SOCIAL STATUS ACROSS CONTEXTS by NICOLE M. LAWLESS DESJARDINS A DISSERTATION Presented to the Department of Psychology and the Graduate School of the University of Oregon in partial fulfillment of the requirements for the degree of Doctor of Philosophy June 2016 ii DISSERTATION APPROVAL PAGE Student: Nicole M. Lawless DesJardins Title: Social Status Across Contexts This dissertation has been accepted and approved in partial fulfillment of the requirements for the Doctor of Philosophy degree in the Department of Psychology by: Sanjay Srivastava Chairperson Sanjay Srivastava Advisor Sara Hodges Core Member Pranjal Mehta Core Member Mark Alfano Institutional Representative and Scott L. Pratt Dean of the Graduate School Original approval signatures are on file with the University of Oregon Graduate School. Degree awarded June 2016 iii © 2016 Nicole M. Lawless DesJardins This work is licensed under a Creative Commons Attribution-NonCommerical-NoDerivs (United States) License. iv DISSERTATION ABSTRACT Nicole M. Lawless DesJardins Doctor of Philosophy Department of Psychology June 2016 Title: Social Status Across Contexts Social groups without formally designated leaders spontaneously form status-based hierarchies in order to facilitate efficient and effective progress toward a common goal. The prevailing theoretical perspectives about who tends to attain status in these groups suggest that status allocation should be context-dependent. That is, the person who is given the most status should have qualities that help the group achieve its goal, and, because goals vary across groups, the characteristics that predict status should also vary. However, most research to date has focused only on the individual differences that predict status across a wide variety of situations, and has largely neglected the role of the specific context in which the group is situated. The primary aim of this dissertation was to investigate the contextualized, interpersonal processes that contribute to status attainment. To this end, I investigated the consistency with which the same people attained status across different groups and relationships, and how stable individual differences and social context interacted to predict status in a variety of situations and relationships. In the first study, N = 346 participants completed up to four activities with four different groups of their peers. Status attainment was moderately consistent across groups. Extraversion and its aspects, assertiveness and enthusiasm, as well as compassion, conscientiousness and intellect predicted status across all four tasks. The largest differences v in the predictors of status attainment appeared to be due to how the task was completed, rather than the goal of the task: generally pro-social attributes predicted status attainment in collaborative tasks, whereas neuroticism and low agreeableness predicted status in more knowledge-based, rote tasks. In the second study, N = 651 informants provided perceptions of N = 267 participants. Status was fairly inconsistent across participants’ relationships with different informants. There was some evidence that different personality traits predicted status in different types of relationships: compared to relationships with friends, agreeable and neurotic participants tended to attain status in their romantic relationships, whereas participants low in dominance tended to attain status with their college friends. Together, these results indicate that different personality traits predict status attainment across situations and relationships. vi CURRICULUM VITAE NAME OF AUTHOR: Nicole M. Lawless DesJardins GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon, Eugene Michigan State University, East Lansing DEGREES AWARDED: Doctor of Philosophy, Social/Personality Psychology, 2016, University of Oregon Master of Science, Social/Personality Psychology, 2011, University of Oregon Bachelor of Arts, Psychology, 2008, Michigan State University AREAS OF SPECIAL INTEREST: Personality Psychology Quantitative Methods PROFESSIONAL EXPERIENCE: Graduate Teaching Fellow, Department of Psychology, University of Oregon, 2011 - 2016 Graduate Research Fellow, Department of Psychology, University of Oregon, 2013 - 2016 GRANTS, AWARDS, AND HONORS: Heritage Dissertation Research Award, Foundation for Personality and Social Psychology, 2015 Gregores Research Award, Department of Psychology, University of Oregon, 2015 Marthe E. Smith Memorial Science Scholarship, College of Arts and Sciences, University of Oregon, 2014 vii Graduate Student Travel Award, Society for Personality and Social Psychology, 2014 Clarence & Lucille Dunbar Scholarship, College of Arts and Sciences, University of Oregon, 2013 Gary E. Smith Summer Professional Development Award, Graduate School, University of Oregon, 2013 University of Oregon Women in Graduate Sciences Travel Award, 2012 Graduate Education Committee Research Travel Award, Department of Psychology, University of Oregon, 2011 - 2015 PUBLICATIONS: Mehta, P.H., Lawless DesJardins, N.M., van Vugt, M., Josephs, R.A. (under review). Testosterone and cortisol predict competition over financial resources in the hawk-dove game. Hormones and Behavior. Lawless, N.M. & Hodges, S.D. (2015). Reading between the lies: Empathic accuracy and deception detection. Social Psychological and Personality Science. DOI: 10.1177/1948550615585829 Carlson, E.N. & Lawless DesJardins, N.M. (2015). Do mean guys always finish first or just say that they do? Narcissists’ awareness of their social status over time. Personality and Social Psychology Bulletin, 41, 901-917. DOI: 10.1177/0146167215581712 Lawless DesJardins, N.M., Srivastava, S., Kufner, A. & Back, M. (2015). Who attains status? Similarities and differences across social contexts. Social Psychological and Personality Science. DOI: 10.1177/1948550615580171 Lucas, R.E., Cheung, F. & Lawless, N.M. (2013). Investigating the happiness of U.S. regions. In P.J. Rentfrow (Ed.) Psychological Geography: Exploring the Interaction of Environment and Behavior. Washington, DC: American Psychological Association. Lucas, R.E. & Lawless, N.M. (2013). Does life seem better on a sunny day? Examining the association between daily weather conditions and life satisfaction judgments. Journal of Personality & Social Psychology, 104, 872- 884. DOI: 10.1037/a0032124. Lawless, N.M. & Lucas, R.E. (2011). Predictors of regional well-being: A county-level analysis. Social Indicators Research, 101, 341-357. viii ACKNOWLEDGMENTS I wish to express my deepest appreciation to my advisor, Sanjay Srivastava, for his mentorship, support, and patience over the last five years. I also want to thank my departmental committee members, Sara Hodges and Pranj Mehta, for their outstanding support throughout this project. Special thanks are due to Psychology Department staff members Michelle Pennoyer—who made data collection for Study 1 possible—and Lori Olsen—who made everything else possible. Thanks to Cassie Brandes and Smrithi Prasad for their immeasurable help in coordinating data collection for Study 1. Thanks, too, to Sarah Norris, who has edited everything I have ever written, including this dissertation. This investigation was made possible by the support of the Foundation for Personality and Social Psychology and the Department of Psychology. ix To Ryan. Let’s do this. x TABLE OF CONTENTS Chapter Page I. INTRODUCTION .................................................................................................... 1 What is “Social Status”? ........................................................................................ 2 How is Status Allocated? ....................................................................................... 4 Characteristics Associated with Status ............................................................ 5 Do the Same Characteristics always Predict Status? ............................................. 6 Contextual Effects on Status Attainment ......................................................... 6 Person by Situation Interactions and Status ..................................................... 9 Current Investigation ............................................................................................. 12 Construct Validation ........................................................................................ 12 Consistency of Status Attainment .................................................................... 13 Person by Situation Interactions ...................................................................... 13 II. STUDY 1 METHODS ............................................................................................ 14 Participants and Procedure ..................................................................................... 14 Participants ....................................................................................................... 14 Informants ........................................................................................................ 14 Procedure ......................................................................................................... 15 Measures ................................................................................................................ 19 Self-Reported Individual Difference Measures ............................................... 19 Perceptions of Informants ................................................................................ 22 Round-Robin Group Ratings ........................................................................... 22 xi Chapter Page Post-Task Ratings ............................................................................................ 23 III. STUDY 1 RESULTS ............................................................................................. 24 Preliminary Analyses ............................................................................................. 24 Data Selection .................................................................................................. 24 Computing Target Effects ................................................................................ 25 Post-Task Manipulation Checks ...................................................................... 25 Construct Validation of Status ............................................................................... 26 Variance Decomposition .................................................................................. 26 Correlations among Status, Influence, Respect, and Liking ............................ 27 Consistency of Status Attainment .......................................................................... 27 Overall Consistency ......................................................................................... 27 Pairwise Consistency ....................................................................................... 28 Personality by Context Interactions ....................................................................... 28 Extraversion ..................................................................................................... 30 Agreeableness .................................................................................................. 30 Conscientiousness ............................................................................................ 31 Neuroticism ...................................................................................................... 32 Openness .......................................................................................................... 32 Dominance, Prestige, and Sense of Power....................................................... 33 Closer Examination of the Problem Solving Activity ..................................... 33 xii Chapter Page Discussion .............................................................................................................. 35 Construct Validation ........................................................................................ 35 Consistency of Status Attainment .................................................................... 35 Person by Situation Interactions ...................................................................... 36 IV. STUDY 2 METHODS ........................................................................................... 38 Participants and Procedure ..................................................................................... 38 Participants ....................................................................................................... 38 Informants ........................................................................................................ 38 Procedure ......................................................................................................... 39 Measures .............................................................................................................. 39 Self-Reported Individual Difference Measures ............................................... 39 Perceptions of Informants ................................................................................ 40 Informant Reports ............................................................................................ 41 V. STUDY 2 RESULTS .............................................................................................. 43 Informant Response Rate ....................................................................................... 43 Relationships with Informants ............................................................................... 43 Construct Validation .............................................................................................. 44 Consistency of Status Attainment .......................................................................... 44 Personality by Relationship Interactions ............................................................... 46 Status across Different Relationships .............................................................. 47 Extraversion ..................................................................................................... 47 Agreeableness .................................................................................................. 47 xiii Chapter Page Conscientiousness ............................................................................................ 48 Neuroticism ...................................................................................................... 48 Openness .......................................................................................................... 48 Dominance, Prestige, and Sense of Power....................................................... 49 Discussion .............................................................................................................. 49 Construct Validation ........................................................................................ 49 Consistency of Status Attainment .................................................................... 50 Person by Relationship Interactions................................................................. 50 VI. GENERAL DISCUSSION .................................................................................... 51 Construct Validation of Status ............................................................................... 51 Consistency of Status Attainment .......................................................................... 53 Person by Situation Interactions ............................................................................ 54 Extraversion ..................................................................................................... 55 Agreeableness .................................................................................................. 56 Conscientiousness ............................................................................................ 57 Neuroticism ...................................................................................................... 58 Openness .......................................................................................................... 59 Dominance, Prestige, and Sense of Power....................................................... 59 Summary .......................................................................................................... 60 Implications............................................................................................................ 61 Limitations and Future Directions ......................................................................... 