task performance and persistence because it allowed for increased focus on

the higher level goal. In addition, because low level action plan strategies are

activated to the extent that the achievement goals require them, the differential

patterns of cognition and behavior derivative from the achievement goals should to

some extent provide the reasons for engaging in the strategies. For example, a

mastery-approach goal such as “I am focused on understanding the content of my

courses as thoroughly as possible” may perhaps provide a reason for doing

extracurricular reading on topics discussed in a course while providing a mechanism

or strategy for achieving the goal of understanding, such as selecting a time-

management action plan of studying at least two hours per every credit taken. A.

Kaplan and Maehr (2007) perhaps state it more succinctly: “Rather than focusing on

the content of what people are attempting to achieve (i.e. objectives, specific

standards), goal orientations define why and how people are trying to achieve various objectives (E. M. Anderman & Maehr, 1994) and refer to overarching purposes of achievement behavior” (p. 142, emphasis in original).

The current conceptualization of goal orientations as resulting patterns of

thought and behavior in the pursuit of goals can also accommodate dispositional

notions of goal orientations. That is, DeShon and Gillespie (2005) view traits as

averages of state-oriented behavior and define dispositional goal orientation as “the

stable pattern of cognition and action that results from the chronic pursuit of a

mastery-approach, performance-approach, or performance-avoid goal in different situations over time” (p. 1115). This is the perspective generally seen in research involving questionnaires focused on mapping goal orientations across time and over

many situations to a set of self-report responses (A. Kaplan & Maehr, 2007). It is

also the perspective taken in this study.

Values

Definition and Relevance.

Implicit in the discussion above is the importance of values in the goal

hierarchy. In addition to the perceived attainability and one’s ability to attain a goal

(Bandura, 1997; Wigfield & Eccles, 2000), the value of a goal will help determine

whether or not one will continue engagement in the pursuit of the goal (Carver &

Scheier, 1998). According to Carver and Scheier (1998), the higher a goal is in the

hierarchy, the more important and valued it is. However, lower level goals also accrue value by their reference to higher level goals because the lower level goals are

instrumental in meeting a higher level goal and/or it possesses the characteristic of

multifinality (Austin & Vancouver, 1996) whereby it achieves multiple goals

simultaneously.

From an expectancy-value point of view, values are an important factor in the

choices students make, providing incentives to approach tasks that are positively

valued and avoid tasks that are negatively valued (Eccles, 2005; Wigfield & Eccles,

1992). Eccles(Parsons) et al. (1983) described academic task values in terms of four dimensions: attainment value, intrinsic value, utility value and cost. Tasks acquire value because they provide a means for confirming aspects of the self (attainment), are inherently interesting (intrinsic value) or are means to a desired end (utility value).

Costs are the negative features of doing the task relative to other tasks. Eccles (2005) suggests that values influence the placement of goals in a hierarchy, and determine the “synchrony” of two or more different goals operating in parallel. The implication is that goals with the lower costs and higher attainment, intrinsic, and utility values will judged more important (Carver & Scheier, 1998) and more desirable than goals with higher costs and lower attainment, intrinsic, and utility values. In addition, this implies that the degree of goal synergy (Ford, 1992) is determined by values to the extent that a valued high level goal can be achieved through multiple goals that may exist in different domains.

As discussed above, synergy that can be achieved between academic and social goals (Urdan & Maehr, 1995; Wentzel, 2005) is one important example.

However, one source of conflict between academic and social goals may occur when academic and social goals have different value levels or achieve different higher level goals that also differ in importance. Thus values place positive or negative valences

on goals that can affect the strength and direction of behavior toward different goals

(Feather, 1988). The idea that students have different goals operating simultaneously

suggests that academic goals must find their place within the goal hierarchy and thus

likely interact with values associated with other goals in this hierarchy.

This suggests that values cannot exist independently of goals. That is, any

statement or indication of a value implies the existence of at least one goal.

Schwartz’s (1994) definition of values clearly includes the notion that values imply

goals: “A value is a (1) belief (2) pertaining to desirable end states or modes of

conduct, that (3) transcends specific situations, (4) guides selection or evaluation of

behavior, people, and events, and (5) is ordered by importance relative to other values

to form a system of value priorities” (p. 20). Furthermore, in addition to values’

relevance to goals (i.e., “end states”) and their importance, this definition also

suggests that values must be trans-situational and guide rather than directly determine

behavior. Presumably, goal importance should be positively related to its trans-

situationality and directiveness. And to the extent that goals are trans-situational and

directive over time they may undergird dispositional entities such as traits and long

term individual interest. Specifically, the contention of the present study is that

values direct and guide self-regulatory choices and, over the long term, the moving

averages or smoothing functions of these behaviors appear as traits and interest.

However, values should guide choice and not actual performance (e.g. Eccles

(Parsons) et al., 1983; Pintrich & De Groot, 1990; Wigfield & Eccles, 1992).

Values and Interest.

This is consistent with a distinction made between individual and situational

interest (e.g. Hidi & Renninger, 2006). Specifically, one defining feature of individual

interest as opposed to situational interest is that individual interest is an enduring

predisposition that requires not only knowledge but also a value component (Hidi &

Renninger, 2006). And while situational interest is a necessary condition for the

development of individual interest, individual interest develops only when a goal becomes valued (Hidi & Renninger, 2006). Recently, Harackiewicz, Durik, Barron,

Linnenbrink-Garcia, and Tauer (2008) showed how values can influence the degree to which different achievement goals are selected. For example, they showed that both individual interest and work-mastery values (i.e., satisfaction with working hard, see

Spence & Helmreich, 1983) positively predicted mastery goals, but negatively predicted work avoidance and performance-avoidance goals and were unrelated to

performance-approach goals. However, value associated with the construct of

competitiveness (Spence & Helmreich, 1983) positively predicted performance-

approach goals as well as performance-avoidance and work-mastery goals but was negatively related to master goals. Also, individual interest did not have a direct relationship with the final grade in an Introductory Psychology course. The effect of individual interest was indirect and mediated positively by mastery goals and situational interest determined early in the academic term and negatively by performance-avoidance goals.

Values and Goal Theory.

What is particularly relevant for the present study is how different goals are

implied by the differential pattern of relationships with values (Eccles, 2005). For example, in the Harackiewicz et al. (2008) study, performance-approach goals were related positively only to competitiveness, whereas mastery goals were related negatively to competitiveness and positively to work-mastery and initial individual interest values. Work-avoidance and performance-avoidance goals had yet another pattern of relationships with these values in that both avoidance goals were associated negatively with initial individual interest and work-mastery values but positively associated with competitiveness.

These differential predictions provided by the differential pattern of relationships goals have with values may account for some of the discrepancies in the literature in the relationships between achievement goals and performance and choice. Specifically, while research has generally shown a positive relationship between mastery goals and some academic outcomes such as effort, persistence, positive feeling about school, and competence beliefs, there does not seem to be a relationship between mastery goals and performance (e.g. grades) (E. M. Anderman

& Wolters, 2006; A. Kaplan & Maehr, 2007). The picture is somewhat muddier for performance goals with some studies suggesting that performance-approach goals are maladaptive (e.g., A. M. Ryan, Hicks, & Midgley, 1997) and other suggesting positive effects (e.g. Elliot, McGregor, & Gable, 1999; Harackiewicz, Barron, &

Elliot, 1998; Urdan, 1997; Wolters, Yu, & Pintrich, 1996). Perhaps the only consensus that has been achieved is the finding that performance-avoidance goals are generally negatively associated with academic performance and positively associated

with maladaptive academic strategies such as self-handicapping (e.g. Elliot &

Church, 1997; Midgley & Urdan, 2001).

Elliot and Church (1997) proposed that perceptions of competence precede

competence goals such that students possessing high perceived competence are more

likely to expect successful task performance and students possessing low perceived

competence are more likely to expect failure. Thus, students with high perceived

competence tend to positively value achievement goals and orient toward an approach

motive and those with low perceived competence tend to negatively value

achievement goals and orient toward an avoidance motive (Elliot, 2005). Elliot

(1999) thus proposed a 2 x 2 framework that places mastery and performance goals

on separate approach and avoidance dimensions. This conceptualization provides

theoretical flexibility in that the interactive and/or additive effects of mastery and

performance goals can be examined in terms of the approach and avoidance

dimensions. Thus, as mastery or intrapersonal goals vary along the approach- avoidance dimensions, their character changes according to striving to improve one’s skills for their own sake (mastery-approach) or striving not to lose one’s skills or knowledge already attained (mastery-avoidance). In a similar way as performance or interpersonal goals vary their character changes according to a desire to do better than others (performance-approach) or striving not to do worse than others (performance-

avoidance) (Pintrich, 2000a).

However, while much progress has been made with the addition of the

valence dimensions of approach and avoidance there are still a number of issues that

need to be resolved. For example, Harackiewicz, Barron, Pintrich, Elliot, & Thrash

(2002) cite evidence showing that performance-approach goals have either no effect or a positive effect on academic outcomes whereas Midgley, A. Kaplan, & Middleton

(2001) maintain that performance-approach goals can be deleterious to academic outcomes and depend on factors such as age, perceived ability, and prior academic performance. Furthermore, sometimes performance-approach and performance- avoidance measures are often highly correlated (e.g., Elliot & Murayama, 2008;

Roeser, 2004; Urdan, 2004) and sometimes they are not (e.g., Cury, Elliot, Da

Fonseca, and Moller, 2006; Elliot and McGregor, 2001).

Furthermore, while Cury, et al. (2006) showed that implicit theories of ability

(Dweck & Leggett, 1988) had the expected effect as predictors of achievement goals, there is still some question about the reasons why students pursue achievement goals or even if students perceive these in the same way researchers intend (Urdan &

Mestas, 2006). Recently, focusing only on performance goals, Urdan & Mestas

(2006) have shown that students have different reasons for pursuing performance goals. The authors claim that this “suggest[s] that performance goals have different meanings for different students, and these meanings may be influenced by a range of factors including culture, achievement history, self-perceptions, and idiosyncratic concerns and prior experiences” (p. 364). These authors found that after completing a standard quantitative achievement goal instrument, high school seniors gave a wide variety of reasons for performance goal pursuit in response to three performance items (two performance-avoidance and one performance-approach). In addition,

Urdan and Mestas found that students often gave different reasons to the same items.

These reasons varied according to social categories such as social comparison, status,

relationships, and teacher perception across dimensions that involved appearance (e.g.

“Do not want parents on my case”; “Do not want to ruin reputation”; “Want to look smart”) and competition (e.g., “Want to keep up with others”; “Do not want to feel stupid”; “Want to prove self to classmate”) (p. 360, Table 4). In addition, students tended to not respond to the putative meaning of the performance goal item in the

way intended. The most blatant contradiction between intended meaning and observed response was seen in the result that students often times did not distinguish between the approach and avoidance aspects of the performance goals items explicitly designed to make this distinction. Obviously, as pointed out by the authors, this result has important implications for the meanings students assign to these goals.

Furthermore, if these multiples reasons do imply different meanings, the effects predicted by the goal orientations can be multiple and unpredictable (Urdan &

Mestas, 2006).

Harackiewicz et al. (2002) can account for these discrepancies first by noting that performance-approach goals have been operationalized differently either as

competitively or normatively based competence goals whereas mastery goals have

been operationalized more consistently with reference to only the development of

competence. In addition, performance-approach goals may be influenced not only by

a need for achievement, but also a fear of failure which may account for the

correlation seen between performance-approach and performance-avoidance goals.

Finally, they contend that the empirical pattern seen from performance-approach

goals is actually quite consistent in that these goals are generally positively related to

performance attainment and unrelated to adaptive outcomes such deep processing and intrinsic motivation (Harackiewicz et al., 2002, pp. 639 – 640).

However, different values may relate in specific ways with achievement goals that may account for some of the differences and similarities between performance- approach, performance-avoidance, and mastery goals. For example, a student with

high academic task values determined by individual interest, high work-mastery and

low competitiveness values may be more likely to pursue mastery goals

(Harackiewicz et al., 2008), show more interest in studying and evidence more effort

and persistence in times of difficulty or potential distractions (Kaplan & Maehr,

2007), yet may not do as well in terms of grades earned relative to a student who is

simply highly competitive (E. M. Anderman & Wolters, 2006). In addition, these

more internal values may be fundamental to the more internal incentives that are the

basis for much self-regulatory behavior (Bandura, 1986). Alternatively, a student who is highly competitive may evidence relatively high grades for a course or semester regardless of individual interest. However, according to the Harackiewicz et al. (2008) results, competitive values may be a two-edged sword in that these values positively predict both performance-approach goals and performance-avoidance goals

(thus, possibly providing at least one more mechanism for the positive relationship between these goals), but also negatively predicts mastery goals. Thus, viewing performance-approach goals solely from the point of view of competitive values may not provide the most complete picture of the functioning of these goals. In fact, Elliot

(2005; Elliot and McGregor, 2001) prefers to define performance-approach goals from a normative competence perspective rather than a self-presentational or

competitive perspective. But this seems to imply only that values other than (or in addition to) competitiveness are involved in performance-approach goals.

In academic contexts, one likely value guiding performance goals is simply valuing receiving high grades. Assuming that many first term freshmen are imbued with an outcome-performance ethos, it might be reasonable to assume that the value charged to the goal of high grades may be rather potent for many of these students. In this case, this more external goal may be able to support the self-regulatory behavior required to forestall interruptions due to distractions or procrastination for example

(Bandura, 1986). In addition, to the extent that getting a high grade is perceptibly

equivalent to not getting a low grade, approach and avoidance performance goals

should be highly correlated and may be supported by similar self-regulatory

behaviors. Bandura (1986) writes that “[p]eople are especially motivated to exercise

self-influence when the behavior they seek to regulate is aversive or potentially so”

(p. 370). Another reason for the similarities sometimes seen between performance-

approach and performance-avoidance goals may be due simply by the fact that these

goals are inherently social in nature. The demonstration of competence implies a

social context and to the extent that this is true, social values like the desire to be

included may be an important factor driving these goals. Thus, in academic contexts

it appears that an appreciation of what students deem important and what they desire

and are interested in over the long term will help determine what higher level goals

they strive to achieve. This, in turn, may allow for a determination of what

achievement goals are selected and how they are differentially utilized as means in

attempts at achieving the higher level goals. However, the determination of the actual

paths between the higher-level principle goals and lower-level achievement goals is

an empirical question and more studies like those of Harackiewicz et al. (2008) are

needed.

Self-Regulation: Linking Values to Achievement Goals, School Belongingness,

Self-Efficacy, and Perceived Stress

The notion that values influence choices suggests that the directive and guidance functions of values thereby contribute in affecting self-regulatory processes.

And while it assumed that self-regulation is active at some level all of the time, it may be lacking below some optimal level (Blunt & Pychyl, 2000; Pychyl, Lee, Thibodeau,

& Blunt, 2000; Steel, 2007) particularly in those situations perceived as tedious, boring, or otherwise aversive – situations where it is needed most. An important issue then is to assess the differential influences of different values on the self- regulatory choices students make. Specifically, given the potential conflicts between academic and social values, how to do these values differentially contribute to self- regulation? Furthermore, to the extent that self-regulation has direct consequences for important outcomes such as self-efficacy (Sirois, 2004), stress (Tice &

Baumeister, 1997), and school belongingness (Anderman & Freeman, 2004), do values have differential indirect influences on these outcomes?

A well-functioning self-regulatory system regulates cognition, motivation, affect, behavior, and to some extent environmental context (Pintrich, 2000b) in a real- time cyclical fashion (Zimmerman, 2000a) where choices are made continuously.

However, in the longer term, high level goals and values can influence how self-

regulatory processes, manifested as dispositions, function. Competition among potentially incongruent values and high level goals as well as inefficient cognitive or metacognitive strategies can disrupt self-regulatory processes resulting in self- defeating behavior patterns such as procrastination (Baumeister, 2002; Baumeister,

Bratslavsky, Muraven, & Tice, 1998; Baumeister & Heatherington, 1996;

Baumeister, Muraven, & Tice, 2000; Muraven & Baumeister, 2000; Pintrich, 2000b;

Steel & König, 2006; Tangney, Baumeister, & Boone, 2004; Tice & Baumeister,

1997) that over the long term will appear as a tendency or disposition to procrastinate.

In the achievement domain competition among goals requires students to make decisions regarding which goal to pursue and these decisions, in turn, reflect how students regulate their performance and their learning. It is often the decisions students make that can determine their academic success. Furthermore, these decisions may have considerable weight in determining success relative to ability.

For example Duckworth and Seligman (2005) showed that measures of delay of gratification and self-discipline predicted academic performance in eighth-graders better than IQ seven months after the measures were taken. In addition “highly self- disciplined adolescents outperformed their more impulsive peers on every academic- performance variable, including report-card grades, standardized achievement-test scores, admission to a competitive high-school, and attendance.” (p. 941)

The distinction seen in the Duckworth and Seligman (2005) study is related to one provided by McCombs & Marzano (1990) who refer to skill (i.e. strategy) and will (desire or choice) as crucial in the academic achievement domain. Skill and will underlie decisions students make with respect to the goals they select, reflect how

students regulate their performance and their learning (Tuckman, 1990), and thus determine in large part the progress they make toward those goals. McCombs and

Marzano (1990) argue that the self is an agent in integrating skill and will. In learning and achievement contexts the integrating activities of the self define how it regulates through choosing cognitive, metacognitive, affective, and behavioral skills.

This act of choice is the “will” which they define as an “innate or ‘self-actualized’ state of motivation, an internal self-generated desire resulting in an intentional choice.

. .” (p. 52). “Skill” is “an acquired cognitive or metacognitive competency that develops with training and/or practice.” (p. 52) Along these lines Tuckman (1990;

1992; Tuckman & Sexton, 1990) distinguished between self-regulated learning and self-regulated performance. Self-regulated learning is comprised of competence and choice. Self-regulated performance reflects only choice. According to Tuckman

(1990) self-regulated performance is “the self-initiated application of effort to perform a task using knowledge and skill that the performer has already acquired.” whereas self-regulated learning is “knowledge and skill acquisition personally initiated and directed by learners themselves” (p. 292). Defined this way, Tuckman’s distinction suggest that self-regulatory processes undergird achievement goal processes.

Thus, disruptions in self-regulation can occur through disruptions in cognitive or metacognitive processes (skill or strategies) mediated by an executive processor

(e.g. Borkowski & Muthukrisha, 1992; Borkowski & Thorpe, 1994), through a depletion of necessary cognitive and motivational processes affecting one’s will or desire (e.g. Baumeister, 2002; Baumeister, Bratslavsky, et al., 1998; Baumeister &

Heatherington, 1996; Baumeister, et al., 2000; Muraven & Baumeister, 2000), and/or

through competition of incongruent values and goals (Steel & König, 2006).

However, while strategic and executive processes affecting self-regulated learning are

clearly important, processes influencing will, desire, or choice of goals are

hypothesized here to have a crucial impact on self-regulation manifested as self-

defeating behavior and often reflected as procrastination.

Procrastination, Value Conflict, and the Potentially Debilitating Effects of Perceived

Social Exclusion

Within a control systems perspective of motivation (Carver and Scheier,

1998), goals are the primary input of the self-regulatory system and are the standard

by which the entire process operates. However, a comparator process that is sensitive

to task aversion (Blunt & Pychyl, 2000; Pychyl, Lee, Thibodeau, & Blunt, 2000), low

conscientiousness (Schouwenburg, 2004; Schouwenburg & Lay, 1995) or

impulsiveness and distraction (Lasane & Jones, 2000; Schouwenburg & Groenewoud,

2001) might be expected to reject some goals in favor of others. The essence of

procrastination is that one or more tasks are deferred in favor of other tasks even if

those other tasks are not productive (e.g. watching television, spending time with

friends rather than studying, etc.). Lack of self-regulation embodied as

procrastination is reflected in choice of goal pursuit, implies a lack of willpower, and

manifests itself as irrationality (Ainslie, 2001). Indeed, Steel’s (2007) definition of

procrastination as a “voluntar[y] delay [of an] intended course of action despite

expecting to be worse off for the delay” (p. 66) encapsulates the constructs of choice,

desire, and irrationality quite well.

Therefore, in addition to disruptions of an executive processor, self-regulatory

failure may occur when incongruent values and goals compete for preference. As

such, procrastination is reflected in the competition among the different goals

students must negotiate (Bembenutty, 2008; Mischel & Baker, 1975; Mischel &

Ebbesen, 1970) and a recently proposed theory of choice, temporal motivation theory

(TMT) (Steel & König, 2006), has been proposed that can account for it.

Specifically, given that perceived utility or desirability (U) of a task or goal is a sufficient condition for the choice a person will make such that for tasks A and B UA >

UB  UA, then the choice for any task or goal is defined in terms of the product of

expectancy and value (i.e. E  V) relative to the product of time to reach the goal (D)

VE and one’s sensitivity to delay ( (i.e. D  ) (i.e., U  , Steel, 2007; see Steel  D

& König, 2006 for a more detailed cumulative cost/benefit basis for this conceptualization). Thus, as one’s expectancy for success and/or value for a task increases, the desire and the likelihood of choice for the task increases for a given delay sensitivity at a given point in time. However, as the delay to goal achievement and/or one’s sensitivity to delay increases, the desire for achieving a goal decreases given a constant expectancy for success and value. There is a considerable amount of empirical evidence supporting the prediction of the time course of dilatory behavior given by this utility function (Gröpel & Steel, 2008; Howell, Watson, Russell,

Powell, & Buro, 2006; König & Kleinmann, 2004; Moon & Illingworth, 2005;

Pychyl, Lee, Thibodeau, Blunt, 2000; Schouwenburg & Groenewoud, 2001).

Tuckman (1996; 1998) for example, showed that students with relatively low

motivational desire as measured by low GPA and high procrastination tendency,

benefited significantly in their performance on large term achievement tests designed

to measure comprehension when they were required to take frequent, but short,

weekly quizzes relative to a condition where they were required to complete

homework assignments designed to help them elaborate the meanings of important

topics. Interestingly, while the frequent quiz condition showed a significant effect on

achievement test scores overall, these differences were not found for students with

high or medium levels of motivation. This effect is predicted by TMT because

dividing larger tasks into smaller parts for students with low motivation decreases the

perceived delay and thereby increases the perceived utility whereas for students with

higher levels of motivation (i.e., high expectancy and/or task value), the external

imposition of dividing larger tasks into smaller ones may have been unnecessary

because these students may have been studying regularly on their own all along.

One important factor that may link goal competition to self-regulatory failure,

which plays out as the utility function described above, is a belief or concern

regarding the potential for social exclusion or rejection. Summarizing a program of

research on the relationship between social exclusion and self-regulatory failure,

Baumeister and DeWall (2005), concluded that in addition to having deleterious

effects on cognitive processes, perceived social exclusion appears to directly affect

the will, as opposed to the ability, to self-regulate. The implied instigator of this effect is a fundamental need to belong (Baumeister & Leary, 1995) that can motivate

self-regulatory behavior in order to gain social acceptance (Baumeister & DeWall,

2005) and may conflict with self-regulatory behavior directed toward academic work.

For example, self-presentational concerns may be a primary mechanism underlying

self-defeating strategies in general (Tice & Baumeister, 1990) and may be one reason

that van Eerde’s (2003) meta-analysis showed that receiving negative performance

feedback was not as important a factor in procrastination as one’s self-image. In

addition, much research on academic procrastination involves the study of conflicts

between studying and social concerns, socializing, or leisure activities involving

socializing (Dietz, Hofer & Fries, 2007; Schouwenburg & Groenewoud, 2001;

Hofer, et al., 2007; Senecal, Julien, and Guay (2003).

These results suggest that value for social belongingness, reflected as a

concern over the potential for social exclusion, may have a debilitating effect on self-

regulation that can be assessed as a significant positive relationship between social

exclusion concerns and procrastination. Social goals will have particularly high value

in situations of perceived social exclusion because, through affiliation, they are the means to the creation and reflection of identity. Specifically, social goals are highly

valued in this situation because, to the extent that perceived social exclusion threatens

identity, they help achieve students’ sense of how the group defines them as well as

how they help define the group. Furthermore, if academic goals are not part of these

social definitions, students will naturally find goals that are. And if academic goals

have a lower expectancy of success (in addition to lower value) and/or a long delay

relative to more diagnostic social goals, the procrastination of starting academic tasks

is a likely outcome. However, procrastination for academic tasks can be attenuated in

situations where the need for belongingness is not a concern or where academic goals are diagnostic of satisfying an active need to belong. In the former situation, the utility function for academic goals is higher than that for social goals and in the latter situation academic and social goals are likely congruent.

Thus, goal competition between social and academic goals such that the utility of social goals is greater than that of academic goals is one possible source of a lack of conscientiousness, perceived task aversion, impulsiveness, and lack of desire to delay gratification. To the extent that the need for belongingness exacerbates or releases these conditions, self-regulatory behavior may suffer. This may then lead to an attenuated desire for achievement goals, a decrease in one’s sense of academic self-efficacy and an increase in perceived stress (Sirois, 2004; Tice & Baumeister,

1997). Furthermore, to the extent that academic procrastination reduces important academic motivational outcomes, it may also have a simultaneous negative influence on one’s perception of belonging to an academic institution or to an “academy” more generally. Thus, in academic contexts, procrastination, as a reflection of choice among values regarding belongingness, academic tasks, and grades may be a linchpin linking these values to achievement goals, academic self-regulatory self-efficacy, perceived stress, perceived belongingness in school, and ultimately, academic performance.

The Consequences of Procrastination: Effects on Learning and Performance Goals

There have not been many studies concerning the relationship between procrastination and achievement goals and those that have been conducted have

conceptualized procrastination as a criterion variable. For example McGregor &

Elliot (2002) found that while performance-avoidance goal positively predicted procrastination, mastery and performance-approach goals were unrelated to procrastination. Wolters (2003) found in his first study that after accounting for work-avoidance orientation (which was significantly positively related to procrastination), both mastery and performance-approach goal orientations were unrelated to procrastination. In his second study, however, performance-approach orientation was weakly positively related to procrastination even after controlling for work-avoidance orientation. Mastery goal orientation was not significantly related to procrastination. Wolters concluded that student expectations about the difficulty of the task and their ability to successfully complete it may be important determinants in procrastination independent of the value students place on the task. More recently,

Howell & Watson (2007) reported a significant negative and a significant positive relationship between a mastery-approach and a mastery-avoidance orientation respectively and procrastination. However, these relationships disappeared after the inclusion of metacognitive and cognitive processing strategies. Procrastination was not related to either of the performance goals.

These results seem to suggest that achievement goals are at best only very weak predictors of procrastination tendency and this may be due to the fact that achievement goals were highly correlated with other predictors. For example,

McGregor & Elliot (2002) reported a significant positive zero-order correlation between procrastination and performance-avoidance goals. However, when performance-avoidance was entered into a simultaneous regressions equation with

perceived control, degree of absorption in studying, and two measures of affect (e.g.

anxiety and hope), it was not predictive of procrastination. One reason for this may have been that performance-avoidance was also highly correlated with every other predictor except hope. Likewise, Howell & Watson (2007) reported a significant zero-order correlation between mastery-approach and procrastination, but the

relationship disappeared after adding metacognitive and cognitive strategy predictors

to the equation. Again, all but one cognitive variable (i.e., surface processing) were

significantly related to mastery-approach.

However, perhaps the significant correlations among the putative predictors of procrastination are instead a reflection of procrastination. That is, perhaps procrastination is not a consequent of achievement goals (or of any of the other variables studied), but rather an antecedent. Specifically, it seems reasonable to

assume that students with a tendency to procrastinate might be less focused on

academic achievement goals. Under the assumption that many study-related

activities will have some task aversiveness for many students, it might be expected that students with a higher tendency to procrastinate will choose more desirable short-

term options in favor of longer-term achievement goals as predicted by TMT. There

is a considerable amount of evidence for this (Blunt & Pychyl, 2000; Lasane & Jones,

2000; Moon & Illingworth, 2005; Pychyl, et al., 2000; Schouwenburg, 2004;

Schouwenburg & Lay, 1995; Schouwenburg & Groenewoud, 2001). It should be expected then that procrastination should have some predictive relationship to achievement goals. Specifically, students with a tendency for procrastination should be expected to show less desire overall for pursuing both mastery- and performance-

approach achievement goals. Furthermore, to the extent that procrastination is an avoidance strategy, it should also be expected that it should be positively related to a

performance-avoidance goal orientation as has been shown in some previous research

(McGregor & Elliot, 2002). While this relationship would be expected to change at

some point in real time in line with TMT, when collapsed across time high

procrastinators are expected to show less preference for achievement goals to the extent that these goals are less attractive on average than other goals. This hypothesis

can be tested at the macro level by measuring procrastination tendency early in an academic term and achievement goals at some date subsequent to this. If the relationships between procrastination and achievement goals are significant in the presence of other predictors of achievement goals, then this should provide evidence that procrastination tendency may be partly responsible for the lack of adoption of achievement goals.

The Consequences of Procrastination: Effects on Self-efficacy

As discussed above, according to Schwartz (1994) values are beliefs about the

desirability and importance of goals that transcend specific situations. Thus, values

are not expected to directly determine behavior in pursuit of lower level, situation-

specific goals, but rather function more as guide for behavior. In academic achievement situations, values guide, rather than direct, behavior through achievement goals (Harackiewicz et al., 2002; Harackiewicz et al., 2008) and self-

regulation (Bandura, 1986). However, beliefs about goals are not limited to

perceived importance, relevance, or costs of the tasks necessary to achieve desired

goals, but are also concerned with the likelihood or expectancy of actually achieving

these goals in two senses: 1) competence beliefs which refer to perceptions of one’s capability to fulfill a certain action or process toward a goal – that is, a focus on the processes or means required and 2) control beliefs which refer to the perception that a

given action can produce a goal (Bandura, 1997; Schunk and Zimmerman, 2006).

Bandura (1997), for example, makes a very clear distinction between competency and

control. In Bandura’s theory competency explicitly refers to self-efficacy beliefs and

is defined as “beliefs in one’s capabilities to organize and execute the courses of action required to produce given attainments.” (Bandura, 1997, p. 3) Control on the other hand refers to outcome expectations or the belief that certain actions, if performed successfully, will result in the desired outcome. Thus, the distinction between values and expectancy beliefs is that values are inextricably linked with goals and determine their perceived importance in the hierarchy while expectancy

beliefs are more directly related to the behaviors necessary to achieve a goal.

Particularly important for the present study are the competency beliefs that contribute

to the causal link between values and one’s behavior (i.e. self-efficacy).

In order for a student to become a self-regulated learner, s/he must first make

choices regarding the amount of effort to expend in order to perform the tasks

necessary to successfully learn – that is, s/he must first focus on self-regulated

performance using the skills currently possessed (Tuckman, 1990) which, in turn, will

be reciprocally determined in the future by the self-regulated learning occurring in the

present. It is in this self-regulatory cycle that self-efficacy plays a very important role

(Bandura, 1997).

There is evidence that points to a strong relationship between self-efficacy and

procrastination. Many studies have shown significant negative relationships between

procrastination and both specific and global measures of self-efficacy (e.g., Haycock,

McCarthy, & Skay, 1998; Klassen, Krawchuk, & Rajani, 2008; Sirios, 2004;

Tuckman, 1991; Wolters, 2003). Wolters (2003) found a strong negative relationship between self-reported procrastination and self-efficacy even after controlling for goal orientation (mastery, performance-approach and avoidance, and work-avoidance) in hierarchical regression analyses. Wolters suggested that these results seem to imply

that regardless of the intrinsic or extrinsic value of an academic task, procrastination

can still occur if it is expected to be too difficult given one’s perceived ability.

However, one problem with this conclusion is that value or importance of the task

was not explicitly measured or defined in this study except to the extent that it was reflected in mastery or performance orientations. Haycock, et al., (1998) also found a significant negative relationship between self-efficacy and procrastination and that

the self-efficacy effect weakened a prior significant bivariate relationship between

both trait- and state-anxiety to non-significance. Finally, meta-analyses conducted by

Steel (2007) and van Eerde (2003) showed very reliable negative relationships

between self-efficacy and procrastination.

However, as was true for achievement goals, self-efficacy has commonly been

viewed as an antecedent to procrastination (e.g. Bandura, 1997; Haycock, et al., 1998;

Klassen, et al., 2008; Wolters, 2003b). One exception to this is a study of the

mediational role of health self-efficacy with in the relationship of procrastination and

health behavior intentions (Sirois, 2004). Health self-efficacy was defined as

“feelings of competence and confidence in being able to carry out actions important for maintaining and taking care of one’s health.” (p. 120) Sirois reported that health self-efficacy completely mediated the effect of procrastination on health behavior intentions with a strong negative relationship between procrastination and health self- efficacy and a strong positive relationship between self-efficacy and health behavior intentions. These findings suggest that measures of procrastination at the beginning of an academic term should predict global measures of self-regulatory self-efficacy near the end of the term. In addition, findings that individual differences related to self-regulation may influence the course and development of self-efficacy over time

(Sexton & Tuckman, 1991; Sexton, Tuckman, & Crehan, 1992; Tuckman & Sexton,

1990) lend further support to the contention that the tendency to procrastinate

assessed at the beginning of an academic term is expected to be negatively related to

regulatory self-efficacy at the end of the term.

The Consequences of Procrastination: Effects on Perceived Stress

There are a number of studies that suggest that a positive relationship exists

between procrastination and perceived stress. For example, Tice and Baumeister

(1997) have shown that while procrastination appeared to have initial beneficial

effects on stress and health at the beginning of an academic term, over the course of

the term procrastinators reported significantly higher levels of stress, more symptoms

of illness, and evidenced lower exam grades. Blunt and Pychyl (2000) asked students

to generate a list of personal projects (Little, 1983) they were currently in engaged in or were thinking about. Personal projects are self-expressed goals people see

themselves caring about and feel a need to achieve (Little, 1983). Blunt and Pychyl

(2000) extracted a number of dimensions from the goals college students were working on. In addition, students indicated the current stage of their projects (i.e., stage of goal completion). These authors reported that for the dimension of task aversion (a demonstrated source of procrastination), in addition to boredom, frustration, and resentment, stress correlated positively during the stages of inception and termination of goal completion. Importantly, stress was more strongly related to the task aversion dimension at the termination of the task relative to the inception of the task. To the extent that task aversion was significantly correlated with procrastination at task inception and termination, these results suggest that stress may be related to putting off aversive tasks.

A number of other studies have shown that procrastination is positively linked to stress and anxiety (Flett, Blankstein, & Martin, 1995; Lay, Edwards, Parker, &

Endler, 1989; Rothblum, Solomon, & Murakami, 1986; Schraw, Wadkins, &

Olafson, 2007; Tice & Baumeister, 1997; Solomon & Rothblum, 1984). For example, a recent qualitative study by Schraw, et al. (2007) showed that, while collectively students reported that “they procrastinated for adaptive reasons and rarely felt that procrastination had a negative impact on learning” and “indicated that they learned more efficiently than they would have had they not procrastinated” (p. 23), procrastination was nevertheless associated with stress, adverse health effects and negative well-being. One possible reason for this discrepancy between what students say and negative health effects reported may be seen in the rationalizations students who procrastinate tend to give (Tuckman, 2005). Indeed, it is possible that

rationalizing may be a surrogate strategy used in lieu of other more optimal and productive self-regulatory strategies and accomplishes little (Ferrari, 2001). To the extent that rationalizations support continued procrastination (Tuckman, 2005), it might be expected that students who truly believed that procrastination was helpful would be at a disadvantage both academically and physically.

Self-Efficacy as a Mediating Influence on Perceived Stress and Academic

Performance

The Role of Self-Efficacy on Academic Outcomes

There is a considerable evidence suggesting that perceived academic self- efficacy is positively related to academic outcomes (Bandura, 1997; Bandura,

Barbaranelli, Caprara, & Pastorelli, 1996; Gore, 2006; Kahn & Nauta, 2001; Pajares,

1996; Schunk, 1991; Pintrich & DeGroot, 1990; Roeser, et al., 1996; Schunk &

Pajares, 2005; Tuckman, 1990;Tuckman & Sexton, 1990; 1991; Zimmerman,

Bandura, & Martinez-Pons, 1992; Zimmerman, 2000b; Zimmerman & Kitsantas,

2005; 2007). However, in addition to direct effects on academic performance, self- efficacy is hypothesized here to play another potential role in academic achievement as a mediator of the influence of achievement goals, procrastination, and a sense of school belonging on academic performance and perceived stress. The relationship between procrastination and self-efficacy has already been discussed. This section therefore discusses the relationships discussed in the literature between achievement

goals and self-efficacy and the effects of self-efficacy on perceived stress in academic

contexts.

Self-efficacy beliefs can change over the course of an academic term (Sexton

& Tuckman, 1991; Sexton et al., 1992) and initial estimates of self-efficacy are poor

estimators of performance (Tuckman & Sexton, 1990). In line with this finding

recent research (Gore, 2006) has shown that while assessment of self-efficacy at the

beginning of college freshmen’s first semester only very weakly predicted semester

grade point average through the third semester, assessment of self-efficacy late in the

first semester was consistently and strongly related to semester grade point average

through the third semester. Kahn and Nauta (2001) reported results similar to this:

measures of self-efficacy assessed late in a term were superior predictors of academic

performance relative to assessments early in a term. A similar pattern was found in predicting first and second year retention. Gore suggested that this finding is

consistent with Bandura’s (1986; 1997) contention that self-efficacy develops over time as a function of prior performance and vicarious learning with prior performance being the most influential, which, as discussed above, should be considerably negatively affected by one’s degree of procrastination over the course of an academic term. Thus, given these findings and a greater interest in assessing the relationship of self-efficacy and academic outcomes as opposed to the change of self-efficacy over the course of the freshmen’s first academic term, the present study assessed self- efficacy once late in the first term.

The Influence of Student’s Achievement Goals on Self-Efficacy

In addition to dispositional procrastination another important factor that is hypothesized to contribute to academic self-regulatory self-efficacy is the degree to which students are focused on mastery and performance achievement goals. Similar to the differential relationship between these achievement goals and academic outcomes and self-regulation, the relationship between mastery or learning goals and self-efficacy appears to be generally positive while the relationship between performance goals and self-efficacy is somewhat more complicated (Pintrich, 2000b).

For example, Wolters et al. (1996) showed that for seventh, eighth, and ninth graders, when learning and performance goals were assessed at the same time as self-efficacy

(late in the school year), learning goals were positively related to self-efficacy in

Mathematics, English, and Social Studies and performance goals were positively related to self-efficacy in English and Social Studies but showed no significant relationship to self-efficacy for Mathematics. However, Middleton and Midgley

(1997) found that while self-efficacy was positively correlated with mastery goals, it was unrelated to performance-approach goals negatively related to performance- avoidance goals. Likewise, Skaalvik (1997) found that self-efficacy was negative related to a performance-avoidance orientation but, unlike Middleton and Midgley, that it was positively related to a performance-approach orientation.

Similar relationships have been found in college age students. For example, in a structural equation model of undergraduates enrolled in introductory management and psychology courses, Phillips and Gully (1997) found that when learning and performance goals were used as predictors of self-efficacy in conjunction with locus of control and ability predictors, learning goals were positively related and

performance goals negatively related to self-efficacy. Breland and Donovan (2005)

found that while dispositional learning and performance goal orientations did not

predict self-efficacy directly their effects were mediated by both situational learning

and performance goal orientations that had a direct positive relationship with self-

efficacy in their second study (but not the first). In the first study only situational

learning goal orientation was positively related to self-efficacy. Performance goal

orientation was unrelated to self-efficacy in this study. Gerhardt and Brown (2006)

looked at the role of goal orientation in self-efficacy development in response to

voluntary participation in “academic self-management” training offered in a college

business course. They found that mastery goal orientation had a positive relationship

to post-training self-efficacy only if self-efficacy was high prior to training. If self-

efficacy was low prior to training post-training self-efficacy did not differ as a

function of mastery goal orientation. Performance goal orientation was positively

related to post-training self-efficacy regardless of pre-training self-efficacy level.

Thus, students with high levels of performance goal orientation tended to benefit from the self-efficacy training. However, students with low levels of performance goal orientation actually evidenced a slight decrease in post-training self-efficacy if pre-training self-efficacy was high relative to low pre-training self-efficacy.

Thus, while the literature appears to paint a relatively consistent picture of the

relationship between mastery goals and self-efficacy, the picture is not so clear for

performance goals. Therefore, this study simply attempts to provide another

descriptive assessment of the relationship between achievement goals and self-

efficacy when the tendency to procrastinate and perceived school belongingness are

taken into account.

The Relationship of Perceived Stress to Self-efficacy

According to Cohen and Wills (1985) perceived stress occurs in situations construed as threatening and where appropriate coping responses are unavailable.

Specifically, construing a situation as threatening refers to a primary appraisal regarding the potential for harm or loss, whereas challenge which refers to a primary appraisal for the potential for growth or mastery (Folkman & Lazarus, 1985). Coping refers to secondary appraisals regarding one’s resources and options as a response to a threatening or challenging event (Folkman & Lazarus, 1985). It appears that self- efficacy may help attenuate the effects of anxiety either through viewing situations as challenging (as opposed to threatening) and/or through secondary appraisals of one’s possession of the necessary resources and decision skills (Bandura, 1997). To the extent that self-efficacy can help students attend to and master skills and knowledge, regulate study behavior and management skills, and see failures as opportunities, it increases the likelihood of believing that one has the necessary coping resources to deal with potentially threatening situations and thus the tendency to view these situations as challenges rather than threats (Chemers, Hu, & Garcia, 2001).

This study assesses the relationship between self-regulatory self-efficacy and general levels of perceived stress (Cohen, Kamarck, & Mermelstein, 1983). It is expected that the perceived stress that students experience during their transition to college can extend beyond that associated specifically with academic work (Hall,

Chipperfield, Perry, Ruthig, & Goetz, 2006; Lumley & Provenzan, 2003).

Furthermore, in line with the finding of Chemers, et al. (2001; see also Bandura,

1997), it is expected that self-efficacy can increase students’ perceptions of their capabilities to deal with challenges and thereby decrease perceived stress. Chemers et al. (2001) found a direct negative relationship between students’ beliefs about their capabilities to deal with academic demands and general perceived stress. In the present study it is expected that high self-efficacy beliefs in one’s capability to handle the rigor of academic studies should decrease the perception of the end-of-term activities as threatening thereby reducing reports of perceived stress.

An attempt at coping with stress is hypothesized to deplete a global regulatory resource that can result in failures in attempts at other self-regulatory functions

(Baumeister & Heatherton, 1996; Baumeister, Heatherington, & Tice, 1994). Thus, if perceived stress culminates near the end of a term, attempts at coping can deplete this resource resulting in decreased attention to studying and an increase of perceptions or beliefs that necessary academic skills and knowledge are not being mastered as expected. Self-efficacy is postulated here to act as a buffer in this regard in that it may have the effect of removing some of the effort in the coping response. It is hypothesized that student’s with high self-efficacy already possess one powerful coping response with the belief that challenges can be overcome as opposed to one where challenges are perceived as threats. Thus, to the extent that self-efficacy leads to the belief that one has the necessary resources to turn threats into challenges the potentially depleting effects of perceived stress may be lessened.

Perceived Stress, the Depletion Resource Model, and the Benefit of Feelings of

School Belonging

Baumeister and colleagues (Baumeister, 2002; Baumeister, et al., 1998;

Baumeister & Heatherington, 1996; Baumeister, et al., 2000; Muraven & Baumeister,

2000) have suggested that the ability to self regulate depends on a resource that

results in self-regulation failure when depleted. This view, referred to as ego or

resource depletion (Baumeister, et al., 1998) is that self-regulation is a limited

resource and when one exerts self-regulatory control on one task, less of that resource

is available for other tasks (Baumeister & Heatherton, 1996). Thus, failure at self-

regulation with respect to a particular goal or task is due to the self’s executive

function loss of “strength” in the process of inhibiting itself or initiating or processing

activities with respect to other goals or tasks. This pool of resources or strength serves

all of the executive function’s processes regardless of whether or not these processes

are related. Therefore, when the executive function is processing for one goal or task, it has less strength available for other, potentially unrelated tasks. With the depletion of this resource, self-control in these other areas will tend to fail. This resource has the characteristics of a muscle in that when it is used it will fatigue producing self- regulation failure, but when used repeatedly becomes stronger and can increase self- regulation potential over time (Muraven & Baumeister, 2000). The multiple goals students are often faced with must be managed over the course of an academic term and can compete with achievement. Over a short-term achievement situation, the multiple goals of achievement, inhibiting disruptive impulses, controlling emotions, etc. all require executive functioning (Schmeichel & Baumeister, 2004) that can

compete for a limited pool of self-regulatory strength (Baumeister & Heatherton,

1996). Resource depletion is seen here as leading to deficits in self-control that can have deleterious effects in academic contexts (Tangney, et al., 2004). Aspects of cognitive, motivational, and behavioral self-regulation involving activities of metacognition, planning, monitoring, and reflecting are core to overall self-regulatory functioning in achievement situations. However, coping with stress is hypothesized to deplete this self-regulatory resource and thereby affects the will to choose appropriate self-regulatory behaviors.

Recently, Oaten and Cheng (2005), using the 10-item version of the Perceived

Stress Scale (PSS) (Cohen, et al., 1983) found that college students having to prepare for exams near the end of a term, reported significantly less daily hygienic behaviors

(e.g. brushing teeth, washing hair, changing cloths, doing laundry, washing dishes) and less general regulatory behavior (impulsive spending, socializing, sleeping in, losing temper) relative to the beginning of a term where the perceived stress of examinations is ostensibly lower. In addition, examination stress even affected short- term regulatory performance as measured in a laboratory Stroop test. Specifically, students preparing for examinations had significantly longer latencies for incongruent color-words combinations relative to control students not preparing for exams. In addition, in a repeated measures assessment, these latencies were longer for students preparing for examinations when assessed during the examination period relative to a period before examinations. Using the resource depletion model, Oaten and Cheng

(2005) concluded that the “results are consistent with a prediction of a limited strength model of self control” and that these “results are expected if coping with

stress consumes some resource required for self-regulation, leaving individuals less able to control and direct a broad spectrum of regulatory behaviors.” (p. 273) These results seem to suggest that increases in perceived global stress experienced by first- quarter freshmen near the end of a term – as their first final examination period is approaching – should be associated with a range of cognitive, metacognitive, motivational self-regulatory deficits resulting in poor academic performance as assessed by first quarter grade point average. And while increases in self-efficacy may attenuate this effect to some extent, it may not be sufficient.

However, as discussed above, social factors such as perceived school belongingness may play a role as an additional buffer against the effects of stress.

Specifically, the importance of perceived school belongingness in motivation, achievement and well-being is seen in its role in promoting congruency between academic and social goals. There has not been much research concerning the relationship of school belongingness to perceived stress. However, there is research that suggests that feelings of school belonging are positively associated with well- being (Anderman & Freeman, 2004) and self-efficacy (Freeman, et al., 2007; Roeser, et al., 1996). For example, feelings of school belonging are negatively related to depression (Anderman, 2002), and positively associated with positive affect

(Anderman, 1999; Roeser, et al., 1996). Anderman (1999) found that for fifth graders transitioning to middle schools and experiencing a sense of school belonging in the new school reported feeling happier and less frustrated and bored relative to the old school. In addition, Freeman et al. (2007) reported significant positive correlations between self-efficacy and class and university belonging.

There is also some research on a construct discussed in the sociological and

higher education literature that bears some resemblance to the construct of school

belongingness. Mattering, defined as the extent to which a person feels as if “he or

she counts” or “makes a difference” (Rosenberg & McCullough, 1981, p. 163).

Recently, Dixon Rayle & Chung (2007) reported correlational evidence that

mattering to one’s friends at college and perceived social support from friends and family were significantly negatively related to perceived academic stress. On this

basis and to the extent that global levels of perceived stress are associated with the

transition to college, it is expected that school belongingness will act as an additional

buffer against perceived stress. That is, higher levels of perceived school belonging

should be negatively related to perceived stress and should thus contribute to the

proposed buffering effect between self-efficacy and perceived stress. Finally, while

there has not been that much research on the relationship between school belonging

and self-efficacy, Anderman and Freeman (2004) suggested a model whereby school

belonging has indirect effects on academic achievement through affective, cognitive,

motivational variables. The model proposed in the current study is similar to this in

that school belonging is proposed to influence academic outcomes indirectly through

perceived stress and self-efficacy.

Summary and Overview of the Model and Hypotheses

Overview of Proposal

The discussion above suggests that there are a number of different

motivational, social, and personality factors involved as potential influences in academic achievement. It also suggests a number of specific hypotheses regarding how these influences relate and how these relationships can potentially affect academic achievement. What follows is a summary of the important implications of the research reviewed and the hypotheses of this study that follow from these implications.

Goals are cognitive structures representing standards (Locke & Latham,

1990), aims (Elliot, 2005), means and ends (Shah & Kruglanski, 2000), or desires

(Austin & Vancouver, 1996) that exist in a hierarchical network (e.g., Carver &

Scheier, 1998; DeShone & Gillespie, 2005, Powers, 1973). The level of a goal in the hierarchy or its degree of complexity or connectedness to other goals higher in the hierarchy reflects its importance or perceived value such that higher level or more connected goals are more fundamental and valued to a greater extent than lower level or less connected goals (Carver & Scheier, 1998). Thus, lower level goals accrue value through their association with higher level goals and/or through association with multiple goals (Carver & Scheier, 1998). It is through these associations or

linkages among goals that behaviors are directed and maintained.

Action hierarchies are postulated to be regulated by needs, emotions, and personality. Of prime importance to this study is the need for belongingness

(Baumeister & Leary, 1995). This need is important in educational contexts because educational achievement is an inherently social process whereby students construct knowledge of the world and of themselves. Students learn through negotiation and

collaboration with others (i.e., teachers, other students, and friends) and it is through

this constant dialectic that transformations of knowledge of the world and of identity

occur. The need for belongingness is assumed to be the driver for this very important

aspect of human existence. Therefore, when affiliation goals in educational

achievement contexts are congruent with academic achievement goals, an important

convergence takes place such that self-regulated choices reflect those of the

educational contexts (Urdan & Maehr, 1995). However, thwarted affiliation goals

manifested as concerns over the potential for social exclusion can be very disruptive

(Baumeister & DeWall, 2005; Baumeister, et al., 2005; Baumeister & Tice, 1990) and

even potentially “painful” (Eisenberger, et al., 2003; McDonald & Leary, 2005).

Therefore, thwarted belongingness needs through failure of the achievement of

affiliation goals in one context will be satiated in other contexts. This may be

problematic if the other contexts are antithetical to educational achievement (Dishion,

et al., 1996; Demerath, 200: 2003; Finn, 1989; Fries, et al, 2007; Fordham & Ogbu,

1986; Hofer, et al., 2007; Hymel, et al., 1996; Juvonen, 2006; Mounts & Steinberg,

1995; Osterman, 2000). Because of the place of social affiliation goals in the

hierarchy and its hypothesized relationship to self and social identity, the programs

and scripts of the lower level goals students attempt to achieve reveal the importance of these goals (Vallacher & Wegner, 1986) and through this what is being self- regulated. Academic achievement and underachievement are the outcomes of the self-regulated choices students make that determine which goals they pursue.

The central issue then is the congruence and conflict among values and goals.

Assuming that certain high-level needs are psychological universals, the

determination of the degree of congruence and conflict in a given context rests with

the choices of certain lower-level goals people feel will help them satisfy one or more

of these needs. These choices are influenced by the subjective importance or values

of the goals. Values are trans-situational beliefs about what is desirable and guide the choices people make. The choice of which goal or goals to attempt is influenced by how useful, interesting, self-confirmational, or costly the goals are (Eccles, 2005;

Eccles(Parsons) et al., 1983; Wigfield & Eccles, 1992).

Harackiewicz et al. (2008) clearly show how values differentially imply different achievement goals. The implication from this work for the present study is that social and non-social values may also be differentially related to mastery and performance goals in academic contexts. Specifically, to the extent that mastery

goals are standards of developing competence and performance goals are standards of

demonstrating competence, it might be expected that academic mastery goals would

have a strong positive relationship with the academic values but no significant

relationship with valuing high grades or values reflecting a desire for social inclusion.

However, academic performance goals are expected to have a strong positive

relationship with values reflecting a desire for social inclusion and getting high grades

but no relationship with academic task values.

DeWall et al. (2008) showed that participants who were led to believe that

they should expect a future of poor social relationships subsequently performed better

on a variety of tasks (e.g. hand-eye coordination, dichotic listening, cold pressor,

persistence on a unsolvable anagram task) when these tasks were framed as

diagnostic of traits desirable for good social relationships but not when framed in

non-diagnostic ways. The experimental research of DeWall, et al., (2008) clearly suggests that since people performed better on non-social tasks framed as diagnostic

of traits desirable for good social relationships when the need for belongingness is

active, academic tasks that have perceived social relevance should have higher

perceived value and thus performance on them should be better in this situation.

Values associated with high level affiliation goals activated by a need for

belongingness, assessed as concerns over the potential for social exclusion at the beginning of an academic term, should have positive indirect effects on academic performance at the end of the term through performance goals and school belongingness to the extent that 1) performance goals involving academic tasks are diagnostic of traits that might be desirable for good social relationships and 2) feelings of school belongingness provide a beneficial outlet for satisfying the need to belong in academic contexts. However, mastery goals are not hypothesized to serve social values associated with the need for belongingness directly, and thus concerns over social exclusion should not have a significant indirect effect on academic performance at the end of the term through mastery goals.

Thus, given the activation of the need to belong, when goal congruence or synergy (Ford, 1992) between academic and social goals is achieved, for example when these goals are achieved as means to the same end, there are many educational benefits (e.g., Urdan & Maehr, 1995; Wentzel, 2005). Problems arise when these goals conflict because, to the extent that values guide the choices people make and contribute to what behaviors are self-regulated, maladaptive self-regulatory behavior become more likely. According to action identification theory (Vallacher & Wegner,

1987) this may be particularly true for lower level achievement and social goals since

it is more likely for people to be distracted and adopt different (and potentially

maladaptive) higher levels of identification. If the need for belongingness is an

important modulating influence on the action hierarchy, social goals may trump

academic goals when this need is active and academic goals are not perceived as

diagnostic of satiating this need.

Thus, behaviors for non-social tasks that are important and have a requirement

for completion, but are nevertheless unpleasant in some way (e.g., boring, arduous,

tedious, etc.), may not be self-regulated in a way that is productive if belongingness

needs are active and the tasks are not perceived as socially relevant or meaningful.

All self-regulation involves a choice (Tuckman, 1990). It is obvious therefore that given limitations of executive processing (Baumeister, 2002; Baumeister, et al., 1998;

Baumeister & Heatherington, 1996; Baumeister, et al., 2000; Muraven & Baumeister,

2000), a choice to self-regulate in one domain implies less willingness to self-regulate in another domain (Ainslie, 2001; Steel & König, 2006). Procrastination, or a lack of self-regulation characterized as the avoidance of activities that are under one’s control

(Tuckman, 1991) is one very common maladaptive choice students can make.

Procrastination as a trait (Schouwenburg, 2004) or behavior episode schema (Ford,

1992) contributes in the modulation of the goal action hierarchy by helping determine

what goal linkages are activated. To the extent that goals are aversive or subjectively

irrelevant, their completion is delayed until their perceived utility surpasses the utility

of all other competing goals (Steel, 2007; Steel & König, 2006).

There is considerable evidence to support the notion that an active need to

belong, manifested as a concern over the potential for social exclusion, disrupts self-

regulation (Baumeister, et al., 2002; Baumeister, & DeWall, 2005; Baumeister, et al.,

2005) and further evidence suggesting a link specifically between the characteristics and effects of perceived social exclusion and procrastination (Bembenutty &

Karbenick, 2004; Senecal, et al., 2003: Twenge, et al., 2003). Thus, it is expected that concerns over social exclusion or, equivalently, the desire to be socially included should be positively related to procrastination in the academic contexts first quarter freshmen are exposed to. And to the extent that academic task values or grade values

(i.e. valuing receiving high grades) can act as buffers against choices to procrastinate,

it is through procrastination that social and academic value incongruence may be seen. Thus, it is expected that academic task and grade values will be negatively related to procrastination. To the extent that the desire to be included socially occurs simultaneously with the ostensible non-social desires for academic challenge and

good grades, the relationships of these values and procrastination represent an

operationalization of goal conflict.

The choices students make should be partly reflected in their degree of procrastination. In turn, the tendency to procrastinate is expected to have deleterious effects on important processes measured at the end of the term. Specifically, it is expected that procrastination measured at the beginning of students’ first academic term will be negatively related to achievement goals, self-efficacy and school belongingness, and positively related to perceived stress measured at the end of the term. First, to the extent that academic exercises are perceived as aversive, students

are more likely to choose more desirable short-term options in favor of achievement goals (Blunt & Pychyl, 2000; Lasane & Jones, 2000; Moon & Illingworth, 2005;

Pychyl, et al., 2000; Schouwenburg, 2004; Schouwenburg & Lay, 1995;

Schouwenburg & Groenewoud, 2001). Thus, a negative relationship is expected between procrastination and mastery and performance goals. Second, there is a considerable amount of evidence that shows that procrastination and self-efficacy are negatively related (Haycock, McCarthy, and Skay, 1998; Klassen, Krawchuk, &

Rajani, 2008; Sirois, 2004; Steel, 2007; Tuckman, 1991; van Eerde, 2003; Wolters,

2003b). However, most research has only looked at the consequences of self-efficacy on procrastination (for an exception see Sirois, 2004). Given the enactive nature of self-efficacy (Bandura, 1997; Sexton & Tuckman, 1991; Sexton, Tuckman, &

Crehan, 1992; Tuckman & Sexton, 1990) it is expected that self-efficacy for college work is something that develops and in the process can be adversely affected by procrastination. Third, it is expected that in addition to being directly and positively affected by concerns over social exclusion and academic and grade values, perceived school belongingness should be adversely affected by procrastination.

Procrastination of academic (school-related) activities reflects interests in diversions and distractions that can disrupt interactions with other students and faculty. To this extent procrastination can disrupt students’ feelings of being “in the school” as well as feelings of the “school in the student”. L. H. Anderman (2003) found that academic task values significantly predicted school belongingness. And while a direct positive relationship between academic values and school belongingness is predicted here as well, given the suggested negative relationship

between academic values and procrastination, it follows that academic values may

have their effects on school belongingness partly through decreasing the tendency to

procrastinate. Finally, there is considerable evidence that shows that procrastination

is positively related to perceived stress (Flett, et al., 1995; Lay, et al., 1989;

Rothblum, et al., 1986; Schraw, et al., 2007; Tice & Baumeister, 1997; Solomon &

Rothblum, 1984). And even in one qualitative study that showed that students who

felt that procrastination was “adaptive” nevertheless reported significant stress and negative health effects (Schraw, et al., 2007). Thus, it is expected that procrastination will have a deleterious effect assessed as a positive relationship with perceived stress.

Procrastination is thus viewed as a reflection of choice that can affect achievement goals, self-efficacy, school belongingness, and perceived stress.

However, the effect of procrastination on academic performance is not hypothesized

to be direct but rather mediated by the motivational, affective, and social factors just

discussed. And given some recent literature on the relationship of these mediating

factors, it is expected that a particular pattern of relationships will exist among achievement goals, self-efficacy, school belongingness, and perceived stress and that these factors will have direct effects on first quarter grade point average three weeks after their assessment.

First, mastery goals have generally been found to be positively related to self- efficacy (e.g., Breland & Donovan, 2005; Gerhardt & Brown, 2006; Middleton &

Midgley, 1997; Pintrich, 2000b; Wolters et al., 1996). However, some studies have found a positive relationship with performance goals (e.g., Skaalvik, 1997

[performance-approach]; Gerhardt & Brown, 2006 [performance] Wolters et al., 1996

[performance]), others have found a negative relationship (e.g., Middleton &

Midgley, 1997 [performance-avoidance]; Phillips & Gully (1997) [performance];

Skaalvik, 1997 [performance-avoidance]), or no relationship (Gerhardt & Brown,

2006 [performance]; Middleton & Midgley, 1997 [performance-approach]). Given these results, it is hypothesized that mastery goals should have a positive influence on self-efficacy but the specification of a hypothesis with respect to performance goals is somewhat more tenuous. Therefore, with respect to the relationship between performance goals and self-efficacy no specific hypothesis is offered. However, an estimate of this relationship will be assessed in an attempt ascertain its strength and direction in the context of the larger model.

Second, there is very little research on the relationship between school belongingness and motivational variables such as achievement goals and self- efficacy. However, L. Anderman & E. Anderman (1999) found a positive relationship between mastery goals and school belonging in sixth graders but no relationship between performance goals and school belonging (L. Anderman & E. Anderman,

1999). Freeman and L. Anderman (2002, cited in L. Anderman & Freeman, 2004) showed that school belongingness predicted intrinsic goal orientation, task value, and self-efficacy in college freshman, but L. Anderman and Hughes (2003, cited in L.

Anderman & Freeman, 2004) reported that school belongingness did not significantly predict any goal orientation outcome in sixth and seventh graders. However, as mentioned above, Anderman and Hughes did find a disturbing interaction that suggested that students reporting low levels of school belongingness were less likely to adopt performance-avoidance goal orientations. Goodenow (1993a) and

Goodenow and Grady (1993) reported significant positive relationships between

school belonging and students expectancy for success in junior high school students

and Roeser, et al. (1996) found that school belongingness predicted academic self-

efficacy in eighth graders. Thus, while school belonging is hypothesized to positively

predict both self-efficacy and mastery goal orientation, it is not clear whether a

relationship between a performance orientation and school belongingness exists and,

if so, what type of relationship it is. Speculatively however, given that a positive relationship is predicted between concerns over social exclusion and performance

goal orientation, it is hypothesized that a positive relationship between school belongingness and performance goal orientation will also exist. That is, to the extent

that performance goals provide information regarding one’s position in the social

milieu of college based relationships, students reporting feelings of belongingness

might be more likely to endorse performance goals.

Third, given the beneficial effects of self-efficacy (Bandura, 1997; Chemers,

et al., 2001) and school belongingness (see L. Anderman & Freeman, 2004) on

helping to reduce perceived stress, it is expected that both self-efficacy and school

belongingness will negatively predict perceived stress reported just prior to final

examinations.

Finally, the wealth of information regarding the effects of self-efficacy on

performance clearly suggests that it will positively predict grade point average. Also,

in line with the general timbre of research on academic goal orientations, it is not

expected that mastery goals will be directly related to academic performance, but a performance (approach) goal orientation should positively predict grade point

average. In addition, consistent with a resource depletion model and the results of

Oaten and Cheng (2005), it is expected that perceived stress will negatively predict academic performance. Finally, it is not clear what relationship to predict between school belongingness and academic performance. While school belongingness clearly has a mediating role in the present model, there is some question as to whether it has direct effects on performance. While at least one study showed that school belonging positively predicted grade point average (E. Anderman, 2002), another has shown that the effect may be confounded by student poverty level (Battistich et al., 1995). Thus, at present no hypothesis is offered regarding the predictive capability of school belongingness on grade point average, but the relationship will be assessed in an exploratory fashion.

This study is an attempt to assess the usefulness of a proposal specifying how procrastination, resulting from choices based on academic and social values measured early in college students’ first quarter, affects subsequent academic goal choice, self- efficacy, school belongingness, and stress and how, in turn, these latter factors affect academic performance at the end of the first term. Given the predictive utility of the first quarter grades for long-term academic performance and graduation relative to standard measures of ability such as high school class rank and standardized test scores, it seems worth assessing the potential psychological factors that may be important in determining performance during students’ first term.

Academic and social values can work in synchrony or compete. These values determine the choices students make and the goals they pursue. Thus values are conceptualized in this study to be exogenous variables directly or indirectly

undergirding decisions regarding self-regulation and achievement goals. The

discussion above implies the following hypotheses.

Hypotheses

Table 1.1 provides a summary of all the major hypotheses of this study. The

criteria variables are presented as columns and each row is an exogenous or

endogenous predictor. Plus and minus signs indicate that a significant positive or negative relationship is expected respectively whereas zeros indicate that a non-

significant relationship is expected. Question marks indicate that the relationship is tested, but no specific hypothesis is presented. Finally ‘N/A’ indicates that the relationship is not of interest, or not possible.

Hypothesis 1 – Differential Relationships between Values and Procrastination and

School Belongingness

The discussion above clearly suggests that while concerns over social

exclusion should positively predict both procrastination and school belongingness,

academic and grade values should negatively predict procrastination and positively

predict school belongingness. Specifically, concerns over social exclusion should

disrupt self-regulation whereas academic and grade values should direct students to

make choices favoring more disciplined, self-regulatory behavior and thus less

procrastination. Also, given the positive relationships found in previous research

between school belonging and academic task values (Anderman, 2003; Anderman &

Freeman, 2004), a positive relationship is expected between these constructs. In

addition, Anderman (2003) also showed that grade point average positively predicted

school belongingness so it seems reasonable to assume that a positive relationship

between grade values and school belongingness will also be seen. Finally, concerns

over social exclusion should motivate students’ attempts to achieve a sense of

belonging in school as they transition into their new environment. As such, a positive

relationship between a desire for social inclusion and school belongingness is

expected.

Hypothesis 2 – Differential Relationships between Values and Achievement Goals:

Table 1.1 shows how achievement goal outcomes may be differentiated by

social, academic, and grade values. Both performance-approach and performance-

avoidance goals are predicted to be positively related to concerns over social exclusion whereas mastery-approach goals are not expected to be related to this social value. And whereas performance-avoidance is expected to be negatively related to academic and grade values, mastery- and performance approach outcomes may be differentiated by these values. Specifically, academic values are expected to be positively related to mastery-approach, but not performance-approach, whereas grade

values are expected to be positively related to performance-approach, but not

mastery-approach goals. Furthermore, no significant relationship between

performance-approach goals and academic values or between mastery-approach goals

and grade values are expected. No specific hypotheses are presented for the

relationship between mastery-avoidance and values.

Hypothesis 3 – Debilitating Consequences of Procrastination:

Procrastination is viewed as a choice, the outcome of which, at least in the

long run, is usually debilitating. Procrastination is hypothesized to be one mediating

variable between values and motivational, affective, social variables. Specifically, it is expected that procrastination measured at the beginning of the term will negatively predict performance-approach and mastery-approach achievement goals, as well as self-efficacy and school belongingness. However it is expected to positively predict perceived performance-avoidance goals and stress near the end of the term. No prediction is made with respect to the relationship between procrastination and mastery-avoidance goals.

Endogenous Criteria Variables* Predictor Variables* PROC SB PAP PAV MAP MAV SE PS GPA CSE + + + + 0 ? 0 0 0 AV - + 0 - + ? 0 0 0 Exogenous GV - + + - 0 ? 0 0 0 Variables ACT N/A N/A N/A N/A N/A N/A N/A N/A + HSCR N/A N/A N/A N/A N/A N/A N/A N/A + PROC N/A - - + - ? - + 0 SB N/A N/A + + + ? + - ? PAP N/A N/A N/A N/A N/A N/A + 0 + Endogenous PAV N/A N/A N/A N/A N/A N/A - 0 - Variables MAP N/A N/A N/A N/A N/A N/A + 0 0 MAV N/A N/A N/A N/A N/A N/A ? 0 ? SE N/A N/A N/A N/A N/A N/A N/A - + PS N/A N/A N/A N/A N/A N/A N/A N/A - †0 = No relationship predicted; ‘+’ = Significant positive relationship predicted; ‘-‘ = Significant negative relationship predicted; ‘?’ = Important but unknown relationship; ‘N/A’ = Not Applicable for this study

*CSE = Social Exclusion Concerns; AV = Academic Values; GV = Grade Values; ACT = Standardized Test Score; HSCR = High School Class Rank; PROC = Procrastination/Lack of Self- regulation; SB = School Belongingness; PAP = Performance-approach; PAV = Performance- avoidance; MAP = Mastery-approach; MAV = Mastery-avoidance; SE = Self-regulatory Self- Efficacy; PS = Perceived Stress; GPA = First Quarter Grade Point Average Table 1.1: Hypothesized Directions of the Structural Relationships†

Hypothesis 4 – Benefits of Perceived School Belongingness:

During the eighth week of the term, school belongingness is expected to be positively associated with performance-approach, performance-avoidance, and mastery-approach achievement goals, and self-efficacy. No specific relationship is predicted for grade point average, but the relationship will be assessed. School belongingness is also expected to be associated with lower levels of perceived stress.

Hypothesis 5 – Differential Relationships between Achievement Goals and Self

Efficacy and GPA:

Mastery-approach achievement goals are expected to positively predict self-

efficacy but are not expected to predict end of term grade point average.

Performance-approach goals are expected to positively predict both self-efficacy and

grade point average and performance-avoidance are expected to negatively predict

self-efficacy and grade point average. It is not clear what relationship to predict

between mastery-avoidance and self-efficacy and grade point average. However, this

path will be estimated to explore potential relationships.

Hypothesis 6 – Benefits of Perceived Self-Efficacy:

Self-efficacy is expected to be associated with lower levels of perceived stress

and positively predict quarter grade point average.

Hypothesis 7 – Debilitating Effects of Perceived Stress on GPA:

Reports of perceived stress experienced over the past month, assessed during

the eighth week of the quarter, are expected to be negatively related to quarter grade

point average.

Hypothesis 8– Predicted Null Relationships:

Because values are expected to guide and not directly determine behavior, it is hypothesized that all three value constructs will have non-significant relationships with self-efficacy, perceived stress, and grade point average. In addition, it is expected that none of the achievement goals will be significantly related to perceived stress.

98 Chapter 2: Methods

Participants

Two-thousand and forty-four first quarter freshmen (mean age at the beginning of

the study = 18.2 [SD = 0.70]; 48.3% women; 14.8% minority; 77.2% White-Non

Hispanic; 8.0% Undesignated Ethnicity) enrolled in a freshmen survey course at a large mid-western university during their first quarter of enrollment (autumn 2008) were requested to participate in the study. At the beginning of the study 50.0% of students had declared a major in the College of Business. The remaining students had undeclared majors.