62 Conclusion ............................................................................................................. 63 xiv Chapter Page APPENDICES ............................................................................................................. 65 A. STUDY 1 TASK MATERIALS ....................................................................... 65 B. STUDY 1 POST-TASK QUESTIONNAIRE ................................................... 82 C. STUDY 2 SINGLE-ITEM MEASURES .......................................................... 86 D. TABLES............................................................................................................ 90 REFERENCES CITED ................................................................................................ 129 xv LIST OF TABLES Table Page 1. Self-report descriptives ......................................................................................... 91 2. Correlations among self-reports ............................................................................. 93 3. Participants and groups by activity ........................................................................ 94 4. Target effect descriptives ....................................................................................... 94 5. Post-task manipulation checks ............................................................................... 95 6. Variance decomposition......................................................................................... 96 7. Correlations among status and related constructs across tasks .............................. 98 8. Consistency of status across tasks.......................................................................... 99 9. Reliability of status target effects .......................................................................... 99 10. Correlations between status and personality by task ............................................. 100 11. Extraversion by task interaction............................................................................. 101 12. Agreeableness by task interaction .......................................................................... 102 13. Conscientiousness by task interaction ................................................................... 103 14. Neuroticism by task interaction ............................................................................. 104 15. Openness by task interaction ................................................................................. 105 16. Dominance, prestige, and power by task interaction ............................................. 106 17. Extraversion by problem solving interaction ......................................................... 108 18. Agreeableness by problem solving interaction ...................................................... 109 19. Conscientiousness by problem solving interaction ................................................ 110 20. Neuroticism by problem solving interaction.......................................................... 111 21. Openness by problem solving interaction .............................................................. 112 xvi Table Page 22. Dominance, prestige, and power by problem solving interaction ......................... 113 23. Participants’ perceptions of informants ................................................................. 114 24. Informant-report descriptives................................................................................. 115 25. Distribution of informants by participant .............................................................. 116 26. Informant relationship type and length .................................................................. 116 27. Quality of relationship with informants ................................................................. 116 28. Correlations among informant reports ................................................................... 117 29. Informant-report variance decomposition .............................................................. 117 30. Status in different relationships ............................................................................. 118 31. Extraversion by relationship interaction ................................................................ 119 32. Agreeableness by relationship interaction ............................................................. 120 33. Conscientiousness by relationship interaction ....................................................... 122 34. Neuroticism by relationship interaction ................................................................. 123 35. Openness by relationship interaction ..................................................................... 125 36. Dominance and prestige by relationship interaction .............................................. 127 37. Sense of and willingness to use power by relationship interaction ....................... 128 1 CHAPTER I INTRODUCTION In small groups, social status is hierarchically distributed amongst group members through a variety of dynamic processes. Status-based hierarchies benefit the group as a whole by facilitating efficient progress towards the group’s goal (Magee & Galinsky, 2008), and the person at the top of the hierarchy reaps the benefits of having high status (e.g., better health, more resources). Because of these benefits, people are typically—and perhaps fundamentally—motivated to seek status (Anderson, Hildreth & Howland, 2015). Yet, given the hierarchical distribution of status, not everyone can attain it. Who attains status? The primary aim of this investigation is to shed light on the contextualized processes that lead to status attainment. Specifically, I examine the nature of status itself, whether the same people tend to consistently attain status, and the extent to which individual differences are associated with status in different situations and relationships. Before exploring the characteristics of people who attain status, we must know what we are measuring when we assess an individual’s status. Status has been defined as the respect with which an individual is regarded as well as the amount of influence that person has over the actions of his or her group (Anderson, John, Keltner & Kring, 2001). Although status is often conflated with respect and influence, these components may not operate in the same way across situations. For instance, someone might be consistently respected but not always have influence over decisions. The first goal of this investigation is to examine how social status is related to respect and influence. The second and third goals of this investigation pertain to the individuals who attain status. A functional perspective of status attainment indicates that whoever is given 2 status should have characteristics that help the group accomplish its goal (Magee & Galinsky, 2008). Because groups may have different goals, it is likely that different people will be given status and different characteristics will be associated with status attainment in different groups. To date, however, researchers have largely focused on characteristics that are robustly associated with status across a wide variety of groups. As a result, an overarching question that remains unanswered is whether the same characteristics—and, perhaps, the same individuals—tend to consistently garner status, or whether status attainment is context-dependent. What is “Social Status”? One important but frequently overlooked problem in investigations of status attainment is defining what, exactly, is meant by “social status.” Social status is often regarded as the amount of respect and influence an individual has in the eyes of his or her fellow group members (Anderson, et al., 2001; Berger, Rosenholtz & Zelditch 1980). Researchers typically measure status with a single item that includes this heterogeneity (e.g., “this person has status (respect, influence)”; Anderson, et al., 2001; Carlson & Lawless DesJardins, 2015), or by aggregating across several items to form a single composite measure (e.g., Anderson, Brion, Moore & Kennedy, 2012). However, this approach obfuscates the possibility that these constructs may operate independently. Indeed, there is some evidence that influence and respect are distinct from one another and should not both be conflated with status. Anderson and colleagues differentiated between rank-order status and respect (Anderson, Willer, Kilduff & Brown, 2012). Rank-order status was more clearly hierarchical in nature, and was defined as similar to leadership or influence. In contrast, respect was not necessarily hierarchical 3 (i.e., everyone in the group could be equally respected), and was not attached to an obligation to lead the group. The authors found that participants consistently desired respect, but would forgo rank-order status if they believed that someone else was better suited to a leadership role. In a different study, participants got to know one another in an unstructured interaction and then rated their group members on a variety of measures conceptually related to status (Lawless DesJardins & Srivastava, unpublished data). These measures included status, which in this study was explicitly defined to participants to include influence, as well as the extent to which group members acted as leaders, liked one another, and respected one another. Status—which was constrained to include influence—strongly correlated with perceived leadership (r = .86). However, status was more weakly associated with respect (r = .38) and liking (r = .44), whereas respect and liking were closely related to each other (r = .64). Because status may be differentially related to influence and respect, it is important to measure and examine these constructs separately, rather than combining them into a single item or scale. This methodological consideration is particularly important when investigating status attainment across contexts. It is possible, for instance, that some people may be well-liked and respected across contexts, but may only influence their group members in contexts where they are able to effectively demonstrate their task-relevant characteristics to others. Moreover, by measuring status, influence, and respect separately, I will gain insight into what individuals mean when they say someone has “status,” and whether participants’ folk concepts match researchers’ theoretical definitions. Given a functional, 4 competency-based model of status attainment (described in detail below), status should be more closely related to influence—to directing the group toward its goal—than to respect. According to these models, status should also be hierarchically structured. That is, everyone in the group should not attain the same amount of status. However, a hierarchical distribution is not a necessary feature of respect, and everyone may be equally respected in a particular group. By asking about status, influence, and respect separately, I can evaluate whether participants’ ratings of “status” corresponds to this theoretical definition—that is, whether it is correlated more strongly with influence than respect, and whether participants make well-differentiated ratings of one another. Finally, although liking is distinct from status, there has been almost no research investigating how the two are related, or whether status might simply be bestowed upon the most well- liked member of the group. How is Status Allocated? When people interact in groups without formally designated leaders (e.g., social groups, students working on a class project), they spontaneously form hierarchies based on social status (Berger, et al., 1980; Anderson, et al., 2001; Magee & Galinsky, 2008). The primary function of a social hierarchy is to reduce intragroup conflict and, in doing so, help the group members achieve their common goal (Milgram, 1974/2009; Magee & Galinsky, 2008). Conflict is reduced, in large part, because the person atop the social hierarchy—the one with the most status—often becomes an informal leader and directs the group’s activities. For example, a student group may have the goal of finishing a class project; the person with the most status might assign tasks (e.g., looking up articles, making a visual aid) to help the group successfully complete the project. Even informal 5 social groups (e.g., groups of friends) may have goals that are more efficiently met when status is hierarchically allocated (e.g., deciding where to eat lunch, resolving an argument amongst group members). Characteristics associated with status. According to theoretical models of status attainment, groups typically have implicit or explicit goals. Group members allocate status to whomever they believe has the characteristics best suited to help the group reach its goal (Berger, et al., 1980; Keltner, Van Kleef, Chen & Kraus, 2008). To do so, the group must know something about its goal and the context in which they are working. Indeed, there is some evidence to suggest that when specific abilities that contribute directly to the group’s goal are known and can be observed by group members, they may be used to confer status (Berger, et al., 1980). This functional theory of status attainment implies that different characteristics may garner status in groups with different goals. However, most empirical work examining status attainment has focused on characteristics that help groups reach their goals across a wide variety of situations. As a result, two very broad characteristics— general intelligence (Lord, De Vader & Alliger, 1986) and social competence (Keltner, et al., 2008)—have emerged as robust predictors of status attainment. In particular, status tends to be allocated to people who are socially engaging. People who readily form social bonds, interact effectively with others, and draw positive attention to themselves often attain status (Keltner, et al., 2008; Anderson & Kilduff, 2009a). Unsurprisingly, then, extraversion—how outgoing, assertive, or sociable someone is (Costa & McCrae, 1995; John & Srivastava, 1999)—is a robust predictor of status attainment (Anderson et al., 2001; Judge, Bono, Ilies & Gerhardt, 2002). Similarly, confident people may tend to 6 attain status across a variety of situations, in part because group members may mistakenly conflate confidence with competence (Anderson & Kilduff, 2009b; Anderson, et al., 2012). Intelligence, social competency, and confidence are associated with status attainment across groups with a wide variety of goals. Indeed, these broad traits are likely to help many groups meet their goals: regardless of the specific task a group needs to complete, giving structural influence to someone who is intelligent and socially engaged is likely to help the group succeed. However, these may not be the only characteristics that are important for status attainment. There remains a paucity of information about whether and how the characteristics that predict status attainment might fluctuate in specific contexts, despite the prominent theoretical position that they must. Do the Same Characteristics always Predict Status? Previous research has identified a handful of broad characteristics that tend to be associated with status attainment, but has done so primarily by examining the personalities of individuals in single, short-term groups. As a result, this work has neglected the interactive relationship between an individual’s characteristics and the group’s goal, and how an individuals’ status might vary across longer-term relationships. Moreover, previous research has made the implicit assumption that the same people (i.e., those who are intelligent, socially competent, and confident) will always attain status. In order to test whether these assumptions hold, it is imperative to examine the same people in multiple groups or across multiple relationships. Contextual effects on status attainment. A functional approach to status attainment indicates that different characteristics should predict status in different 7 situations (Berger, et al., 1980; Fiske & Berdahl, 2007; Keltner, et al., 2008). Despite this perspective, investigations of status attainment have prioritized examination of the general role of broad traits over examination of the specific contexts in which status hierarchies form. Because goals can vary among groups, it is likely that the predictors of status also vary, at least to some extent. That is, if a person does not have the qualities required to help their group reach its goal, that person is not likely to obtain status in that group. In the current investigation, I focus on the personality traits that may differentially predict status across contexts. Possession of