There were two rounds of questionnaires (the first during the second week of the quarter and the second during the eighth week of the quarter) and participation in the study was defined as the completion of both questionnaires and student consent to the

access of high school class rank, standardized test scores, and demographic variables of

sex, and ethnicity available from university records. Of the 2,044 students initially requested to participate, 923 completed the first round questionnaire and 672 completed both the first and second round questionnaires. No student completed only the second round questionnaire. Only students completing both rounds were included in the analysis of the study’s full model and core hypotheses. Students completing only the first round questionnaire and not the second round questionnaire were considered to have dropped

99 out of the study. One student completing only the first round questionnaire explicitly requested removal from the study. In addition, one student completing both questionnaires requested removal late in the quarter. This student’s data is not included in the analyses that follow. Thus, the data from 671 students was used to test the core hypotheses of the study. The mean age of this sample at the beginning of the study was

18.2 years (SD = 0.75) and was composed of 60.4% women, 10.3% minority, 78.2%

White-Non Hispanic, and 11.5% Undesignated Ethnicity. At the beginning of the study

61.4% of the sample had declared a major in the College of Business. The remaining

students had undeclared majors.

Procedure

Data Collection

Students were recruited during a meeting of their freshmen survey course in the first

week of autumn quarter 2008 and were given bonus credit for the course for participating

in the study. Students were given other bonus credit options if they chose not to

participate in the study. All students were introduced to the study by their survey course

instructor who handed out a cover letter of the study (Appendix A). The instructor’s role

was to inform students that participation in the study was just one of a number of extra

credit bonus options for the course. In addition, instructors informed students that they

needed to complete two web-based questionnaires available through the university’s

computerized course management system (CARMEN) in order to receive the bonus

credit for participating in the study. The cover letter students received from the instructor

described the details of the study and students were referred to the experimenter if they

100 had any comments, concerns, or questions. After students had received the cover letter

from their instructors, an e-mail notification from the experimenter (Appendix B) was sent to all students at the end of the first week of the quarter and again at the end of the seventh week of the quarter to remind them of the study and bonus credit, to give details

regarding the instructions for completing the online questionnaire, and to notify students

who had possibly missed the class in which the cover letter was handed out. Just prior to

completing both web-based questionnaires, all students were required to read and

electronically sign a release form (see Appendix C) permitting the experimenter access to

their grades and demographic information available from university records. If students

did not electronically sign the release form they were prevented from completing the

questionnaires. Upon agreeing with the conditions of the study and signing the form, the

questionnaire was made immediately available. Finally, all students who completed a

questionnaire were sent an e-mail acknowledgment (Appendix D) thanking them for their

participation, reminding them that they had given their consent authorizing the

experimenter access to certain demographic information, and that they were free to leave

the study at any time without penalty.

Analyses

The hypotheses above imply a nomological network that is best assessed using a

path analytic procedure utilizing the methodology of structural equation modeling

(SEM). SEM allows for an assessment of the fit between the covariance matrix generated from the simultaneous equations representing the hypothesized structure and relations among constructs and the covariance matrix that exists in the data sampled. In addition,

101 it permits the analysis of a measurement model that explicitly takes into account the

measurement error of indicators in measuring a construct in addition to the structural

model that represents hypotheses of the relationships among the constructs (Schumacher

& Lomax, 2004). Given the complexity of the hypotheses of the present study, SEM

overcomes the limits of having to use only a small subset of variables, allowing for the

multidimensional and complex nature of the phenomenon to be taken into account

(Schumacher & Lomax, 2004).

Except where otherwise noted, Mplus 5.2 statistical software (Muthén & Muthén,

2007) was used in all exploratory and confirmatory factor analyses and the analysis of the

structural models. Unreduced correlation matrices were analyzed in the exploratory

factor analyses (EFA) and covariance matrices were analyzed in the confirmatory

measurement (CFA) and structural models (SEM). In addition, since the individual item

variables and parcel indicator variables were nonnormal (see Appendices E and F

respectively) and some variables had missing values (a maximum of 3.1% missing for

high school class rank), the MLR estimator and a sandwich estimator for estimating

standard errors available in Mplus 5.2 was used for all EFA, CFA, and SEM models. The

MLR estimator is asymptotically equivalent to the rescaled T2* statistic of Yuan and

Bentler (2000) (Muthén & Muthén, 2007). Yuan and Bentler (2000) recommend using this test statistic and a sandwich-type estimator for standard errors for medium size data sets with missing data that are not normally distributed. The details of these analyses and other ancillary analyses are described in more detail in the appropriate sections below.

Measurement of Constructs

102 Description of Instruments

The items analyzed and the parcels to which they were assigned are presented in

Appendices E and F respectively.

Concerns: Social Exclusion (CSE): This is a six item scale developed for this study intended as a measure of concern over social exclusion and desire for belongingness. The six items that make up this scale are part of a larger 18-item instrument. There is no published reliability or validity data on this instrument.

However, the internal consistency reliability of these six items for the 671 students was

0.82.

Academic Values (AV): This construct is based on an adaptation of the 12 items of

the Motivated Strategies for Learning Questionnaire (MSLQ) Task Value scale which is

intended to operationally define “judgments of how interesting, useful, and important the

course content is to the student” (Pintrich, Smith, Garcia, and McKeachie, 1993, p. 802).

The adaptation for the present study involved asking students about the challenge,

interest, importance, and utility of all their classes. It has a published internal consistency

of 0.90 (Pintrich, et al., 1993). The internal consistency for the current sample was 0.86.

Grade Values (GV): This construct is measured using two items adapted from

Zimmerman, Bandura, & Martinez-Pons (1992) that asks students what grade point

average would be minimally satisfying for the current quarter and for the current

academic year. The internal consistency of these two items was 0.92 in this study.

Because these items request information specifically about a grade point average, they

represent the higher level goal of demonstrating success in college defined specifically in

terms in terms of grades.

103 Procrastination/Lack of Self-Regulation (PROC)): Two separately developed

scales were used to measure this construct. The Self Control Scale (SCS) was developed

by Tangney, Baumeister & Boone (2004) to measure aspects of self-control such as

restraint or impulse control, self-discipline, and distractibility. The current research uses the 13-item Brief Form of this instrument. The authors reported a test-retest reliability

coefficient of 0.89 and internal consistency reliabilities ranging from 0.83 to 0.85 for the

13-item version. The second used to measure procrastination is The Tuckman

Procrastination Scale (TPS) (Tuckman, 1991). This is a 16-item scale designed to detect the tendency to procrastinate in completing college assignments. The author reported an internal consistency coefficient of 0.86. This scale is a widely used measure of procrastination and was designed specifically for college student populations.

Prior pilot research suggested that the separate latent constructs of self-control based on the SCS and procrastination based on the TPS were highly correlated (r =

-0.741). As such, the items from both instruments were combined in a factor analysis in order to extract dimensions as a preliminary step in creating parcel indicators for this construct. (More detail regarding the construction of parcel indicators is discussed in more detail below.) Preliminary factor analysis of these items resulted in the combination of the 12 of the original 13 items from the SCS (all items reverse scored) and 14 of the original 16 items from the TPS. One item from SCS and two items TPS were removed from further analyses because their similar wordings resulted in dimensions including only those items. The internal consistency of the 26 items was 0.93.

This construct represents procrastination as a lack of self-regulation associated with weak impulse control, distractibility, and lack of work discipline.

104 School Belongingness (SB): This construct is based on an adaptation of

Goodenow’s (1993b) The Psychological Sense of School Membership (PSSM) Scale.

This is an 18-item instrument that measures what Goodenow refers to as “the extent to

which students feel personally accepted, respected, included, and supported by others in

the school social environment”. (p. 80) Internal reliabilities reported by the author ranged

from 0.82 to 0.88 for fifth through eight graders. The internal reliability for the present

study of college freshman was 0.89.

Achievement Goal Orientation: This is an adapted version of a 12-item scale

developed by Elliot and McGregor (2001) that is intended to measure the dimensions of

approach and avoidance of mastery and performance goal orientations. The four

subscales in this instrument, intended to measure the 2 x 2 dimension x goal structure,

have reported reliabilities ranging from 0.83 to 0.92.

The adaptation of this instrument for this study involved rewording the items so

that students would attend to a given achievement goal specifically in the context of their

first quarter at the university. Thus all twelve items began with the phrase “In setting my academic goals for the end of this quarter, I am focused on . . .” This phrase was then followed by an achievement goal (see Appendix E for all twelve items).

Self-Regulatory Self-Efficacy (SE): The current research adapts 13 of the 19 items of the Self-Efficacy of Learning Form (SELF-A) (Zimmerman & Kitsantas, 2007). This scale assesses beliefs regarding their use of self-efficacy regulatory processes in academic settings. The authors report an internal consistency of 0.97. However, preliminary factor analyses of the full 19-item version showed that six of the items loading together as a separate dimension had an unacceptable internal consistency (alpha

105 = 0.687) and factor determinacy (r = 0.873) and were thus removed from further analyses. The internal reliability for the 14 adapted items in the present sample was 0.89.

Perceived Stress (PS): The Perceived Stress Scale developed by Cohen, Kamarck,

and Mermelstein (1983) is used in this study to measure the construct of perceived stress defined as “the degree to which situations in one’s life are appraised as stressful” (Cohen et al., 1983, p. 394). The internal consistency reported in the original report averaged 0.85 over three samples. The 14-item version use in this study had an internal reliability of

0.88.

High School Class Rank (HSCR) and Standardized Test Scores (ACT): These

scores, pulled from the university’s central records database, were used as measures of

ability. HSCR is measured as a percentile. Standardized test scores were the highest

ACT composite or SAT verbal/math composite a student supplied to the university. SAT

composite scores were converted to ACT scores using a standard concordance table and

thus all standardized test scores are in ACT units with a range of 10 to 36.

Academic Achievement. Achievement was measured as the first quarter grade

point average (GPA) at the end of the autumn quarter measured on a scale from 0.00 to

4.00.

All questionnaire items were scored on an 11-point scale. The two GV items,

which reflect grade point averages, were anchored at 0.0 at the low end and 4.0 at the

high end. The first value above 0.0 was 1.0 and the remaining 8 values increased from 1.0

to 4.0 in 0.3 point units. The other items were anchored at 0 as “That Is Not Me For Sure”

and at 10 as “That Is Me for Sure”. The items were presented in random order in an

106 attempt to attenuate method variance due to item similarity. HSCR, ACT, and GPA were scored in their own units of 0 to 100, 10 to 36 and 0.00 to 4.00 respectively.

The Construction of Parcel Indicators: Rationale

Given the large number of variables used in this study (103 items plus two ability measures and one achievement measure), the first concern in constructing a measurement model is the very large correlation matrix that would result and the number of parameters that need to be estimated. The problem with a correlation matrix of this size is that, despite the fact that more indicators may provide for a better represented construct, a model is less likely to fit as the number of items in a correlation increases even for a model that provides a good description of a given phenomenon (Coffman & MacCallum,

2005). One reason for this is that the likelihood of finding spurious correlations increases as the number of items increases putting the researcher in the bind of either estimating them, leading to potentially incorrect conclusions, or not estimating them, leading to the rejection of a model that otherwise may represent the phenomenon under consideration

(Little, Cunnignham, Shahar, & Widaman, 2002; Marsh, Hau, Balla, and Grayson, 1998).

Therefore, it is desirable to seek some balance between the number of indicators per construct and the reduction in the order of the correlation matrix (Little, Lindenberger, &

Nesselroade, 1999).

One technique that can reduce the number of indicators while utilizing the information from all of the items is to create parcels (Cattell & Burdsal, 1975; Coffman

& MacCallum, 2005; Kishton & Widaman, 1994; Little, et al., 2002; MacCallum, &

Austin, 2000; Schumacker & Lomax, 2004). A parcel is an aggregated indictor created by

107 summing or averaging two or more items (Little, et al., 2002). Marsh, et al. (1998) found

in a simulation study that increasing the number of indicators per factor from 2 through

12 increased the likelihood of convergence to proper solutions, but model fit declined

even when the model was correctly specified. Also, even though the 12-item solutions

were more likely to converge than parceled solutions, parceling resulted in smaller parameter estimates for unique variances and less variability of the factor loadings when the number of items in the parceled solutions increased providing more precise estimates.

Furthermore, the increase in parsimony has the additional advantage of reducing the likelihood of correlated residual error and dual factor loadings (Little, et al., 2002). Thus, apart from merely reducing the number of parameters that need to be estimated and reducing the order of the correlation matrix, parceling may be advocated based on other psychometric properties. Specifically, parcels tend to have higher reliability and communality, can deviate less from a normal distribution as the number of items in a parcel increase, and are more efficient than individual items because they tend be closer to the true construct centroid (Little, et al., 1999; Little, et al., 2002). In addition, models using parcel indicators increases the degrees of freedom of the model (Coffman &

MacCallum, 2005) and thereby increases the power of the tests assessing model fit

(MacCallum, Browne, & Sugawara, 1996).

The practice of parceling is not without its critics however. The primary objection to the use of parceling is that it can mask a multidimensional factor structure (Bandalos,

2002: Bandalos & Finney, 2001; Hall, Snell, & Singer Foust, 1999). In addition to

possibly biasing parameter estimates (Hall et al., 1999), this problem represents the other

side of the issue in that parceling may make it more likely to not reject a misspecified

108 model. Multidimensionality occurs when two or more latent constructs influence one item and can thus be the source of model misspecification. Bandalos (2002) and Hall et al. (1999) showed that parceling can result in unspecified secondary and unmodeled factors being spread across the parcel indicators and their effects getting absorbed into the model parameters. Specifically, misspecified models including shared secondary influences would be typically represented as correlated residual errors. However, if the secondary influences on items are included along with the modeled influences, the correlated uniquenesses are absorbed into the shared common variance thus becoming a part of the modeled variance. These unmodeled factors then become a part of the latent construct confounding the interpretation of the constructs of interest. Therefore, Bandalos

(2002; Bandalos & Finney, 2001) has recommended that parceling be done only on unidimensional constructs.

Little et al. (2002) maintain, however, that these arguments are valid only when analyses are done at the item level. That is, if the goal of the research is to specifically represent dimensionality at the item level, then parceling should not be done. However, they contend that another approach to latent variable modeling is to represent relationships among latent variables. According to these authors, since parcels remove hidden sources of error (resulting in parcel indicators deviating less from the construct centroid) a “model fitted to parcel-level data would . . . no longer be misspecified.” (p.

164) Specifically, they claim that

item indicators are merely tools that allow one to build a measurement model for a desired latent construct. Once built, the item indicators become less consequential. With such an approach, if a dual loading were eliminated through aggregating items in order to specify a clean latent construct, then the goals of the researcher are realized through parceling, not hindered by it. The dual loading is

109 unimportant and can be effectively minimized during an initial construct-building phase. The same logic applies to correlated residuals. If items are only building blocks, estimating the additional shared relationships is unimportant to the theory building in latent space. Eliminating the residual through parceling is as effective as explicitly correlating the residual. (p. 169)

An argument for parceling multidimensional constructs can also be made on intuitive

grounds. For example, in an attempt to determine students’ knowledge of statistics, one

does not construct a test consisting of only one type of problem. A well-constructed

statistics test might include, for example, short word problems and possibly short proofs

to be solved in class as closed-book exercises as well as longer take-home portion where

real-world data problems are worked out. If a test included only a take-home exercise,

for example, the test would be unidimensional but also very narrow and not entirely

representative of students’ knowledge of statistics. If “statistics knowledge” is

considered a latent construct, then a more complete representation is achieved by

assessing many dimensions of the construct, not just one1. Likewise, when the concern is with the relationship among latent variables, it would seem that multidimensionality of the constructs, if it exists, needs to be taken into account. Thus, while the multidimensionality must be thoroughly examined prior to parceling (Little, et al., 2002), it is not necessarily an undesirable feature at the latent construct level of analysis.

One particularly useful method for creating parcels is referred to as the domain- representative approach (Kishton & Widamen, 1994). This method explicitly takes multidimsionality into account by creating parcels from items from all of the dimensions or facets of a construct (Little, et al., 2002) such that all of the facets of a construct are represented in each parcel. In this approach items from each dimension are summed or

averaged to create the parcel. An alternative approach is to create homogenous or

110 internally consistent parcels such that only the items of a dimension are summed or

averaged to create a parcel. Thus, a homogeneous parcel contains only the items within a

given dimension. Empirically, the domain-representative approach has produced results superior to those of the homogeneous approach in providing better fitting models

(Coffman & MacCallum, 2005; Kishton & Widamen, 1994). This is most likely due to the fact that domain-representative parcels better represent a construct because they have more variability in common whereas homogeneous parcels will more likely have higher specific variances (Coffman & MacCallum, 2005). Thus, the domain-representative approach was used to create parcel indicators in this study for all multidimensional latent constructs. For unidimensional constructs, parcel indicators were created by randomly assigning items to parcels. The achievement goals and GV constructs were measured by the individual items since there were only three item indicators for each of the achievement goal constructs and two item indicators for the GV constructs.

The Construction of Parcel Indicators: Procedure

Based on the recommendations of Little et al. (2002), the dimensions of the five

latent constructs (i.e., AV, PROC, SB, SE, and PS) were first determined separately for

each construct. For each instrument representing a construct, a series of oblique rotation

(Geomin) maximum likelihood exploratory factor analyses were carried out on the items.

The dimensions are determined using a number of criteria to be described below involving examining the eigenvalues, various fit indices, factor determinancies, strength

of relationship of items with a dimension, and interpretability of dimensions. After the

dimensions of each construct were determined, items from these dimensions were

111 combined through averaging the item scores using the domain-representative approach to

create the parcel indicators.

The first problem that needs to be dealt with in constructing parcels is identifying the correct number of dimensions of a construct. There is consistent and widespread agreement that no one method is sufficient for determining the number of factors or dimensions underlying a correlation matrix (e.g., Nunnally & Bernstein, 1994; Fabrigar,

Wegener, MacCallum, & Strahan, 1999; Rummel, 1970; Zwick & Velicer, 1986). This is problematic because of the importance in determining the correct number of dimensions.

For example, it is generally recommended to avoid underfactoring because of the general distortion of the factor structure and the loss of important information due to ignoring important dimensions (Rummel, 19770; Zwick & Velicer, 1986). However, there is also concern that overfactoring can also lead to minor factors that are not interpretable and are unlikely to replicate (Comrey, 1978; Zwick & Velicer, 1986) as well as being poorly determined (Schönemann & Wang, 1972). In addition, there is recent evidence that overfactoring can be problematic in maximum likelihood estimation in that the likelihood-ratio test statistic can deviate from the presumed chi-square distribution due to problems of rank deficiency of the factor loading matrix (Hayashi, Bentler, & Yuan,

2007). Therefore, a number of criteria were used to determine the number of dimensions indicating a construct. These criteria were used in a convergent manner such that the decision on the number of dimensions of a construct was determined by an emergent property of the results of the implementation of all of the criteria and not by the domination of one criterion over the others. However, the implementation of the criteria was ordered systematically as described below.

112 The first criterion for determining the number of dimensions utilized the method

of parallel analysis (Horn, 1969; Montanelli & Humphreys, 1976; O’Conner, 2000). This

method compares some pre-specified percentile (e.g. 50th or 95th percentile) of a

distribution of eigenvalues of a series of correlation matrices of randomly generated data

(using the same number of subjects and variables as in the real sample data) with the

eigenvalues of the correlation matrix of the real sample data. The crossover point, where

the eigenvalues of the sample correlation matrix equal the eigenvalues of the random

data, indicates where random and insignificant dimensions of the sample data begin

(Nunnally & Bernstein, 1994). This study used O’Conner’s (2000) implementation where the 95th percentile of the distribution of eigenvalues from 1000 randomly generated correlation matrices is compared to the eigenvalues of the real data based on exploratory

factor analyses of the items of the instruments representing each of the five constructs.

The dimensions of a latent construct retained are those with eigenvalues of the correlation

matrix of the real sample data greater than the eigenvalues of the randomly generated

data.

A second criterion determining the number of factors to retain was based on

examining the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root-Mean-

Square Error of Approximation (RMSEA), and Standardized Root-Mean-Square

Residual (SRMR) of different models with varying number of dimensions generated from

exploratory factor analyses using maximum likelihood estimation and Mplus

implementation of oblique Geomin rotation. After the eigenvalues were examined in the

parallel analysis phase, two different exploratory factor analysis models were run for

each instrument representing a latent construct. Each factor analysis extracted a different

113 number of dimensions that varied as a function of the number of dimensions determined in the parallel analysis. Thus, if p dimensions were determined by a parallel analysis of a given instrument, two exploratory factor analyses were run extracting p and p+1 dimensions and the fit indices for each of the three exploratory factor analyses were examined. Indices indicating the best fit for a given number of dimensions, in combination with a high degree of factor determinacy, interpretability, and low factor specificity were used as criteria in selecting the number of dimensions.

Factor determinacy was used as a third criterion. Factor determinacy refers to the uniqueness of the factor scores (Rummel, 1970) and involves the problem in common factor analysis that an infinite number of factor scores can be derived from a given pattern of factor loadings (Grice, 2001). Even though the calculation of factor scores, per se, is not necessary for the selection of the number of factors, factor indeterminacy can create conceptual obstacles in the interpretation of the derived factors. That is, the issue of factor indeterminacy goes beyond merely calculating factor scores to the deeper issue of what factor scores represent conceptually (Schönemann & Wang, 1972). Specifically, factor determinacy reflects the multiple correlation between each factor and the measured variables (Rummel, 1970) and is an indication of the correlation between the estimated factor scores with the true factor scores (Grice, 2001; Rummel, 1970). Furthermore, letting  represent the multiple correlation between each factor and the measured variables, the degree to which competing orthogonal factor scores can be computed for

2 the same factor is rmin = 2 – 1 and represents the smallest correlation of all of the factor

2 score estimates and the true factor scores (Rummel, 1970). If rmin = 2 – 1 = 1, then the estimated factor scores are said to be unique. Thus, high positive values of rmin are 114 necessary if meaningful factor interpretation is to be achieved. Notice that  has to be

very close to 1 in order for the factor score estimates to be unique. In fact, even a

multiple correlation of  = 0.80 (which represents 36 percent indeterminacy), implies that

the smallest correlation between all the factor score estimates and the true factor score is

only 0.28. This means that the maximally different estimates do not correlate highly with

the true factor score (Rummel, 1970) and implies that considerable variability may exist

in the interpretation of the factor extracted and its real meaning. There are no known

standards or rules of thumb as to a minimal cutoff value for factor determinacy, but

clearly any  less than 0.90 implies that the smallest correlation between all of the factor

score estimates and the true factor score is less than 0.62. Thus, while any cutoff value is

arbitrary, any factor with a factor determinacy less than 0.90 was considered to be not

unique. This criterion can help temper the tendency of maximum likelihood estimation to

extract too many factors leaving either trivial or poorly determined factors (Schönemann

& Wang, 1972). Also, while not used as a criterion for dimension extraction, Cronbach’s alpha will also be calculated for the variables of the extracted dimensions. Internal

reliability relates only to the variables with the highest loadings on a dimension and not

to the determinacy of that dimension. Nevertheless, if these variables represent an

internally consistent dimension then Cronbach’s alpha should approach acceptable levels

(e.g., greater than 0.80).

A fourth criterion is parsimony or a lack of highly specific dimensions. Extracted

factors with only one or two variables loading highly and the remaining variables

showing low loadings are considered to be too highly specific. The search at this phase

of the analysis is for common dimensions. Overextraction of dimensions can represent a 115 violation of parsimony if a solution with one less extracted dimension shows an acceptable fit.

The final criterion is interpretability. This may help prevent underextraction of small, yet meaningful dimensions. Rummel (1970) claims that “smaller factors . . . may tap conceptually new or unsuspected influences in a domain” (p. 362). As such, a small meaningful dimension with only two or three items loading heavily on it with an acceptable factor determinacy would be considered if a significant increase in fit relative to a simpler model was evidenced.

After the dimensions were determined, the items were assigned to parcels using the domain-representative approach by assigning items to each parcel such that the dimensions are as equally represented in each parcel as possible. Once achieved, the parcels indicator values for each student were created by averaging the items assigned to each parcel.

Estimation of the Measurement Model

In line with the recommendations of Anderson and Gerbing (1988), a confirmatory measurement model was estimated separately and before the estimation of the structural model. The rationale for this procedure is to ascertain the stability of the measurement of the latent constructs. That is, the measures should indicate the same unidimensional construct for both the measurement and structural models. This is assessed by noting the changes in the pattern coefficients from the measurement and structural model. Specifically, the coefficients estimated in the measurement model should not change or “should change only trivially” (Anderson & Gerbing, 1988, p. 418)

116 in the structural model. The confirmatory measurement model used the domain- representative parcels as indicators of the latent constructs (with the exception of grade value and goal orientation constructs which used individual items). The primary hypothesis of the confirmatory factor model is that the indicators load only on the constructs they are intended to measure (i.e. simple structure with no cross loadings) and

that there is no significant correlation among any of the indicator errors. As this is a

saturated model all of the latent constructs were correlated.

Estimation of the Structural Model and Testing of the Structural Hypotheses

The hypotheses of this study are shown in Table 1.1 which is reproduced here as

Table 2.1 for convenience.

To test these hypotheses a number of structural models were tested. First, a model (Model S1) was run that estimated all of the structural paths not identified as

“N/A”. As such, this may be considered to be a saturated or full model with respect to

the hypotheses.

After Model S1 was run a series of planned tests, based on the recommendations of Gonzalez & Griffen (2001), were implemented involving the placement of constraints on all structural path coefficients one at a time to the freely estimated parameters.

Constraints specified by each hypothesis were forced on the model one at a time and chi- square likelihood ratio tests2 with one degree of freedom of the differences between the

full and reduced models were used to assess the reasonableness of each imposed

constraint. After each individual constraint was tested and assessed, it was removed (i.e.,

freely re-estimated) while the next individual constraint was forced and tested. After

117 testing all the individual constraints, all structural paths assessed to be not significantly different from zero were constrained to zero and the new model (S2) retested freely estimating the remaining parameters. This model was then compared to the theoretical model specified by the eight hypotheses.

*CSE = Social Exclusion Concerns; AV = Academic Values; GV = Grade Values; ACT = Standardized Test Score; HSCR = High School Class Rank; PROC = Procrastination/Lack of Self- regulation; SB = School Belongingness; PAP = Performance-approach; PAV = Performance- avoidance; MAP = Mastery-approach; MAV = Mastery-avoidance; SE = Self-regulatory Self- Efficacy; PS = Perceived Stress; GPA = First Quarter Grade Point Average Table 2.1: Hypothesized Directions of the Structural Relationships (Reproduced from Table 1.1)†

118 Chapter 3: Results

General Descriptive Statistics of Gender, Ethnicity, and Ability

Table 3.1 provides the distributions, by gender and ethnicity, of students in the

total sample who did not respond to the questionnaires (Non-Responders), responded to

only the first questionnaire (Partial Responders), or responded to both questionnaires

(Complete Responders).

Second Week Second and Total Response Eighth Week Non Response Responses Only Response Requested (Partial) (Complete) Gender Ethnicity N Percent N Percent N Percent N Percent AM IND/ESK/A 4 0.4% 0 0.0% 0 0.0% 4 0.2% ASIAN/PAC IS 29 2.6% 10 4.0% 23 3.4% 62 3.0% BLACK 46 4.1% 11 4.4% 22 3.3% 79 3.9% Female HISPANIC 14 1.2% 5 2.0% 6 0.9% 25 1.2% OTHER 21 1.9% 5 2.0% 44 6.5% 70 3.4% WHITE,NON-HS 355 31.7% 80 31.9% 313 46.6% 748 36.6% Female Subtotal 469 41.8% 111 44.2% 408 60.7% 988 48.3%

AM IND/ESK/A 2 0.2% 0 0.0% 1 0.1% 3 0.1% ASIAN/PAC IS 42 3.7% 9 3.6% 6 0.9% 57 2.8% BLACK 29 2.6% 7 2.8% 5 0.7% 41 2.0% Male HISPANIC 21 1.9% 5 2.0% 6 0.9% 32 1.6% OTHER 47 4.2% 12 4.8% 33 4.9% 92 4.5% WHITE,NON-HS 511 45.6% 107 42.6% 213 31.7% 831 40.7% Male Subtotal 652 58.2% 140 55.8% 264 39.3% 1056 51.7% Grand Total 1121 251 672 2044

Table 3.1: Gender and Ethnicity Distributions of All Students, Non-Responders, Partial Responders, and Complete Responders

119 It is the latter group (minus one student who opted out of the study [N = 671]) whose data is reported on below. It can be seen that for the sample to be reported on

(Complete Responders) gender differs by 12.4% relative to the total sample (N = 2,044).

A proportions test showed that women in general were clearly overrepresented relative to the population (z = 5.67, p < 0.0001). In addition, ethnic minorities comprised 10.3% of the sample of students completing both questionnaires relative to 14.8% in the total sample. A proportions test confirmed that minorities were underrepresented in the sample to be analyzed (z = -3.23, p = 0.001).

With respect to ability measures (ACT and HSCR), there were no significant differences in median levels of ACT (Kruskal-Walllace H = 3.49, p = 0.175) among Non-

Responders, Partial Responders, and Complete Responders, but there were significant differences in median levels of HSCR (H =51.82, p < 0.0001). Descriptive statistics of the ability measures are shown in Table 3.2.

1st 3rd ACT N M SD Md Quartile Quartile Non Response 249 26.5 3.4 24.0 27.0 29.0 Second and Eighth Week Response (Complete) 662 27.0 3.0 25.0 27.0 29.0 Second Week Response Only (Partial) 1112 26.8 3.4 25.0 27.0 29.0 Total Responses Requested 2023 26.9 3.3 25.0 27.0 29.0

1st 3rd HSCR N M SD Md Quartile Quartile Non Response 247 84.5 13.0 79.0 86.9 93.8 Second and Eighth Week Response (Complete) 651 87.8 9.6 82.0 90.1 94.8 Second Week Response Only (Partial) 1111 83.6 12.7 77.7 85.9 93.0 Total Responses Requested 2009 85.1 12.0 79.4 87.6 93.8

Table 3.2: Descriptive Statistics of Ability Measures for All Students, Non-Responders, Partial Responders, and Complete Responders

120 Preliminary Exploratory Factor Analyses Determining Dimensionality of Constructs

The correlation matrixes of the items for each construct whose dimensionality was to be determined through the preliminary factor analyses is shown in Appendix G. A summary of the results of the preliminary factor analyses used to determine the dimensionality of the six constructs (i.e., CSE, AV, PROC, SE, SB, and PS) are shown in

Tables 3.3 through 3.5 which show the results of the parallel analyses, maximum likelihood factor analyses, and factor determinacy values respectively. Appendix H provides the factor pattern loadings and factor structure coefficients for each item. Also shown in Appendix H for convenience are the final factor determinacy values, internal reliability coefficients (Cronbach’s alpha) for items defining a dimension, and correlations between the dimensions. The items used to calculate internal reliabilities for a given dimension are identified by bolded and italicized pattern and structure coefficients.

Differences Sample Eigenvalues Random Eigenvalues (Sample Minus Random) Construct 123412341234 CSE 3.21 0.81 0.67 0.52 1.18 1.11 1.05 1.00 2.03 -0.30 -0.38 -0.48 AV 5.24 1.51 0.97 0.86 1.28 1.20 1.15 1.11 3.96 0.31 -0.18 -0.25 PROC 9.14 2.16 1.31 1.08 1.43 1.36 1.31 1.27 7.71 0.80 -0.01 -0.19 SB 6.53 2.13 1.17 0.99 1.35 1.28 1.23 1.19 5.18 0.85 -0.06 -0.20 SE 5.51 1.02 0.90 0.78 1.30 1.23 1.18 1.14 4.21 -0.21 -0.28 -0.36 PS 5.71 1.72 0.94 0.82 1.30 1.23 1.82 1.39 4.41 0.48 -0.88 -0.58

Table 3.3: Parallel Analyses of Dimensional Structure: Eigenvalues for the Sample and Randomly Generated Correlation Matrices

121 The parallel analyses suggested one dimension for CSE and SE and two dimensions for AV, PROC, SB, SE, and PS. Table 3.3 shows the first four eigenvalues for the correlation matrices of the individual items for each construct in the sample of 671 students (sample eigenvalues), the 95th percentiles of eigenvalues from 1,000 randomly

generated correlation matrices, and the differences between the sample and random

eigenvalues. The differences between the eigenvalues are calculated by subtracting the

random eigenvalue from the sample eigenvalue. A differences less than or equal to zero

suggest insignificant or random dimensions.

As is shown in Table 3.3, only the first sample eigenvalue for CSE and SE is

greater than randomly generated eigenvalues suggesting a unidimensional structure for

these two constructs. However, the parallel analyses suggest that AV, PROC, SB, and PS

are multidimensional with two dimensions for each construct.

Based on the dimensional structure suggested by the parallel analyses, maximum

likelihood factor analyses were conducted for each construct. Two maximum likelihood

analyses were conducted for all constructs. That is, for each construct, p and p+1

dimensions were extracted where p is the number of dimensions suggested by parallel

analyses. Table 3.4 summarizes the results of the maximum likelihood factor analyses extracting p and p+1 dimensions.

Looking first at CSE in Table 3.4, it is clear that a model extracting one dimension results in acceptable fit indices. Unfortunately, a model attempting to extract two dimensions using the six items of this scale failed to converge. However, given the acceptable fit indices and factor determinacy of 0.920 (see Table 3.5) in addition to the

122 results of the parallel analysis, it was concluded that a unidimensional structure underlies this construct.

RMSEA Constructb 2 df 2/df CFI TLI SRMR EST 90%CI p  0.05

CSE (1) 32.733 9 3.637 0.969 0.948 0.063 (0.040, 0.086) 0.161 0.030 AV (2) 181.189 43 4.214 0.944 0.915 0.069 (0.059, 0.080) 0.001 0.035 AV (3) 76.997 33 2.333 0.982 0.965 0.045 (0.032, 0.058) 0.739 0.022 PROC (2) 834.578 274 3.046 0.897 0.878 0.055 (0.051, 0.060) 0.022 0.041 PROC (3) 703.807 250 2.815 0.917 0.892 0.052 (0.047, 0.057) 0.228 0.035 SB (2) 573.514 118 4.860 0.868 0.828 0.076 (0.070, 0.082) 0.000 0.047 SB (3) 360.650 102 3.536 0.925 0.887 0.061 (0.055, 0.068) 0.003 0.035 SE (1) 188.110 65 2.894 0.938 0.925 0.053 (0.044, 0.062) 0.269 0.040 SE (2) 106.204 53 2.004 0.974 0.960 0.039 (0.028, 0.049) 0.960 0.027 PS (2) 216.196 64 3.378 0.938 0.912 0.060 (0.051, 0.068) 0.035 0.036 PS (3) 110.872 52 2.132 0.976 0.958 0.041 (0.030, 0.052) 0.915 0.023 aChi-square values are re-scaled to account for non-normal missing data. The statistic is based on the Yuan- Bentler T2* statistic. bNumbers in parentheses next to the construct label represent the number of dimensions extracted. Bolded and italicized rows indicate the models representing the number of dimensions used

Table 3.4: Maximum Likelihood Analyses of Dimensional Structure: Goodness of Fit Measuresa from the Extraction of the Number of Dimensions Determined by Parallel Analyses

Second, examining the fit statistics for AV, it can be seen in Table 3.4 that a model extracting three dimensions appears to be a better fitting model than the one with two dimensions suggested by the parallel analyses. However, Table 3.5 shows that the factor determinacy value for the third dimension in the three dimensional model is less than 0.900 (r = 0.779). In addition, the pattern loadings for this third dimension are all less than 0.500. Finally, an examination of the ratios of the pattern loadings to their

123 standard errors for the third dimension revealed that they were less than the corresponding ratios for the first two dimensions for all items. Thus, even though the

three-dimensional model results in better fit indices relative to the two-dimensional

model, all other indices point to a two-dimensional structure underlying AV. Finally, these dimensions are easily interpretable. As seen in Appendix H, the first dimension appears to reflect utility whereas the second dimension appears to reflect academic challenge.

p CSE AV PROC SB SE PS Dimensions 1 0.920 0.950 0.965 0.929 0.952 0.930 2 0.936 0.912 0.946 0.920 (a)

p +1 CSE AV PROC SB SE PS Dimensions 1 0.956 0.964 0.923 0.942 0.936 2 0.938 0.906 0.915 0.893 0.928 3 0.779 0.737 0.935 0.825 (b) Panel (a) shows factor determinacy values for the p dimensions suggested by the parallel analysis. Panel (b) shows these values for p+1 dimensions.