specific, task-relevant abilities (e.g., training in negotiation, quantitative skills) may help an individual attain status in highly specific situations. However, personality traits may differ in their relevance to status in contexts that vary in a more abstract way (e.g., those that require competition or cooperation). These abstract variations may more closely approximate the types of situations people frequently encounter. Moreover, personality incorporates broad patterns of social behavior, and researchers have found that specific abilities are less important to status attainment than are interpersonal skills (Anderson, et al., 2012; Anderson & Kilduff, 2009b). Although some traits (in particular, social competency) predict status across a wide variety of situations, emerging evidence suggests that personality traits relevant to the group’s goal may also play a role in status attainment. For instance, although kindness and social warmth may contribute to the positive social engagement that is theoretically important for status attainment, researchers have uncovered a weak and unreliable association between agreeableness and status (e.g., Anderson, et al., 2001). My 8 colleagues and I recently replicated this small main effect of agreeableness. However, when we accounted for context, we found that agreeableness contributed to status attainment in groups with cooperative and affiliative goals, but not in groups with competitive goals (Lawless DesJardins, Srivastava, Küfner & Back, 2015). Similarly, in organizational settings, people tend to attain status in their workplaces when they have characteristics that align with the goals of their organization (Anderson, Spataro & Flynn, 2008). The role of situation-specific predictors has received more attention in studies of leadership. Leadership is similar to (but distinct from) social status, as individuals who attain status often embody leadership roles in informal groups. Early investigators expressed frustration with a trait-based approach to predicting leadership emergence and effectiveness because the results were often inconsistent (e.g., Stogdill, 1948; Yukl & Van Fleet, 1992). In response to this frustration, investigators focused on situational differences but obtained mixed results. They found leadership had both stable and transient components, but did not examine when particular traits mattered (e.g., Barnlund, 1962; Kenny & Zaccaro, 1983). In a more recent meta-analysis, Judge and colleagues (2002) examined the relationship between leadership and Big Five personality traits in three contexts: business, military/government, and academia (elementary through college students). The authors found that the correlates of leadership differed somewhat across these three broad categories. For instance, openness—which includes creativity and divergent thinking—was moderately associated with leadership in business (r = .23) and academic (r = .28) settings, but not in the military (r = .06). Together, these findings lend 9 credence to the theory that the predictors of status might shift across situations, but more investigation is required to determine how and when these shifts occur. Person by situation interactions and status. In order to directly study the interactive effect of situational characteristics and individual differences on status attainment, researchers must examine the same people in different contexts. In this project I examined four distinct situations that often arise in research on status and group dynamics: (1) unstructured social tasks in which the goal is to have group members get to know one another (e.g., Vazire, 2010); (2) competitive tasks in which group members must compete with each other for limited resources (e.g., Anderson, Srivastava, Beer, Spataro & Chatman, 2006; Lawless DesJardins, et al., 2015); (3) cooperative tasks in which group members must cooperate with one another to maximize the success of the group as a whole (e.g., Anderson, et al., 2006; Lawless DesJardins, et al., 2015); and (4) problem solving tasks, where the goal is to solve a variety of concrete problems that have correct answers (e.g., Anderson & Kilduff, 2009b; Anderson, et al., 2012). The goals in each of these tasks differ, so people with different traits might attain status in each of them. I also investigated the predictors of status attainment in different types of longer- term relationships. This examination was exploratory, and I did not have specific hypotheses about the effects of different traits in different relationships. However, it is likely that these longer-term relationships would involve cooperation and affiliation. As a result, I expected agreeableness to predict status attainment across relationships. Trait-specific hypotheses. Past research demonstrates that extraversion consistently predicts higher status (Anderson, et al., 2001; Judge, et al., 2002, Lawless DesJardins, et al., 2015), and I expect it to do so across all four types of situations. 10 Extraverts are, by definition, outgoing and assertive, but not necessarily brusque or rude (DeYoung, Weisberg, Quilty & Peterson, 2013; John & Srivastava, 1999). Because extraverts can effectively draw positive attention to themselves and form positive social bonds, they are likely to attain status across many social situations (Keltner, et al., 2008). Likewise, individuals with low extraversion tend to be withdrawn and submissive, and are not expected to gain status. The remaining Big Five traits have been inconsistently linked to status in previous research (Anderson, et al., 2001; Judge, et al., 2002; Lord, et al., 1986). It is possible that this inconsistency is because the effect of context has not been taken into account. Highly agreeable people are generally compassionate, warm, and pro-socially oriented (DeYoung, et al., 2013; John & Srivastava, 1999). Moreover, people with a tendency to act warmly or compassionately toward others tend to invite similarly warm and compassionate behavior in return (Horowitz et al., 2006; Markey, Funder & Ozer, 2003; Tiedens & Fragale, 2003). In groups where members need to cooperate or affiliate with one another, allocating status to an agreeable group member may help the group achieve their cooperative goal (Lawless DesJardins, et al., 2015). As a result, I expect high agreeableness to predict status in unstructured social tasks and cooperative tasks (where agreeableness could encourage group cohesion), but not in competitive or problem solving tasks (Lawless DesJardins, et al., 2015). People who are neurotic are prone to unstable, negative moods (John & Srivastava, 1999), and may be particularly reactive to stressful situations, like status competitions (Bolger & Schilling, 1991). Because of these tendencies, highly neurotic individuals may be especially prone to withdraw from their groups. Just as extraverted 11 individuals may attain status by speaking up and engaging with their group members, neurotic individuals may be especially likely to lose status as they disengage from their group members. Indeed, neuroticism may be particularly detrimental for status attainment in unstructured and cooperative tasks, where social engagement might be particularly beneficial. Conscientiousness and openness may both be especially important in tasks that require problem solving, where organization and divergent thinking may help the group complete a variety of problems. Conscientiousness is defined by tendencies to be diligent, organized, and detail-oriented (John & Srivastava, 1999). These qualities may help conscientious individuals organize their group members, especially in clearly delineated tasks. Additionally, highly open individuals tend to be intelligent and imaginative (John & Srivastava, 1999), and often attain positions of leadership (Judge, et al., 2002). These characteristics might be particularly useful for status attainment in situations that require creative problem solving. In addition to examining the Big Five traits, I explored the extent to which their aspects differentially predicted status in different contexts. These provide a more nuanced look at the aspects of each trait that might be associated with status. For instance, the two aspects of neuroticism are withdrawal and emotional volatility (DeYoung, Quilty & Peterson, 2007). If, as I hypothesize, neurotic individuals fail to attain status, these exploratory aspect-level analyses may help determine whether low status is due to a tendency to withdraw from the group versus a tendency to be emotionally volatile. Finally, I investigated the roles of dominance, prestige, and sense of power in status attainment. Dominance and prestige are different strategies individuals may use to 12 attain status (Cheng, Tracy, Foulsham, & Kingstone, 2012). Individuals who use dominance strategies tend to be overbearing and aggressive, and use assertiveness and intimidation to force their way to the top of the hierarchy. In contrast, people who use prestige strategies tend to use their expertise to gain respect and, in turn, status (Cheng, et al., 2012). Both of these strategies are viable ways to attain status, and, like extraversion, may be expected to predict status across a variety of situations. Similarly, a person who has a high sense of power—and willingness to use that power—may be likely to attain status because they generally feel confident and self-assured (Anderson, John, & Keltner, 2012). Current Investigation Status is a multifaceted construct with many social benefits. The primary aim of this dissertation is to examine the extent to which specific characteristics of particular situations moderate the effect of stable individual differences on status attainment. In the first study, participants completed up to four tasks with different goals and different group members. In the second study, informants provided reports about their relationships with the participants. In both studies, I investigated the relationships among status, respect, influence, and liking, the consistency with which the same individuals attained status, and the extent to which the relationships among personality traits and status were moderated by context. Construct validation. First, I examined the extent to which status was associated with influence, respect, and liking. In the first study, I obtained group members’ perceptions of these constructs and examined the extent to which participants were rated similarly on these dimensions when they interacted with different groups. In the second 13 study, I inspected the associations among these dimensions across different types of long- term relationships outside the laboratory. Consistency of status attainment. Prior to investigating whether different traits predicted status in different situations, I assessed the consistency with which the same individuals attained status across contexts. In the first study, I examined whether the same people attained status in different groups comprised of different members. In the second study, I explored the extent to which individuals consistently attained status across several long-term relationships. Person by situation interactions. Finally, I investigated how specific contextual factors (e.g., competition, cooperation) moderated the effect of stable individual differences on status attainment. That is, did the same traits predict status across situations, or did different situations favor different traits? In the first study, I examined whether the same characteristics predicted status in groups with different members and goals. In the second, more exploratory study, I examined the characteristics that were associated with status in different types of relationships (e.g., roommates vs. romantic partners). 14 CHAPTER II STUDY 1 METHODS In the first study, participants completed up to four tasks with four different groups with non-overlapping membership. The objectives of this study were to investigate the associations among status-relevant constructs, the consistency with which the same individuals attained status, and the extent to which different characteristics predicted status attainment in groups with social, competitive, cooperative, and problem solving goals. Participants and Procedure Participants. I recruited 346 participants (68% female; Mage = 19, SDage = 2.29; 69% White) during Spring Term and Fall Term 2015 from an introductory psychology course at the University of Oregon. Participants received extra course credit for their participation. The target minimum number of groups was k = 70 in order to obtain power of .80 for small correlations (ρ = .2) among group members’ ratings of each other (Lashley & Kenny, 1998). Informants. Participants provided valid email addresses for N = 1,159 informants (247 participants nominated at least one informant). Of those nominated, 38% (N = 443; 61% female; 73% White; Mage = 30, SDage = 15.12) completed the online survey. As a result, 180 participants had at least one informant. Among participants who had at least one informant report, they had, on average, two informants (M = 2.5; SD = 1.5; range = 1 - 7). On average, informants had known participants for 12 years (M = 11.69, SD = 7.65; range = 1 – 27). Informants who were nominated by participants recruited in Spring 2015 15 did not receive compensation. Informants nominated by participants recruited in Fall 2015 were entered into a drawing to receive $50. Informants were contacted via email after the in-person data were collected. These emails contained a brief description of the study, the name of the participant who nominated the informant, and a link to the online survey. Informants were asked to complete the survey within one month of receipt of the initial email and were sent two reminder emails. After providing consent, informants described the nature of their relationship with the participant and the participant’s personality. Informant reports were not analyzed for Study 1, but were combined with the reports obtained in Study 2. Informants completed a subset of the personality measures that participants completed, and these are described in the Methods section of Study 2. All results pertaining to informant reports are also reported in Study 2. Procedure. After signing up for an in-person session, participants were invited via email to complete an online questionnaire. Participants were asked to complete the questionnaire prior to attending the in-person session. After providing consent, participants answered questions about themselves and then nominated up to eight informants who knew them well and who could answer questions about them. Fourteen in-person sessions were conducted in a large classroom furnished with round tables to facilitate group interactions. Six sessions occurred in May 2015; eight sessions occurred in October and November 2015. There were 315 who participants completed the in-person session. Upon arriving to the session, participants received a folder labeled with their name and participant identification number. These folders contained (a) the consent form, (b) materials for saliva collection, (c) four round-robin 16 rating forms, (d) the Lost on the Moon ranking sheet, and (e) the Problem Solving test packet. Participants also received a bottle of water and a pen. Participants were directed to sit at any table for the first portion of the study. At the beginning of the session, participants read and signed the consent form (this was the same consent form they had already completed online). Then, participants were instructed to provide a saliva sample. While providing this sample, participants completed a questionnaire about their activities during the day and provided a brief medical history.1 After participants completed the questionnaire, they were randomly assigned to their first group. Groups were assigned at random, with the contingency that all groups would have non-overlapping membership (i.e., no two participants would interact with one another more than once). This feature of the design ensured that status attainment across tasks would not be due to any carry-over effects from group members previously working with one another. The number of groups was based on the number of participants present for the session. Participants were directed to look at a projection screen, where they could see their participant identification number as well as the table to which they were assigned. 2 Participants then went to their assigned tables, and I provided verbal instructions for the first task. Each of the four tasks is described in detail below. Participants were given approximately 10 minutes to complete the task. Participants were then asked to stop their activity, get out their first round-robin rating sheet, and to prop up their folders 1 This questionnaire is relevant to the analyses of neuroendocrine markers in saliva. Neither the questionnaire nor the saliva samples were analyzed for this dissertation, and they will not be discussed further. 2 This procedure was followed for all except the first session of data collection. In the first session, participants had been assigned to groups prior to their arrival (rather than during the session). Because of the high number of no-shows, however, this strategy was abandoned for later sessions. 17 as dividers. Participants were given approximately 12 minutes to complete their round- robin forms. They were then assigned to their next group, and this procedure was repeated up to four times. Participants in six sessions completed all four activities; participants in six sessions completed three of the activities; participants in two sessions completed just two activities. The number of activities completed in a session was determined by the number of participants who arrived for that session: for example, sessions comprised of 20 or 21 participants could only complete three activities because there were not enough participants to create a fourth set of non-overlapping groups. The order of activities was randomized, and participants in different sessions completed the activities in a different order. All participants in the same session completed the activities in the same order and at the same time. After participants completed their final set of round-robin ratings, they were debriefed as a group, given an opportunity to ask questions about the study, and were thanked for their participation. Activities. Every participant completed up to four activities that had different goals. Participants completed tasks with (1) a broadly defined social goal; (2) a competitive goal; (3) a cooperative goal; and (4) a knowledge-based problem solving goal. The goals of these tasks were intended to differentiate between the characteristics that are required to obtain status. Participants completed each task with different groups that had non-overlapping membership; that is, any two participants only interacted with each other once. In some sessions, participants were unable to complete all four tasks because too few participants attended to allow for four non-overlapping groups. In these 18 sessions, participants always completed the competitive and problem solving activities, because these tasks were designed to have the most distinct goals. When participants completed three activities, the third activity alternated between the social and cooperative tasks. The full instructions and materials for each task can be found in Appendix A. Social task. Participants engaged in a loosely structured interaction in which they were simply asked to get to know one another as well as possible in 10 minutes. To ensure that each group member had an opportunity to speak, participants were instructed to first take turns introducing themselves, and then to use the rest of the time to get acquainted with their group members. Competitive task. Participants engaged in a leaderless group discussion (LGD; adapted from John & Robins, 1994 and Staw & Barsade, 1993) in which they were asked to take on the role of a regional representative of the University of Oregon alumni association. As a group, the representatives formed an awards committee charged with distributing $100,000 of prize money to the nominees. Participants were instructed to advocate for and to get the largest allocation of the available funds for their nominee. Each participant represented one of five nominees; assignment of the nominee was random, with the condition that no two members of the group could represent the same nominee. Participants were given two minutes to individually read the information about their own nominee. Then, as a group, participants had 10 minutes to select a first-prize winner. The first-prize winner was to receive the largest monetary award as well as the most public acclaim. They then determined how to divide the remaining prize money amongst the rest of the nominees. Participants were instructed to give a different amount of money to each nominee (i.e., the prize money could not be divided evenly). As a 19 result, participants competed over the allocation of limited resources, and had to work both for the interest of their chapter’s nominee and the broader interest of the committee. Cooperative task. Participants completed a modified Lost on the Moon (LOM) task (adapted from Robins & Beer, 2001), in which they were asked to imagine that they had just crash landed on the moon and needed to prioritize a limited set of items to help them get back to the mother ship. In order to incentivize cooperation, groups were instructed to work together to develop a ranked list of items that would optimize the entire group’s ability to get back to the ship. Participants were given two minutes to think about the items and to create their own rankings of the items’ importance. Groups then had 10 minutes to work together to develop their final list of rankings. Problem solving. Participants were asked to solve, as a group, 30 verbal and quantitative problems drawn from practice SAT questions (Kahn Academy, 2015). Each participant was given a copy of the question booklet. Every booklet contained the same multiple-choice questions in the same order. Participants were instructed to work together to solve as many of the questions as they could in 10 minutes. Although participants were welcomed to work in their own booklet, each group was asked to produce a single answer sheet at the end of the allotted time. Measures Self-reported individual difference measures. Participants completed a series of questionnaires online. These questionnaires assessed individual differences that may predict status attainment. Means, standard deviations, and reliabilities (where appropriate) for all self-report measures can be found in Table 1 (see Appendix D for all tables). Correlations among the self-reported measures can be found in Table 2. 20 Big Five Personality. Participants completed both factor- and aspect-level measures of the Big Five personality traits. To assess factor-level personality, participants completed the 44-item Big Five Inventory (BFI; John & Srivastava, 1999). The BFI measures extraversion (e.g., is outgoing, sociable), agreeableness (e.g., is helpful and unselfish with others), conscientiousness (e.g., makes plans and follows through with them), neuroticism (e.g., worries a lot), and openness (e.g., has an active imagination). Participants also completed the 50-item Big Five Aspect Scale (BFAS; DeYoung, Quilty & Peterson, 2007). The BFAS provides a more nuanced measurement of the Big Five by assessing two aspects of each superordinate trait. The two facets of extraversion are enthusiasm (e.g., has a lot of fun) and assertiveness (e.g., takes charge). The two facets of agreeableness are compassion (e.g., sympathizes with others’ feelings) and politeness (e.g., avoids imposing their will on others). The two facets of conscientiousness are industriousness (e.g., carries out my plans) and orderliness (e.g., keeps things tidy). The two facets of neuroticism are withdrawal (e.g., easily discouraged) and volatility (e.g., gets angry easily). Finally, the two facets of openness are intellect (e.g., quick to understand things) and openness (e.g., see beauty in things that others might not notice). Both measures were completed on a scale from 1 (strongly disagree) to 5 (strongly agree). Power. Participants completed the Personal Sense of Power Scale (PSPS; Anderson, John & Keltner, 2012). The 16-item PSPS assesses an individual’s general sense of power in their relationships (e.g., I can get people to listen to what I say; I think I have a great deal of power) as well as their willingness to use power (e.g., I use my 21 power whenever I can; If I can get what I want, I’ll take it). The PSPS was completed on a scale from 1 (strongly disagree) to 7 (strongly agree). Dominance-Prestige. Participants completed the Dominance-Prestige scale (Cheng, Tracy & Henrich, 2010). This 17-item scale assesses the extent to which participants use dominance (e.g., I am willing to use aggressive tactics to get my way) and prestige (e.g., Members of my peer group respect and admire me) strategies to gain status. This measure was completed on a scale from 1 (not at all like me) to 7 (very much like me). Narcissism. Participants completed the 16-item version of the Narcissistic Personality Inventory (NPI-16; Ames, Rose & Anderson, 2006). This scale uses binary response options (e.g., I really like to be the center of attention vs. It makes me uncomfortable to be the center of attention) to assess sub-clinical narcissism. The number of narcissistic responses are summed for a final score that can range from 0 – 16. Need to Belong. Participants completed the 10-item Need to Belong Scale (Leary, Kelly, Cottrell & Schreindorfer, 2013). This scale assesses participants’ need to be around (e.g., Being apart from my friends for long periods of time does not bother me) and accepted by (e.g., I want other people to accept me) other people. All items were completed using a scale from 1 (strongly disagree) to 5 (strongly agree). Demographics. Participants completed several demographic questions. In addition to basic demographic questions (i.e., age, sex, race, whether English was their first language), participants also answered questions about their education (i.e., year in school, major) and educational performance (i.e., college GPA, high school GPA, standardized test scores). Finally, participants answered several questions about their 22 socioeconomic status. They responded to three versions of the MacArthur Scale of Subjective Social Status (Adler, Epel, Castellazzo & Ickovics, 2000). This measure asks participants to rate their social status, relative to other people in the United States, on a scale from 0 (lowest status) to 10 (highest status). Participants were asked to rate their social status (1) when they were growing up, (2) now, and (3) where they aspire to be upon graduation. Participants also indicated their current social class (upper class, upper- middle class, middle class, lower-middle class, and lower class) and their social class while they were growing up. Perceptions of informants. Participants were asked to nominate up to eight informants. They then reported basic information about the nature and length of their relationship with each informant. Participants also reported on the quality of their relationships with each informant. Questions about relationship quality included closeness, liking, similarity, and the extent to which the participant felt like their “true selves” with the informant. Finally, participants indicated how much status they had in their relationships.3 Round-robin group ratings. Following each task, participants rated themselves and their group members on a variety of dimensions. Participants used a scale ranging from 0 (disagree strongly) to 10 (agree strongly) to record their impressions. These included the 10-item Big Five Inventory (Rammstedt & John, 2007) as well as items about each group members’ interaction with the group (e.g., Had the right skills or expertise for this particular task; Was difficult to work with). Participants also indicated the extent to which they knew each of their group members prior to the interaction on a 3 The question about status with informants was omitted from the questionnaire completed by participants who were recruited during Spring 2015, and is only available for participants who were recruited during Fall 2015. 23 scale from 0 (I have never met this person before today) to 4 (We were already friends before today). The full questionnaire can be found in Appendix B. Status and related constructs. Participants used the round-robin questionnaire to rate their own and their group members’ status. Importantly, the item assessing status did not define it (e.g., as being related to either influence or respect). To indicate status, participants responded to the item, this person had high status in the group. Separately, participants rated how much influence their group members had during the activity (this person had influence over decisions). Participants also indicated the extent to which they liked (I like this person) and respected (I respect this person) their group members. Post-task ratings. Following each activity, participants completed eight items that assessed their perceptions of the task and of their group as a whole. These questions included the extent to which group members cooperated and competed with one another, the extent to which the task required special knowledge, and how well group members got to know one another. These questions also assessed the extent to which each participant enjoyed the task and liked working with their group. All questions were answered using a scale from 0 (disagree strongly) to 10 (agree strongly). The full text of the questions can be found in Appendix B. 24 CHAPTER III STUDY 1 RESULTS Preliminary Analyses Data selection. Because of the round-robin design of the group activities, Social Relations Modeling (SRM; Kenny, 1994) was used to analyze the group data. SRM provides estimates of the proportion of variance in the ratings that are attributable to the target (i.e., the person who is being rated) and the perceiver (i.e., the person doing the rating), and allows for computation of unbiased mean ratings. However, SRM requires that groups be comprised of four or more individuals. With fewer than four participants, the proportion of variance attributable to perceivers versus targets cannot be estimated and the unbiased ratings cannot be computed. As a result, small groups were dropped from analyses using these ratings (i.e., all analyses of the round-robin ratings). Two sessions had exceptionally low attendance, so participants completed activities in dyads and triads. These participants were dropped from analyses (n = 18). Data were also excluded if participants had previously worked with any member of their group. As a result of the unexpected no-shows in the first session, participants completed two tasks with one group, and then switched and completed two tasks with a second group. The first task with each group (i.e., their first and third activities) were retained for analyses, while the second task with each group (i.e., the second and fourth activities) were dropped. Two additional groups from later sessions were dropped from analyses because two of their members had worked with each other on a previous activity. The final analyses are based on N = 297 participants interacting in k = 240 groups. The breakdown of participants and groups by activity can be seen in Table 3. 25 Although participants were recruited from the same, large introductory course, most did not know each other prior to the experimental session. Ninety percent of the dyadic combinations of participants reported that they had never met; 4% reported that they had seen each other but never spoken; and just 2% reported that they were friends prior to the study. Because the overwhelming majority of participants were strangers, all participants were analyzed together, regardless of their previous relationship with other participants. Computing target effects. To assess group members’ perceptions of one another (including their perceptions of status), I computed target effects for each variable with the TripleR package in R (Schönbrodt, Back, & Schmukle, 2012). The target effect is a bias- corrected average rating (Kenny, 1994). It reflects the average rating a participant receives from his or her group members, weighted by the number of group members, and adjusted for systematic missingness in the data (i.e., every member in the group is rated by a slightly different set of perceivers, because members’ self ratings are not included in the calculation). Target effects are often centered about the group mean in order to account for differences across groups, but the uncentered effects might also be used when group differences are accounted for by a random effect of group. The means and standard deviations of the uncentered effects can be found in Table 4. Post-task manipulation checks. Descriptive statistics for each of the post-task manipulation check items can be found in Table 5. Overall, participants cooperated and got along well with one another. Importantly, however, participants reported being most competitive in the LGD (M = 4.18, SD = 3.27), and least competitive in the Social task (M = 1.35, SD = 2.03). In general, participants reported that they cooperated with their 26 group members. However, they cooperated less in the Problem Solving task (M = 8.20, SD = 1.96) than in the LGD (M = 8.59, SD = 1.64), and cooperated most during the Social task (M = 8.88, SD = 1.63). Construct Validation of Status Variance Decomposition. SRM analyses were conducted with the TripleR package in R (Schönbrodt, et al., 2012). Variance was decomposed into three components: target variance, perceiver variance, and error variance. The proportion of variance explained by targets provides an estimate of the extent to which group members agree about particular others (i.e., it is a measure of consensus). The proportion of variance explained by perceivers provides an estimate of the extent to which different perceivers use different baseline levels in their ratings across targets (i.e., it is a measure of the influence of individual differences on the tendency to see others in a particular way). The decomposition for status, liking, respect, and influence can be seen in Table 6. Consistent with previous studies, agreement about status was quite high (whole sample, 29%). As might be expected, agreement about status was lowest in the social task, though it was still substantial (20%). Participants also agreed with each other about who had the most influence in their groups (whole sample, 29%). Agreement about liking was somewhat lower (whole sample, 14%), which is in line with what has been reported elsewhere (e.g., Carlson & Lawless DesJardins, 2015). Unexpectedly, agreement was exceptionally low for respect (whole sample, 6%). This is likely a result of ceiling effects, as participants tended to rate one another very highly on this item (M = 8.2, SD = 1.11; scale: 0 - 10). 