Table 3.5: Factor Determinacy Values Resulting from the Extraction of p and p+1 Dimensions

Third, the results for PROC were similar to those for AV in that a three-

dimensional model resulted in better fit indices relative to the two-dimensional model

suggested by the parallel analysis, the additional third dimension had a very low factor

determinacy (r = 0.737, see Table 3.5), and none of the pattern loading to standard error

124 ratios for any item loading on this dimension were greater than the ratios for the first two dimensions. As such, it was concluded that the items measuring PROC reflect a two- dimensional structure. As shown in Appendix H, these dimensions appear to reflect lack of self-discipline and impulsiveness respectively.

Fourth, the results for SB show that a three-dimensional structure better describes this construct than the two-dimensional structure suggested by the parallel analysis.

Specifically, Table 3.4 shows that the fit indices for the two-dimensional model are generally not acceptable, whereas a three-dimensional structure provides considerably better indices. In addition, the better fitting three-dimensional model had factor determinacy values greater than 0.900 for all three dimensions (Table 3.5). Finally, the three-dimensional model was easily interpretable with the three dimensions reflecting instructor/student support, general perceived acceptance, and feelings of school belongingness respectively.

Fifth, Table 3.4 shows that while the two-dimensional model for SE resulted in very good fit indices, the one-dimensional model indices were acceptable. In addition, while three items loaded on the second dimension, this dimension had a low factor determinacy (r = 0.893, see Table 3.5). It was thus concluded that a unidimensional structure underlies SE that reflects self-regulatory self-efficacy.

Finally, Table 3.4 shows that a two-dimensional structure for PS suggested by the parallel analysis appears to fit the data acceptably. In addition, a three-dimensional model, while resulting in better fit indices yielded a third dimension that had an unacceptable factor determinacy (r = 0.825, see Table 3.5) and pattern loading to standard error ratios less than those for the other two dimensions for all items. It thus appears that

125 PS is represented by a two-dimensional structure with one dimension representing a

primary appraisal of outer demands and a secondary appraisal of how these demands are

being dealt with (Lazarus, 2001).

In summary, with the exception of SB and PROC, the results of the parallel

analyses were corroborated by the maximum likelihood analyses. The dimensional

structure for SB appears to have one more dimension than that suggested by the parallel

analysis. The maximum likelihood fit indices for the two-dimensional structure

suggested by the parallel analysis for PROC (particularly CFI and TLI) were less than

acceptable. However, given that likelihood ratio test for maximum likelihood factor

analysis can overestimate the number of dimensions (Schönemann & Wang, 1972) in

addition to the fact that the extraction of p+1 dimensions produced one dimension that

had very low pattern loadings and had low factor determinacy, it is possible that the extraction of one more dimension than that suggested by the parallel analyses may have

been susceptible to a rank deficiency of the factor loading matrix whereby at least one

column of the population matrix is composed of all zeros. Hayashi et al. (2007) have

shown that in cases such as this, the likelihood ratio test statistic does not follow the chi-

square distribution and may lead to the extraction of too many factors. Thus, despite the

lower than desirable CFI and TLI statistics, the tendency of the likelihood ratio test

statistic to overextract dimensions, the general concordance the parallel analyses,

RMSEA and SRMR statistics, factor determinacy values, and parsimony all seem to point

to a two-dimensional structure for PROC.

Confirmatory Factor Analysis: Measurement Model

126 Appendix F shows the assignment of the individual items to the parcel indicators.

Appendix I shows the correlation matrix of the parcel indicators.

The results of the measurement model (M1) using the parcels as indicators are

presented in Appendix J which shows the unstandarized and standardized robust loadings

and standard errors for the indicators. This model, with no cross loadings or correlated

indicator errors and all latent constructs correlated, resulted in an acceptable fit to the

data. Table 3.6 provides the fit indices for model M1.

RMSEA Model  df cfb CFI TLI SRMR EST 90%CI p  0.05

M1 1015.832 409 1.175 0.951 0.940 0.047 (0.043, 0.051) 0.910 0.042 M2 585.680 263 1.152 0.968 0.960 0.043 (0.038, 0.047) 0.995 0.038 aT2* values are re-scaled test statistics to account for non-normal missing data. The statistic is based on the Yuan- Bentler T2* statistic. The actual chi-square value is  2 *2  cfT .

bcf = scaling correction factor

Table 3.6: Fit Indicesa for Measurement Models

Having determined that the hypothesized measurement model adequately fit the

data, an examination of the correlations among the latent variables revealed an extremely

high positive correlation (0.859) between PAP and PAV (see Table 3.7). Given that the

performance achievement goal variables are conceptualized as predictors of end of

quarter GPA, the potential for problems due to the collinearity between them needed to

be addressed. The effects of collinearity of the magnitude seen here are well known,

127 among them being very large standard errors of the structural path coefficients (Marsh,

Dowson, Pietsch, & Walker, 2004) and unacceptably high Type II error rates (Grewal,

Cote, & Baumgartner, 2004).

Latent 1 2 3 4 5 6 7 8 9 10 Variables* 1 PAP 2 PAV .859 3 MAP .330 .355 4 MAV .227 .271 .379 5 CSE .144 .155 .033 .055 6 AV .179 .174 .640 .258 -.003 7 GV .159 .084 .190 .152 .086 .212 8 PROC -.134 -.139 -.416 -.123 .098 -.512 -.200 9 SB .238 .219 .464 .193 .150 .342 .054 -.330 10 SE .362 .317 .724 .322 -.003 .506 .241 -.579 .548 11 PS -.149 -.100 -.316 -.243 .026 -.221 -.038 .391 -.572 -.521 *PAP = Performance-approach; PAV = Performance-avoidance; MAP = Mastery-approach; MAV = Mastery-avoidance; CSE = Social Exclusion Concerns; AV = Academic Values; GV = Grade Values; PROC = Procrastination/Lack of Self-regulation; SB = School Belongingness; SE = Self-regulatory Self-Efficacy; PS = Perceived Stress r >= |0.092|, p<=0.05

Table 3.7: Correlation Matrix of Latent Variables from Model M1

It appears that students did not distinguish among the items measuring these

putatively distinct constructs and it was thus necessary to either allow the items to

measure one performance goal orientation construct as opposed to two distinct constructs

or drop one of the constructs. At present there is no theoretical or empirical justification

for combining performance-approach and performance-avoidance constructs. It is known

that students may not distinguish between the approach and avoidance dimensions of performance goals in the same ways that researchers intend (Urdan & Mestas, 2006). In

128 addition, there is at least one other study that found performance-approach and performance-avoidance goals to be highly correlated (Elliot & Murayama, 2008).

However, other studies have found considerably lower correlations between these constructs (Cury, et al., 2006; Elliot & McGregor, 2001; McGregor & Elliot, 2002).

Indeed, the study (Elliot & McGregor, 2001) from which the present research adapted the goal orientation items found a correlation of only 0.18 between the latent performance- approach and performance-avoidance constructs in a confirmatory factory analysis.

Furthermore, as reviewed above there is a considerable amount of research suggesting that a performance-avoidance orientation tends to have reliably detrimental effects on motivation and achievement.

As such it was decided to narrow the scope of the hypotheses to mastery and performance distinctions along the approach dimension. Mastery and performance goals in general are the most frequently discussed (Deshon & Gillespie, 2005). The approach dimension was chosen because there is considerably more research studying mastery- approach relative to mastery-avoidance. Furthermore, this research clearly demonstrates the beneficial and desirable academic effects of adopting a mastery-approach orientation.

And while there is much research attesting to the general detrimental effects of adopting a performance-avoidance orientation, the high correlation found in this study between performance-approach and performance-avoidance orientations and the possibility, shown by some other researchers, of students not distinguishing between them requires that a performance-approach goal be studied to avoid the confounding of the approach- avoidance distinction in studying the potential differences between developing and demonstrating competence.

129 The confirmatory factor analysis was re-run as model M2 with PAV and MAV

removed. The results of the measurement model (M2) showing the unstandarized and standardized robust loadings and standard errors are presented in Appendix J along side

those of model M1. This model, with no cross loadings or correlated indicator errors and

all latent constructs correlated, resulted in an acceptable fit to the data. Table 3.6

provides the fit indices for model M2. The correlation matrix of the latent variables for

model M2 is shown in Table 3.8. Removing the avoidance goals orientations did not

change the pattern loadings or the latent variable correlations.

Latent 1 2 3 4 5 6 7 8 Variables* 1 PAP 2 MAP .328 3 CSE .142 .033 4 AV .178 .639 -.003 5 GV .158 .199 .086 .211 6 PROC -.134 -.416 .098 -.511 -.200 7 SB .236 .464 .150 .342 .055 -.330 8 SE .360 .724 -.003 .506 .242 -.580 .548 9 PS -.150 -.317 .026 -.222 -.041 .392 .573 -.523 *PAP = Performance-approach; MAP = Mastery-approach; CSE = Social Exclusion Concerns; AV = Academic Values; GV = Grade Values; PROC = Procrastination/Lack of Self-regulation; SB = School Belongingness; SE = Self-regulatory Self-Efficacy; PS = Perceived Stress r >= |0.092|, p<=0.05

Table 3.8: Correlation Matrix of Latent Variables from Model M2

Structural Model: Direct Effects

The full model (S1) estimating all the structural paths fit the data well (see Table

3.9). The results for likelihood ratio tests of the individual parameter estimates for models nested within S1 are shown in Table 3.10 which reproduces Table 1.1 but also 130 shows the hypothesized relationship (top symbol), standardized parameter estimate for

each structural path estimated (middle number) of model S1, and the likelihood ratio test

statistic (bottom number). Shown in bold italics are the parameter estimates that did not

conform to the hypothesized relationships. Specifically, one relationship hypothesized to

be positive (SB on GV) was found to be non-significant. In addition, two relationships

hypothesized to be negative (PAP on PROC and MAP on PROC) were also found to be

non-significant. Finally, one relationship hypothesized to be non-significant (GPA on

GV) was found to be significantly positive.

RMSEA

Model  df cfb BIC CFI TLI SRMR EST 90%CI p  05.0

S1 801.385 327 1.136 61996 0.955 0.944 0.046 (0.042, 0.051) 0.920 0.045 S2 853.106 351 1.138 61976 0.952 0.945 0.046 (0.042, 0.050) 0.945 0.055 S3 845.238 348 1.137 61977 0.953 0.945 0.046 (0.042, 0.050) 0.946 0.052 S4 857.873 349 1.137 61988 0.951 0.944 0.047 (0.043, 0.051) 0.920 0.053 S2B 857.265 352 1.137 61977 0.952 0.945 0.046 (0.042, 0.050) 0.942 0.055 aT2* values are re-scaled test statistics to account for non-normal missing data. The statistic is based on the Yuan- Bentler T2* statistic. The actual chi-square value is  2 *2  cfT .

bcf = scaling correction factor

Table 3.9: Fit Indicesa for Structural Models

Table 3.11 shows the correlation coefficients of the exogenous variables

estimated in model S1 (top number) and the likelihood ratio test statistics (bottom

numbers) from individual tests of the coefficients. Only four correlations were significant. These correlations are bolded and italicized in Table 3.11. 131 Endogenous Criteria Variables* Predictor Variables* PROC SB PAP MAP SE PS GPA + + + 0 0 0 0 CSE 0.11 0.17 0.11 -0.01 -0.07 0.07 -0.1 (4.06) (13.6) (4.56) (0.46) (1.59) (1.13) (2.37) - + 0 + 0 0 0 AV -0.49 0.19 0.11 0.52 -0.05 0.06 0.08 (114.7) (11.4) (2.47) (144.3) (1.79) (1.14) (2.05) - + + 0 0 0 0 GV -0.09 -0.08 0.11 0.05 0.07 0.05 0.15 (4.40) (3.06) (5.31) (1.69) (3.52) (1.59) (10.4) + ACT N/A N/A A N/ N/A N/A N/A 0.13 (9.08) + HSCR N/A N/A N/A N/A N/A N/A 0.15 (13.1) - - - - + 0 PROC N/A -0.23 -0.02 -0.07 -0.32 0.15 -0.05 (17.3) (0.13) (1.72) (132.6) (10.3) (1.31) + + + - ? SB N/A N/A 0.17 0.25 0.19 -0.43 -0.11 (11.3) (23.8) (34.4) (541.1) (6.83) + 0 + PAP N/A N/A N/A N/A 0.14 0.03 0.08 (16.2) (0.69) (5.32) + 0 0 MAP N/A N/A N/A N/A 0.5 0.13 0.06 (82.9) (2.22) (1.18) - + SE N/A N/A N/A N/A N/A-0.36 0.16 (27.7) (4.68) - PS N/A N/A N/AN/A N/A N/A -0.14 (9.00) †For each cell the top symbol is the hypothesized relationship given in Table 1.1, the number in the middle is the standardized parameter estimate for the given structural path, and the number on the bottom in parentheses is the likelihood ratio test statistic. Since each chi- square test statistic is evaluated at one degree of freedom, chi-square values less than 3.84 indicate parameter estimates that are not significant at  = 0.05.

*CSE = Social Exclusion Concerns; AV = Academic Values; GV = Grade Values; ACT = Standardized Test Score; HSCR = High School Class Rank; PROC = Procrastination/Lack of Self-regulation; SB = School Belongingness; PAP = Performance-approach; MAP = Mastery-approach; SE = Self-regulatory Self-Efficacy; PS = Perceived Stress; GPA = First Quarter Grade Point Average

Table 3.10: Hypothesized Direction and Standardized Coefficients of Direct Effects†

132 A second model (Model S2) was run constraining to zero all structural parameters found to be zero in S1 including the three parameters originally hypothesized to be significantly different from zero. In addition, Model S2 included the estimation of the parameter from GV to GPA that was originally hypothesized to be zero. Finally, six of the correlations among the exogenous variables found to be non-significant in S1 were constrained to zero in S2. All other structural parameters that were free in S1 were free in S2. As such, S2 is nested within S1 and the differences between the models may be assessed by a chi-square difference test with 24 degrees of freedom. Table 3.9 shows that the model fits the data adequately, but, as expected, is significantly different from the full model (  2 )24( = 51.887, p = 0.0008). It must be noted here that models S1 and S2 are not the theoretical models. Model S1 was the full model from which model S2 was empirically created based on a series of likelihood ratio tests of the parameter estimates.

To test the theoretical model it is necessary to compare it against the empirical model. In testing the theoretical model, the pattern of correlations among the exogenous variables found in model S1 and maintained in model S2 were also maintained in the theoretical model. The interest in assessing the theoretical model rests only with the structural (one- directional) paths between the constructs.

In comparing the theoretical model with model S2 it was noted that there were only four paths that differed between the originally proposed theoretical model and the empirically derived model. The theoretical model hypothesized that a significant positive relationship would exist between SB and GV and that significant negative relationships would exist between PAP and PROC and MAP and PROC respectively. However, the empirical model suggests that these relationships are not significantly different from zero. 133 Finally, the theoretical model hypothesized that the relationship between GV and GPA would be non-significant but the empirical model suggests that these constructs are positively related.

Variables* 1 2 3 4 1 1 CSE

-0.003 1.000 2 AV (0.000) 0.084 0.211 1.000 3 GV (3.379) (23.247) 0.106 0.037 0.298 1.000 4 ACT (5.568) (1.030) (48.53) -0.031 0.055 0.093 0.063 5 HSCR (0.000) (1.522) (4.421) (2.782) †For each cell the top number is the correlation estimate and the number on the bottom in parentheses is the likelihood ratio test statistic. Since each chi-square test statistic is evaluated at one degree of freedom, chi-square values less than 3.84 indicate parameter estimates that are not significant at  = 0.05.

*CSE = Social Exclusion Concerns; AV = Academic Values; GV = Grade Values; ACT = Standardized Test Score; HSCR = High School Class Rank.

Table 3.11: Correlations between Exogenous Variables in the Structural Model (S1)†

In order to test the theoretical model against the empirically derived model, the following testing procedure was used. Model S2 was nested in a third model, model S3, which was identical to model S2 except the three structural parameter found to be zero in

S2 (i.e. PAP on PROC, MAP on PROC, and SB on GV) were freed. Freeing these parameters partly conforms to the theoretical model. However, model S3 is not the theoretical model because it includes estimating the structural path of GPA on GV. The

134 reasoning behind this procedure is that if model S3 is significantly different from model

S2 it would suggest that freeing these three paths may be warranted. Further, if this were the case, then a fourth model (S4, the theoretical model) could be compared against model S3 by constraining the path of GPA on GV to zero. This fourth model would be nested in model S3 differing by one degree of freedom. However, if models S2 and S3 are not different then model S2 should be chosen as the “best” model because it is the most parsimonious.

Table 3.9 shows the fit indices for model S3 where it can be seen that the model fits the data adequately and the addition of three extra parameters resulted in a trivial penalty according the BIC criterion. However the difference between models S2 and S3 was only marginally significant (  2 3( ) = 7.814, p = 0.05).

Even though this result, in combination with the fact that S2 is more parsimonious than S3, suggests that S2 should be accepted, model S4 was run to see if the path from

GV to GPA should be dropped. Constraining this path produces a model (S4) that is significantly different from S3 (  2 1( ) = 12.635, p = 0.0004). Furthermore the BIC value for model S4 (see Table 3.9) is considerably higher than the BIC values of either

S2 or S3 suggesting that S4 is inferior to models S2 and S3. Given these results and the fact that model S2 is more parsimonious, model S4 (the proposed theoretical model) is rejected and model S2 tentatively accepted as the best model.

One final model (Model S2B) was run to assess the importance of the path from

SB to GPA. It was hypothesized that SB would positively predict GPA. However, a significant negative relationship between these constructs was found. This may be due to

135 a suppression effect (Cohen & Cohen, 1975), and implies that for any level of self- efficacy, performance-approach orientation, grade value, ACT, and high school class rank, GPA tends to decrease with increasing level of perceived school belongingness. In particular, the relatively large correlations between SB and SE and between SB and PS

(0.548 and -0.572 respectively; see Table 3.8) suggest that the zero order positive zero order relationship between SB and GPA may be due to strong relationships among the predictor variables and that removing the influence these variables from SB reveals a slight debilitating influence of school belongingness on GPA. It turns out, however, that constraining the path from SB to GPA creates a model (S2B) that differs significantly from model (S2) (  2 1( ) = 4.931, p = 0.0264). Based on this finding, it is concluded that the path with a negative structural coefficient from SB to GPA may not be removed from the final model. The final model is shown in Figure 1.

Appendix K shows the unstandardized and standardized factor loadings for the final model along side the measurement model (M2). As can be seen the loadings in the measurement and structural models differ only trivially.

Structural Model: Total and Indirect Effects

Of interest in this study is an assessment of the total and indirect effects of values on the endogenous variables. Table 3.12 shows the total and total indirect effects of values on all the endogenous variables. In addition, the total and indirect effects of school belongingness and procrastination on GPA are also shown.

First, with the exception of the insignificant total effect of grade values on mastery-approach all of the total effects of values on achievement goals were significant 136 (although the total effect of social exclusion concerns on mastery-approach was small).

However, it is clear that the magnitudes of the total effects were in the direction of the hypotheses. That is, the strongest total effects were from social exclusion concerns and grade values to performance-approach and from academic task values to mastery- approach.

Second Week Eighth Week End of Term

CSE .10 PAP

.17 .15 .21

.11 .09 SE .13 -.33 .19 .19 -.21

GV -.09 PROC -.23 SB -.11 GPA

-.42 .15 -.14 -.49 PS

.21 .16 .26 .47 .17 .13 (ACT)

MAP .15 (HSCR) AV .56

ACT & HSCR

CSE = Concern Over Social Exclusion MAP = Mastery Goal Orientation GPA = End of Quarter Grade Point Average AV = Academic Value – Challenge PAP = Performance Goal Orientation GV = Academic Value – Grade Setting SB = School Belongingness PROC = Procrastination/Lack of Self-Regulation SE = Self-Efficacy ACT = Standardized Test Scores PS = Perceived Stress HSCR = High School Class Rank

Correlations not shown: rGV.ACT = 0.298; rGV.HACR = 0.093; rCSE.ACT = 0.106

Figure 1: Path Diagram of the Final Structural Model (S2)

137

Endogenous Criteria Variables* Predictor GPA Effect Variables*  EST/SE p CSE Total 0.01 0.776 0.438 Indirect 0.01 0.776 0.438 AV Total 0.11 4.478 0.000 Indirect 0.11 4.478 0.000 GV Total 0.18 3.858 0.000 Indirect 0.03 2.940 0.003 SB Total 0.05 1.179 0.238 Indirect 0.15 5.283 0.000 PROC Total -0.10 -4.779 0.000 Indirect -0.10 -4.779 0.000

Predictor PAP MAP Effect Variables*  EST/SE p  EST/SE p CSE Total 0.13 2.688 0.007 0.04 2.799 0.005 Indirect 0.03 2.531 0.011 0.04 2.799 0.005 AV Total 0.06 3.416 0.001 0.64 17.113 0.000 Indirect 0.06 3.416 0.001 0.07 4.474 0.000 GV Total 0.14 3.194 0.001 0.01 1.837 0.066 Indirect 0.01 1.777 0.076 0.01 1.837 0.066

Predictor SB SE PS Effect Variables*  EST/SE p  EST/SE p  EST/SE p CSE Total 0.14 3.116 0.002 0.03 1.096 0.273 -0.05 -1.833 0.067 Indirect -0.03 -2.037 0.042 0.03 1.096 0.273 -0.05 -1.833 0.067 AV Total 0.28 5.794 0.000 0.52 16.286 0.000 -0.30 -8.910 0.000 Indirect 0.11 4.163 0.000 0.52 16.286 0.000 -0.30 -8.910 0.000 GV Total 0.02 1.930 0.054 0.06 2.720 0.007 -0.03 -2.259 0.024 Indirect 0.02 1.930 0.054 0.06 2.720 0.007 -0.03 -2.259 0.024 *CSE = Social Exclusion Concerns; AV = Academic Values; GV = Grade Values; PROC = Procrastination/Lack of Self-regulation; SB = School Belongingness; PAP = Performance-approach; MAP = Mastery-approach; SE = Self- regulatory Self-Efficacy; PS = Perceived Stress; GPA = First Quarter Grade Point Average

Table 3.12: Standardized Total and Indirect Effects for Values, School Belongingness, and Procrastination

Second, while there was a significantly negative (albeit small) indirect effect of

social exclusion concerns on perceived school belonging, the total effect was positive as

predicted. However, while the direct effects of social exclusion concerns and academic

138 task values on perceived school belonging were similar, the total effect of academic task values was twice that of social exclusion concerns. This is probably due to the large negative relationship between academic task values and procrastination in addition to the positive relationship between social exclusion concerns and procrastination.

Third only academic task and grade values had significant total effects on self- efficacy and perceived stress. It is noteworthy that both academic task and grade values had positive total effects on self-efficacy and negative total effects on perceived stress.

Finally, there were significant positive total effects of academic task and grade values, but not social exclusion concerns, on GPA. In addition, procrastination had a significant negative total effect on GPA. Interestingly, the total effect of perceived school belonging on GPA was not significant, but the indirect effect was. This corroborates the notion that school belonging may have its beneficial effects on academic performance indirectly through its positive association with self-efficacy and its negative association with perceived stress.

139 Chapter 4: Discussion

Summary and Interpretation of Results

Hypothesis 1 predicted that academic task, grade, and social values would differentially influence the tendency to procrastinate and student’s sense of school belonging. The results of this study partially support this hypothesis. First, academic values positively predicted school belongingness but the relationship between grade values and school belongingness was not significant directly or totally. Given the

positive relationships found in previous research between school belonging and academic

task values (Anderman, 2003; Anderman & Freeman, 2004), a positive relationship was

also expected between these constructs in the present research and the results bear this

out. In addition, Anderman (2003) also showed that grade point average positively

predicted school belongingness so it was assumed that a positive relationship between

grade values and school belongingness would also be seen. However, this did not occur

in the present study. While the significant positive relationship between academic task

values and school belonging are perhaps non-controversial, the lack of relationship

between grade values and school belongingness requires some explanation. Anderman

(2003) found a positive relationship between grade point average and school

belongingness, but not between expectancy for academic success and school

belongingness. The measure of grade value in the present study was intended to

140 explicitly measure students’ desire for a grade by asking what grade point average would be “minimally acceptable”. However, given the results of this study, it is highly likely that this construct may have in fact been measuring students perceived expectancy for a grade point average rather than a value per se. If this is the case, the lack of relationship between this construct and school belongingness is understandable given Anderman’s

(2003) result.

Second, the positive relationship hypothesized between concern over social exclusion and school belongingness was found in this study and corroborates the notion that a sense of school belongingness is one particular way in which students can achieve the higher-level goal of affiliation.

Third, evidence was found for the specific predictions regarding the relationships between values and procrastination: academic and grade values negatively predicted and social exclusion concerns positively predicted procrastination respectively. With respect to academic task and grade values, this finding is perhaps unremarkable in that high levels of these values imply high attainment, intrinsic, and/or utility value (Eccles, 2005) for academic pursuits and should be negatively associated with procrastination.

The positive relationship between procrastination and concerns over social exclusion is also predicted based on studies by Baumeister and his co-worker (e.g.,

Baumeister, et al., 2002; Baumeister, 2005) where it has been shown experimentally that perceived social exclusion impairs self-regulation. Other research suggests potential mechanisms for the relationships between social exclusion and procrastination. First, there is evidence that social exclusion has effects that resemble those seen in procrastination. For example, Twenge, Catanese, & Baumeister (2003) showed that

141 participants who were led to believe that other participants did not want them in their

group (i.e. socially excluded) displayed a distorted perception of time such that they

overestimated the duration of time intervals relative to controls. This implies that they

perceived time as passing more slowly. In addition, these socially excluded participants

tended to favor immediate gratification in that a sizeable number of them claimed that

they would advise someone choosing between two jobs, one with short-term and the other with long-term benefits, to choose the job that was better in the short run.

However, most of the control participants (nearly all of them, in fact) favored the job

with the long-term benefits. Procrastinators can also show distorted time perception

(Lay, 1988) and tend to work more slowly than non-procrastinators (Ferrari, 2001). This

implies that procrastinators view time as passing more slowly that it actually is. Given

this, it is reasonable to assume that increases in perceived social exclusion results in

increases in D in the TMT utility equation thereby lowering the utility of a task relative to

what would be expected given a more accurate perception of the delay. Thus, distorted

perceptions of time might be one mechanism in the positive relationship between

perceived social exclusion and procrastination.

Second, the desire for immediate gratification seen in the socially excluded

participants of Twenge et al. (2003) resembles the inability to delay short-term

gratification often seen in procrastinators. Bembenutty and Karabenick (2004) maintain

that delay of gratification is a self-regulatory strategy that depends on a person’s

perceived orientation toward temporarily distant goals or future time perspective (FTP).

They review correlational evidence that suggests that FTP, operationally defined as

outcome expectancy for a future academic goal, is associated with delay of gratification

142 and self-regulation. Schouwenburg and Groenewoud (2001) showed a hyperbolic pattern in the number of hours students reported they would study, resistance to social temptation, and general motivation for studying as of function of time to exam. That is, students generally reported that they would rarely study and would be more susceptible to social temptation far in advance of an exam but then would rapidly increase studying and be more likely to resist temptation as the time of the exam draws near. This hyperbolic pattern was shown for all students, but was seen more strongly (especially for hours of studying) in high trait procrastinators. Delay of gratification, to the extent that it reflects degree of distractibility, impulsiveness, and lack of self-discipline and self-control, exemplifies sensitivity to delay (). The Bembenutty and Karbenick (2004) conceptualization of FTP reflects E/.

A third reason why academic procrastination might mediate social exclusion is that academic procrastination might be directly influenced by the interference of interpersonal relationships. Senecal, et al. (2003) found a reasonably good fit of a structural equation model that showed that low levels of self-determined motivation

(Deci, Vallerand, Pelletier, & R. M. Ryan, 1991) for academic and interpersonal roles increased the likelihood of role conflict that directly and positively predicted academic procrastination. The importance of this result, as they point out, is that academic procrastination can depend to some extent on how students organize the roles and priorities in their life. This role conflict, in turn, reflects a willingness to defer long-term goals (i.e. the “academic role”) in favor of short-term goals (i.e. the “interpersonal role”).

Role conflict in the context of TMT reflects an aspect of value (V) that along with

143 expectancy (E) determines initial utility for tasks as well as the length of time it takes

before the utility of one task surpasses another.

Finally, as discussed above, social exclusion has been found to affect self-

defeating behavior generally. Twenge, et al. (2002) showed that a belief in a future alone

increased the likelihood of participants taking riskier gambling options, making less

healthy choices, and procrastinating. Specifically, in their Study 4 they showed that

when participants were made to believe that the results of a personality survey indicated

they would likely be alone in life, they tended to spend more time procrastinating during

a fifteen minute study period provided to them to prepare for a complex arithmetic test.

Participants had been told that practicing these equations was shown to significantly

improve performance on the upcoming test. Regardless, participants who believed they

would likely spend their life alone procrastinated more than participants who believed

they would likely not be alone in life as well as a “misfortune” control group who were

told that their personality profile indicated that they were accident prone.

Hypothesis 2 dealt with the pattern of relationships between values and

achievement goal outcomes. Specifically, and looking only at the approach dimension studied in the structural model, the hypothesized relationships patterns between values and performance goals are essentially the mirror image of those between values and mastery goals. This pattern was supported in this study. First, this study corroborated

prior research (see E. M. Anderman & Wolters, 2006 and A. Kaplan & Maehr, 2007 for

reviews) by showing that mastery-approach, but not performance-approach, was directly

positively related to academic task values whereas performance-approach, but not

mastery-approach, was positively related to grade values. Furthermore, even though

144 academic task values had a significant total effect on performance-approach goal orientation it was quite small ( = 0.06) relative to the total effect of academic task values on mastery-approach goals ( = 0.64). In addition, the total effect of grade values on mastery-approach was not significant. Second, students were significantly more likely to endorse performance-approach goals if they also expressed a concern about social exclusion (desire for social inclusion). Third, there was no significant direct relationship found between the desire for inclusion and mastery-approach goal endorsement although the total effect was significant ( = .04, p = 0.005) but still considerably smaller that the total effect of social exclusion concerns on performance-approach goal orientation ( =

0.13, p = 0.007).

These last two findings are important because they suggest that students endorsing mastery-approach goals are guided more by a reflective sense of identity (i.e., common-identity) where a student can incorporate the “school” or “academy” within oneself and develop a sense of agency and competence without having to divulge one’s personal experiments and failures. Mastery-approach goals are private and can be achieved in ways other than public demonstration. In this sense, there is no obvious motive to “do” anything other than enjoy the challenge of understanding. Performance- approach goals, however, can be more exposed because of the more public values associated with them. This suggests that failure at these goals could reduce one’s relational value, especially in an academic setting. Social values involving a desire for inclusion as well as grade values which can be made public, are based more on a common-bond, creative identity where a student is more concerned with incorporating oneself within the group by demonstrating agency and competence that necessarily 145 entails exposing one’s successes and failures. Thus “doing” can be one of many means

of forming common-bonds. To the extent that the need for belongingness is active – as it

is expected to be in a population of students making a transition from old to new

associations – high-level affiliation goals and the links from them to lower-level goals

like achievement goals are also expected to be active. Therefore, performance goals, to the extent that they are diagnostic of fulfilling the desire for inclusion, can be potent motivators for disciplined academic work when the aim is for high grades.

Hypothesis 3 was also partially supported. Specifically, procrastination measured in the second week of the term negatively predicted self-efficacy and school belongingness and positively predicted perceived stress respectively in the eighth week of the term. However, the relationship between both achievement goals and procrastination was not significant. In addition, there was a significant negative total effect of procrastination on end of term grade point average. It is clear that procrastination can have debilitating influences on motivational, affective, academic achievement outcomes despite students’ claims to the contrary (Schraw, et al., 2007; Tuckman, 2007).

However, the lack of a predictive relationship of procrastination on achievement goals was contrary to expectations. It was initially proposed that, since students with a higher tendency to procrastinate should choose more desirable short-term options in favor of longer-term achievement goals, a negative relationship between procrastination and achievement goals would result. Specifically, when collapsed across time, high procrastinators are expected to show less preference for achievement goals to the extent that these goals are less attractive on average than other goals. This contention was not supported in the present study. Instead, the results support the distinction between goals

146 and plans (Gollwitzer, 1999). That is, goals reflect intentions whereas plans reflect implementation of those intentions. Achievement goals are hypothesized to be mid-level goals in the goal hierarchy (DeShon & Gillespie, 2005) linked to a number of lower behavioral or implementation goals. At the lower level the possibility for distraction is higher (Vallacher & Wegner, 1987) and it is at this level that procrastination is mot likely to be seen (Steel, 2007). Thus, the results of this study seem to suggest that it is possible for a student at a given level of procrastination tendency to express varying levels of desire for achievement goals because the tendency to procrastinate does not affect the expressed intention to attempt goals at this level. Rather, it is the actual implementation of these goals that is most likely affected by a tendency to procrastinate.

The other predictions regarding the influence of procrastination were supported.

First, in line with other evidence (Haycock, McCarthy, and Skay, 1998; Klassen,

Krawchuk, & Rajani, 2008; Martin, Flett, Hewitt, Krames, & Szanto, 1996; Steel, 2007;

Tuckman, 1991; van Eerde, 2003; Wolters, 2003b) procrastination and self-efficacy were negatively related. However, this study shows that procrastination may function as a partial mediator between values and self-efficacy and supports not only Sirois’ (2004) study on procrastination as a mediator and antecedent of self-efficacy but also the contention of a reciprocal relationship between self-efficacy and self-regulation (Sexton

& Tuckman, 1991; Sexton, Tuckman, & Crehan, 1992; Tuckman, 1990; Tuckman &

Sexton, 1990). Specifically, self-efficacy develops over time with enactive experience

(Bandura, 1997) and to the extent that a student procrastinates that experience is not as likely forth coming and, if it is, is likely less than satisfying given the debilitation in performance that often results because of procrastination.

147 Second, in line with previous research (Flett, et al., 1995; Lay, et al., 1989;

Rothblum, et al., 1986; Schraw, et al., 2007; Tice & Baumeister, 1997; Solomon &

Rothblum, 1984), procrastination positively predicted perceived stress near the end of students’ first academic term. These results show that a tendency to procrastinate can have deleterious effects on students’ perception of stressors and their ability to cope with them.

Third, procrastination was negatively related to perceived school belongingness.

This is perhaps the first demonstration of this relationship and these results suggest that procrastination affects not only academic tasks per se, but also the willingness to engage in activities that might foster a sense of school belongingness.

These last three results support the resource depletion model (Baumeister, 2002;

Baumeister, et al., 1998; Baumeister & Heatherington, 1996; Baumeister, et al., 2000;

Muraven & Baumeister, 2000). Resource depletion is seen here as leading to deficits in self-regulation due to the presence of multiple demands and can have deleterious effects in academic contexts (Tangney, et al., 2004). Procrastination, as a representative measure of the lack of self-regulation, should inhibit one’s sense of self-efficacy and ability to cope with stress. That is, during students’ first weeks at a university, multiple demands are expected to require some significant amount of self-regulation. But as these demands increase, students with a predilection for procrastination may be more severely affected by these multiple demands and may be more likely to opt to put work off to a later time. This will only have the effect of increasing the demands and further reduce one’s resources to cope. By the end of the term, students in this situation would thus be more likely to experience higher levels of stress and lower self-efficacy. The irony of

148 course is that procrastination as coping strategy simply exacerbates the demands and further reduces one’s ability to effectively cope with them.

The interesting relationship here is between procrastination and school belongingness because while the desire for social inclusion is positively related to procrastination it is also positively related to perceived school belongingness. However, procrastination was found to be negatively related to school belongingness. These relationships can be accounted for in part by the resource depletion model and by the

notion of synchrony or congruence among values. Specifically, students with a relatively high desire for social inclusion may be more likely to devote their self-regulatory

“muscle” to social concerns leaving less of that resource for academic tasks. As such, these students may be more likely to report having a sense of school belongingness and a higher tendency to procrastinate. The upshot of course is that as these students procrastinate, they are less likely to be engaged in academic endeavors or with those who are and thereby actually lessen their sense of school belongingness. In fact the indirect effect of social exclusion concerns on perceived school belongingness through procrastination is small but significantly negative ( = -.03, p = 0.042). However, even

though the total effect of social exclusion concerns on school belongingness is positive (

= .14, p = 0.002), looking at the direct standardized coefficients for academic task values

predicting procrastination from Table 3.10 as well as the indirect effect of academic task values on school belongingness from Table 3.12 (i.e.,  = .28, p < 0.0001), students with

high academic and social values might have a buffer against procrastination – academic

task values in particular are over four times greater in absolute magnitude than social

concern values in predicting procrastination (Table 3.10). Thus, it is likely that students 149 who desire to be included and have relatively high academic values (i.e. congruency)

might have a lower tendency to procrastinate relative to students who desire social

inclusion and have low academic values. Thus, academic task values, by minimizing the

influence of procrastination, may increase the potential for a positive influence of the

desire to be socially included on school belongingness. Note however that if academic

task values are low, a concern over social exclusion may increase one’s tendency to

procrastinate and thereby potentially negate any beneficial influences social values might

otherwise have regarding a sense of school belongingness.

Hypothesis 4 was concerned with the beneficial influences of a sense of school

belongingness on mastery-approach goal orientation, self-efficacy, perceived stress, and

grade point average. Save the significant positive relationship between school belonging

and performance-approach orientation and the unforeseen significant negative

relationship between school belongingness and grade point average, these hypothesized

relationships were supported and supports prior research (Anderman & Freeman, 2004).

However, the negative relationship between school belongingness and grade point average was unexpected. The most likely explanation here is a net suppression effect

(Cohen & Cohen, 1975). Specifically, referring back to Table 3.8, it can be seen that

school belongingness is highly correlated with the other three predictors of grade point

average (self-efficacy, performance-approach goal orientation, and perceived stress).

Therefore, the positive correlation between school belongingness and grade point average

may be due to the beneficial relationship school belongingness has with the other

predictors. Specifically, the suppression effect indicates that holding self-efficacy,

performance-approach goal orientation, and perceived stress constant, school

150 belongingness may actually have a deleterious influence on grade point average.

However, why this would be true is not entirely clear. Perhaps for a given level of self-

efficacy, performance goals, and perceived stress, students with a lower sense of school

belonging are simply taking classes to achieve good grades to perhaps transfer to another

school or to simply graduate. This conjures up an image of a student, detached from

others and one’s school environment, with a steadfast and resolute determination to

simply get through. On the other hand, at a given level of self-efficacy, performance

goals, and perceived stress, a student expressing a higher degree of school belongingness

may be involved in school activities to an extent that it disrupts effective studying for

exams and other assignments. This may imply the potentially disruptive effects of

concerns over social exclusion that reflects a need for belongingness. Regardless, this

does not imply that a sense of school belongingness necessarily has a detrimental

influence on academic performance. While the total effect of school belongingness on

grade point average was not significant ( = .05, p = 0.239), the total indirect effect was

significant ( = .15, p < 0.0001). This implies that the beneficial influences of a sense of

school belongingness on grade point average work through increasing self-efficacy and

performance-approach goals and decreasing perceived stress.

Hypothesis 5 involves the relationship of achievement goals to self-efficacy and

grade point average. In line with previous research on the relationship between self-

efficacy and mastery-approach goal orientation Gerhardt and Brown, 2006; Pintrich,

2000b; Middleton and Midgley, 1997; Phillips and Gully, 1997; Wolters et al., 1996),

mastery approach positively predicted self-efficacy. In addition, performance-approach

goal orientation positively predicted self-efficacy in this study. As reviewed above, there 151 does not seem to be a consistent finding regarding the relationship between performance goals and self-efficacy. However, in this study since the performance goal construct measured the degree to which students were focused on demonstrating to others that they could get good grades and the self-efficacy construct measured self-regulatory self- efficacy with the implied goal of getting good grades this positive relationship is not surprising.

Hypothesis 6, the benefits of perceived self-efficacy, was also supported. Self- efficacy negatively predicted perceived stress and positively predicted grade point average. Apparently, perceived self-efficacy of one’s ability to overcome problems of self-regulation may have helped students cope with the rigor of their first quarter and in doing so may have turned potential threats into challenges (Bandura, 1997; Chemers, et al., 2001).

In addition, and in support of research showing that academic self-efficacy measured at the end of the term is superior to that measured at the beginning of the term

(Kahn & Nauta, 2001; Sexton & Tuckman, 1991; Sexton et al., 1992; Tuckman &

Sexton, 1990) self-efficacy measured three weeks prior to final examinations was a significant predictor of grade point average in this study. The reason for the superior performance of self-efficacy as a predictor of academic achievement measured near the time performance is to be assessed is that the perception of one’s ability is simply more accurate due to constant revisions through enactive experience and vicarious learning

(Bandura, 1986; 1997). These results imply that through self-regulatory self-efficacy students, even in their first academic term, have a relatively good sense of how well they will achieve academically as measured by grade point average.

152 Hypothesis 7 regarding the debilitating influence of perceived stress on grade

point average was supported – perceived stress negatively predicted grade point average.

This result is predicted by the resource depletion model and the significant positive

relationship between procrastination and perceived stress. First, given that students are attempting to cope with the multiple demands of their first academic term, it might be expected that, in the absence of relatively high self-regulatory self-efficacy, many of the necessary cognitive, metacognitive, motivational, and behavioral self-regulatory functions required for successful negotiation of exams may suffer due to competition with these demands. This may result in poorer grades and the perception of the demands reported as perceived threats and a decreased sense of one’s ability to cope with the demands. Second, reduced self-regulation as manifested in a tendency to procrastinate can further reduce one’s ability to cope. Indeed, it is likely that, given the positive relationship between procrastination and perceived stress, procrastination itself reflects failed attempts at coping with demands early in the academic term. Thus students with a tendency to procrastinate may be putting themselves at a disadvantage even before they enter situations such as their first term at a university. As pointed out above, while procrastination was not hypothesized to be directly related to grade point average, the total effect of procrastination on grade point average was significant. Thus, procrastinating students, through decreased coping mechanisms such as self-efficacy and increased perceived stress, are significantly less likely to be academically successful and most likely experience less subjective well being and satisfaction with their university experience.

153 Finally, Hypothesis 8 predicting certain null relationships was supported with the

exception of the relationship between grade values and grade point average. On the

surface, this relationship may not seem surprising, but this construct was initially

conceived of as representing a value that guide and directs, but does not directly

determine behavior and as such this finding was unexpected. As it turns out however

asking students to report their desire regarding what grade point average is minimally

acceptable is apparently tantamount to asking then to set a goal for a grade at the end of the quarter as well as at the end of the year. One consistent result predicted by goal

setting theory (Locke & Latham, 1990; 2002) is that setting specific, challenging goals

provides precision regarding goal focus and standards of performance. If students

interpreted the items as setting goals as opposed or in addition to what they value

regarding an end-of-quarter or end-of-year grade point average this result is not

surprising.

Significance of the Study

First, the results of this study corroborate findings of previous research (see E. M.

Anderman & Wolters, 2006 and A. Kaplan & Maehr, 2007 for reviews) showing the differences between mastery- and performance-approach goals. Mastery-approach orientation was found to be positively related to academic values but not grade values (or perhaps grade goal setting) and had no significant relationship to end of term grades.

Performance-approach orientation was found to be significantly related to grade values

(or perhaps grade goal setting) and significantly (and positively) predicted end of quarter grade point average. Also, the relationships between school belongingness and academic

154 values, mastery-approach orientation, and self-efficacy found in prior research were also

found in the present study (Anderman, 2003; Anderman & Freeman, 2004). This study

also supported prior research in showing a significant negative relationship between

procrastination and self-efficacy (Bandura, 1997; Haycock, et al., 1998; Klassen, 2008;

Sirois, 2004; Steel, 2007; van Eerde, 2003; Wolters, 2003b) and a significant positive

relationship between procrastination and perceived stress (Flett, et al., 1995; Lay, et al.,

1989; Rothblum, et al., 1986; Schraw, et al., 2007; Tice & Baumeister, 1997; Solomon &

Rothblum, 1984). This corroboration is important in that it further bolsters our

understanding of key processes of achievement motivation.

Second, this study has also shown how a concern over social exclusion can

negatively influence self-regulation. While experimental research has clearly

demonstrated the deleterious effects of social exclusion on self-regulation (Baumeister, et

al., 2002; Baumeister, et al., 2005; DeWall & Baumeister, 2005) this is perhaps the first

study that has corroborated this relationship in the field.

Third, this study has shown how individual difference variables such as values

and the tendency to procrastinate can have significant influences on social, motivational,

and affective processes over the long term. Specifically, these individual difference

measures assessed early in an academic term were significantly related to school

belongingness, achievement goal orientations, self-efficacy, and perceived stress assessed six weeks later. In addition, this is the first study to show the potentially deleterious influence of procrastination on perceived school belongingness.

Fourth, and related to the third point, this study has demonstrated the relevance of

school belongingness, self-efficacy, and perceived stress to predicting grade point

155 average even after ability measures are controlled. Given the relevance of first quarter

grade point average in predicting the likelihood of graduation, understanding factors

beyond ability measures related to grade point average can be highly useful. This study

showed that social, motivational, and affective measures assessed three weeks prior to

first quarter freshmen’s first final examinations contributed significantly in predicting grade point average beyond that of ability measures alone. And while the total percentage of variability in grade point average accounted for in this study is not particularly high (only 16.7%) it is clearly better than the ability measures currently in use.

Finally, and perhaps most significantly, this study has shown the relevance of offering courses and programs in the first term designed to help students overcome procrastination and manage their time (e.g. Tuckman, Abry, & Smith, 2008). This study is founded on the assumption of the cyclical nature of self-regulatory, goal, and behavioral processes (e.g. Carver & Scheier, 1998; Vallacher & Wegner, 1987;

Zimmerman, 2000a). Clearly, programs designed to help students learn mastery skills

that can be utilized at a performance level will not only improve performance

(Zimmerman, & Kitsantas, 1997; 1999) but will also foster in students an increased sense

of self-efficacy and motivation (Zimmerman, 2000b). Increased self-efficacy will, in

turn, likely lead to perceptions of competence and expectations of successful task

performance (Bandura, 1997) promoting even further valuing of positive approach

mastery and performance orientation (Elliot, 2005; Elliot & Church, 1997). With an

increase in academic values, students are more likely to develop an intrinsic interest in

academic achievement goals further promoting the use of the deep processing strategies

156 characteristic of mastery goals (Harackiewicz, et al., 2008) and required for successful academic performance.

The current study has shown a strong negative relationship between academic values and procrastination. In addition, academic values can buffer the potentially deleterious effects of perceived social exclusion. Furthermore, the influence of academic values can have a positive cascading effect in students’ first college term by attenuating the effects of procrastination early in the term resulting in increased self-efficacy and higher levels of perceived school belongingness and decreased levels of perceived stress later in the term.

However, while the results of this study suggest that procrastination is less likely for students entering their first term already possessing high academic values, the cyclical nature of self-regulatory processes suggests that providing students with a pedagogy that fosters academic goal pursuit in small, yet challenging, doses may be able to increase academic task value for students whose valuing of academic tasks risks competition with other social/leisure values. This idea is based on the notion that task value accrues when tasks are specific, challenging (Locke & Latham, 1990) and successful thereby increasing self-efficacy (Bandura, 1997). Simply put, academic task value in addition to learning per se accrues by actually doing and succeeding at academic work. As noted earlier, action identification theory (Vallacher & Wegner, 1987) specifies that as people become comfortable at the lower performance level of the goal hierarchy, the more abstract levels of identification tend to emerge. Furthermore, as higher level goals tend to emerge, difficulty at this level provokes lower level solutions. This cyclical process can be manifested in many ways (e.g. Zimmerman, 2000a), but it is clearly important to foster in

157 students the competence and self-efficacy required for comfortable lower level performance.

A beneficial aspect of this notion has been demonstrated in a hybrid instructional approach combining traditional and computer-aided instruction (ADAPT - Active

Discovery And Participation through Technology) (Tuckman, 2002). This approach involves direct traditional instruction with over 200 computer-mediated activities designed to integrate instruction and feedback that “serves to make the latter a seamless part of the former and one that has relevance to the student far beyond earning a grade”

(Tuckman, 2002, p. 263). With instruction and assessment done on a daily basis, structure and discipline are fostered through active participation in learning that can promote transfer to other courses and situations (Tuckman, 2002).

A particular example of the benefit of the pedagogical approach for first quarter freshmen is provided in a recent study by Tuckman and Kennedy (2009). This study showed that students taking the Learning and Motivation Strategies course utilizing the

ADAPT instructional approach maintained a higher mean quarter grade point average throughout their first year and into their second year of study, were more likely to be retained during this period, and had higher graduation rates relative to students not enrolled in this program but of comparable ability and demographic makeup.

Furthermore, it is noteworthy that students making up the population of interest in this study tended to be below average in academic ability as assessed by standardized test scores and high school class rank. These results clearly show that the structure and discipline provided by this instructional approach transferred more generally and implies that a more positive intrinsic outlook on academic work was promoted.

158 The results of the current study perhaps provide one explanation for the benefits

of the ADAPT approach. Specifically, the implication of the results found using this

pedagogical approach is that fostering structure and discipline by integrating active

participation and feedback early in students’ tenure in college may help promote

academic values. That is, reducing the tendency to procrastinate increases the likelihood

of successful enactive experience. Furthermore, the integration of instruction and

feedback provide students with information regarding self-efficacy fostering further

successful performance. This likely increases value in these tasks thereby promoting a

more permanent tendency to not procrastinate over the long term. With a decrease in the tendency to procrastinate, further increases in self-efficacy are likely to result. In addition, the results of the current study have shown that long term decreases in procrastination are also likely to foster increases in feelings of school belongingness and

decreased perceptions of perceived stress that were clearly shown to have beneficial

academic results.

Limitations of the Study

This study has a number of limitations that need to be addressed. First, the data

reported in this study is based on a convenience sample that is not entirely representative

of the population requested to participate. Specifically, this sample of students who

agreed to participate was overrepresented with respect to high-ability students and

women and underrepresented with respect to minorities. Samples such as this always

require caution when generalizing the results to the desired population.

159 Second, since this study was correlational causal statements cannot be implied by

the directionality of the hypotheses. At best, the relationships between constructs

measured at different times are predictive. In addition, there is always the potential for

method variance to confound interpretations of relationships between constructs

measured at the same time. Thus, an assessment of causal relationships among these

constructs awaits experimental evidence.

Finally, and perhaps resulting in the most chagrin, is the inability of this study to

measure distinct performance-approach and performance-avoidance constructs. Much of

the literature in this area has shown that these constructs are related but distinct.

However, they were very highly positively correlated in this study. One reason for this may have been the wording of the items that resulted in students interpreting them as essentially equivalent (e.g., avoiding low grades is equivalent to approaching high grades). However, despite a general consensus regarding the distinction between these constructs, some other research has shown the potential for research participants either conflating or equating the constructs. As discussed in some detail above Urdan and

Mestas (2006) showed that students may not distinguish the constructs. In addition, even

Elliot and Murayama (2008), in a revision of the instrument used in this study, showed an attenuated correlation of 0.68 between performance-approach and performance- avoidance. That is, the correlation was between summed scores, not latent constructs. It is reasonable to assume that a disattenuated correlation would be even higher.

However, there may be even bigger problems for the performance goal construct in general. Brophy (2005) has pointed out that adopting performance-approach goals, while showing potentially beneficial results, may have detrimental effects in the long run.

160 Pointing to a study by Middleton, Kaplan, and Midgley (2004) that showed that sixth- grade students expressing high mathematics self-efficacy and performance-approach goals tended to shift to performance-avoidance goals in the seventh grade, his primary point is that we shouldn’t be recommending that teachers endorse performance goal orientation in their classes. But this seems to imply that performance-approach and performance-avoidance goals may not really be different goals, but rather different expressions of the same goal that depend on other factors (such as perceived self- efficacy) for the particular form it takes. Perhaps even more devastating to the performance goal construct is Brophy’s contention that it is epiphenomenal to the extent that its relationship with performance depends on students’ reflection on their past achievements. Specifically, students with a history (or at least belief) of good academic performance, may be more likely to express a desire to academically “best” other students. Notice, however, that as in the Middleton, et al. (2004) study, this epiphenomenal relationship seems dependent on perceived self-efficacy. The fate of the performance goal construct, when it is useful, and when it is not awaits further research.

Suggestions for Future Research

First, it is necessary to replicate this study and assess the relationships between values and self-regulation without the potential confounding influence of method variance as may have occurred in this study. That is, values and procrastination measures were taken at the same time and the relationships that were found may have been due in part to this confound. The same applies to the social, motivational, and affective

161 measures taken during the eighth of the term. Thus, it is now necessary to analyze parts

of this model in more detail.

Second, the relationship of school belonging to motivational, affective, and behavioral processes in educational contexts needs more study. Specifically, what are the

conditions under which perceived school belonging helps or hinders academic

performance? This study implies that students with a concern over social exclusion may

tend to focus on common-bond and relax common-identity group identification. This

may be problematic because common-bonds, while increasing one’s sense of being in a

group, may have the potential for distraction of the solitary academic work that is often

required for a university education. However, school belongingness also promotes mastery-approach goal orientations perhaps beyond that of academic values alone. This orientation is reminiscent of a common-identity group identification where students can feel a sense of the “school-in-them” where interest in academic pursuits are perhaps

individual and intrinsic.

Third, this also brings up the question of the relationship between school

belongingness and a performance-approach orientation. If the positive relationship

between these constructs is replicated (and in the case of performance-approach, this may

be questionable), is this due to a common-bond or common-identity identification? Thus

further research is needed to assess not only the replicability of this relationship, but also

to assess the extent to which school belongingness plays any role in the beneficial or

detrimental aspects of a performance-approach orientation. For example, performance

goals, to the extent that a need for belongingness is active and the tasks are in some way

diagnostic of satiating this need, may have beneficial effects on academic performance.

162 However, performance goals may have detrimental effects when the need to belong is active, the tasks are not diagnostic of satiating the need, and academic values related to grades are deemphasized. This may occur in situations where there are few performance markers (e.g. grades) to use as would be the case in courses with only one or two assessments per term. The upshot, as implied in DeWall, et al. (2008), is that students emphasizing performance goals may experience poorer performance when they feel included in classroom contexts where the demonstration of grades is emphasized. In this

case, the need to belong is satiated or otherwise deactivated and if students are truly more

interested in demonstrating rather than developing competence, feelings of inclusion may

ironically remove a primary driver for achievement

Finally, research is needed to replicate the finding of the deleterious influence of

procrastination on school belonging and, if successful, to address the question of why this

relationship occurs. Speculatively, it was hypothesized that procrastination may result in

distractions away from academic pursuits with a concomitant loss of university based

relationships. Research could address this phenomena by examining potential differences

in this relationship as a function of where students live (e.g., on campus, near campus,

with parents/guardians, etc). It would seem that if this hypothesis is correct, students

living off campus might show a stronger negative relationship between procrastination

and school belonging relative to students living on campus because these former students

may have potentially more distractions pulling them away from the college or university.

Conclusion

163 Education is an inherently social process primarily because the formation of self is based on collaborative and oppositional constructions with other people. Academic achievement, then, will depend to a large extent on how these social processes unfold over time. The things that interest students, the efforts they endure, and the things they attend to in order to achieve are ultimately rooted in these processes because of their intimate connection to one’s being. John Dewey (1913) wrote that

. . . when we recognize there are certain powers within the child urgent for development, needing to be acted out in order to secure their own efficiency and discipline, we have a firm basis upon which to build. Effort arises normally in the attempt to give full operation, and thus growth and completion, to these powers. Adequately to act upon these impulses involves seriousness, absorption, definiteness of purpose; it results in formation of steadiness and persistent habit in the service of worthy ends. But this effort never degenerates into drudgery, or mere strain of dead lift, because interest abides – the self is concerned throughout. Our first conclusion is that interest means unified activity. (pp. 14 – 15, emphasis in the original)

Effort toward the achievement of a goal depends on action bound to self. Without interest, effort degenerates into the “mere strain of dead lift”.

Anything that disrupts or promotes self-formation and self-transformation will ultimately disrupt or promote effort. Perceptions of potential exclusion or rejection from social relationships can be a potential source of (the lack of) effort and (the lack of) interest. Perceived social exclusion can be a source of stress and can influence achievement in the academic domain because it is when one is under the potential threat of rejection that one is motivated to attend to “unifying activity” in those places where the self is concerned.

164 Juvonen’s (2006) review of the relationship between belongingness and school functioning she concluded that “unmet or threatened sense of belonging is not only an

outcome of relationships, but also motivates changes in student behavior” (p. 669, emphasis in the original). A fuller understanding of Juvonen’s conclusion in the college student population needs to include an understanding of the values (interests) and self- regulatory tendencies (effort) students enter a university with and the potential

disruptions of interest and effort that a threatened sense of belonging can engender.

Threats to students’ perception of their relational value may have undesirable

consequences as they negotiate the various goals of their desire – the goals that make

them who they are.

165

References

Ainslie, G. (2001). Breakdown of will. Cambridge, UK: Cambridge University Press.

Ames, C., (1992). Classrooms: goals, structures, and student motivation. Journal of Educational Psychology, 84, 261 – 271.

Ames, C., & Archer, J. (1988). Achievement goals in the classroom: students’ learning strategies and motivation processes. Journal of Educational Psychology, 80, 260 – 267.

Ames, G., & Murray, F. B. (1982). When two wrongs make a right: promoting cognitive changes by social conflict. Developmental Psychology, 18, 894 – 898.

Anderman, E. M. (2002). School effects on psychological outcomes during adolescence. Journal of Educational Psychology, 94, 795 – 809.

Anderman, E. M., Austin, C. C., & Johnson, D. M. (2002). The development of goal orientation. In A. Wigfield & J. S. Eccles (Eds.) Development of achievement motivation. (pp. 197 – 220) San Diego, CA: Academic Press.

Anderman, E. M. & Maehr, M. L. (1994). Motivation and schooling in the middle grades. Review of Educational Research, 64, 287 – 309.

Anderman, E. M., & Wolters, C. A. (2006). Goals, values, and affect: influences on student motivation. In P. A Alexander & P. H. Winne (Eds.), Handbook of educational psychology (2nd ed. pp. 369 – 389). Mahwah, NJ: Lawrence Erlbaum Associates.

Anderman, L. H. (1999). Classroom goal orientation, school belonging and social goals as predictors of students’ positive and negative affect following the transition to middle school. Journal of Research and Development in Education, 32, 89 – 103.

Anderman, L. H. (2003). Academic and social perceptions as predictors of change in middle school students’ sense of school belonging. Journal of Experimental Education, 72, 5 – 22.

166

Anderman, L. H., & Anderman, E. M. (1999). Social predictors of changes in students’ achievement goal orientations. Contemporary Educational Psychology, 25, 21 – 37.

Anderman, L H., & Freeman, T. M. (2004). Students’ sense of belonging in school. In P. R. Pintrich & M. L. Maehr (Eds.), Advances in motivation and achievement: Vol 13. Motivating students, improving schools: the legacy of Carol Midgley (pp. 27 – 63). Amsterdam: Elsevier.

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: a review and recommended two-step approach. Psychological Bulletin, 103, 411 – 432.

Asparouhov, T. and Muthén, B. (2005). Multivariate Statistical Modeling with Survey Data. Proceedings of the Federal Committee on Statistical Methodology (FCSM) Research Conference. Retrieved May 6, 2009 from http://www.fcsm.gov/05papers/Asparouhov_Muthen_IIA.pdf

Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: structure, process, and content. Psychological Bulletin, 120, 338 – 375.

Bandalos, D. L. (2002). The effects of item parceling on goodness-of-fit and parameter estimate bias in structural equation modeling. Structural Equation Modeling, 9, 78 – 102.

Bandalos, D. L., & Finney, S. J. (2001). Item parceling issues in structural equation modeling. In G. A. Marcoulides & R. E. Schumacker (Eds.), Advanced structural equation modeling: new developments and techniques. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Bandura, A. (1986). Social foundations of thought and action: a social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.

Bandura, A. (1997). Self-efficacy: the exercise of control. New York: W. H. Freeman.

Bandura, A., Barbaranelli, C., Caprara, G.V., & Pastorelli, C. (1996). Multifaced impact of self-efficacy beliefs on academic functonning. Child Development, 67, 1206-1222.

Battistich, V., & Hom, A. (1997). The relationship between students’ sense of their school as a community and their involvement in problem behaviors. American Journal of Public Health, 87, 1997 – 2001.

Battistich, V., Solomon, D., Kim, D., Watson, M., & Schaps, E. (1995). Schools as communities, poverty levels of student populations, and students’ attitudes,

167

motives, and performance: a multilevel analysis. American Educational Research Journal, 32, 627 – 658.

Berndt, T. J. (1999). Friends’ influence on students’ adjustment to school. Educational Psychologist, 34, 15 – 28.

Berndt, T. J., & Keefe, K. (1995). Friends' influence on adolescents' adjustment to school. Child Development, 66, 1312 – 1329.

Baumeister, R. F. (2002). Ego depletion and self-control failure: an energy model of the self’s executive function. Self and Identity, 1, 129 – 136.

Baumeister, R. F., & DeWall, C. N (2005). The inner dimension of social exclusion: intelligent thought and self-regulation among rejected persons. In K. D. Williams, J. P. Forgas, & W. von Hippel, (Eds.), The social outcast: ostracism, social exclusion, rejection, and bullying (pp. 53 – 73). New York: Psychology Press.

Baumeister, R. F., Bratslavsky, E., Muraven, M., Tice, D. M. (1998). Ego depletion: is the self a limited resource Journal of Personality and Social Psychology, 74, 1252 – 1265.

Baumeister, R. F., DeWall, C. N., Ciarocco, N. J., & Twenge, J. M. (2005). Social exclusion impairs self-regulation. Journal of Personality and Social Psychology, 88, 589 – 604.

Baumeister, R. F. & Heatherington, T. F. (1996). Self-regulation failure: an overview. Psychological Inquiry, 7, 1 – 15.

Baumeister, R. F. & Leary, M. R. (1995). The need to belong: desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497 – 529.

Baumeister, R. F., Muraven, M., & Tice, D. M. (2000). Ego depletion: a resource model of volition, self-regulation, and controlled processing. Social Cognition, 18, 130 – 150.

Baumeister, R. F. & Tice, D. M. (1990). Anxiety and social exclusion. Journal of Social and Clinical Psychology, 9, 165 – 195.

Baumeister, R. F., Twenge, J. M., & Nuss, C. K. (2002). Effects of social exclusion on cognitive processes: Anticipated aloneness reduces intelligent thought. Journal of Personality and Social Psychology, 83, 817 – 827.

168

Bembenutty, H. (2008). Academic delay of gratification and expectancy-value. Personality and Individual Differences, 44, 193 – 202.

Bembenutty, H., & Karabenick, S. A. (2004). Inherent association between academic delay of gratification, future time perspective, and self-regulated learning. Educational Psychology Review, 16, 35 – 57.

Ben-Ze’ev, A. (2000). The subtlety of emotions. Cambridge, MA: MIT Press.

Blunt, A., & Pychyl, T. A. (2000). Task aversiveness and procrastination: a multidimensional approach to task aversiveness across stages of personal projects. Personality and Individual Differences, 28, 153–167.

Bolger, N., DeLongis, A., Kessler, R. C., & Schilling, E. A. (1989). Effects of daily stress on negative mood. Journal of Personality and Social Psychology, 57, 808 – 818.

Borkowski, J. G., & Muthukrishna, N. (1992). Moving metacognition into the classroom: “Working models” and effective strategy teaching. In M. Pressley, K. R. Harris, & J. T. Guthrie (Eds.), Promoting academic literacy: Cognitive research and instructional innovation. (pp 477 – 501). Orlando, FL: Academic Press.

Borkowski, J. G., & Thorpe, P. K. (1994). Self-regulation and motivation: A life-span perspective on underachievement. In D. H. Schunk & B. J. Zimmerman (Eds.), Self-regulation of learning and performance: Issues and Educational Applications. (pp 45 – 73). Hillsdale, NJ: Lawrence Erlbaum Associates.

Breland, B. T., & Donovan, J. J. (2005). The role of state goal orientation in the goal establishment process. Human Performance, 18, 23 – 53.

Brewer, M. B. (1991). The social self: on being similar and different at the same time. Personality and Social Psychology Bulletin, 17, 475 – 482.

Brewer, M. B. (2003). Optimal distinctiveness, social identity, and the self. In M. Leary and J. Tangney (Eds.), Handbook of Self and Identity. (pp 480 – 490). New York: Guilford.

Brewer, M. B., & Gardner, W. (1996). Who is this “we”? Levels of collective identity and self representation. Journal of Personality and Social Psychology, 71 83 – 93.

Brophy, J. (2005). Goal theorists should move on from performance goals. Educational Psychologist, 40, 167 – 176.

169

Cairns, R. B., Cairns, B. D., & Neckerman, H. J. (1989). Early school dropout: configurations and determinants. Child Development, 60, 1437 – 1452.

Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. Cambridge UK: Cambridge University Press.

Cattell, R. B., & Burdsal, C. A. (1975). The radial parcel double factoring design: a solution to the item-vs-parcel controversy. Multivariate Behavioral Research, 10, 165 – 179.

Chemers, M. M., Hu, L., & Garcia, B. F. (2001). Academic self-efficacy and first- year college student performance and adjustment, Journal of Educational Psychology, 93, 55 – 64.

Coffman, D. L., & MacCallum, R. C. (2005). Using parcels to convert path analysis models into latent variable models. Multivariate Behavioral Research, 40, 235 – 259.

Cohen, J., & Cohen, P. (1975). Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum.

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385 – 396.

Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310 – 357.

Comrey, A. L. (1978). Common methodological problems in factor analytic studies. Journal of Consulting and Clinical Psychology, 46, 648 – 659.

Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-processing mechanisms in children’s social development. Psychological Bulletin, 115, 74 – 101.

Cury, F., Elliot, A. J., Da Fonseca, D., & Moller, A. C. (2006). The social-cognitive model of achievement motivation and the 2  2 achievement goal framework. Journal of Personality and Social Psychology, 90, 666 – 679.

Deci, E. L., Vallerand, R. J., Pelletier, L. G., & Ryan, R. M. (1991). Motivation and education: the self-determination perspective. Educational Psychologist, 26, 325 – 346.

Demerath, P. (2000). The social cost of acting “extra”: Students’ moral judgments of self, social relations, and academic success in Papua New Guinea. American Journal of Education, 108, 196 – 235.

170

Demerath, P. (2003). Negotiating individualis and collectivist futures: emerging subjectivities and social forms in Pupua New Guinean high schools. Anthropology & Education Quarterly, 34, 136 – 157.

DeShon R. P., & Gillespie, J. Z. (2005). A motivated action theory account of goal orientation. Journal of Applied Psychology, 90, 1096 – 1127.

DeWall, C. N., & Baumeister, R. F. (2006). Alone but feeling no pain: Effects of social exclusion on physical pain tolerance and pain threshold, affective forecasting, and interpersonal empathy. Journal of Personality and Social Psychology, 91, 1 – 15.

DeWall, C. N., Baumeister, R. F., & Vohs, K. D. (2008). Satiated with belongingness? Effects of acceptance, rejection, and task framing of self- regulatory performance Journal of Personality and Social Psychology, 95, 1367 – 1382.

Dewey, J. (1913). Interest and effort in education. Boston: Houghton Mifflin.

Dietz, F., Hofer, M., & Fries, S. (2007). Individual values, learning routines and academic procrastination. British Journal of Educational Psychology, 77, 893 – 906.

Dishion, T. J., Spracklen, K. M., Andrews, D. W., & Patterson, G. R. (1996). Deviancy training in male adolescent friendships. Behavior Therapy, 27, 373 – 390.

Dixon Rayle, A., & Chung, K-Y. (2007). Revisiting first-year college students’ mattering: social support, academic stress, and the mattering experience. Journal of College Student Retention, 9, 21 – 37.

Dweck, C. S., & Leggett, E. L. (1988). A social cognitive approach to motivation and personality. Psychological Review, 95, 256 – 273.

Duckworth, A. L., & Seligman, M. E. P. Self-Discipline Outdoes IQ in Predicting Academic Performance of Adolescents. Psychological Science, 16, 939 – 944.

Eccles (Parsons), J. S., Adler, T. F., Futterman, R., Goff, S. B., Kaczala, C. M., Meece, J. L., & Midgley, C. (1983). Expectancies, values, and academic behaviors. In J. T. Spence (Ed.), Achievement and achievement motivation: psychological and sociological approaches (pp. 75 – 146). San Francisco: Freeman.

171

Eccles, J. S. (2005). Subjective task value and the Eccles et al. model of achievement- related choices. In A. J. Elliot & C. S. Dweck (Eds.), Handbook of competence and motivation. (pp. 105 – 122). New York: Guilford.

Eisenberger, N. I., Lieberman, M. D., & Williams, K. (2003). Does rejection hurt? an fMRI study of social exclusion. Science, 302, 290 – 292.

Elliot, A. J. (1999). Approach and avoidance motivation and achievement goals. Educational Psychologist, 34, 149 – 169

Elliot, A. J. (2005). A conceptual history of the achievement goal construct. In A. J. Elliot & C. S. Dweck (Eds.), Handbook of competence and motivation. (pp. 52 – 72). New York: Guilford.