27 Correlations among status, influence, respect, and liking. The correlations among status, influence, respect, and liking can be found in Table 7. Confirming the assumption that lay ideas about status might be most related to influence (vs. respect), status and influence were strongly related to each other (r = .85). As expected, liking and respect were also strongly related to one another (r = .71). The remaining correlations were moderate (.26 < r > .38). Taken together, these results indicate that participants interpret “status” to be more closely related to influence than to respect or liking. Additionally, the presence of ceiling effects and the extremely low degree of consensus for respect indicate that respect is not hierarchically distributed. As a result, the focus of subsequent analyses will be on the predictors of status. Consistency of Status Attainment Overall consistency. To examine the relative consistency of status across tasks, I obtained the pairwise correlations of the group-mean centered target effects, which account for group dependencies. Although the focus was on whether participants consistently attained status, I included the correlations among the target effects of respect, liking, and influence for comparison. I then used the Fisher’s r-to-z transformation to compute an average association across all tasks, and this average was transformed back to r. Full results can be found in Table 8. Overall, consistency across tasks was highest for status (r = .30). Unsurprisingly, influence was similarly consistent (r = .27). Consistency was somewhat lower for liking (r = .19), and was lowest for respect (r = .10). However, these low estimates are, at least in part, due to the low degree of consensus for both liking 28 and respect. As mentioned above, the low degree of consensus is partially explained by the ceiling effects present for both variables. To help account for differences in reliability, I computed disattenuated correlations using the reliability estimates of the target effects for each variable provided by TripleR. Reliability estimates are available in Table 9. The disattenuated correlations indicated that respect was the most consistent (r = .55), followed by status, liking, and influence (all rs = .42). Pairwise consistency. To obtain a finer-tuned look at consistency, I examined the pairwise correlations between tasks. Status (raw r = .42; disattenuated r = .60), influence (raw r = .35; disattenuated r = .54), and liking (raw r = .30; disattenuated r = .66) were most consistent between the Social and LOM activities. Respect was most consistent between the LGD and the LOM task (raw r = .27; disattenuated r = 1.044). Averaging across all four characteristics, consistency was highest between the LGD and LOM task (raw r = .30; disattenuated r = .80). Status was least consistent between the Social and Problem Solving tasks (raw r = .20; disattenuated r = .29), and between the Social task and the LGD (raw r = .21; disattenuated r = .32). Personality by Context Interactions To determine the extent to which the task moderated the effect of personality on status, I fit a series of multi-level regressions in which participants (i) were partially crossed with group (j). I first fit a main-effects model (Equation 1, below) that included just the personality variable predicting status attainment. I then fit a model (Equation 2, below) that included the personality variable and its interactions with task. For each 4 Disattenuated correlations can be greater than 1 for a variety of reasons, including non-normally distributed errors and underestimation of the reliability coefficient (Charles, 2005). 29 regression, the target effect of status obtained from the SRM analyses was used as the dependent variable. The uncentered effects were used in these models to aid interpretation. Group dependencies were accounted for with the inclusion of the random effect of group membership. Self-reported personality variables were included at the participant level, and task was included at the group level. The personality measures were those from the pre- interaction online questionnaire and were grand-mean centered prior to inclusion in the models. Custom, hypotheses-driven contrast codes were created to test the differences among tasks for each personality variable. Only random intercepts were estimated. The basic form of the models are included in Equations 1 and 2 below, where TaskCn indicates a contrast code for task. status.tij = g00 + g01Personality + eij + u0i + u0j (1) status.tij = g00 + g01Personality + g10TaskC1 + g11TaskC1*Personality + (2) g20TaskC2 + g21TaskC2*Personality + g30TaskC3 + g31TaskC3*Personality + eij + u0i + u0j In addition to these more formal analyses, I also ran pairwise correlations between self-reported personality and the group-mean centered target effects of status in each task. These correlations can be found in Table 10. Although these correlations can provide a descriptive look at the relationships among status and personality in each task, they do not appropriately account for the repeated-measures nature of the data. 30 Extraversion. I expected a main effect of extraversion to emerge, and did not expect this effect to differ across tasks. As a result, for all models in which extraversion and its facets predicted status, task was dummy coded with the LGD as the referent. As anticipated, the main effect model revealed a large main effect of extraversion on status (b = 0.42, SE = 0.08, t(285.88) = 5.35, p < .001), such that extraverted participants were given more status (i.e., a 1-point increase on the 5-point extraversion scale resulted in a 0.42-point increase on the 10-point status scale when the model did not account for task) . This was also the case for the two aspects of extraversion, enthusiasm (b = 0.48, SE = 0.10, t(289.99) = 4.67, p < .001) and assertiveness (b = 0.49, SE = 0.09, t(281.79) = 5.37, p < .001). Interestingly, the interaction model revealed a small interaction between extraversion and the Problem Solving task (simple b = 0.28; interaction b = -0.26, SE = 0.15, t(660.7) = -1.78, p = .08), indicating that extraversion was less strongly related to status in this task than in the LGD (simple b = 0.53, SE = 0.11, t(245.8) = 4.42, p < .001). This interaction was more pronounced with enthusiasm (Problem Solving simple b = 0.17; interaction b = -0.39, SE = 0.19, t(658.4) = -2.03, p = .04), but did not appear with assertiveness. The full results can be found in Table 11. Agreeableness. I expected agreeableness to predict status in cooperative or affiliative tasks, but not competitive or problem solving tasks. Although participants reported that they generally cooperated with one another, the LGD was designed to elicit the most competition and least cooperation. Agreeableness might be especially important for status in the Social task because it was meant to elicit cooperation along with affiliation, and both the LOM and Problem Solving activities might be similarly cooperative. To test these hypotheses, task was contrast coded to capture the differences 31 between (a) the LGD (1) and the Social task (-1), (b) the Problem Solving (1) and LOM (-1) tasks, and (c) the Social task (1) and all other tasks (-1). In line with the hypotheses, there was a somewhat larger effect of agreeableness in the Social task (simple b = .65) compared to the rest of the tasks (simple b = -.34; interaction b = 0.36, SE = 0.22, t(656.9) = 1.65, p = .10), though this difference was not statistically significant. Contrary to expectations, however, there was no difference in the effect of agreeableness on status between the Social and LGD groups. There was some indication that agreeableness was more important for status in the LOM (simple b = .28) than in the Problem Solving group (simple b = .03; interaction b = -0.17 SE = 0.10, t(620.9) = -1.67, p = .10), but this difference was not statistically significant. Similarly, the politeness aspect of agreeableness did not predict status. However, there was a main effect of compassion on status attainment (b = 0.42, SE = 0.11, t(283.09) = 3.86, p < .001). The full results can be found in Table 12. Conscientiousness. I anticipated that conscientiousness would be most strongly related to status in the Problem Solving task, and weakly related in the social task. The LOM task involved some degree of problem solving (i.e., finding the optimal rank-order of the available items), so conscientiousness might also predict status in this task. I did not have hypotheses about the effect of conscientiousness in the LGD. To test these hypotheses, task was contrast coded to examine the differences between (a) the Problem Solving (1) and the Social task (-1); (b) the LGD (1) and LOM (-1) tasks; and (c) the Problem Solving task (1) and the rest of the tasks (-1). Contrary to these expectations, conscientiousness did not differentially influence status attainment across the four activities. Unexpectedly, a main effect of 32 conscientiousness did emerge, such that more conscientious individuals tended to attain status across activities (b = 0.26, SE = 0.12, t(275.7) = 2.19, p = .03). Similarly, a main effect of the industriousness aspect of conscientiousness emerged (b = 0.33, SE = 0.11, t(275.79) = 2.89, p < .001), and this was qualified by an interaction that indicated that industriousness was more weakly associated with status in the Problem Solving task (simple b = -0.01) compared to the other tasks (simple b = 0.66; interaction b = -0.46, SE = 0.23, t(649.3) = -2.06, p =.04). The order aspect did not predict status. The full results can be found in Table 13. Neuroticism. I anticipated that neuroticism would be negatively associated with status across all four tasks, and that this might be especially pronounced in the Social activity. To test this, task was contrast coded with the same codes used for the agreeableness analyses. Although there was no main effect of neuroticism, an unexpected interaction emerged that indicated that neuroticism was positively associated with status in the Problem Solving task (simple b = 0.16) compared to the LOM task (simple b = -0.27; interaction b = 0.18, SE = 0.08, t(673.6) = 2.11, p = .04). This effect did not emerge for either the withdrawal or emotional volatility aspects of neuroticism. However, the expected main effect did emerge for the withdrawal aspect (b = -0.17, SE = 0.09; t(284.03) = -1.88, p = .06), indicating that people who scored highly in withdrawal did not attain status (i.e., a 1-point increase on the 5-point withdrawal scale resulted in a 0.17 decrease on the 10-point status scale). Full results can be found in Table 14. Openness. Like with conscientiousness, I anticipated that openness, and especially the intellect facet of openness, would predict status in the Problem Solving task. As a result, the same set of contrast codes was used for theses analyses as for the 33 analyses of conscientiousness. Though a small, positive effect of openness emerged in the main effects model (b = 0.18, SE = 0.11, t(273.58) = 1.72, p = .09), it did not interact with task. The openness aspect of openness was not associated with status. There was, however, a positive main effect of the intellect aspect (b = 0.34, SE = 0.11, t(279.68) = 3.11, p < .001). The interaction model revealed that intellect was more strongly associated with status in the Problem Solving task (simple b = 0.66) than in the Social task (simple b = -0.003; interaction b = 0.46, SE = 0.18, t(677.5) = 2.51, p = .01). Full results can be found in Table 15. Dominance, prestige, and sense of power. Like extraversion, I did not have hypotheses beyond main effects of dominance, prestige, and sense of power on status attainment. As with the analyses for extraversion, task was dummy coded with the LGD as the referent for these analyses. Main effects of both prestige (b = 0.22, SE = 0.09, t(283.66.9) = 2.48, p = .01) and sense of power (b = 0.22, SE = 0.08, t(283.83) = 2.83, p = .01) emerged, though they were not statistically significant when task was included in the model. Dominance did not predict status attainment, and there were no interactions with task. Full results can be found in Table 16. Closer examination of the problem solving activity. In the course of examining the interactive effects of task and personality on status attainment, it appeared that processes guiding the hierarchy formation in the Problem Solving activity may have been rather different than in the other three activities. Indeed, participants reported getting to know each other the least in this task (M = 2.92, SD = 2.45), and that it required the most specialty knowledge (M = 8.85, SD = 1.83). Moreover, a large portion of the groups spent the majority of their time independently working on problems. These groups only 34 engaged with one another at the beginning of the task, when they determined who would work on which problems, and the end, when they reported their answers. This is quite unlike the other tasks, which were primarily spent in discussion. As a result of both these quantifiable and observed differences, I re-ran the above analyses with task dummy coded with the Problem Solving activity as the referent group. The full results of these analyses can be seen in Tables 17 to 22. With regard to extraversion, the effect of enthusiasm was markedly smaller in the Problem Solving task (b = 0.16, SE = 0.16, t(833.4) = 1.04, p = .30) than in the LGD (simple b = 0.54; interaction b = 0.39, SE = 0.19, t(658.4) = 2.03, p = .04) and LOM (simple b = 0.68; interaction b = 0.63, SE = 0.20, t(668.9) = 3.11, p < .001) tasks. Compared with the Problem Solving task (b = -0.14, SE = 0.16, t(782.6) = -0.87, p = .39), agreeableness was more strongly associated with status in the Social task (simple b = 0.39; interaction b = 0.46, SE = 0.21, t(664.5) = 2.21, p = .03), and, to a lesser extent, in the LGD (simple b = 0.23; interaction b = 0.38, SE = 0.20, t(649.6) = 1.90, p = .06) and LOM task (simple b = 0.10; interaction b = 0.35, SE = 0.21, t(620.9) = 1.67, p = .10). The industriousness aspect of conscientiousness was somewhat more strongly associated with status in the LOM task (simple b = .42; b = 0.36, SE = 0.21, t(648.2) = 1.72, p = .09) than in the Problem Solving task (b = 0.15, SE = 0.16, t(775.1) = 0.93, p = .35). Compared to the Problem Solving task, neuroticism was particularly detrimental to status in the LOM (simple b = -0.32; b = -0.35, SE = 0.17, t(673.6) = -2.11, p = .04). As with the primary analyses, the intellect aspect of openness was positively associated with status in the Problem Solving task (b = 0.50, SE = 0.16, t(801.9) = 3.14, p < .001). Interestingly, compared to the Problem Solving task (b = 0.50, SE = 0.16, t(801.9) = 3.14, 35 p < .001), intellect was unrelated to status in the Social task (simple b = 0.04; interaction b = -0.53, SE = 0.20, t(669) = -2.62, p = .01). Finally, dominance had a small positive effect on status in the Problem Solving task (b = 0.16, SE = 0.08, t(782.7) = 1.94, p = .05). In comparison to the Problem Solving task, however, dominance was particularly detrimental in both the LOM (simple b = -0.16; interaction b = -0.23, SE = 0.11, t(661.9) = -2.15, p = .03) and Social (simple b = -0.01; interaction b = -0.25, SE = 0.10, t(615.3) = -2.49, p = .01) activities. Taken together, these results indicate that, compared to other tasks, participants tended to attain status in the Problem Solving task when they were low in extraversion and agreeableness, but high in neuroticism, intelligence, and dominance. Discussion Construct validation. In the first study, participants completed up to four tasks with different groups. The first goal of this study was to separate status from influence, respect, and liking. The results supported the hypothesis that status—which is theoretically hierarchical in nature—was more closely related to influence than to respect. Additionally, the ceiling effects and low degree of consensus that emerged for respect indicated that respect was not hierarchically distributed. As a result, subsequent analyses focused on status in different contexts. Consistency of status attainment. I next examined whether the same individuals consistently attained status across tasks. The results indicated that, on average, status was moderately consistent across tasks. Status was most consistent between social and cooperative tasks, and least consistent between the social and competitive tasks. These moderate correlations indicate that there may be some individuals who consistently attain 36 status, but this is not universally true. That is, I did not find strong evidence to suggest that the same people always attained status. The finding that status was most consistent in the social and cooperative tasks suggests that similar personality traits might predict status in these tasks. Likewise, because status was least consistent in the social compared to the competitive and problem solving tasks, different personality traits might be important for status attainment across these tasks. Person by situation interactions. Finally, I examined whether the same characteristics predicted status in different contexts. I expected that extraversion, neuroticism, and dominance-related measures would predict status across tasks, but that other traits (agreeableness, conscientiousness, and openness) would differentiate between tasks with different goals. The results provided mixed support for these hypotheses. Extraversion predicted status attainment across all four tasks. Agreeableness did not predict status attainment, though there was a small, non-significant interaction that suggested it may have been most important in the Social task compared to the rest of the tasks. Unexpectedly, a main effect of compassion emerged, indicating that more compassionate participants attained status across the four tasks. Moreover, conscientiousness predicted status across tasks, though the industriousness aspect was somewhat less important in the Problem Solving task than in the other tasks. Neuroticism was not associated with status. However, the expected negative relationship between status and withdrawal did emerge, indicating that people who were more withdrawn did not attain status. Surprisingly, a small positive effect of neuroticism emerged in the Problem Solving task, indicating that more neurotic individuals attained status in this activity. Although openness was not associated with status, intellect was, and this 37 relationship was particularly strong in the Problem Solving task. Interestingly, neither dominance nor prestige significantly predicted status in these activities, though there was some evidence of marginal effect of sense of power on status attainment. 38 CHAPTER IV STUDY 2 METHODS In Study 2, I obtained information about participants’ status in their relationships with different people. The primary aims of this exploratory study were (1) to examine the associations among status and related constructs in long-term relationships, and (2) to examine whether different personality traits predicted status across different types of relationships (e.g., parents, friends, romantic partners). Status might serve a similar function in these relationships as it does in short-term groups. However, the associations among status-relevant constructs and among personality and status might be different from those in task-based groups, given that these relationships are longer-term and may have nebulous, frequently changing goals. Participants and Procedure Participants. I recruited 234 (68% female; 72% White; Mage = 20, SDage = 3.39) participants through the Department of Psychology’s Human Subjects Pool. Participants received partial course credit for their participation. To obtain power of .80 with an expected effect size of ρ = .2, the target number of participants was 200. Informants. Participants provided valid email addresses for N = 610 informants (N = 138 participants nominated at least one informant). To incentivize informant responses, informants who completed the survey were entered into a drawing to win $50. Of the informants who were nominated, 34% (N = 208; 65% female; 80% White; Mage = 31, SDage = 16.25) completed the online survey. As a result, 96 participants had at least one informant report. Among participants who had at least one informant report, participants had, on average, two informant reports (M = 2.17, SD = 1.2; range = 1 - 5). 39 On average, participants knew informants for 12 years (M = 12.07, SD = 7.94, range = 1 – 26). Procedure. The data for this study were collected entirely online. Participants were recruited through the Human Subjects Pool and signed up for the experiment via the Psychology Department’s online SONA system. Once they signed up, participants were directed to the online survey in Qualtrics. After providing consent, they completed the self-report measures described above. Participants were then asked to nominate up to eight people from at least three contexts (e.g., friends from home, roommates, romantic partners) who could provide reliable information about the participant. Participants described their relationships with each informant, provided their perceptions of each informant, and then provided their meta-perceptions for each informant. Finally, participants were asked to provide email addresses for each of the informants that they had nominated, were thanked for their participation, and were debriefed. After data collection for participants was completed, informants were emailed a link to a brief survey asking them to describe the personality of the participant who nominated them. Informants provided consent and described the nature of their relationship with the participant and the participant’s personality. Measures Self-reported individual difference measures. Participants completed a number of individual difference measures to assess stable differences that may predict status attainment. Participants completed the same scales as in Study 1, as well as a series of single-item measures. Means, standard deviations, and reliabilities (where applicable) for 40 all self-reported measures can be found in Table 1. Correlations among the self-reported scales can be found in Table 2. Personality scales. Participants completed the same scales described in Study 1, above. These included the BFI, BFAS, PSPS, Dominance-Prestige and NPI. Single-item measures. Participants completed 20 single-item measures that assessed, for example, their self-esteem, need to belong, and life satisfaction. All items were completed using a scale from 1 (strongly disagree) to 5 (strongly agree). The full list of items can be found in Appendix C. Demographics. Participants completed the same demographic questions as in Study 1, described above. Perceptions of Informants. Participants provided information about their relationship with each informant, their perceptions of informants’ personalities, and meta- perceptions of informants’ perceptions of them. Descriptive statistics for participants’ perceptions and meta-perceptions of informants can be found in Table 23. Relationship. Participants reported basic information about the nature and length of their relationship with the informant. Participants also reported on the quality of their relationships with each informant. Questions about relationship quality included closeness, liking, similarity, and the extent to which the participant felt like their “true selves” with the informant. Finally, participants indicated how much status they had in their relationships. Informant personality. Participants completed a 10-item measure of the BFI (Rammstedt & John, 2007) for each of their informants. They also reported on their informants’ self-esteem, intelligence, and status. All of the personality measures were 41 completed on a scale ranging from 1 (strongly disagree) to 5 (strongly agree), and started with the prompt, “INFORMANT is someone who…” Meta-perceptions. Finally, participants reported their meta-perceptions—how they thought informants perceived them—of each informant. Participants reported meta- perceptions for the Big Five (measured with the BFI-10) and the 20 single-item measures that were included as self-reports. These measures were completed on a scale ranging from 1 (strongly disagree) to 5 (strongly agree), and started with the prompt, “INFORMANT sees me as someone who…” Informant Reports. Descriptive statistics of informant reports can be found in Table 24. Informants recruited for Study 1 and Study 2 reported their impressions of the participants’ personalities using the same online measures. Informants completed a subset of the personality measures that participants completed: BFI, BFAS, PSPS, Dominance- Prestige Scale, and eight single-item measures (self-esteem, intelligence, trustworthy, funny, arrogant, has influence, need to belong, and respect). The single-item measures were answered on a scale from 1 (strongly disagree) to 5 (strongly agree). Informants also reported on the length, nature, and quality of their relationship with the participant. These questions were answered on a scale from 1 (not at all) to 9 (very much). To further elucidate the nature of their relationship, informants completed an open-ended question asking them to describe the types of situations that they are typically in with the participants. Status and related constructs. Informants completed two measures of status. In the first, they indicated “how much status does [the participant] have in your relationship” on a scale from 1 (very little) to 9 (a lot). In the second, they indicated 42 whether the participant “had status,” in general, on a scale from 1 (strongly disagree) to 5 (strongly agree). Informants indicated how much they liked the participant (How much do you like this person?) using the 1 (very little) to 9 (a lot) scale. Informants also provided reports of how much they respected the participant (I respect [Participant]) and how much influence the participant had in the relationship ([Participant] has a lot of influence in our relationship) using the 1 (strongly disagree) to 5 (strongly agree) scale. Informant demographics. Finally, informants self-reported demographic information, including age, sex, race, and whether English was their first language. Demographic information was the only self-reported information obtained from informants. 43 CHAPTER V STUDY 2 RESULTS Informant Response Rate Participants recruited for Study 2 provided their perceptions of 1,375 informants. However, email addresses were provided for just 716 informants, and, of these, only 610 email addresses were valid (i.e., recruitment emails sent to them did not bounce). Of the 610 informants who were contacted, N = 208 completed the online survey. Because of the relatively low informant response rate in Study 2, these data were combined with the informant reports from Study 1. Although participants in Study 1 completed only a subset of the survey used in Study 2, informants from both studies completed identical surveys. As a result, analyses are based on the responses of N = 651 informants reporting on N = 276 participants. In this combined sample, participants had reports from an average of three informants (M = 3.19, SD = 1.58). The distribution of informants across participants is available in Table 25. Relationships with Informants Both participants and informants provided information about their relationships with one another. A subset of participant-informant dyads (N = 113) disagreed about their relationship classification (e.g., one indicated they were friends from college, the other indicated that they were roommates). For these cases, the participant’s response was treated as the default “correct” response, because participants had less missing data for this question. When possible, discrepancies were resolved using the open-ended responses from informants. For instance, one participant indicated that their informant was their roommate, whereas the informant indicated that they were friends from college. 44 In their open-ended response, however, the informant said that they lived with the participant. In this case, the relationship type was recoded to roommate. To determine which relationship category was most accurate, six independent coders read each situation description and indicated whether they felt the relationship category provided by the informant was more accurate than that provided by the participant. Relationship classifications were changed to reflect what the informant said when it was selected by four or more of the coders. The number of informants of each relationship type can be found in Table 26. Among participants who had more than one informant, the majority (97%) had reports from at least two different types of informants. Overall, participants and informants thought highly of each other and of their relationships with one another. Descriptive statistics for relationship quality variables can be found in Table 27. Construct Validation Correlations among informant-reports of status, influence, respect, and liking can be found in Table 28. The strongest associations were among liking and status in the relationship (r = .51) and liking and respect (r = .52). Status in the relationship was also strongly associated with influence (r = .40). Interestingly, however, ratings of the participants’ status, in general, were only moderately related to liking (r = .21), influence (r = .24), respect (r = .22), and status in the relationship (r = .11). Consistency of Status Attainment To assess whether informants agreed with one another about participants’ status, I fit multi-level models in which informants were nested within participants. These models 45 predicted informant-reported status-related variables from a random intercept, as can be seen in Equation 3, below. Personality_Informantij = g00 + u00 (3) I then computed the intra-class correlation (ICC) for each facet of status to determine the percentage of variance that was attributable to participants. These ICCs reflect the extent to which different informants agreed with one another about a given participants’ personality. Higher ICCs indicate more agreement. Variance decomposition, including the ICCs, can be found in Table 29. This analysis is parallel to the variance decomposition in Study 1, in which I examined the extent to which group members agreed about one another’s status. The majority of participants only had one informant, and those who had more than one informant often did not have the same kind of informants (e.g., one participant may have reports from a friend and a mother, whereas another participant has reports from a romantic partner and a roommate). This resulted in extremely low power for the pairwise correlations across relationships (average n for each comparison = 8), so I did not examine the correlations of status ratings across different relationships. Informants agreed with one another about how much status the participants had, in general (ICC = .17). In contrast, they did not agree with each other about how much status participants had in their relationships (ICC = .05). They also did not agree with one another with regard to how much they liked (ICC = .06) or respected (ICC = .07) the participant, nor about how much influence (ICC = .08) participants had. However, as in 46 Study 1, this low agreement may, in part, be due to ceiling effects: overall, informants did like (M = 8.39, SD = 1.16) and respect (M = 4.75, SD = 0.50) participants, participants had status in their relationships (M = 7.88, SD = 1.49), and participants were seen as having influence (M = 3.95, SD = 0.80).5 Personality by Relationship Interactions To examine the role of different personality traits in attaining status in different types of relationships, I first fit main effects models in which self-reported personality predicted informant-reported status in their relationship (Equation 4, below). I then fit a second set of models that included self-reported personality, type of relationship, and the interaction between the two as predictors of status ratings (Equation 5, below). Informants (i) were nested within participants (j) in both models. For the interaction models, relationship type was dummy-coded. Mothers and fathers were grouped into a single relationship type (“parent”), and brothers and sisters were also grouped together (“siblings”). Because exceptionally few aunts (n = 7), cousins (n = 5), coworkers (n = 15), and teammates (n = 3) responded, these informants were grouped together to form a single category (“other”). Because they comprised the largest group, friends (n = 175) were used as the referent group. Each model was fit with each of the dummy codes, the grand-mean centered self-reported personality variable, and the interactions between the personality variable, each dummy code, and a random intercept. The dependent measure was the informant’s rating of the participant’s status in their relationship. Full results can be found in Tables 30 to 37. 5 Liking and status in relationship were rated on a 1-9 scale; respect and influence were rated on a 1-5 scale. 47 Status_in_Relationshipij = g00 + g01Personality (4) Status_in_Relationshipij = g00 + g01Personality + g10Parent + g20Sibling (5) + g30Other + g40CollegeFriend + g50Roommate + g60RomanticPartner + g70Parent*Personality + g80Sibling*Personality + g90Other*Personality + g100CollegeFriend*Personality + g110Roommate*Personality + g120RomanticPartner*Personality + uij Status across different relationships. Compared to their friends, participants’ parents ascribed more status to participants (b = 0.99, SE = 0.16, t(438.80) = 6.13, p < .001), whereas participants were given less status by their college friends (b = -1.06, SE = 0.22, t(454.10) = -4.74, p < .001) and roommates (b = -0.56, SE = 0.23, t(455.9) = -2.41, p = .02). Extraversion. Given that extraversion robustly predicts status attainment in small groups, I anticipated that extraversion would also predict status in long-term relationships. Interestingly, however, there was no main effect of extraversion (b = - 0.05, SE = 0.08, t(150.78) = - 0.63, p = .53) on status attainment, and this effect was not moderated by relationship type. Similarly, neither enthusiasm (b = 0.02, SE = 0.11, t(142.05) = 0.18, p = .86) nor assertiveness (b = - 0.10, SE = 0.11, t(171.48) = - 0.93, p = .35) predicted status attainment in these relationships. Agreeableness. Because long-term relationships often require affiliation and cooperation, I anticipated that agreeableness would predict status attainment. In line with this hypothesis, there was a main effect of agreeableness on status attainment (b = 0.27, 48 SE = 0.13, t(162.84) = 2.17, p =.03), indicating that a 1-point increase in agreeableness (scale: 1 - 5) resulted in a 0.27 increase in status (scale: 1 – 9). Compared to friends (b = -0.13, SE = 0.16, t(408.20) = -0.80, p =.42), agreeableness had a somewhat larger effect on status with participants’ romantic partners (simple b = 1.76; interaction b = 1.05, SE = 0.57, t(436.70) = 1.85, p = .06). There were no effects of either compassion or politeness on status attainment. Conscientiousness. There was no effect of conscientiousness or its aspects, industriousness and order, on status attainment in relationships. There were, however, small interaction effects that indicated that both industriousness (simple b = -1.21; interaction b = -0.65, SE = 0.36, t(448.3) = - 1.83, p = .07) and order (simple b = -1.00; interaction b = -0.57, SE = 0.33, t(446.2) = - 1.71, p = .09) were less important for status attainment with roommates compared to friends, though neither effect reached significance. Neuroticism. Compared to friends, neuroticism was a somewhat stronger predictor of status attainment in relationships with romantic partners (simple b = 0.75; interaction b = 0.72, SE = 0.40, t(443.20) = 1.80, p = .07). Surprisingly, this effect was positive: romantic partners tended to give more status to participants who were more neurotic. This effect did not emerge with the withdrawal aspect, but was apparent with the volatility aspect (simple b = 0.54; interaction b = 0.81, SE = 0.44, t(446.8) = 1.85, p = .06). Openness. I did not have hypotheses about whether openness would contribute to status attainment in long-term relationships. There was no effect of openness (b = 0.05, SE = 0.12, t(161.71) = 0.39, p = .70) on status attainment, and this effect was not 49 moderated by status. Similarly, neither intellect (b = 0.02, SE = 0.12, t(148.66) = 0.16, p = .87) nor the openness aspect (b = 0.03, SE = 0.12, t(180.71) = 0.27, p = .79) contributed to status attainment. Dominance, prestige, and sense of power. As with extraversion, there was no main effect of dominance (b = - 0.07, SE = 0.07, t(192.22) = - 1.13, p = .26) on status attainment in long-term relationships. However, compared to friends, low dominance was a somewhat stronger predictor of status attainment among college friends (simple b = -1.54; interaction b = -0.38, SE = 0.22, t(448.90) = -1.75, p = .08). Similarly, there was no main effect of willingness to use power (b = - 0.03, SE = 0.14, t(190.35) = -0.64, p = .52), but, compared to friends, low willingness to use power was a stronger predictor of status among college friends (simple b = -1.70; interaction b = -0.54, SE = 0.25, t(449.6) = -2.17, p = .03). Interestingly, in both cases, participants who were less dominant attained status with their college friends. There were no effects of prestige or sense of power on status. Discussion Construct validation. Interestingly—and in contrast to the findings in Study 1— status in participants’ relationships with their informants was more closely related to how well-liked they were than to how much influence they had in the relationship. However, as in Study 1, liking and respect were highly related to each other, and influence was still rather strongly associated with status in the relationship. The higher correspondence between status and liking in these relationships may be due to the longer-term nature of these relationships. That is, as people get to know one another (i.e., for longer than 15 minutes in a laboratory setting), status and liking may begin to converge. 50 Consistency of status attainment. Overall, informants and participants thought very highly of one another. In general, informants agreed with each another about participants’ status, but did not agree about how much they liked or respected participants, or about how much status participants had in their relationships. However, low agreement across these measures was likely a consequence of ceiling effects. Person by relationship interactions. Finally, I examined whether status attainment in relationships was moderated by personality. Although there were relatively few differences among the predictors of status attainment in different types of relationships, these results do give some insight into the differences between friends, romantic partners, and college friends. Specifically, compared to friends, individuals who were agreeable and neurotic tended to attain status in their romantic relationships. Additionally, low dominance and low willingness to use power was more important for status attainment among college friends compared to hometown friends. In part, the difference between hometown and college friends may have to do with the length of their relationships. Participants were primarily young college students, and their college friendships were just beginning to form, whereas they often knew hometown friends for much longer. At the beginning of these (potentially long-term) relationships, it might be particularly important to use low-dominance tactics (e.g., not being aggressive or rude) so as not to alienate others. 51 CHAPTER VI GENERAL DISCUSSION The primary aims of this investigation were to examine the extent to which status is consistent or inconsistent across different situations and relationships, and whether specific characteristics of particular situations moderate the effect of stable individual differences on status allocation. An additional aim included addressing the construct validity of status itself. In two studies, I examined the relationships among the various constructs associated with status and the extent to which different characteristics were associated with status in different types of situations and relationships. Construct Validation of Status Status has been defined to include both influence and respect (Anderson, et al., 2001). However, some research has indicated that respect may differ from status, at least insofar as people are motivated to attain it (Anderson, et al., 2012). Moreover, functional theories of status attainment indicate that status is hierarchically distributed and, as such, fosters more efficient decision-making in the group (Magee & Galinsky, 2008). Whereas influence may meet these criteria, respect is not necessarily hierarchical, nor does it necessarily result in an individual guiding group decisions. Additionally, it is not yet clear whether status is simply bestowed upon someone who is well-liked. I tested the relationships among status, influence, respect, and liking in two studies. In the first study, I obtained group members’ perceptions of influence, respect, status, and liking, and examined the extent to which these dimensions correlated with each other across several different groups. These groups were small, short-lived, task- oriented (i.e., they had a clear goal), and composed primarily of people who had never 52 met before the interaction. In the second study, I examined the correlations among these dimensions in longer-term relationships. I hypothesized that status would be more closely related to influence than to either respect or liking. This was supported in the first study. Whereas status was moderately related to respect and liking, it was nearly indistinguishable from influence. These results indicate that, when participants are answering a question about who has status in their group, they are basing this information on who has the most influence. The amount of influence someone has is highly observable, and, as a result, group members agreed with one another about both how much influence and, in turn, how much status each member had during the activity. In contrast, ratings of respect and liking both had substantial perceiver variance. That is, ratings of respect and liking were idiosyncratic, and based on individual preferences. These results also indicate that, unlike influence, respect is not hierarchically distributed. Both the presence of ceiling effects and the low agreement about who was respected indicated that perceivers did not distinguish among each other with regard to respect in the same way that they did with regard to status and influence. In contrast, the pattern of associations among these dimensions was different in long-term relationships. Although the correlation between status in these relationships and influence was still rather high, status in long-term relationships was more closely related to liking than to either influence or respect. It is possible that as relationships develop over time, liking and status begin to converge. This is consistent with previous reports of the relationship between status and liking, which indicate the association 53 between the two increases as groups get to know each other better over a three-month period (Carlson & Lawless DesJardins, 2015). Consistency of Status Attainment Before investigating whether different characteristics predicted status attainment in different situations and relationships, I first tested the extent to which individuals consistently attained status. In the first study, I examined the correlations among status ratings across different groups. I found that status attainment was moderately consistent across groups. That is, although some individuals may have attained status in multiple groups, many people had different levels of status across their groups. Looking at the pairwise consistency across tasks, status was most consistent between the social and cooperative tasks, and least consistent between the social and competitive and problem solving tasks. This pattern of correlations indicates that status allocation processes were more similar in the social and cooperative tasks than in the other tasks. The social and cooperative tasks were similar in that both involved cooperation and affiliation, whereas the social, competitive, and problem solving tasks were designed to be distinct from one another. Taken further, this pattern suggests that similar personality traits might predict status in the social and cooperative tasks. Likewise, because status was least consistent in the social compared to the competitive and problem solving tasks, different personality traits might be important for status attainment across these tasks. In the second study, I examined the extent to which informants agreed about participants’ status. Interestingly, agreement about status-relevant dimensions was particularly low among informants. Although informants did agree with one another 54 about participants’ status, in general, agreement was considerably lower than that found in the small groups in Study 1. Informants did not agree with one another about participants’ status in their relationships. Relatively low agreement about status across informants may, in part, be due to the different relationships participants had with their informants. That is, they know their informants from different contexts, and have different levels of status in each of those contexts. Unfortunately, I was not able to examine the correlations of status across different types of relationships. Person by Situation Interactions The results discussed above indicated that, while status was somewhat consistent, it was not perfectly stable across groups or relationships—the same individuals were not always attaining status. To help explain why different people attained status, and to address the primary question about whether the relationship between personality and status attainment was moderated by situational factors, I investigated how specific goals and relationships moderated the effect of stable individual differences on status attainment. In the first study, participants completed up to four activities with different groups of people. Each of these activities had different goals—affiliation, competition, cooperation, and problem solving. Because of these distinct goals, I expected different personality traits to predict status in each task. In the second, more exploratory study, I examined the characteristics that predicted status in different types of relationships (e.g., roommates vs. romantic partners). Unlike the experimental groups, these relationships were longer-term, self-selected, and were not necessarily formed to meet a specific goal. These relationships existed prior to the study and likely persisted beyond it, and, as discussed above, status was fairly 55 strongly associated with liking in these relationships. Because these relationships presumably require cooperation and affiliation to maintain, I expected that agreeableness would predict status, but did not have specific hypotheses about the effects of different traits in different relationships. Extraversion. In line with previous work (Anderson, et al., 2001; Judge, et al., 2002; Keltner, et al., 2008), I expected extraversion to predict status in all four tasks. This hypothesis was supported: factor-level extraversion, as well as its two aspects, enthusiasm and assertiveness, were associated with status attainment in all four tasks. The effect of extraversion did not significantly vary across tasks. However, there was some indication that extraversion, and its enthusiasm aspect in particular, was less important for status attainment in the problem solving task. Interestingly, however, extraversion was not associated with status attainment in existing relationships. There are a few possible explanations for this discrepancy. First, extraversion might be particularly important during task-based activities, when groups need someone to speak up or step up to direct the group. Because informants were describing their relationships, in general, rather than recalling a specific time when participants exercised their status, extraversion may be less functionally important. Second, extraversion might be more important when a hierarchy is being established than it is once the hierarchy has been established. This possibility, however, is not generally supported in previous work (e.g., Anderson, et al., 2001). A similar, but perhaps more likely, explanation is that extraversion is more important in more clearly hierarchical social structures than it is in more nebulous situations, like long-term relationships with friends and parents. 