Elliot, A. J., & Church, M. A. (1997). A hierarchical model of approach and avoidance achievement motivation. Journal of Personality and Social Psychology, 72 218 – 232.

Elliot, A. J., & McGregor, H. A. (2001). A 2  2 achievement goal framework. Journal of Personality and Social Psychology, 80, 501 – 519.

Elliot, A. J., & McGregor, H. A., & Gable, S. (1999). Achievement goals, study strategies, and exam performance: a mediational analysis. Journal of Educational Psychology, 91, 549 – 563.

Elliot, A. J., & Murayama, K. (2008). On the measurement of achievement goals: critique, illustration, and application, Journal of Educational Psychology, 100, 613 – 628.

Elliot, A. J., & Thrash, T. M. (2001). Achievement goals and the hierarchical model of achievement motivation. Educational Psychology Review, 12 139 – 156.

Elliot, E. S., & Dweck, C. S. (1988). Goals: an approach to motivation and achievement. Journal of Personality and Social Psychology, 54, 5 – 12.

Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research, Psychological Methods, 4, 272 – 299.

Ferrari, J. R. (2001). Procrastination as self-regulation failure of performance: effects of cognitive load, self-awareness, and time limits on ‘working best under pressure’. European Journal of Personality, 15, 391 – 406.

172

Farrari, J. R., & Emmons, R. A. (1995). Methods of procrastination and their relation to self-control and self-reinforcement. Journal of Social Behavior and Personality, 10, 135 – 142.

Feather, N. T. (1988). Values, valences, and course enrollment: testing the role of personal values within an expectancy-value framework. Journal of Educational Psychology, 80, 381 – 391.

Finn, J. D. (1989). Withdrawing from school. Review of Educational Research, 59, 117 – 142.

Fitzsimons, G. M., & Bargh, J. A. (2003). Thinking of you: nonconscious pursuit of interpersonal goals associated with relationship partners. Journal of Personality and Social Psychology, 84, 148 – 164.

Flett, G. L., Blankstein, K. R., & Martin, T. R. (1995). Procrastination, negative self- evaluation, and stress in depression and anxiety. In J. R. Ferrari, J. L. Johnson, & W. McCown (Eds.), Procrastination and task avoidance: Theory, research, and treatment (pp. 137 – 167). New York: Plenum.

Folkman, S., & Lazarus, R. S. (1985). If it changes it must by a process: study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology, 48, 150 – 170.

Ford, M. E. (1992). Motivating humans: Goals, emotions, and personal agency beliefs. Newbury Park, CA: Sage Publications.

Fordham, S., & Ogbu, J. U. (1986). Black students’ school success: coping with the “burden of ‘acting White’”. The Urban Review, 18, 176 – 205.

Freeman, T. M., Anderman, L. H., & Jensen, J. M. (2007). Sense of belonging in college freshmen at the classroom and college levels. The Journal of Experimental Education, 75, 203 – 220.

Fries, S., & Dietz, F. (2007). Learning in the face of temptation: the case of motivation interference. Journal of Experimental Education, 76, 93 – 112.

Garcia, T., & Pintrich, P. R. (1994). Regulating motivation and cognition in the classroom: the role of self-schemas and self-regulatory strategies. In D. H. Schunk & B. J. Zimmerman (Eds.). Self-regulation of learning and performance: Issues and educational applications. (pp. 127 – 153). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

173

Gardner, W. L., Pickett, C. L., & Brewer, M. B. (2000). Social exclusion and selective memory: how the need to belong influences memory for social events. Personality and Social Psychology Bulletin, 26, 486 – 496.

Gerhardt, M. W., & Brown, T. K. G. (2006). Individual differences in self-efficacy development: the effects of goal orientation and affectivity. Learning and Individual Differences, 16, 43 – 59.

Gollwitzer, P. M. (1999). Implementation intentions: strong effects of simple plans. American Psychologist, 54, 493 – 503.

Gonzalez, R., & Griffin, D. (2001). Testing parameters in structural equation modeling: every “one” matters. Psychological Methods, 6, 258 – 269.

Goodenow, C. (1993a). Classroom belonging among early adolescent students: relationships to motivation and achievement. Journal of Early Adolescence, 13, 21 – 43.

Goodenow, C. (1993b). The psychological sense of school membership among adolescents: scale development and educational correlates. Psychology in the Schools, 30, 79 – 90.

Goodenow, C., & Grady, K. E. (1993). The relationship of school belonging and friends’ values to academic motivation among urban adolescent students. Journal of Experimental Education, 62, 60 – 71.

Gore, P. A. (2006). Academic self-efficacy as a predictor of college outcomes: two incremental validity studies. Journal of Career Assessment, 14, 92 – 115.

Gutman, L. M., & Midgley, C. (2000). The role of protective factors in supporting the academic achievement of poor African American students during the middle school transition. Journal of Youth and Adolescence, 29, 223 – 248.

Gröpel, P., & Steel, P. (2008). A mega-trial investigation of goal setting, interest enhancement, and energy of procrastination. Personality and Individual Differences, 45, 406 – 411.

Grewal, R., Cote, J. A., & Baumgartner, H. (2004). Multicollinearity and measurement error in structural equation models: implications for theory testing. Marketing Science, 23, 519 – 529.

Grice, J. W. (2001). Computing and evaluating factor scores. Psychological Methods, 6, 430 – 450.

174

Hall, N. C., Chipperfield, J. G., Perry, R. P., Ruthig, J. C., & Goetz, T. (2006). Primary and secondary control in academic development: gender-specific implications for stress and health in college students. Anxiety, Stress, and Coping, 19, 189 – 210.

Hall, R. J., Snell, A. F., & Singer Foust, M. (1999). Item parceling strategies in SEM: Investigating the subtle effects of unmodeled secondary constructs. Organizational Research Methods, 2, 233–256.

Harackiewicz, J. M., Barron, K. E., & Elliot, A. J. (1998). Rethinking achievement goals: when are they adaptive for college students and why? Educational Psychologist, 33, 1 – 21.

Harackiewicz, J. M., Barron, K. E., Pintrich, P. R., Elliot, A. J., & Thrash, T. M. (2002). Revision of achievement goal theory: necessary and illuminating. Journal of Educational Psychology, 94, 638 – 645.

Harackiewicz, J. M., Durik, A. M, Barron, K. E., Linnenbrink-Garcia, L, & Tauer, J. M. (2008). The role of achievement goals in the development of interest: reciprocal relationship between achievement goals, interest, and performance. Journal of Educational Psychology, 100, 105 – 122.

Hayashi, K., Bentler, P. M., & Yuan, K-H. (2007). On the likelihood ratio test for the number of factors in exploratory factor analysis. Structural Equation Modeling, 14, 505 – 526.

Haycock, L. A., McCarthy, P., & Skay, C. L. (1998). Procrastination in college students: the role of self-efficacy and anxiety. Journal of Counseling and Development, 76, 317 – 324.

Heaven, P. C. L. (2001). The social psychology of adolescence. New York: Palgrave.

Henderson-King, D., & Smith, M. N. (2006). Meanings of education for university students: academic motivation and personal values as predictors. Social Psychology of Education, 9, 195 – 221.

Hidi, S., & Renninger, K. A. (2006). The four-phase model of interest development. Educational Psychologist, 41, 111 – 126.

Hofer, M., Schmid, S., Fries, S., & Dietz, F., Clausen, M., & Reinders, H. (2007). Individual values, motivational conflicts, and learning for school. Learning and Instruction, 17, 17 – 28.

Hogg, M. A. (2005). All animals are equal but some are more equal than others: social identity and marginal membership. In K. D. Williams, J. P. Forgas, &

175

W. von Hippel, (Eds.), The social outcast: ostracism, social exclusion, rejection, and bullying (pp. 243 – 261). New York: Psychology Press.

Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis. Psychometrica, 30, 179 – 185.

Howell, A. J., & Watson, D. C. (2007). Procrastination: associations with achievement goal orientation and learning strategies. Personality and Individual Differences, 43, 167 – 178.

Howell, A. J., Watson, D. C., Powell, R. A., & Buro, K. (2006). Academic procrastination: the pattern and correlates of behavioral postponement. Personality and Individual Differences, 40, 1519 – 1530.

Hymel, S., Comfort, C., Schonert-Reichl, K., & Mcdougall, P. (1996). Academic failure and school dropout: the influence of peers. In J. Juvonen, & K. R. Wentzel (Eds.), Social motivation: Understanding children’s school adjustment (pp. 313 – 345). New York: Cambridge University Press.

Juvonen, J. (2006). Sense of belonging, social bonds, and school functioning. In P. A Alexander & P. H. Winne (Eds.), Handbook of educational psychology (2nd ed. pp. 655 – 674). Mahwah, NJ: Lawrence Erlbaum Associates.

Kahn, J. H., & Nauta, M. M. (2001). Social-cognitive predictors of first-year college persistence: the importance of proximal assessment. Research in Higher Education, 42, 633-652.

Kaplan, A., & Maehr, M. L. (2007). The contributions of prospects of goal orientation theory. Educational Psychology Review, 19, 141 – 184.

Kaplan, D. S., Peck, B. M., & Kaplan, H. B. (1997). Decomposing the academic failure-dropout relationship: a longitudinal analysis. Journal of Educational Research, 90, 331 – 343.

Kennedy, G. J. (2008). [Relationship between first quarter grade point average and graduation rate for three cohorts.] Unpublished raw data.

Kindermann, T. A. (1993). Natural peer groups as contexts for individual development: the case of children’s motivation in school. Developmental Psychology, 29, 970 – 977.

Kindermann, T. A., McCollam, T. L., & Gibson, E. (1996). Peer networks and students’ classroom engagement during childhood and adolescence. In J. Juvonen, & K. R. Wentzel (Eds.), Social motivation: Understanding

176

children’s school adjustment (pp. 279 – 312). New York: Cambridge University Press.

Kishton, J. M., & Widaman, K. F. (1994). Unidimensional versus domain representative parceling of questionnaire items: an empirical example. Educational and Psychological Measurement, 54, 757 – 765.

Klassen, R. M., Krawchuk, L. L., & Rajani, S. (2008). Academic procrastination of undergraduates: low self-efficacy to self-regulate predicts higher levels of procrastination. Contemporary Educational Psychology, 33, 915 – 931.

König, C., & Kleinmann, M. (2004). Business before pleasure: no strategy for procrastinators? Personality and Individual Differences, 37, 1045 – 1057.

Kopp, C. B. (1982). Antecedents of self-regulation: a developmental perspective. Developmental Psychology, 18, 199 – 214.

Kuhl, J. (1994). Motivation and volition. In G. d’Ydewalle, P. Eelen, & P. Bertelson (Eds.), International perspectives on psychological science (Vol. 2, pp. 311 – 340). Hillsdale, NJ: Erlbaum.

Lasane, T. P., & Jones, J. M. (2000). When socially induced temporal myopia interferes with academic goal-setting. Journal of Social Behavior and Personality, 15, 75 – 86.

Lazarus, R. S. (2001). Relational meaning and discrete emotions. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion (pp. 37 – 67). Oxford: Oxford University Press.

Lay, C. H. (1988). The relation of procrastination and optimism to judgements of time to complete an essay and anticipation of setbacks. Journal of Social Behavior and Personality, 3, 201 – 214.

Lay, C.H., Edwards, J. M., Parker, J. D., & Endler, N. S. (1989). An assessment of appraisal, anxiety, coping, and procrastination during an examination period. European Journal of Personality, 3, 195 – 208.

Leary, M. R., & Downs, D. L. (1995). Interpersonal functions of the self-esteem motive: The self-esteem system as a sociometer. In M. H. Kernis (Ed.). Efficacy, agency, and self-esteem. (pp. 123 – 144). New York: Plenum Press.

Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an interpersonal monitor: The sociometer hypothesis. Journal of Personality and Social Psychology, 68, 518 – 530.

177

Little, B. R. (1983). Personal projects: A rationale and method for investigation. Environment and Behavior, 15, 273 – 309.

Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel: exploring the question, weighing the merits. Structural Equation Modeling, 9, 151 – 173.

Little, T. D., Lindenberger, U., Nesselroade, J. R. (1999). On selecting indicators for multivariate measurement and modeling with latent variables: when “good” indicators are bad and “bad” indicators are good. Psychological Methods, 4, 192 – 211.

Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance. Englewood Cliffs, NJ: Prentice Hall.

Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: a 35-year odyssey. American Psychology, 57, 705 – 717.

Lumley, M. A., & Provenzano, K. M. Stress management through written emotional disclosure improves academic performance among college students with physical symptoms. Journal of Educational Psychology, 95, 641 – 649.

MacCallum, R. C., & Austin, J. T. (2000). Applications of structural equation modeling in psychological research. Annual Review of Psychology, 51, 201 – 226.

MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods, 1, 130 – 149.

MacDonald, G. & Leary, M. R. (2005). Why does social exclusion hurt? The relationship between social and physical pain. Psychological Bulletin, 131, 202 – 223.

Markus, H., & Wurf, E. (1987). The dynamic self-concept: a social psychological perspective. Annual Review of Psychology, 38, 299 – 337.

Marsh, H. W., Dowson, M., Pietsch, J., & Walker, R. (2004). Why multicollinearity matters: a reexamination of relationsh between self-efficacy, self-concept, and achievement. Journal of Educational Psychology, 96, 518 – 522.

Marsh, H. W., Hau, K-T, Balla, J. R., & Grayson, D. (1998). Is more ever too much? The number of indicators per factor in confirmatory factor analysis. Multivariate Behavioral Research, 33, 181 – 220.

178

McClelland D. C. (1985). Human motivation. Glenview, IL: Scott, Foresman and Company.

McClelland, D. C., Koestner, R., & Weinberger, J. (1989). How do self-attributed and implicit motives differ? Psychological Review, 96, 690 – 702.

McCombs, B. L, & Marzano, R. J. (1990). Putting the self in self-regulated learning: the self as agent in integrating will and skill. Educational Psychologist, 25, 51 – 69.

McCrae, R. R., & Costa, P. T. (1988). Personality trait structure as a human universal. American Psychologist, 52, 509 – 516.

McGregor, H. A., & Elliot, A. J. (2002). Achievement goals as predictors of achievement-relevant processes prior to task engagement. Journal of Educational Psychology, 94, 381 – 395

Metcalfe, J., & Mischel, W. (1999). A hot/cool-system analysis of delay of gratification: dynamics of willpower. Psychological Review, 106, 3 – 19.

Middleton, M., Kaplan, A., Midgley, C. (2004). The change in middle school students’ achievement goals in mathematics over time. Social Psychology of Education, 7, 127 – 159.

Middleton, M., & Midgley, C. (1997). Avoiding the demonstration of lack of ability: an underexplored aspect of goal theory. Journal of Educational Psychology, 89, 719 – 718.

Midgley, C., Anderman, E. M., & Hicks, L. (1995). Differences between elementary and middle school teacher and students: a goal theory approach. Journal of Early Adolscence, 15, 90 – 113.

Midgley, C., Kaplan, A., Middleton, M. (2001). Performance-approach goals: Good for what, for whom, under what circumstances, and at what cost? Journal of Educational Psychology, 93, 77 – 86.

Midgley, C., Kaplan, A., Middleton, M., Maehr, M. L., Urdan, T., Hicks Anderman, L., et al. (1998). The development and validation of scales assessing students’ achievement goal orientations. Contemporary Educational Psychology, 23, 113 – 131.

Midgly, C., & Urdan, T. (2001). Academic self-handicapping and achievement goals: a further examination. Contemporary Educational Psychology, 26, 61 – 75.

179

Millar, K. U., Tesser, A., & Millar, M. G. (1988). The effects of a threatening life event on behavior sequences and intrusive thought: a self-disruption explanation. Cognitive Therapy and Research, 12, 441 – 458.

Mischel, W., & Baker, N. (1975). Cognitive appraisals and transformations in delay behavior. Journal of Personality and Social Psychology, 31, 254 – 261.

Mischel, W., & Ebbesen, E. B. (1970). Attention in delay of gratification. Journal of Personality and Social Psychology, 16, 329 – 337.

Montanelli, R. G., & Humphreys, L. G. (1976). Latent roots of random data correlation matrices with squared multiple correlations on diagonal – Monte- carlo study. Psychometrika, 41, 341 – 348.

Moon, S. M., & Illingworth, A. J. (2005). Exploring the dynamic nature of procrastination: a latent growth curve analysis of academic procrastination. Personality and Individual Differences, 28, 297 – 309.

Mounts, N. S., & Steinberg, L. (1995). An ecological analysis of peer influence on adolescent grade point average and drug use. Developmental Psychology, 31, 915 – 922.

Muraven, M., & Baumeister, R. F. (2000). Self-regulation and depletion of limited resources: does self-control resemble a muscle? Psychological Bulletin, 126, 247 – 259. Muthén, L. K. & Muthén, B. O. (2007). Mplus (Version 5.2) [Computer software and manual]. Los Angeles, CA: Muthén & Muthén.

Nicholls, J. G. (1984). Achievement motivation: conceptions of ability, subjective experience, task choice, and performance. Psychological Review, 91, 328 – 346.

Newman, B. M., Lohman, B. J., & Newman, P. R. (2007). Peer group membership and a sense of belonging: their relationship to adolescent behavior problems. Adolescence, 42, 241 – 263.

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory. (3rd ed.). New York: McGraw-Hill.

O’Connor, B. P. (2000). SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behavior Research Methods, Instruments, & Computers, 32, 396 – 402.

Oaten, M., & Cheng, K. (2006). Inproved self-control: the benefits of a regular program of academic study. Basic and Applied Social Psychology, 28, 1 – 16.

180

Osterman, K. F. (2000). Students’ need for belonging in the school. Review of Educational Research, 70, 323 – 367.

Pajares, F. (1996). Self-efficacy beliefs in academic settings. Review of Educational Research, 66, 453 – 578.

Patrick, H. (1997). Social self-regulation: exploring the relations between children’s social relationship, academic self-regulation, and school performance. Educational Psychologist, 32, 209 – 220.

Piaget, J. (1977/1985). The equilibration of cognitive structures: The central problem of intellectual development (T. Brown & K. J. Thampy, Trans.). Chicago: The University of Chicago Press. (Original work published 1975 as L’équilibration des structures cognitives: Problème central du developpement)

Pintrich, P. R. (2000a). An achievement goal theory perspective on issues in motivation terminology, theory, and research. Contemporary Educational Psychology, 25, 92 – 104.

Pintrich, P. R. (2000b). The role of goal orientation in self-regulated learning. In M. Boekaerts, P. Pintrich, & M. Zeidner (Eds.), Handbook of self-regulation: theory, research, and applications (pp. 451 – 502). San Diego, CA: Academic Press.

Pintrich, P. R., & De Groot, E. V. (1990). Motivational and self-regulated learning components of classroom academic performance. Journal of Educational Psychology, 82, 33 – 40.

Pintrich, P. R., Smith, D., Garcia, T., & McKeachie, W. J. (1993). Reliability and predictive validity of the Motivated Strategies for Learning Questionnaire (MSLQ). Educational and Psychological Measurement, 53, 810 – 813.

Pittman, L. D., & Richmond, A. (2007). Academic and psychological functioning in late adolescence: the importance o school belonging. Journal of Experimental Education, 75, 270 – 290.

Pittman, L. D., & Richmond, A. (2008). University belonging, friendship quality, and psychological adjustment during the transition to college. Journal of Experimental Education, 76, 343 – 361.

Phillips, J. M., & Gully, S. M. (1997). Role of goal orientation, ability, need for achievement and locus of control in the self-efficacy and goal-setting process. Journal of Applied Psychology, 82, 792 – 802.

181

Postmes, T., Spears, R., Lee, A. T., & Novak, R. J. (2005). Individuality and social influence in groups: inductive and deductive routes to group identity. Journal of Personality and Social Psychology, 89, 747 – 763.

Powers, W. T. (1973). Behavior: the control of perception. Chicago: Aldine.

Prentice, D. A., Miller, D. T., & Lightdale, J. R. (1994). Asymmetries in attachments to groups and to their members: distinguishing between common-identity and common-bond groups. Personality and Social Psychology Bulletin, 20, 484 – 493.

Pychyl, T. A., Lee, J., Thibodeau, R., & Blunt, A. (2000). Five days of emotion: an experience-sampling study of undergraduate student procrastination. Journal of Social Behavior and Personality, 15, 239–254.

Roeser, R. W. (2004). Competing schools of thought in achievement goal theory? In P. R. Pintrich & M. L. Maehr (Eds.), Advances in motivation and achievement: Vol. 13. Motivating students, improving schools: The legacy of Carol Midgley (pp. 361 – 392). Oxford, England: Elsevier.

Rothblum, E. D., Solomon, L. J., & Murakami, J. (1986). Affective, cognitive, and behavioral differences between high and low procrastinators. Journal of Counseling Psychology, 33, 387 – 394.

Ryan, A. M. (2001). The peer group as a context for the development of young adolescent motivation and achievement. Child Development, 72, 1135 – 1150.

Ryan, A. M. Hicks, L., & Midgley, C. (1997). Social goals, academic goals, and avoiding seeking help in the classroom. Journal of Early Adolescence, 17, 152 – 171.

Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: classic definitions and new directions. Contemporary Educational Psychology, 25 54 – 67.

Roeser, R. W., Eccles, J. S., & Sameroff, A. J. (1998). Academic and emotional functioning in early adolescence: longitudinal relations, patterns, and prediction by experience in middle school. Development and Psychopathology, 10, 321 – 352.

Roeser, R. W., Midgley, C., & Urdan, T. C. (1996). Perceptions of the school psychological environment and early adolescents’ psychological and behavioral functioning in school: the mediating role of goals and belonging. Journal of Educational Psychology, 88, 408 – 422.

182

Rosenberg, M., & McCullough, B. (1981). Mattering: inferred significance and mental health among adolescents. Research in Community and Mental Health, 2, 163 – 182.

Rothblum, E. D., Solomon, L. J., & Murakami, J. (1986). Affective, cognitive, and behavioral differences between high and low procrastinators. Journal of Counseling Psychology, 33, 387 – 394.

Rummel, R. J. (1970). Applied factor analysis. Evanston, IL: Northwestern University Press.

Schmeichel, B. J., & Baumeister, R. F. (2004). Self-regulatory strength. In R. F. Baumeister, & K. D. Vohs, (Eds.), Handbook of Self-regulation. (pp. 84 – 98). New York: Guilford Press.

Schönemann, P. H., & Wang, M-W (1972). Some new results on factor indeterminacy. Psychometrika, 37, 61 – 91.

Schouwenburg, H. C. (2004). Procrastination in academic settings: general introduction. In H. C. Schouwenburg, C. H. Lay, T. A. Pychyl, & J. R. Ferrari (Eds.), Counseling the procrastinator in academic settings (pp. 3 – 17). Washingtion, DC: American Psychological Association.

Schouwenburg, H. C., & Groenewoud, J. T. (2001). Study motivation under social tempation: effects of trait procrastination. Personality and Individual Differences, 30, 229 – 240.

Schouwenburg, H. C., & Lay, C. H. (1995). Trait procrastination and the big-five factors of personality. Personality and Individual Differences, 18, 481 – 490.

Schraw, G., Wadkins, T., & Olafson, L. (2007). Doing the things we do: a grounded theory of academic procrastination. Journal of Educational Psychology, 99, 12 – 25.

Schumacker, R. E., & Lomax, R. G. (2004). A beginner’s guide to structural equation

modeling. Mahwah, NJ: Lawrence Erlbaum Associates.

Schunk, D. H. (1991). Self-efficacy and academic motivation. Educational Psychology, 26, 207 – 231.

Schunk, D. H., & Pajares, F. (2005). Competence beliefs and academic functioning. In A. J. Elliott & C. S. Dweck (Eds.), Handbook of competence and motivation, (pp. 85 – 104). New York, NY: Guilford Press.

183

Schunk, D. H., Pintrich, P. R., & Meece, J. L. (2008). Motivation in education (3rd ed.). Upper Saddle River, NJ: Pearson.

Schunk, D. H., & Zimmerman, B. J (2006). Competence and control beliefs: distinguishing the means and ends. In P. A. Alexander, & P. H. Winne (Eds.), Handbook of educational psychology, (p. 349 – 367). Washington, DC: American Psychological Association.

Schwartz, S. H. (1994). Are there universal aspects in the structure and contents of human values? Journal of Social Issues, 50, 19 – 45.

Seijts, G. H., & Latham, G. P. (2005). Learning versus performance goals: When should each be used? Academy of Management Executive, 19, 124 – 131.

Senecal, C., Julien, E., & Guay, F. (2003). Role conflict and academic procrastination: a self-determination perspective. European Journal of Social Psychology, 33, 135 – 145.

Sexton, T. L, & Tuckman, B. W. (1991). Self-beliefs and behavior: the role of self- efficacy and outcome expectation over time. Personality and Individual Differences, 12, 725 – 736.

Sexton, T. L, & Tuckman, B. W., & Crehan, K. (1992). An investigation of the patterns of self-efficacy, outcome expectation, outcome value, and performance across trials. Cognitive Therapy and Research, 16, 329 – 348.

Shah, J. Y. (2003). The motivational looking glass: how significant others implicitly affect goal appraisals. Journal of Personality and Social Psychology, 84, 1109 – 1121.

Shah, J. Y., & Kruglanski, A. W. (2000). Aspects of goal networks. In M. Boekaerts, P. R. Pintrich, & M. Zeidner (Eds.), Handbook of self-regulation (pp. 86 – 110). San Diego, CA: Academic Press.

Shah, J. Y., & Kruglanski, A. W. (2003). When opportunity knocks: bottom-up priming of goals by means and its effects on self-regulation. Journal of Personality and Social Psychology, 84, 1109 – 1121.

Shoda, Y., Mischel, W., & Peake, P. (1990). Predicting adolescent cognitive and self- regulatory competencies from preschool delay of gratification: identifying diagnostic conditions. Developmental Psychology, 26, 978 – 986.

184

Sirois, F. (2004). Procrastination and intentions to perform health behaviors: the role of self-efficacy and the consideration of future consequences. Personality and Individual Differences, 37, 115 – 128.

Skaalvik, E. (1997). Self-enhancing and self-defeating ego orientation: relationsh with task avoidance orientation, achievement, self-perceptions, and anxiety. Journal of Educational Psychology, 89, 71 – 81.

Solomon, L. J., & Rothblum, E. D. (1984). Academic procrastination: frequency and cognitive-behavioral correlates. Journal of Counseling Psychology, 31, 503 – 509.

Spears, R., Scheepers, D., Jetten, J., Doosje, B., Ellemers, N., & Postmes, T. (2004). Entitativity, group distinctiveness, and social identity. In V. Yzerbyt, C. M. Judd, & O. Corneille (Eds.), The psychology of group perception: perceived variability, entitativity, and essentialism. Psychology Press: New York, NY

Spears, R., Ellemers, N., Doosje, B., & Branscombe, N. R. (2006). The individual with the group: Respect!. In T. Postmes & J. Jetten (Eds.), Individuality and the group: advances in social identity. London: Sage Publications.

Spence, J. T., & Helmreich, R. L. (1983). Achievement-related motives and behaviors. In J. T. Spence (Ed.), Achievement and achievement motivation: psychological and sociological approaches (pp. 7 – 74). San Francisco: Freeman.

Steel, P. (2007). The nature of procrastination: a meta-analytic an theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133, 65 – 94.

Steel, P., & König, C., (2006). Integrating theories of motivation. Academy of Management Review, 31, 889 – 913.

Steinberg, L., Brown, B. B., & Dornbusch, S. M. (1996). Beyond the classroom. New York: Simon & Schuster.

Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good adjustment, less pathology, better grades, and interpersonal success. Journal of Personality, 72, 271 – 324.

Tice, D. M., & Baumeister, R. F. (1990). Self-esteem, self-handicapping, and self- presentation: the strategy of inadequate practice. Journal of Personality, 58, 443 – 464.

185

Tice, D.M., & Baumeister, R.F. (1997). Longitudinal study of procrastination, performance, stress, and health: The costs and benefits of dawdling. Psychological Science, 8, 454-458.

Tice, D. M., Bratslavsky, E., & Baumeister, R. F. (2001). Emotional distress regulation takes precedence over impulse control: if you feel bad, do it!. Journal of Personality and Social Psychology, 80, 53 – 67.

Tinto, V. (1993). Leaving college: rethinking the causes and cures of student attrition. (2nd ed.). Chicago, IL: University of Chicago Press.

Tuckman, B. W. (1990). Group versus goal-setting effects on the self-regulated performance of students differing in self-efficacy. Journal of Experimental Education, 58, 291 – 298.

Tuckman, B. W. (1991). The development and concurrent validity of the procrastination scale. Educational and Psychological Measurement, 51, 473 – 480.

Tuckman, B. W. (1992). The effect of student planning and self-competence on self- motivated performance. Journal of Experimental Education, 60, 119 – 127.

Tuckman, B. W. (2003). The effect of learning and motivation strategies training on college students’ achievement. Journal of College Student Development, 44, 430 – 437.

Tuckman, B. W. (1996). The relative effectiveness of incentive motivation and prescribed learning strategy in improving college students’ course performance. Journal of Experimental Education, 64, 197 – 210.

Tuckman, B. W. (1998). Using frequent testing to increase students’ motivation to achieve. Journal of Experimental Education, 66, 141 – 147.

Tuckman, B. W. (2002). Evaluating ADAPT: a hybrid instructional model combining Web-based and classroom components. Computers and Education, 39, 261 – 269.

Tuckman, B. W. (2005). Relations of academic procrastination, rationalizations, and performance in a web course with deadlines. Psychological Reports, 96, 1015 – 1021.

Tuckman, B. W., Abry, D. A., & Smith, D. R. (2008). Learning and Motivation Strategies: Your Guide to Success. Upper Saddle River, NJ: Pearson.

186

Tuckman, B. W., & Kennedy, G. J. (2009). Teaching Learning and Motivation Strategies to enhance the success of first-term college students. Paper presented at the AERA Annual Meeting, San Diego, CA, 2009.

Tuckman, B. W., & Sexton, T. L. (1990). The relation between self-beliefs and self- regulated performance. Journal of Social Behavior and Personality, 5, 465 – 472.

Twenge, J. M., Catanese, K. R., & Baumeister, R. F. (2002). Social exclusion causes self-defeating behavior.. Journal of Personality and Social Psychology, 83, 606 – 615

Urdan, T. C. (1997). Examining the relations among early adolescent students’ goals and friends’ orientation toward effort and achievement in school. Contemporary Educational Psychology, 22, 165 – 191.

Urdan, T. C. (2004). Can achievement goal theory guide school reform? Advances in motivation and achievement: Vol. 13. Motivating students, improving schools: The legacy of Carol Midgley (pp. 361 – 392). Oxford, England: Elsevier.

Urdan, T. C., & Maehr, M. L. (1995). Beyond a two-goal theory of motivation and achievement: a case for social goals. Review of Educational Research, 65, 213 – 243.

Urdan, T. C., & Mestas, M. (2006). The goals behind performance goals. Journal of Educational Psychology, 98, 354 – 365.

Vallacher, R. R., Wegner, D. M. (1987). What do people think they’re doing? Action identification and human behavior. Psychological Review, 94, 3 – 15.

van Eerde, W. (2003). A meta-analytically derived nomological network of procrastination. Personality and Individual Differences, 35, 1401 – 1418.

Wentzel, K. R. (1993). Social and academic goals at school: motivation and achievement in early adolescence. Journal of Early Adolescence, 13, 4 – 20.

Wentzel, K. R. (2005). Peer relationships, motivation, and academic performance at school. In A. J. Elliot, & C. S. Dweck (Eds.), Handbook of competence and motivation (pp. 279 – 296). New York: Guilford Press.

White, P. H., Sanbonmatsu, D. M., Croyle, R. T., & Smittipatana, S. (2002). Test of socially motivated underachievement: “Letting up” for others. Journal of Experimental Social Psychology, 38, 162 – 169.

187

Wigfield, A., & Eccles, J. S. (1992). The development of achievement task values: a theoretical analysis. Developmental Review, 12, 265 – 310.

Wigfield, A., & Eccles, J. S. (2000). Expectancy-value theory of achievement motivation. Contemporary Educational Psychology, 25, 68 – 81.

Wolters, C. A. (2003). Understanding procrastination from a self-regulated learning perspective. Journal of Educational Psychology, 95, 179 – 187.

Wolters, C. A., Yu, S. L., & Pintrich, P. R. (1996). The relation between goal orientation and students’ motivational beliefs and self-regulated learning. Learning and Individual Differences, 8, 211 – 238.

Yuan, K-H, & Bentler, P. M. (2000). Three likelihood-based methods for mean and covariance structure analysis with nonnormal missing data. Sociological Methodology, 30, 165 – 200.

Zimmerman, B. J. (1989). A social cognitive view of self-regulated academic learning. Journal of Educational Psychology, 81, 329 – 339.

Zimmerman, B. J. (2000a). Attaining self-regulation: a social-cognitive perspective. In M. Boekaerts, P. R. Pintrich, & M. Zeidner (Eds.), Handbook of self- regulation (pp. 13 – 39). San Diego: Academic Press.

Zimmerman, B. J. (2000b). Self-efficacy: an essential motive to learn. Contemporary Educational Psychology, 25, 82 – 91.

Zimmerman, B. J. (2008). Goal setting: a key proactive source of academic self- regualtion. In D. H. Schunk, & B. J. Zimmerman (Eds.), Motivation and self- regulated learning (pp. 267 – 295). San Diego: Academic Press.

Zimmerman, B. J., Bandura, A., & Martinez-Pons, M. (1992). Self-motivation for academic attainment: The role of self-efficacy beliefs and personal goal setting. American Educational Research Journal, 29, 663 – 676.

Zimmerman, B. J., & Kitsantas, A. (1997). Developmental phases in self-regulation: shifting from process to outcome goals. Journal of Educational Psychology, 89, 29 – 36.

Zimmerman, B. J., & Kitsantas, A. (1999). Acquiring writing revision skill: shifting from process to outcome self-regulatory goals. Journal of Educational Psychology, 91, 1 – 10.

Zimmerman, B.J., & Kitsantas, A. (2005). The Hidden dimension of personal competence: Self-Regulated Learning and Practice. In A. J. Elliot and C. S.

188

Dweck (Eds.), Handbook of Competence and Motivation (pp. 204 – 222). New York: Guilford Press.

Zimmerman, B. J., & Kitsantas, A. (2007). Reliability and validity of Self-efficacy for Learning Form (SELF) scores of college students. Zeitschrift für Psychologie /Journal of Psychology, 215 157 – 163.

Zwick, W. R., & Velicer, W. F. (1986). Comparison of 5 rules for determining the number of components to retain. Psychological Bulletin, 99, 432 – 442.

189 Footnotes

1I would like to thank Dr. Richard G. Lomax for suggesting this analogy.

2The chi-square value for the MLR estimator in Mplus can not be used for chi-square difference tests because they are weighted or scaled chi-squares (Muthén & Muthén,

2008). This weight is referred to as a scaling correction factor (cf). In order to compute chi-square difference statistics using the MLR, Muthén & Muthén (2008) have provided a formula (available at http://www.statmodel.com/chidiff.shtml).

2 2 Let n and c be values from a chi-square with dfn and dfc degrees of freedom for

* * the nested and comparison models respectively. Further, let Tn andTc be the Yuan-

Bentler scaled chi-square statistics with scaling factors cfn and cfc and dfn and

dfc degrees of freedom for the nested and comparison models respectively. The

difference test statistic is distributed as chi-sqaure with dfn – dfc degrees of freedom

(Asparouhov & Muthén, 2005) and may be calculated as:

* * 2     TcfTcf ccnn   dfdf cn  .  cfdfcfdf ccnn      dfdf cn 

190 Appendix A: Cover Letter to Students

Hello!

I am requesting your participation in a study for my dissertation. This cover letter is a request for your participation in this research project and contains important information about the study and what to expect if you decide to participate. The purpose of this study is to assess the relationship between social and motivational factors in students’ lives and academic performance and is attempting to understand how life outside of the classroom affects performance in the classroom.

If you choose to participate you will receive bonus credit for this course and the only thing you have to do is fill out two questionnaires that will be available through Carmen. The first questionnaire will be available from October 6, 2008 through October 12, 2008. The second questionnaire will be available from November 10, 2008 through November 16, 2008. Each questionnaire will take approximately 40 minutes to complete.

Within the next day or two you will receive an e-mail from me reminding you of this bonus opportunity. This e-mail will include a link to Carmen where you may complete the first questionnaire anytime up until October 12, 2008. I will send you an e-mail reminder during the first week of November informing you that the second questionnaire is available. When you receive these e-mails from me all you have to do is follow the link to the Carmen site. Once there you will read a brief statement outlining the procedure of the research project and another asking for your permission to certain aspects of your student record. If you agree to participate, simply click the “I agree” button and the questionnaire will appear. If you do not agree to participate simply do nothing and close your web browser. Please note that once you receive the e-mail from me you are in no way obligated to click the link to Carmen for this particular bonus opportunity. If you do not want to participate in this research, simply do nothing. Please note that there may or may not be other assignments in this course that may require you to access Carmen. Please talk to your instructor about whether or not these other opportunities and assignments require access to Carmen.

In order for the relationship between your responses on the questionnaire and classroom performance to be understood, I will need to check your grades at the end of the next four quarters including this one. In addition, I will need to access the university database system to get your demographic information such as sex, ethnicity, and date of birth and I will need to know your high school rank and standardized tests scores (ACT or SAT). I cannot check any of this information without your consent. If you choose not to participate I will not have access to any of this information. However, if you do choose to participate you will be able to indicate your consent online before completing the questionnaires and will be identified by your university assigned e-mail address only. This address is in the form of your name.#. In no way will you be able to identify yourself by social security number or full name.

Your information (i.e., standardized test scores, high school rank, grades, sex, ethnicity, and date of birth) will be downloaded from Ohio State University’s data storage servers. Your data will be joined to the responses you give on the questionnaire and will stored electronically in a secure environment on an

191 encrypted computer sitting behind a university provided secure firewall. No data will be copied to laptop computers or in any way removed from behind the firewall. Your identifying information (i.e. your e-mail address) will be permanently removed from the electronic data and your record will be associated with an arbitrary numerical identification number. Your questionnaire responses will be destroyed at the termination of the study. The risks in participating in this study are minimal. Your information is stored in extremely secure environments at the university. Your only identification will be based on your university assigned e-mail address.

You will receive bonus credit for this course by completing these questionnaires. As such, completing the questionnaires is completely voluntary. The questionnaires are not tests. If you choose not to complete the questionnaires, but do want bonus credit for this course, there are other assignments you can complete. Please ask your instructor about these other optional bonus assignments. Please note that you may leave the study at any time. If you decide not to participate or to stop participating in the study, there will be no penalty to you, and you will not lose any benefits to which you are otherwise entitled. Your decision will not affect your future relationship with The Ohio State University. By agreeing to participate you do not give up any personal legal rights you may have as a participant in this study.

Confidentiality: Efforts will be made to keep your study-related information confidential. However, there may be circumstances where this information must be released. For example, personal information regarding your participation in this study may be disclosed if required by state law. Also, your records may be reviewed by the following groups (as applicable to the research):  Office for Human Research Protections or other federal, state, or international regulatory agencies;  The Ohio State University Institutional Review Board or Office of Responsible Research Practices;  The sponsor, if any, or agency (including the Food and Drug Administration for FDA-regulated research) supporting the study.

Contacts and Questions: For questions, concerns, or complaints about the study you may contact Gary Kennedy at [email protected].

For questions about your rights as a participant in this study or to discuss other study-related concerns or complaints with someone who is not part of the research team, you may contact Ms. Sandra Meadows in the Office of Responsible Research Practices at 1-800-678-6251.

Your participation in this research project is greatly appreciated. If you have any questions please send me an e-mail at [email protected].

Thank you and have a great first quarter at Ohio State!

Gary Kennedy

192 Appendix B: E-mail Notifications to Students

First Notification

Hello!

I hope your quarter is off to a good start! I am sending you this e-mail to remind you of a bonus credit opportunity in your freshmen survey course and to provide the link to Carmen where you will have access to the first questionnaire. I do need to remind you of the conditions of participation. First, if you choose to participate the only thing you have to do is fill out two questionnaires that will be available through Carmen. The first questionnaire will be available from Monday October 6, 2008 through Sunday October 12, 2008. The second questionnaire will be available from November 10, 2008 through November 16, 2008. I will send you an e-mail reminder during the first week of November informing you that the second questionnaire is available. Second, you will receive bonus credit for completing these questionnaires. As such, completing the questionnaires is completely voluntary. If you choose not to complete the questionnaires, but do want bonus credit for this course, please talk to your instructor. Third, in order for the relationship between your responses on the questionnaire and classroom performance to be understood, I will need to check your grades at the end of the next four quarters including this one. In addition, I will need to access the university database system to get your demographic information such as sex, ethnicity, and date of birth and I will need to know your high school grades and standardized tests scores (ACT or SAT). I cannot check any of the information without your consent. If you choose not to participate I will not have access to any of this information. However, if you do choose to participate you will be able to indicate your consent online before completing the questionnaires.

If you choose to participate you will be identified by your university assigned e-mail address only. This address is in the form of your name.#. PLEASE do not identify yourself by social security number or full name.

When you enter Carmen, scroll down until you find the “Ongoing” link. Look for “Education: Education Policy and Leadership”. Click this and then click “Kennedy Study”. Locate the “Surveys” link and then click “Questionnaire 1”.

Please read the consent form very carefully. At the bottom of the consent form, you will find two radio buttons. If you want to participate click the “I Consent” radio button. Otherwise, click the “I do not Consent” radio button if you do not want to participate in this bonus assignment. You will need to type in your university assigned name.# (for example “Yourlastname.9999”) so that I can let your instructor know that you completed the questionnaire. If you do not identify yourself in this way I cannot let your instructor know you completed this assignment and I cannot use the information you submit.

When you finish answering all the questions, click the “Submit” button. If you do not click “Submit” I will not know you completed the questionnaire.

Your participation in this research project is greatly appreciated!

193 If would like to participate in this study for bonus credit in your freshmen survey course, please click the link below: https://carmen.osu.edu/

If you have any questions please send me an e-mail at [email protected].

Thank you! Gary Kennedy

Second Notification

Hello again!

I am sending you this e-mail to let you know that the second questionnaire that is part of the bonus credit opportunity in your freshmen survey course will be available starting November 10, 2008 and will run through November 16, 2008.

Please remember that in order to receive the bonus credit for your survey course you must complete this second questionnaire. However, completing the questionnaire is completely voluntary. If you choose not to complete the questionnaire, but do want bonus credit for this course, please talk to your instructor.

Also, I want to provide some reminders that I gave you in the first e-mail and that you will see again in the consent form for the second questionnaire. In order for the relationship between your responses on the questionnaire and classroom performance to be understood, I will need to check your grades at the end of the next four quarters including this one. In addition, I will need to access the university database system to get your demographic information such as sex, ethnicity, and date of birth and I will need to know your high school grades and standardized tests scores (ACT or SAT). I cannot check any of the information without your consent. If you choose not to participate I will not have access to any of this information. However, if you do choose to participate you will be able to indicate your consent online before completing the questionnaires.

When you enter Carmen, scroll down until you find the “Ongoing” link. This may appear as a separate course. Look for “Education: Education Policy and Leadership”. Click this and then click “Kennedy Study”. Locate the “Surveys” link and then click “Questionnaire 2”. Note that Questionnaire 1 is no longer available.

When you finish answering all the questions, click the “Submit” button. If you do not click “Submit” I will not know you completed the questionnaire.

Your participation in this research project is greatly appreciated!

194 If would like to participate in this study for bonus credit in your freshmen survey course, please click the link below: https://carmen.osu.edu/

If you have any questions please send me an e-mail at [email protected].

Thank you! Gary Kennedy

195 Appendix C: Online Consent to Participate in Research

This is a consent form for research participation. It contains important information about this study and what to expect if you decide to participate. Your participation is voluntary. Please consider the information carefully. Feel free to ask questions before making your decision whether or not to participate. If you decide to participate, you will be asked to click an “I Consent” button before you proceed to the questionnaire. Purpose: The purpose of this study is to assess the relationship between social and motivational factors in students’ lives and academic performance and is attempting to understand how life outside of the classroom affects performance in the classroom.

Procedures/Tasks: You are being requested to complete a questionnaire that will be made available immediately upon clicking the “I Consent” button at the bottom of this form. In addition, upon clicking the “I Consent” button you are giving the researchers of this study permission to access certain information from your university record. Specifically, this information includes the courses you are taking and the grades you receive in these courses at Ohio State University during your first year of attendance, your high school class rank, your standardized test scores, your ethnicity, and your sex.

Duration: This questionnaire will take approximately 40 minutes to complete.

You may leave the study at any time. If you decide to stop participating in the study, there will be no penalty to you, and you will not lose any benefits to which you are otherwise entitled. Your decision will not affect your future relationship with The Ohio State University.

Risks and Benefits: Your information (i.e., standardized test scores, high school class rank, courses taken and grades, sex, and ethnicity) will be downloaded from Ohio State University’s data storage servers. Your data will be joined to the responses you give on the questionnaire and will stored electronically on a secure, encrypted computer sitting behind a university provided secure firewall. No data will be copied to laptop computers or in any way removed from behind the firewall. Your identifying information (i.e. your e-mail address) will be permanently removed from the electronic data and all data will be destroyed when the study terminates. The risks in participating in this study are minimal. Your information is stored in extremely secure environments at the university. Your only identification will be based on your university assigned e-mail address. The direct benefit to you is that you will receive bonus credit in your freshmen survey course for your participation in this study. In addition, the information gained could potentially help future students attending this university.

Incentives: As an incentive to participate, you will receive bonus credit in your survey course this quarter.

196 Participant Rights: You may refuse to participate in this study without penalty or loss of benefits to which you are otherwise entitled. If you are a student or employee at Ohio State, your decision will not affect your grades or employment status.

If you choose to participate in the study, you may discontinue participation at any time without penalty or loss of benefits. By clicking the “I Consent” on the bottom of this form, you do not give up any personal legal rights you may have as a participant in this study.

An Institutional Review Board responsible for human subjects research at The Ohio State University reviewed this research project and found it to be acceptable, according to applicable state and federal regulations and University policies designed to protect the rights and welfare of participants in research. Contacts and Questions: For questions, concerns, or complaints about the study you may contact Gary Kennedy at [email protected].

I have read (or someone has read to me) this form and I am aware that I am being asked to participate in a research study that includes completing a questionnaire. In addition, I am aware that as a participant in this study I am giving my consent to the researchers to access certain aspects of my record including only the following: standardized test scores, high school class rank, courses taken and grades in these courses, sex, and ethnicity. I have had the opportunity to ask questions and have had them answered to my satisfaction. I voluntarily agree to participate in this study.

I am not giving up any legal rights by clicking either the “I Consent” or “I do not Consent” buttons.

I Consent I do not Consent

197 Appendix D: E-mail Notification to Students Verifying Submission of Questionnaire and Consent to Records Access

Dear ,

Thank you for submitting your responses to the first (second and final) bonus assignment questionnaire!

I am also writing to remind you that as a participant in this research project, you have authorized the researchers to access only certain aspects of your record. These include only your standardized test scores, high school class rank, courses taken and grades, sex, and ethnicity.

Finally, I would also like to remind you that you may leave this study at any time. If you no longer wish to be a participant in this study simply reply to this e-mail or contact Gary Kennedy at [email protected] and let me know. If I receive an e-mail from you indicating that you do not want to participate in the research your data will be removed from the study.

If you no longer want to participate in this study, please talk to your instructor about other bonus opportunities for your survey class.

Thank you again and have a great quarter!

Gary Kennedy

198 Appendix E: Descriptive Statistics and Multivariate Skewness and Kurtosis of Items By Construct †

Construct N M SD Skewness Kurtosis SOCIAL EXCLUSION CONCERNS (CSE) I think it would be hurtful for my friends or people close to me to exclude me from being or doing 671 7.15 2.28 -0.906 0.475 things with them. I think it is painful to be without friends or social 671 7.81 2.14 -1.223 1.435 relationships. I am a person who likes to have a lot of friends. 671 7.68 2.29 -0.913 0.171 I can't imagine anything more painful than my friends excluding me from being or doing things 671 6.05 2.63 -0.483 -0.608 with them. I don't really care if my friends or people close to R 671 7.26 2.22 -0.878 0.425 me exclude me from doing things with them. I would be upset if I lose friends because something 671 8.71 1.75 -2.003 5.137 I said or did made them angry or sad. Mardia Skewness and Kurtosis 1259 36.5

ACADEMIC TASK VALUES (AV) It is important for me to learn what is being taught 671 7.96 1.71 -0.835 0.9 in my classes. R I would like to find the easiest major possible. 671 8.15 2.09 -1.481 2.188 I think that what I’m learning from my classes is 671 7.27 2.03 -0.808 0.59 important. I like to choose academic projects that are 671 5.80 2.15 -0.144 -0.379 challenging, even if they require more work. I think that most of the classes that I’m taking are a R 671 7.19 2.10 -0.827 0.425 waste of time. I prefer to take classes that are challenging so I can 671 6.27 2.04 -0.333 0.09 learn new things. I think what is being taught in my classes is useful 671 7.05 1.99 -0.700 0.437 know. I believe that most of what is taught in my classes is R 671 6.94 2.20 -0.756 0.147 not useful.* I really enjoy the academic challenges my classes 669 6.06 2.13 -0.397 0.013 provide. R I find that most of my college classes are boring. 671 5.77 2.37 -0.274 -0.409 If I could get rich without going to college, I would R 671 5.13 3.30 -0.097 -1.201 drop all my classes today. I am in college because I think learning is useful and 671 7.94 1.94 -1.234 2.04 important in its own right. Mardia Skewness and Kurtosis 39051 264.7

199 Construct N M SD Skewness Kurtosis ACADEMIC GRADE VALUES (GV) What grade point average would you regard as 671 3.34 0.35 -0.221 -0.182 minimally satisfying for this quarter? What grade point average would you regard as 671 3.33 0.34 -0.176 -0.284 minimally satisfying for this year? Mardia Skewness and Kurtosis 36.85 21.26

PROCRASTINATION/LACK OF SELF- REGULATION (PROC)* I needlessly delay finishing jobs, even when they are 671 3.76 2.44 0.324 -0.648 important (PR) I postpone starting in on things I don't like to do. 671 5.87 2.36 -0.460 -0.243 When I have a deadline, I wait till the very last 671 4.37 2.68 0.139 -0.894 minute (PR) I delay making tough decisions. (PR) 671 5.20 2.60 -0.094 -0.769 I keep putting off improving my work habits. (PR) 671 4.05 2.44 0.359 -0.543 I find an excuses for not doing something. (PR) 671 4.67 2.44 -0.036 -0.630 I put the necessary time into even the most boring 671 3.97 2.20 0.293 -0.392 R tasks. (PR) When something is too tough to handle, I believe in 670 4.36 2.28 0.147 -0.717 postponing it. (PR) I promise myself I'll do something and then drag my 670 4.66 2.39 0.105 -0.516 feet. (PR) R Whenever, I make a plan of action, I follow it. (PR) 671 3.32 1.79 0.484 0.673 I'm a time waster and I can't seem to do anything 670 3.44 2.55 0.603 -0.360 about it. (PR) I always finish important jobs with time to spare. 671 3.71 2.22 0.390 -0.542 R (PR) I get stuck in neutral even though I know how 670 4.82 2.51 0.003 -0.833 important it is for me to get started. (PR) Even though I hate myself if I don't get started, it 671 4.41 2.67 0.092 -0.932 doesn't get me going. (PR) R I am good at resisting temptation. (SCS) 671 3.57 2.34 0.511 -0.515 I have a hard time breaking bad habits. (SCS) 671 5.47 2.58 -0.241 -0.710 I am lazy. (SCS) 671 4.39 2.70 0.223 -0.840 I say inappropriate things. (SCS) 671 4.52 2.93 0.216 -1.041 I do certain things that are bad for me, if they are 671 4.65 2.81 -0.025 -1.058 fun. (SCS) R I refuse things that are bad for me. (SCS) 671 4.05 2.66 0.282 -0.954 I wish I had more self-discipline. (SCS) 671 5.99 2.88 -0.409 -0.852 Pleasure and fun sometimes keep me from getting 671 6.03 2.30 -0.428 -0.275 work done. (SCS) I have trouble concentrating. (SCS) 671 4.87 2.61 0.099 -0.924 I am able to work effectively toward long-term 670 2.73 1.90 0.788 0.934 R goals. (SCS) Sometimes I can’t stop myself from doing 671 4.30 2.61 0.167 -0.910 something, even if I know it is wrong. (SCS) I often act without thinking through all the 670 4.10 2.57 0.260 -0.817 alternatives. (SCS) Mardia Skewness and Kurtosis 228000 661.8

200 Construct N M SD Skewness Kurtosis SCHOOL BELONGINGNESS (SB) I feel like a real part of Ohio State. 671 7.10 2.49 -0.755 -0.045 People here notice when I’m good at something. 671 6.51 2.21 -0.552 0.135 R It is hard for people like me to be accepted here. 671 7.83 2.35 -1.267 1.159 Others students at this university take my opinions 671 6.75 1.84 -0.470 0.261 seriously. Most of my instructors at Ohio State are interested 671 5.30 2.20 -0.099 -0.365 in me. R Sometimes I feel as if I don’t belong here. 671 6.81 2.92 -0.673 -0.652 There’s at least one instructor or other adult at this 671 6.58 2.93 -0.591 -0.771 university I can talk to if I have a problem. People at this university are friendly to me. 671 7.73 1.80 -1.103 1.737 R Instructors here are not interested in people like me. 671 6.65 2.26 -0.421 -0.395 I am included in lots of activities at Ohio State. 671 5.02 2.58 -0.013 -0.726 I am treated with as much respect as other students. 671 7.82 1.80 -0.999 0.976 R I feel very different from most other students here. 671 6.11 2.66 -0.395 -0.713 I can really be myself at this university. 671 7.64 2.21 -1.158 1.097 The instructors here respect me. 671 7.22 1.99 -0.602 0.178 People here know I can do good work. 671 7.22 1.94 -0.672 0.218 R I wish I were in a different school. 671 7.96 2.62 -1.468 1.485 I feel proud of belonging to Ohio State. 671 8.56 2.08 -1.841 3.423 Other students here like me the way I am. 671 7.46 1.86 -0.691 0.399 Mardia Skewness and Kurtosis 5077 60.7

SELF-REGULATORY SELF-EFFICACY

(SE) If a lecture is especially boring, I can motivate 671 5.70 2.39 -0.235 -0.569 myself to keep good notes. If another student asks me to study with her/him, I 671 7.31 2.06 -0.909 0.708 can be an effective study partner. If problems with friends and peers conflict with 671 7.40 1.88 -0.747 0.572 schoolwork, I can keep up with my assignments. If I feel moody or restless during studying, I can focus my attention well enough to finish my 671 6.25 2.36 -0.526 -0.295 assigned work. If I find myself getting increasingly behind in a course, I can increase my study time sufficiently to 671 6.89 1.93 -0.554 0.319 catch up. If I discover that my homework assignments for the semester are much longer than expected, I can 671 7.09 1.82 -0.702 1.106 change my other priorities to have enough time for studying. If I have trouble recalling an abstract concept, I can think of a good example that will help me remember 671 7.09 1.89 -0.477 0.161 it on the test. If I have to take a test in a school subject I dislike, I can find a way to motivate myself to earn a good 671 6.86 2.03 -0.599 0.391 grade. If I am feeling depressed about a forthcoming test, I 671 6.63 2.19 -0.735 0.411 can find a way to motivate my self to do well. If my last test results were lower than I would have liked, I can figure out potential questions before the 671 6.55 2.15 -0.550 0.127 next test that will improve my score greatly. 201 Construct N M SD Skewness Kurtosis

SELF-REGULATORY SELF-EFFICACY

(SE) (Continued)

If I am struggling to remember technical details of a concept for a test, I can find a way to associate them 671 6.74 1.85 -0.295 0.011 together that will ensure recall. If I think I did poorly on a test I just finished, I can go back to my notes and locate all the information I 671 7.07 2.18 -0.759 0.383 had forgotten. If I find that I had to “cram” at the last minute for a test, I can begin my test preparation much earlier so 671 6.46 2.41 -0.539 -0.166 I won’t need to cram the next time. Mardia Skewness and Kurtosis 4805 63.09

PERCEIVED STRESS (PS) In the last month, I have often been upset because of 671 5.28 2.56 0.006 -0.643 something that happened unexpectedly? In the last month, I have often felt that I was unable 671 3.82 2.57 0.398 -0.596 to control the important things in my life? In the last month, I have often felt nervous and 671 6.89 2.57 -0.647 -0.342 “stressed”? In the last month, I have often dealt successfully 671 2.97 2.03 0.800 0.771 R with day to day problems and annoyances? In the last month, I have often felt that I was effectively coping with important changes that were 671 3.16 2.22 0.701 0.223 R occurring in your life? In the last month, I have often felt confident about 671 3.08 2.24 0.857 0.454 R my ability to handle my personal problems? In the last month, I have often felt that things were 671 4.25 2.31 0.420 -0.118 R going my way? In the last month, I have often found that I could not 671 4.46 2.53 0.199 -0.638 cope with all the things that I had to do? In the last month, I have often been able to control 671 3.78 2.28 0.509 -0.268 R irritations in my life? In the last month, I have often felt that I was on top 671 3.77 2.23 0.366 -0.430 R of things? In the last month, I have often been angered because of things that happened that were outside of my 671 5.27 2.70 -0.083 -0.772 control? In the last month, I have often found myself thinking 671 8.30 1.83 -1.281 1.692 about things that I have to accomplish? In the last month, I have often been able to control 671 3.33 2.13 0.558 0.208 R the way I spend my time? In the last month, I have often felt difficulties were 671 4.19 2.61 0.338 -0.743 piling up so high that I could not overcome them? Mardia Skewness and Kurtosis 2436 49.4

202 Construct N M SD Skewness Kurtosis PERFORMANCE-APPROACH GOAL ORIENTATION (PAP) In setting my academic goals for the end of this quarter, I am focused on demonstrating to others 671 6.21 2.60 -0.514 -0.453 that I can do better academically than other students. In setting my academic goals for the end of this quarter, I am focused on demonstrating to others 671 6.47 2.54 -0.604 -0.354 that I can do well compared to other students in my classes. In setting my academic goals for the end of this quarter, I am focused on demonstrating to others 671 6.28 2.61 -0.564 -0.394 that I can get better grades than most of the other students. Mardia Skewness and Kurtosis 181.4 24.36

PERFORMANCE-AVOIDANCE GOAL

ORIENTATION (PAV) In setting my academic goals for the end of this quarter, I am focused on demonstrating to others 671 6.50 2.67 -0.662 -0.308 that I can avoid doing poorly in my classes. In setting my academic goals for the end of this quarter, I am focused on demonstrating to others 671 5.56 2.74 -0.261 -0.742 that I can avoid not meeting their academic standards. In setting my academic goals for the end of this quarter, I am focused on demonstrating to others 671 6.43 2.66 -0.675 -0.293 that I can avoid getting poor grades. Mardia Skewness and Kurtosis 260.9 19.98

MASTERY-APPROACH GOAL ORIENTATION

(MAP) In setting my academic goals for the end of this quarter, I am focused on learning as much as 671 7.65 1.85 -0.847 0.77 possible from my classes. In setting my academic goals for the end of this quarter, I am focused on understanding the content 671 7.58 1.79 -0.611 0.032 of my courses as thoroughly as possible. In setting my academic goals for the end of this quarter, I am focused on completely mastering the 671 7.01 1.99 -0.610 0.174 material presented in my classes. Mardia Skewness and Kurtosis 282.2 20.31

203 Construct N M SD Skewness Kurtosis MASTERY-AVOIDANCE GOAL ORIENTATION

(MAV) In setting my academic goals for the end of this quarter, I am focused on avoiding not learning all 671 6.14 2.52 -0.553 -0.229 that I possibly could in my classes. In setting my academic goals for the end of this quarter, I am focused on avoiding not understanding 671 6.20 2.70 -0.653 -0.290 the content of my classes as thoroughly as I’d like. In setting my academic goals for the end of this quarter, I am focused on avoiding not mastering the 671 5.66 2.99 -0.388 -0.946 material presented in my classes. Mardia Skewness and Kurtosis 315.8 19.98

ABILITY MEASURES Standardized Test Score (ACT Equivalent) 661 27.00 2.98 -0.127 0.300 High School Class Rank (HSCR) 650 87.83 9.64 -1.207 1.558

First Term Grade Point Average (GPA) 671 3.40 0.55 -1.671 4.835

ITEM SKEWNESS AND KURTOSIS RANGE Low -2.003 -1.201 High 0.857 5.137 †Mardia’s skewness and kurtosis estimates for the items could not be calculated due to a singular covariance matrix. *Items indentified as ‘PR’ are from the Tuckman Procrastination Scale (Tuckman, 1991) and items identified as ‘SCS’ are from the Self Control Scale (Tangney, Baumeister & Boone, 2004). R = Item is reverse scored.

204 Appendix F: Descriptive Statistics and Multivariate Skewness and Kurtosis of Parcel Indicators by Construct and Assigned Item

Item Assigned to Parcel Parcel N M SD Skewness Kurtosis SOCIAL EXCLUSION CONCERNS (CSE) I think it is painful to be without friends or

social relationships. CSE1 671 7.75 1.88 -0.880 0.505 I am a person who likes to have a lot of

friends. I can't imagine anything more painful than my friends excluding me from being or doing things with them. CSE2 671 6.66 2.14 -0.578 -0.121 I don't really care if my friends or people R close to me exclude me from doing things with them. I think it would be hurtful for my friends or people close to me to exclude me from being or doing things with them. CSE3 671 7.93 1.68 -1.074 1.189 I would be upset if I lose friends because something I said or did made them angry or sad. Mardia Skewness and Kurtosis 285.2 13.49

ACADEMIC TASK VALUES (AV) It is important for me to learn what is being taught in my classes. I would like to find the easiest major R possible. AV11 671 7.29 1.33 -0.468 0.176 I think that what I’m learning from my classes is important. I like to choose academic projects that are challenging, even if they require more work. I think that most of the classes that I’m R taking are a waste of time. I prefer to take classes that are challenging so I can learn new things. AV12 671 6.64 1.59 -0.375 0.406 I think what is being taught in my classes is useful know. I really enjoy the academic challenges my classes provide.

205 Item Assigned to Parcel Parcel N M SD Skewness Kurtosis ACADEMIC TASK VALUES (AV) I believe that most of what is taught in my R classes is not useful. I find that most of my college classes are R boring. AV13 671 6.44 1.75 -0.218 0.045 If I could get rich without going to college, I R would drop all my classes today. I am in college because I think learning is useful and important in its own right. Mardia Skewness and Kurtosis 71.03 4.03

PROCRASTINATION/LACK OF SELF- REGULATION (PROC)* I needlessly delay finishing jobs, even when they are important (PR) When I have a deadline, I wait till the very last minute. (PR) I keep putting off improving my work habits. (PR) I put the necessary time into even the most R boring tasks. (PR) SR1 671 4.16 1.54 0.132 -0.057 Whenever, I make a plan of action, I follow R it. (PR) R I am good at resisting temptation. (SCS) I have a hard time breaking bad habits. (SCS) I am lazy. (SCS) I say inappropriate things. (SCS) I postpone starting in on things I don't like to do. (PR) I delay making tough decisions. (PR)

I find an excuse for not doing something. (PR) When something is too tough to handle, I believe in postponing it. (PR) SR2 671 4.95 1.53 -0.369 -0.049 I always finish important jobs with time to R spare. (PR) I do certain things that are bad for me, if they are fun. (SCS) R I refuse things that are bad for me. (SCS) I wish I had more self-discipline. (SCS) Pleasure and fun sometimes keep me from getting work done. (SCS)

206 Item Assigned to Parcel Parcel N M SD Skewness Kurtosis PROCRASTINATION/LACK OF SELF- REGULATION (PROC) (Continued) I promise myself I'll do something and then drag my feet. (PR) I'm a time waster and I can't seem to do anything about it. (PR) I get stuck in neutral even though I know how important it is for me to get started. (PR) Even though I hate myself if I don't get SR3 671 4.17 1.66 -0.074 -0.41 started, it doesn't get me going. (PR) I have trouble concentrating. (SCS) I am able to work effectively toward long- R term goals. (SCS) Sometimes I can’t stop myself from doing something, even if I know it is wrong. (SCS) I often act without thinking through all the alternatives. (SCS) Mardia Skewness and Kurtosis 57.41 3.91

SCHOOL BELONGINGNESS (SB) I feel like a real part of Ohio State. It is hard for people like me to be accepted R here. Others students at this university take my opinions seriously. BLNG1 671 7.48 1.45 -0.711 0.522 People at this university are friendly to me. I am treated with as much respect as other students. I can really be myself at this university. Most of my instructors at Ohio State are interested in me. R Sometimes I feel as if I don’t belong here. Instructors here are not interested in people BLNG2 671 6.9 1.51 -0.451 -0.044 R like me. The instructors here respect me. R I wish I were in a different school. Other students here like me the way I am. People here notice when I’m good at something. There’s at least one instructor or other adult at this university I can talk to if I have a problem. I am included in lots of activities at Ohio BLNG3 671 6.67 1.46 -0.425 0.33 State. I feel very different from most other students R here. People here know I can do good work. I feel proud of belonging to Ohio State. Mardia Skewness and Kurtosis 112.3 5.57

207 Item Assigned to Parcel Parcel N M SD Skewness Kurtosis SELF-REGULATORY SELF-EFFICACY

(SE) If I feel moody or restless during studying, I can focus my attention well enough to finish my assigned work. If I have trouble recalling an abstract concept, I can think of a good example that will help me remember it on the test. If my last test results were lower than I SE1 671 6.42 1.4 -0.025 0.172 would have liked, I can figure out potential questions before the next test that will improve my score greatly. If I find that I had to “cram” at the last minute for a test, I can begin my test preparation much earlier so I won’t need to cram the next time. If another student asks me to study with her/him, I can be an effective study partner. If I find myself getting increasingly behind in a course, I can increase my study time sufficiently to catch up. If I have to take a test in a school subject I SE2 671 6.6 1.37 -0.053 0.262 dislike, I can find a way to motivate myself to earn a good grade. If I am struggling to remember technical details of a concept for a test, I can find a way to associate them together that will ensure recall. If a lecture is especially boring, I can motivate myself to keep good notes. If problems with friends and peers conflict with schoolwork, I can keep up with my assignments. If I discover that my homework assignments for the semester are much longer than expected, I can change my other priorities to SE3 671 6.56 1.42 -0.118 -0.102 have enough time for studying. If I am feeling depressed about a forthcoming test, I can find a way to motivate my self to do well. If I think I did poorly on a test I just finished, I can go back to my notes and locate all the information I had forgotten. Mardia Skewness and Kurtosis 42.92 8.44

208 Item Assigned to Parcel Parcel N M SD Skewness Kurtosis PERCEIVED STRESS (PS) In the last month, I have often been upset because of something that happened unexpectedly? In the last month, I have often dealt R successfully with day to day problems and annoyances? In the last month, I have often felt that things R PSS1 671 5.05 1.45 0.383 0.451 were going my way? In the last month, I have often found that I could not cope with all the things that I had to do? In the last month, I have often found myself thinking about things that I have to accomplish? In the last month, I have often felt that I was unable to control the important things in my life? In the last month, I have often felt that I was R effectively coping with important changes that were occurring in your life? In the last month, I have often been able to R PSS2 671 3.87 1.64 0.143 -0.068 control irritations in my life? In the last month, I have often been angered because of things that happened that were outside of my control?

In the last month, I have often been able to R control the way I spend my time?

In the last month, I have often felt nervous

and “stressed”? In the last month, I have often felt confident R about my ability to handle my personal problems? PSS3 671 4.48 1.78 0.172 -0.105 In the last month, I have often felt that I was R on top of things? In the last month, I have often felt difficulties were piling up so high that I could not overcome them? Mardia Skewness and Kurtosis 37.57 7.45

Overall Mardia Skewness and Kurtosis 2351 31.57 *Items indentified as ‘PR’ are from the Tuckman Procrastination Scale (Tuckman, 1991) and items identified as ‘SCS’ are from the Self Control Scale (Tangney, Baumeister & Boone, 2004). R = Item is reverse scored.

209 Appendix G: Correlation Matrices of Items Used in Preliminary Factor Analysis

CSE 1 2 3 4 5 6 I think it would be hurtful for my friends or 1 people close to me to exclude me from 1 being or doing things with them. I think it is painful to be without friends or 2 .529 1 social relationships. I am a person who likes to have a lot of 3 .405 .437 1 friends. I can't imagine anything more painful than 4 my friends excluding me from being or .566 .567 .327 1 doing things with them. I don't really care if my friends or people 5 close to me exclude me from doing things .584 .498 .369 .545 1 with them.* I would be upset if I lose friends because 6 something I said or did made them angry or .392 .388 .319 .267 .317 1 sad. *Reversed Scored

210 AV 1 2 3 4 5 6 It is important for me to learn what is being 1 1 taught in my classes. I would like to find the easiest major 2 .195 1 possible.* I think that what I’m learning from my 3 .472 .133 1 classes is important. I like to choose academic projects that are 4 challenging, even if they require more .216 .259 .296 1 work. I think that most of the classes that I’m 5 .456 .274 .534 .257 1 taking are a waste of time.* I prefer to take classes that are challenging 6 .331 .299 .391 .669 .331 1 so I can learn new things. I think what is being taught in my classes is 7 .506 .178 .720 .318 .614 .412 useful know. I believe that most of what is taught in my 8 .457 .244 .619 .231 .671 .336 classes is not useful.* I really enjoy the academic challenges my 9 .384 .225 .373 .584 .305 .685 classes provide. I find that most of my college classes are 10 .321 .171 .411 .315 .525 .377 boring.* If I could get rich without going to college, 11 .114 .206 .174 .304 .305 .323 I would drop all my classes today.* I am in college because I think learning is 12 .411 .230 .380 .373 .425 .472 useful and important in its own right.

7 8 9 10 11 I think what is being taught in my classes is 7 1 useful know. I believe that most of what is taught in my 8 .683 1 classes is not useful.* I really enjoy the academic challenges my 9 .422 .355 1 classes provide. I find that most of my college classes are 10 .468 .431 .407 1 boring.* If I could get rich without going to college, I 11 .221 .210 .252 .331 1 would drop all my classes today.* I am in college because I think learning is 12 .446 .412 .477 .349 .326 useful and important in its own right. *Reversed Scored

211 PROC 1 2 3 4 5 6 I needlessly delay finishing jobs, even when they are 1 1 important 2 I postpone starting in on things I don't like to do. .458 1 3 When I have a deadline, I wait till the very last minute .597 .496 1 4 I delay making tough decisions. .313 .348 .272 1 5 I keep putting off improving my work habits .470 .432 .436 .248 1 6 I find an excuses for not doing something. .422 .446 .437 .320 .352 1 I put the necessary time into even the most boring 7 .280 .331 .324 .141 .343 .221 tasks.* When something is too tough to handle, I believe in 8 .399 .472 .369 .279 .436 .339 postponing it. I promise myself I'll do something and then drag my 9 .494 .388 .392 .291 .448 .416 feet. 10 Whenever, I make a plan of action, I follow it.* .381 .311 .382 .203 .349 .335 I'm a time waster and I can't seem to do anything about 11 .617 .410 .605 .302 .521 .463 it. 12 I always finish important jobs with time to spare.* .450 .423 .522 .226 .319 .277 I get stuck in neutral even though I know how important 13 .512 .519 .481 .347 .514 .447 it is for me to get started. Even though I hate myself if I don't get started, it doesn't 14 .553 .434 .526 .394 .459 .445 get me going. 15 I am good at resisting temptation.* .246 .247 .242 .205 .220 .239 16 I have a hard time breaking bad habits. .309 .249 .301 .239 .303 .329 17 I am lazy. .417 .366 .441 .269 .439 .455 18 I say inappropriate things. .284 .223 .273 .108 .277 .268 19 I do certain things that are bad for me, if they are fun. .274 .234 .249 .096 .234 .297 20 I refuse things that are bad for me.* .251 .193 .207 .104 .173 .243 21 I wish I had more self-discipline. .364 .325 .281 .332 .416 .399 Pleasure and fun sometimes keep me from getting work 22 .431 .461 .354 .268 .337 .346 done. 23 I have trouble concentrating. .411 .348 .363 .253 .347 .325 24 I am able to work effectively toward long-term goals.* .337 .227 .295 .200 .377 .276 Sometimes I can’t stop myself from doing something, 25 .257 .298 .246 .207 .316 .348 even if I know it is wrong. 26 I often act without thinking through all the alternatives. .278 .240 .234 .183 .379 .241

212 PROC 7 8 9 10 11 12 7 I put the necessary time into even the most boring tasks.* 1 When something is too tough to handle, I believe in 8 .229 1 postponing it. 9 I promise myself I'll do something and then drag my feet. .144 .386 1 10 Whenever, I make a plan of action, I follow it.* .404 .252 .336 1 11 I'm a time waster and I can't seem to do anything about it. .301 .383 .499 .382 1 12 I always finish important jobs with time to spare.* .281 .298 .289 .365 .445 1 I get stuck in neutral even though I know how important 13 .292 .445 .525 .327 .505 .381 it is for me to get started. Even though I hate myself if I don't get started, it doesn't 14 .301 .393 .529 .392 .654 .401 get me going. 15 I am good at resisting temptation.* .261 .155 .199 .280 .264 .269 16 I have a hard time breaking bad habits. .166 .238 .322 .245 .365 .175 17 I am lazy. .295 .311 .422 .333 .532 .268 18 I say inappropriate things. .211 .206 .223 .153 .273 .163 19 I do certain things that are bad for me, if they are fun. .164 .175 .150 .141 .198 .196 20 I refuse things that are bad for me.* .217 .128 .129 .248 .201 .197 21 I wish I had more self-discipline. .201 .293 .433 .331 .384 .230 Pleasure and fun sometimes keep me from getting work 22 .246 .366 .336 .206 .383 .297 done. 23 I have trouble concentrating. .266 .247 .376 .323 .455 .304 24 I am able to work effectively toward long-term goals.* .350 .295 .190 .465 .377 .277 Sometimes I can’t stop myself from doing something, 25 .157 .293 .259 .257 .298 .181 even if I know it is wrong. 26 I often act without thinking through all the alternatives. .200 .202 .256 .185 .308 .136

13 14 15 16 17 18 I get stuck in neutral even though I know how important 13 1 it is for me to get started. Even though I hate myself if I don't get started, it doesn't 14 .547 1 get me going. 15 I am good at resisting temptation.* .206 .282 1 16 I have a hard time breaking bad habits. .334 .420 .306 1 17 I am lazy. .385 .477 .249 .324 1 18 I say inappropriate things. .268 .220 .278 .248 .298 1 19 I do certain things that are bad for me, if they are fun. .229 .180 .405 .219 .205 .430 20 I refuse things that are bad for me.* .176 .183 .510 .229 .170 .327 21 I wish I had more self-discipline. .423 .425 .268 .370 .341 .195 Pleasure and fun sometimes keep me from getting work 22 .380 .408 .355 .325 .298 .331 done. 23 I have trouble concentrating. .435 .407 .279 .294 .307 .267 24 I am able to work effectively toward long-term goals.* .321 .354 .233 .202 .371 .145 Sometimes I can’t stop myself from doing something, 25 .295 .329 .433 .330 .240 .342 even if I know it is wrong. 26 I often act without thinking through all the alternatives. .304 .287 .353 .259 .246 .328

213 PROC 19 20 21 22 23 24 25 I do certain things that are bad for me, if they are 19 1 fun. 20 I refuse things that are bad for me.* .607 1 21 I wish I had more self-discipline. .206 .182 1 Pleasure and fun sometimes keep me from 22 .324 .292 .355 1 getting work done. 23 I have trouble concentrating. .283 .239 .394 .398 1 I am able to work effectively toward long-term 24 .090 .167 .247 .219 .236 1 goals.* Sometimes I can’t stop myself from doing 25 .505 .488 .299 .451 .288 .165 1 something, even if I know it is wrong. I often act without thinking through all the 26 .358 .344 .253 .320 .265 .218 .469 alternatives. *Reversed Scored

214 SB 1 2 3 4 5 6 1 I feel like a real part of Ohio State. 1 2 People here notice when I’m good at something. .370 1 3 It is hard for people like me to be accepted here. * .406 .235 1 Others students at this university take my opinions 4 .290 .335 .204 1 seriously. Most of my instructors at Ohio State are interested 5 .245 .324 .096 .282 1 in me. 6 Sometimes I feel as if I don’t belong here.* .625 .247 .577 .202 .147 1 There’s at least one instructor or other adult at this 7 .305 .353 .177 .254 .368 .207 university I can talk to if I have a problem. 8 People at this university are friendly to me. .494 .404 .433 .344 .225 .412 Instructors here are not interested in people like 9 .275 .319 .219 .206 .392 .276 me.* 10 I am included in lots of activities at Ohio State. .434 .281 .143 .312 .258 .221 11 I am treated with as much respect as other students. .249 .238 .345 .416 .150 .271 12 I feel very different from most other students here.* .331 .113 .551 .145 .098 .510 13 I can really be myself at this university. .520 .356 .436 .222 .116 .511 14 The instructors here respect me. .359 .475 .181 .297 .473 .219 15 People here know I can do good work. .443 .497 .247 .324 .275 .297 16 I wish I were in a different school.* .569 .210 .472 .144 .055 .626 17 I feel proud of belonging to Ohio State. .624 .231 .349 .161 .123 .491 18 Other students here like me the way I am. .466 .357 .490 .366 .212 .462

7 8 9 10 11 12 There’s at least one instructor or other adult at this 7 1 university I can talk to if I have a problem. 8 People at this university are friendly to me. .312 1 Instructors here are not interested in people like 9 .319 .298 1 me.* 10 I am included in lots of activities at Ohio State. .365 .267 .191 1 11 I am treated with as much respect as other students. .205 .385 .220 .167 1 12 I feel very different from most other students here.* .077 .310 .140 .143 .193 1 13 I can really be myself at this university. .261 .527 .206 .203 .366 .356 14 The instructors here respect me. .370 .432 .412 .275 .320 .075 15 People here know I can do good work. .265 .375 .354 .290 .366 .144 16 I wish I were in a different school.* .170 .374 .172 .207 .206 .419 17 I feel proud of belonging to Ohio State. .186 .383 .178 .254 .222 .271 18 Other students here like me the way I am. .272 .537 .310 .270 .414 .352

13 14 15 16 17 13 I can really be myself at this university. 1 14 The instructors here respect me. .386 1 15 People here know I can do good work. .387 .507 1 16 I wish I were in a different school.* .514 .152 .218 1 17 I feel proud of belonging to Ohio State. .475 .236 .290 .670 1 18 Other students here like me the way I am. .564 .398 .420 .384 .362 *Reversed Scored

215 SE 1 2 3 4 5 6 7 If a lecture is especially boring, I can motivate 1 1 myself to keep good notes. If another student asks me to study with her/him, I 2 .208 1 can be an effective study partner. If problems with friends and peers conflict with 3 .344 .238 1 schoolwork, I can keep up with my assignments. If I feel moody or restless during studying, I can 4 focus my attention well enough to finish my .363 .278 .393 1 assigned work. If I find myself getting increasingly behind in a 5 course, I can increase my study time sufficiently to .412 .325 .483 .379 1 catch up. If I discover that my homework assignments for the semester are much longer than expected, I can 6 .429 .252 .460 .356 .635 1 change my other priorities to have enough time for studying. If I have trouble recalling an abstract concept, I 7 can think of a good example that will help me .290 .267 .288 .219 .443 .356 1 remember it on the test. If I have to take a test in a school subject I dislike, 8 I can find a way to motivate myself to earn a good .410 .374 .445 .418 .583 .529 .375 grade. If I am feeling depressed about a forthcoming test, 9 .359 .223 .357 .336 .461 .462 .337 I can find a way to motivate my self to do well. If my last test results were lower than I would have 10 liked, I can figure out potential questions before .401 .266 .368 .270 .470 .432 .411 the next test that will improve my score greatly. If I am struggling to remember technical details of 11 a concept for a test, I can find a way to associate .293 .249 .323 .227 .455 .407 .551 them together that will ensure recall. If I think I did poorly on a test I just finished, I can 12 go back to my notes and locate all the information .263 .226 .301 .238 .422 .384 .339 I had forgotten. If I find that I had to “cram” at the last minute for a 13 test, I can begin my test preparation much earlier .462 .221 .303 .291 .482 .509 .371 so I won’t need to cram the next time.

216 SE 8 9 10 11 12 If I have to take a test in a school subject I dislike, 8 I can find a way to motivate myself to earn a good 1 grade. If I am feeling depressed about a forthcoming test, 9 .454 1 I can find a way to motivate my self to do well. If my last test results were lower than I would have 10 liked, I can figure out potential questions before .416 .410 1 the next test that will improve my score greatly. If I am struggling to remember technical details of 11 a concept for a test, I can find a way to associate .454 .418 .429 1 them together that will ensure recall. If I think I did poorly on a test I just finished, I can 12 go back to my notes and locate all the information .394 .273 .337 .359 1 I had forgotten. If I find that I had to “cram” at the last minute for a 13 test, I can begin my test preparation much earlier .397 .320 .428 .339 .346 so I won’t need to cram the next time.

217 PS 1 2 3 4 5 6 7 In the last month, I have often been upset 1 because of something that happened 1 unexpectedly. In the last month, I have often felt that I 2 was unable to control the important .454 1 things in my life. In the last month, I have often felt 3 .445 .341 1 nervous and “stressed”. In the last month, I have often dealt 4 successfully with day to day problems .253 .436 .199 1 and annoyances.* In the last month, I have often felt that I was effectively coping with important 5 .263 .377 .216 .547 1 changes that were occurring in your life.* In the last month, I have often felt 6 confident about my ability to handle my .302 .454 .244 .599 .565 1 personal problems.* In the last month, I have often felt that 7 .361 .439 .386 .459 .443 .529 1 things were going my way.* In the last month, I have often found that 8 I could not cope with all the things that I .541 .528 .474 .403 .417 .428 .461 had to do. In the last month, I have often been able 9 .317 .355 .247 .457 .435 .449 .398 to control irritations in my life.* In the last month, I have often felt that I 10 .265 .347 .317 .358 .384 .424 .524 was on top of things.* In the last month, I have often been 11 angered because of things that happened .561 .442 .426 .250 .230 .306 .329 that were outside of my control. In the last month, I have often found 12 myself thinking about things that I have .082 -.003 .252 -.187 -.198 -.135 -.055 to accomplish. In the last month, I have often been able 13 .254 .388 .192 .394 .373 .427 .381 to control the way I spend my time.* In the last month, I have often felt 14 difficulties were piling up so high that I .403 .512 .493 .346 .343 .383 .455 could not overcome them.

218 PS 8 9 10 11 12 13 14 In the last month, I have often found that 8 I could not cope with all the things that I 1 had to do? In the last month, I have often been able 9 .373 1 to control irritations in my life? In the last month, I have often felt that I 10 .430 .371 1 was on top of things? In the last month, I have often been 11 angered because of things that happened .425 .281 .179 1 that were outside of my control? In the last month, I have often found 12 myself thinking about things that I have .016 -.103 -.121 .075 1 to accomplish? In the last month, I have often been able 13 .402 .334 .463 .204 -.164 1 to control the way I spend my time? In the last month, I have often felt 14 difficulties were piling up so high that I .603 .293 .482 .343 .057 .374 could not overcome them? *Reversed Scored

219 Appendix H: Pattern Loading and Factor Structure Coefficients Defining the Dimensionality of Six Latent Constructs

Pattern CSE Weights I think it would be hurtful for my friends or people close to me to exclude me from being or doing things 0.773 with them. I think it is painful to be without friends or social 0.731 relationships. I am a person who likes to have a lot of friends. 0.530 I can't imagine anything more painful than my friends 0.727 excluding me from being or doing things with them. I don't really care if my friends or people close to me 0.724 exclude me from doing things with them. I would be upset if I lose friends because something I 0.475 said or did made them angry or sad. Factor Determinancy 0.920 Cronbach's Alpha 0.822

220 Pattern Weights Factor Structure AV Utility Challenge Utility Challenge

It is important for me to learn what is 0.546 0.088 0.594 0.385 being taught in my classes. I would like to find the easiest major 0.121 0.264 0.265 0.330 possible.* I think that what I’m learning from my 0.764 0.014 0.772 0.430 classes is important. I like to choose academic projects that are challenging, even if they require more -0.081 0.803 0.356 0.759 work. I think that most of the classes that I’m 0.790 -0.045 0.766 0.384 taking are a waste of time.* I prefer to take classes that are challenging 0.001 0.874 0.476 0.874 so I can learn new things. I think what is being taught in my classes 0.841 0.013 0.848 0.470 is useful know. I believe that most of what is taught in my 0.867 -0.091 0.818 0.380 classes is not useful.* I really enjoy the academic challenges my 0.084 0.735 0.484 0.781 classes provide. I find that most of my college classes are 0.462 0.200 0.571 0.452 boring.* If I could get rich without going to college, I would drop all my classes 0.125 0.312 0.295 0.380 today.* I am in college because I think learning is 0.338 0.367 0.538 0.551 useful and important in its own right. Factor Determinancy 0.950 0.936 Cronbach's Alpha 0.867 0.760 Factor Correlation 0.544 *Item is reversed scored.

221 Pattern Weights Factor Structure PROC Lack of Lack of Self- Impulsive Self- Impulsive Discipline Discipline I needlessly delay finishing jobs, even when they 0.730 0.014 0.736 0.314 are important I postpone starting in on things I don't like to do. 0.617 0.049 0.637 0.303 When I have a deadline, I wait till the very last 0.712 -0.015 0.706 0.278 minute I delay making tough decisions. 0.460 -0.018 0.452 0.171 I keep putting off improving my work habits 0.649 0.032 0.662 0.299 I find an excuse for not doing something. 0.559 0.124 0.610 0.354 I put the necessary time into even the most boring 0.383 0.100 0.424 0.258 tasks.* When something is too tough to handle, I believe 0.559 -0.003 0.558 0.227 in postponing it. I promise myself I'll do something and then drag 0.679 -0.068 0.651 0.212 my feet. Whenever, I make a plan of action, I follow it*. 0.504 0.058 0.528 0.265 I'm a time waster and I can't seem to do anything 0.804 -0.056 0.781 0.275 about it. I always finish important jobs with time to spare.* 0.544 0.008 0.547 0.232 I get stuck in neutral even though I know how 0.723 -0.017 0.715 0.280 important it is for me to get started. Even though I hate myself if I don't get started, it 0.787 -0.055 0.764 0.269 doesn't get me going. I am good at resisting temptation.* 0.143 0.550 0.370 0.609 I have a hard time breaking bad habits. 0.403 0.177 0.476 0.343 I am lazy. 0.616 0.009 0.620 0.263 I say inappropriate things. 0.194 0.414 0.364 0.494 I do certain things that are bad for me, if they are -0.003 0.757 0.308 0.755 fun. I refuse things that are bad for me. -0.037 0.774 0.282 0.759 I wish I had more self-discipline. 0.518 0.086 0.554 0.299 Pleasure and fun sometimes keep me from getting 0.428 0.291 0.548 0.468 work done. I have trouble concentrating. 0.496 0.151 0.558 0.355 I am able to work effectively toward long-term 0.475 -0.002 0.474 0.193 goals.* Sometimes I can’t stop myself from doing 0.171 0.613 0.423 0.683 something, even if I know it is wrong. I often act without thinking through all the 0.226 0.421 0.399 0.514 alternatives. Factor Determinancy 0.965 0.912 Cronbach's Alpha 0.919 0.806 Factor Correlation 0.411 *Item is reversed scored.

222 Pattern Weights Factor Structure SB School School Support Acceptance Support Acceptance Belonging Belonging I feel like a real part of 0.267 0.012 0.663 0.533 0.527 0.776 Ohio State. People here notice when 0.614 0.018 0.054 0.642 0.270 0.308 I’m good at something. It is hard for people like 0.010 0.804 -0.025 0.285 0.792 0.490 me to be accepted here.* Others students at this university take my 0.474 0.170 -0.086 0.500 0.283 0.209 opinions seriously. Most of my instructors at Ohio State are interested in 0.584 -0.089 -0.029 0.540 0.099 0.145 me. Sometimes I feel as if I 0.007 0.456 0.414 0.333 0.722 0.707 don’t belong here.* There’s at least one instructor or other adult at 0.504 -0.039 0.081 0.522 0.191 0.255 this university I can talk to if I have a problem. People at this university 0.424 0.334 0.104 0.584 0.551 0.484 are friendly to me. Instructors here are not interested in people like 0.488 0.091 -0.015 0.515 0.255 0.236 me.* I am included in lots of 0.386 -0.119 0.263 0.448 0.184 0.340 activities at Ohio State. I am treated with as much 0.373 0.384 -0.140 0.454 0.427 0.251 respect as other students. I feel very different from -0.110 0.683 0.030 0.144 0.664 0.421 most other students here.* I can really be myself at 0.251 0.325 0.300 0.485 0.604 0.606 this university. The instructors here 0.762 -0.046 0.003 0.747 0.226 0.275 respect me. People here know I can do 0.609 0.041 0.090 0.659 0.314 0.357 good work. I wish I were in a different -0.140 0.179 0.757 0.223 0.610 0.815 school.* I feel proud of belonging 0.019 -0.113 0.868 0.322 0.445 0.803 to Ohio State. Other students here like 0.391 0.482 0.014 0.567 0.630 0.475 me the way I am. Factor Determinancy 0.923 0.915 0.935 Cronbach's Alpha 0.818 0.816 0.824 Support Accept. Factor Accept. 0.354 Correlation Schl. Blng. 0.395 0.635 *Item is reversed scored.

223 Pattern SE Weights If a lecture is especially boring, I can motivate myself to keep good notes. 0.573 If another student asks me to study with her/him, I can be an effective study 0.415 partner. If problems with friends and peers conflict with schoolwork, I can keep up 0.592 with my assignments. If I feel moody or restless during studying, I can focus my attention well 0.504 enough to finish my assigned work. If I find myself getting increasingly behind in a course, I can increase my 0.787 study time sufficiently to catch up. If I discover that my homework assignments for the semester are much longer than expected, I can change my other priorities to have enough time for 0.741 studying. If I have trouble recalling an abstract concept, I can think of a good example 0.571 that will help me remember it on the test. If I have to take a test in a school subject I dislike, I can find a way to motivate 0.725 myself to earn a good grade. If I am feeling depressed about a forthcoming test, I can find a way to motivate 0.607 my self to do well. If my last test results were lower than I would have liked, I can figure out 0.628 potential questions before the next test that will improve my score greatly. If I am struggling to remember technical details of a concept for a test, I can 0.611 find a way to associate them together that will ensure recall. If I think I did poorly on a test I just finished, I can go back to my notes and 0.529 locate all the information I had forgotten. If I find that I had to “cram” at the last minute for a test, I can begin my test 0.615 preparation much earlier so I won’t need to cram the next time. Factor Determinancy 0.948 Cronbach's Alpha 0.881

224 Pattern Weights Factor Structure

PS Primary Secondary Primary Secondary Appraisal Appraisal Appraisal Appraisal

In the last month, I have often been upset because of 0.694 0.001 0.695 0.364 something that happened unexpectedly. In the last month, I have often felt that I was unable 0.456 0.326 0.626 0.564 to control the important things in my life. In the last month, I have often felt nervous and 0.739 -0.102 0.686 0.285 “stressed”. In the last month, I have often dealt successfully -0.033 0.762 0.365 0.745 with day to day problems and annoyances.* In the last month, I have often felt that I was effectively coping with important changes that were -0.010 0.719 0.366 0.714 occurring in your life.* In the last month, I have often felt confident about 0.040 0.749 0.431 0.770 my ability to handle my personal problems.* In the last month, I have often felt that things were 0.291 0.494 0.549 0.646 going my way.* In the last month, I have often found that I could not 0.618 0.242 0.744 0.565 cope with all the things that I had to do. In the last month, I have often been able to control 0.125 0.520 0.397 0.586 irritations in my life.* In the last month, I have often felt that I was on top 0.216 0.468 0.461 0.581 of things.* In the last month, I have often been angered because of things that happened that were outside of my 0.618 0.005 0.620 0.328 control. In the last month, I have often found myself -0.424 0.474 -0.176 0.253 thinking about things that I have to accomplish.* In the last month, I have often been able to control 0.115 0.515 0.384 0.575 the way I spend my time.* In the last month, I have often felt difficulties were 0.585 0.201 0.690 0.507 piling up so high that I could not overcome them? Factor Determinancy 0.920 0.930 Cronbach's Alpha 0.839 0.827 Factor Correlation 0.522 *Item is reversed scored.

225 Appendix I: Correlation Matrix of the Parcel Indicators

1 2 3 4 5 6 7 8 9 10 1 GPA 1 2 ACT .178 1 3 HSCR .180 .064 1 4 PAP1 .154 -.024 -.031 1 5 PAP2 .150 -.037 -.029 .798 1 6 PAP3 .160 -.040 -.082 .899 .790 1 7 PAV1 .078 -.098 .000 .687 .723 .672 1 8 PAV2 .141 .033 -.016 .677 .622 .665 .573 1 9 PAV3 .091 -.074 -.018 .721 .749 .701 .827 .630 1 10 MAP1 .104 -.079 .038 .252 .277 .244 .263 .217 .271 1 11 MAP2 .178 -.092 .080 .220 .207 .206 .218 .192 .232 .600 12 MAP3 .193 -.067 .073 .278 .270 .266 .255 .242 .283 .637 13 MAV1 .121 .156 .033 .204 .143 .163 .162 .417 .172 .335 14 MAV2 .130 .134 .056 .190 .182 .163 .183 .379 .185 .212 15 MAV3 .157 .182 .034 .166 .136 .156 .132 .372 .142 .230 16 CSE1 .084 .068 -.058 .118 .153 .115 .110 .116 .140 .054 17 CSE2 .052 .111 -.058 .067 .124 .073 .065 .085 .093 -.061 18 CSE3 .063 .070 .049 .104 .146 .102 .102 .079 .117 .022 19 AV1 .124 .052 .072 .163 .170 .149 .152 .119 .164 .518 20 AV2 .113 .048 .023 .172 .198 .161 .163 .138 .144 .473 21 AV3 .114 .058 .023 .038 .048 .053 .038 .059 .067 .422 22 GS1 .258 .234 .083 .134 .142 .156 .076 .104 .065 .113 23 GS2 .247 .302 .086 .122 .130 .131 .045 .107 .055 .082 24 SR1 -.202 .065 -.156 -.135 -.137 -.133 -.126 -.084 -.140 -.297 25 SR2 -.170 .053 -.153 -.083 -.082 -.086 -.090 -.058 -.105 -.268 26 SR3 -.173 -.022 -.144 -.108 -.119 -.103 -.097 -.052 -.115 -.305 27 BLNG1 .062 .023 .003 .171 .189 .195 .207 .150 .168 .317 28 BLNG2 .143 -.032 .075 .162 .185 .176 .161 .122 .131 .345 29 BLNG3 .137 .012 .029 .215 .242 .238 .203 .169 .177 .343 30 SE1 .201 -.082 .040 .270 .270 .273 .224 .184 .238 .475 31 SE2 .228 .017 .051 .304 .286 .298 .236 .189 .236 .451 32 SE3 .223 -.066 .073 .286 .311 .286 .277 .221 .260 .480 33 PSS1 -.151 -.073 .013 -.054 -.053 -.079 -.035 -.066 .015 -.138 34 PSS2 -.159 -.016 -.001 -.149 -.153 -.175 -.153 -.141 -.113 -.300 35 PSS3 -.206 -.027 -.014 -.112 -.120 -.155 -.103 -.106 -.055 -.209

226 11 12 13 14 15 16 17 18 19 20 11 MAP2 1 12 MAP3 .684 1 13 MAV1 .325 .334 1 14 MAV2 .231 .219 .602 1 15 MAV3 .190 .225 .620 .692 1 16 CSE1 .063 .093 .055 .070 .057 1 17 CSE2 -.065 -.027 .006 .034 .039 .587 1 18 CSE3 .034 .057 .031 -.003 .022 .586 .612 1 19 AV1 .476 .481 .223 .158 .193 .077 -.063 .033 1 20 AV2 .440 .448 .218 .163 .169 .043 -.051 -.009 .786 1 21 AV3 .410 .361 .192 .171 .191 .027 -.028 -.036 .623 .736 22 GS1 .188 .164 .168 .088 .115 .107 .028 .055 .204 .162 23 GS2 .175 .133 .146 .086 .100 .094 .029 .064 .211 .167 24 SR1 -.310 -.315 -.122 -.018 -.116 .012 .054 .008 -.431 -.405 25 SR2 -.260 -.282 -.122 -.012 -.060 .169 .126 .122 -.394 -.388 26 SR3 -.307 -.311 -.188 -.065 -.107 .033 .075 .073 -.409 -.400 27 BLNG1 .302 .302 .198 .120 .114 .192 .028 .109 .240 .194 28 BLNG2 .370 .313 .208 .120 .112 .136 .027 .063 .317 .297 29 BLNG3 .329 .307 .190 .087 .116 .195 .053 .132 .272 .277 30 SE1 .512 .519 .261 .159 .166 .059 -.101 -.012 .408 .443 31 SE2 .463 .492 .306 .199 .224 .085 -.088 -.004 .397 .413 32 SE3 .501 .520 .300 .208 .203 .057 -.092 .018 .392 .401 33 PSS1 -.150 -.133 -.188 -.141 -.133 -.044 .006 .081 -.090 -.132 34 PSS2 -.327 -.290 -.231 -.160 -.161 -.013 .059 .048 -.222 -.202 35 PSS3 -.198 -.199 -.191 -.165 -.170 -.051 .003 .040 -.156 -.182

21 22 23 24 25 26 27 28 29 21 AV3 1 22 GS12 .151 1 23 GS34 .150 .856 1 24 SR1 -.408 -.173 -.161 1 25 SR2 -.376 -.165 -.140 .771 1 26 SR3 -.366 -.179 -.155 .813 .778 1 27 BLNG1 .181 .021 .017 -.230 -.164 -.250 1 28 BLNG2 .284 .048 .037 -.280 -.220 -.312 .765 1 29 BLNG3 .260 .081 .058 -.278 -.201 -.287 .733 .738 1 30 SE1 .337 .169 .160 -.453 -.398 -.457 .397 .419 .450 31 SE2 .312 .219 .188 -.384 -.327 -.430 .442 .438 .447 32 SE3 .292 .194 .166 -.438 -.378 -.432 .428 .423 .452 33 PSS1 -.119 -.029 .022 .181 .246 .282 -.374 -.393 -.314 34 PSS2 -.217 -.046 .013 .332 .324 .384 -.509 -.494 -.450 35 PSS3 -.128 -.067 -.041 .258 .295 .344 -.408 -.437 -.385

30 31 32 33 30 SE1 1 31 SE2 0.755 1 32 SE3 0.770 0.731 1 33 PSS1 -0.265 -0.345 -0.316 1 34 PSS2 -0.433 -0.470 -0.510 0.739 35 PSS3 -0.364 -0.393 -0.391 0.752

227 Appendix J: Factor Loadings for Measurement Models

Measurement Model Loadings With Avoidance Goals Without Avoidance Goals

(Model M1) (Model M2) Unstandardized Standardized Unstandardized Standardized Factor Factor Factor Factor Loadings Loadings Loadings Loadings Parcel Construct EST SE EST SE EST SE EST SE Indicator CSE1 1.000 0.000 0.759 0.028 1.000 0.000 0.759 0.028 Social Exclusion CSE2 1.158 0.079 0.773 0.030 1.159 0.079 0.772 0.030 Concerns (CSE) CSE3 0.925 0.065 0.782 0.031 0.927 0.065 0.783 0.031 Procrastination-Lack SR1 1.000 0.000 0.896 0.012 1.000 0.000 0.896 0.012 of Self-Regulation SR2 0.949 0.033 0.859 0.014 0.949 0.032 0.858 0.014 (PROC) SR3 1.090 0.033 0.908 0.011 1.092 0.034 0.909 0.011 SE1 1.000 0.000 0.833 0.016 1.000 0.000 0.833 0.016 Self-Efficacy (SE) SE2 0.951 0.044 0.841 0.018 0.949 0.044 0.840 0.018 SE3 0.976 0.036 0.866 0.015 0.976 0.036 0.867 0.015 BLNG1 1.000 0.000 0.869 0.014 1.000 0.000 0.869 0.014 School Belongingness BLNG2 1.055 0.041 0.879 0.015 1.056 0.041 0.879 0.015 (SB) BLNG3 0.979 0.037 0.842 0.016 0.979 0.037 0.842 0.016 PSS1 1.000 0.000 0.855 0.017 1.000 0.000 0.853 0.018 Perceived Stress (PS) PSS2 1.145 0.051 0.866 0.018 1.151 0.051 0.869 0.018 PSS3 1.216 0.046 0.845 0.018 1.216 0.046 0.844 0.018 AV1 1.000 0.000 0.851 0.016 1.000 0.000 0.851 0.016 Academic Task AV2 1.296 0.053 0.922 0.013 1.297 0.053 0.923 0.013 Values (AV) AV3 1.200 0.064 0.780 0.019 1.200 0.064 0.779 0.019 GS1 1.000 0.000 0.945 0.038 1.000 0.000 0.952 0.044 Grade Values (GV) GS2 0.926 0.082 0.905 0.039 0.913 0.095 0.899 0.046 PAP1 1.000 0.000 0.947 0.008 1.000 0.000 0.95 0.01 Performance PAP2 0.884 0.027 0.857 0.018 0.862 0.026 0.840 0.019 Approach (PAP) PAP3 0.995 0.016 0.938 0.010 0.996 0.019 0.943 0.011 MAP1 1.000 0.000 0.767 0.022 1.000 0.000 0.765 0.022 Mastery Approach MAP2 1.015 0.065 0.803 0.024 1.019 0.064 0.803 0.024 (MAP) MAP3 1.171 0.062 0.835 0.022 1.177 0.062 0.837 0.023 PAV1 1.000 0.000 0.876 0.018 Performance PAV2 0.835 0.048 0.713 0.032 Avoidance (PAV) PAV3 1.044 0.024 0.919 0.017 MAV1 1.000 0.000 0.753 0.034 Mastery Avoidance MAV2 1.165 0.081 0.820 0.026 (MAV) MAV3 1.302 0.083 0.828 0.026

228 Appendix K: Factor Loadings for Final Structural Model S2 Compared to Measurement Model M2

Measurement Model Loadings Structural Model Loadings Without Avoidance Goals Without Avoidance Goals

(Model M2) (Model S2) Unstandardized Standardized Unstandardized Standardized Factor Factor Factor Factor Loadings Loadings Loadings Loadings Parcel Construct EST SE EST SE EST SE EST SE Indicator CSE1 1.000 0.000 0.759 0.028 1.000 0.000 0.757 0.028 Social Exclusion CSE2 1.159 0.079 0.772 0.030 1.167 0.077 0.776 0.029 Concerns (CSE) CSE3 0.927 0.065 0.783 0.031 0.926 0.063 0.781 0.030 Procrastination-Lack of SR1 1.000 0.000 0.896 0.012 1.000 0.000 0.896 0.012 Self-Regulation SR2 0.949 0.032 0.858 0.014 0.949 0.033 0.858 0.014 (PROC) SR3 1.092 0.034 0.909 0.011 1.092 0.034 0.909 0.011 SE1 1.000 0.000 0.833 0.016 1.000 0.000 0.832 0.016 Self-Efficacy (SE) SE2 0.949 0.044 0.840 0.018 0.946 0.044 0.835 0.018 SE3 0.976 0.036 0.867 0.015 0.970 0.036 0.861 0.016 BLNG1 1.000 0.000 0.869 0.014 1.000 0.000 0.901 0.012 School Belongingness BLNG2 1.056 0.041 0.879 0.015 1.103 0.041 0.848 0.017 (SB) BLNG3 0.979 0.037 0.842 0.016 0.922 0.038 0.823 0.018 PSS1 1.000 0.000 0.853 0.018 1.000 0.000 0.853 0.018 Perceived Stress (PS) PSS2 1.151 0.051 0.869 0.018 1.149 0.050 0.867 0.018 PSS3 1.216 0.046 0.844 0.018 1.219 0.046 0.846 0.018 AV1 1.000 0.000 0.851 0.016 1.000 0.000 0.851 0.016 Academic Task Values AV2 1.297 0.053 0.923 0.013 1.296 0.052 0.922 0.013 (AV) AV3 1.200 0.064 0.779 0.019 1.197 0.063 0.777 0.019 GS1 1.000 0.000 0.952 0.044 1.000 0.000 0.885 0.031 Grade Values (GV) GS2 0.913 0.095 0.899 0.046 1.056 0.074 0.967 0.030 PAP1 1.000 0.000 0.952 0.009 1.000 0.000 0.951 0.009 Performance Approach PAP2 0.862 0.026 0.840 0.019 0.863 0.026 0.839 0.019 (PAP) PAP3 0.996 0.019 0.943 0.011 0.999 0.019 0.945 0.011 MAP1 1.000 0.000 0.765 0.022 1.000 0.000 0.763 0.023 Mastery Approach MAP2 1.019 0.064 0.803 0.024 1.027 0.066 0.808 0.025 (MAP) MAP3 1.177 0.062 0.837 0.023 1.177 0.063 0.836 0.023

229