56 Agreeableness. Agreeable people are typically compassionate and warm (DeYoung, et al., 2013; John & Srivastava, 1999), and these qualities are associated with inviting others to be compassionate and warm (Horowitz et al., 2006; Markey, Funder & Ozer, 2003; Tiedens & Fragale, 2003). As a result, I expected to replicate previous work indicating that agreeableness would predict status in the social and cooperative tasks, where the groups’ goals involved cooperation and affiliation (Lawless DesJardins, et al., 2015). Similarly, I expected agreeableness to predict status in relationships with informants, which presumably require some degree of affiliation to maintain. Support for these hypotheses was somewhat mixed. In line with the hypotheses, agreeableness was more closely associated with status in the social than in the problem solving task. Descriptively, agreeableness was more important for status in the social compared to the rest of the tasks, and in the cooperative compared to the problem solving task, but neither of these interactions reached significance. In contrast with the primary hypothesis that agreeableness would predict status in the social task but not the competitive task, there was no difference in the effect of agreeableness on status between these two activities. Moreover, an unanticipated main effect of the compassion aspect of agreeableness emerged, indicating that more compassionate individuals attained status, and this effect did not vary across tasks. This effect may be a result of the social pressures of the experimental session: participants were classmates, and may have been particularly motivated to get along with one another, regardless of the goal of the task. Overall, agreeableness was not associated with participants’ status in their relationships with informants. However, it was more closely associated with status with romantic partners than with friends. 57 Conscientiousness. In previous work, conscientiousness has been inconsistently associated with status attainment (e.g., Judge, et al., 2002). Given that conscientious individuals tend to be diligent and organized, I expected conscientiousness to predict status in the problem solving task and, perhaps to a lesser extent, in the cooperative task. In line with these hypotheses, conscientiousness predicted status attainment in groups completing both types of tasks. However, conscientiousness also predicted status in the competitive and social tasks. That is, the conscientiousness consistently predicted status attainment, and was not moderated by task. Although surprising, these results replicate previous work (Lawless DesJardins, et al., 2015, Study 1) that revealed a main effect of conscientiousness across competitive and cooperative tasks in a sample extremely similar to the one in the current study. The industriousness aspect of conscientiousness also predicted status across tasks, but this effect was weaker in the problem solving task compared to the rest of the activities. Conscientiousness was not associated with status in relationships with informants. This null effect may be the result of the nature of the relationship. In particular, because these long-term relationships are less likely to be specifically task-based, conscientiousness may be less important for status in these relationships. Interestingly, however, both industriousness and order were negatively associated with status attainment among roommates, though these effects were not significant. Although these results were surprising—it would not be unreasonable to expect a conscientious housemate who helps to maintain a tidy living space to be both well-liked and to attain status—they are not wholly unprecedented. One possibility is that roommates who are not conscientious demonstrate their influence over the living space, and, in turn, attain 58 status by leaving their belongings around and disregarding other people’s space (Galinsky, Gruenfeld & Magee, 2003; Galinsky, Magee, Inesi & Gruenfeld, 2006). Neuroticism. Neuroticism is associated with unstable, negative moods (John & Srivastava, 1999), and, as a result, neuroticism is often negatively associated with status attainment (e.g., Anderson, et al., 2001; Judge, et al., 2002). I anticipated that people who scored highly in neuroticism would be less likely to attain status, and that this effect might be particularly strong in the social and cooperative tasks, where positive social engagement might be particularly beneficial. As expected, the withdrawal aspect of neuroticism was negatively associated with status across tasks. This effect is consistent with the positive effect of extraversion: participants needed to engage with their group members in order to attain status. The effects of factor-level neuroticism, however, were surprising. Across tasks and relationships, the effect of neuroticism was slightly negative, but was not statistically distinct from zero. Unexpectedly, neuroticism was positively associated with status in the problem solving task, especially when compared to the cooperative task, where neuroticism was (as expected) particularly detrimental to status. Moreover, neuroticism was positively associated with status in participants’ relationships with their romantic partners. There may be a few explanations for theses unexpected positive associations. The problem solving task was a timed math and verbal quiz. It is possible that neurotic participants were more aware of and stressed by the time pressure, and, in pointing this out to their group members, may have helped the group complete more problems. Individuals who are neurotic tend to be extremely sensitive to rejection, and react strongly to real or imagined rejection in their social relationships (Denissen & Penke, 59 2008; Downey & Feldman, 1996). In their relationships with romantic partners, neurotic participants may attain status and influence the relationship in an effort to avoid rejection. Openness. Because openness is associated with intelligence and divergent thinking, I expected it to be associated with status in the problem solving task. Neither factor- nor aspect-level openness was associated with status attainment across the four tasks or in existing relationships. As with conscientiousness, openness may not be important in long-term relationships because they are not generally task-based (e.g., maintaining a friendship does not necessarily require creative thinking). Although there were no effects of openness, there was a main effect of intellect, indicating that it predicted status in all four tasks. This effect was moderated by task: as anticipated, the intellect aspect of openness was a stronger predictor of status in the problem solving compared to the social task. This result supports previous work showing a general effect of intelligence on status attainment (Lord, et al., 1986). It also supports the hypothesis that intellect is especially important in groups performing intelligence- based tasks. Dominance, prestige, and power. Like extraversion, I expected dominance, prestige, and sense of power to predict status attainment across the four tasks. Small main effects of prestige and sense of power emerged, and, though they were in the predicted direction, they were not statistically distinct from zero. Dominance was associated with status attainment in the problem solving task. However, when compared to the problem solving task, dominance was particularly detrimental to status attainment in both the cooperative and social tasks. The dominance scale assesses the extent to which individuals use dominance strategies—including aggression, force, and self-interest—to 60 attain status. Because the cooperative and social tasks were cooperative and affiliative in nature, it makes sense that dominance strategies would work against the interest of the group and would be negatively associated with status. Similarly, dominance was detrimental to status attainment among college friends. In the problem solving task, however, dominance—in particular, a willingness to coerce answers out of group members—may have been an effective strategy to get through as many problems as possible. Summary. Overall, support was mixed for the hypotheses regarding when personality traits would predict status attainment. In large part, these results seem to be driven by the nature of the experimental session. Participants were classmates, and generally seemed to want to get along with each other. Indeed, though participants rated the LGD as the most competitive task, its rating for competitiveness (M = 4.18, SD = 3.27) was below the mid-point of the 0 – 10 scale. Participants also tended to rate each other generally positively, as evidenced by the ceiling effects for respect (M = 8.03, SD = 1.11) and liking (M = 7.59, SD = 1.25). This general social pressure to get along may explain the unexpected main effects of compassion on status attainment, as well as the effect of prestige – but not dominance – on status attainment. The differences that did emerge seemed to be more related to how the task was accomplished rather than the goal of the task itself. The largest differences in the predictors of status attainment were between the problem solving task and the social task: agreeableness was important for status in the social but not problem solving task; intellect was important in the problem solving but not social task; dominance was positively associated with status in the problem solving task but negatively associated in the social 61 task. Similar effects emerged when comparing the problem solving to the cooperative task. Taken together, these results suggest that more pro-social qualities—agreeableness, low neuroticism, low dominance—are important for status attainment when group members are engaged in more socially oriented tasks that involve a high degree of discussion, cooperation, and interaction. When groups engage in less social, more task- driven activities, a different pattern emerges, and less agreeable, more dominant, neurotic, and intelligent individuals attain status. It may be, then, that the predictors of status attainment have less to do with the nature of the group’s goal, and more to do with how that goal is accomplished. Implications Taken together, these results provide some support for a functional theory of status attainment that indicates status should be allocated to individuals who can help groups meet their goals. In short-term, task-focused groups, participants interpreted “status” to be more similar to influence than to respect or liking. This finding lends credence to the theory that status is functional: status attainment has more to do with helping the group reach its goal than it does with being liked or respected. Across tasks and relationships, status attainment was somewhat—but not perfectly—consistent, indicating that the same individuals did not always attain status. To help explain this variation, I examined the extent to which the predictors of status attainment varied across different groups and relationships. Descriptively, the results indicated that groups with different goals ascribed status to people who had characteristics that helped the goal succeed. For instance, agreeableness was somewhat 62 more important in affiliative tasks than in competitive tasks, and intelligence was more important in knowledge-based tasks than in affiliative tasks. However, the results also revealed that differences among groups may be more closely related to how the group accomplishes their goal than to the goal itself. Larger differences in the predictors of status were evident when comparing tasks that involved group discussion to tasks that did not than when comparing tasks with different explicit goals. In particular, more pro-social attributes (e.g., agreeableness, low neuroticism) were associated with status in discussion-based groups than in groups where members worked largely independently of one another. Limitations and Future Directions Despite the strengths of this investigation, it is not without limitation. First, ceiling effects were present for many of the measures of interest. The tendency for participants and informants to rate each other exceptionally highly reduces the reliability of the estimates of these measures, and can distort the relationships among measures. The ceiling effects among informants were not entirely unexpected—participants were likely to nominate informants of whom they thought highly and who thought highly of them. I did not, however, expect ceiling effects to be as pervasive as they were in the group ratings in Study 1. As noted above, participants were enrolled in the same class, and completed the study in large group sessions. Although steps were made to ensure that participants had as much privacy as possible when making their ratings, it is certainly possible that they were concerned their group members might see their ratings. This particular feature of the study sessions, coupled with the relatively high probability that participants would interact with one another again after the session, may have provided 63 an exceptionally good environment for ceiling effects to emerge. It is also possible, however, that participants and informants actually thought very highly of each other. In future studies, it will be important to try to separate these two possibilities by maximizing privacy while ratings are made and by studying populations (e.g., coworkers, strangers) who may be less inclined to produce such uniformly positive ratings. Second, this study did not examine whether the traits of individuals who attained status impacted group efficacy. I theorized that people might attain status when their characteristics align with their group’s goal. To take this further, when these individuals attain status, their groups should perform better than groups who ascribe status to individuals who do not have these characteristics. I did not investigate these outcomes in the current study. However, given the moderate support for the hypothesis that different traits predict status in different situations, this is a promising area for future research. Finally, the investigation into informant reports of status was exploratory. As a result, the findings should be treated as preliminary. Importantly, it was not clear how informants were interpreting the questions about status. As might be expected, informants who had different relationships with participants indicated that participants had different levels of status in those relationships. However, it is not clear how they came to these conclusions, or whether those differences represent a difference in understanding of how status might operate in that relationship. For instance, a child’s status in their relationship with a parent may be very different from their status with a close friend. Though both the parent and the friend may provide similar ratings, they could have different meanings within the context of the specific relationship. 64 Conclusion Social status helps groups meet their goals efficiently and effectively. When people in small, informal groups form a status-based hierarchy, they tend to ascribe status based on group members’ influence over the group’s decisions. In longer-term relationships, however, status is more closely related to liking. Although status attainment is somewhat consistent across contexts, the same individuals do not always attain status, nor do the same characteristics consistently predict status attainment. In this particular study, there were strong social pressures for participants to get along with one another. These pressures revealed themselves in the characteristics that predicted status (e.g., compassion) in group activities. However, the characteristics that predicted status also aligned with the goals of the group. 65 APPENDIX A STUDY 1 TASK MATERIALS Social Task Verbal instructions. This task just involves talking with your group members. During this part of the study, your task is simply to get to know one another. We believe the best way for you to do this is for you all to share information about yourselves with each other. Please start by going around the table and introducing yourselves. Then, spend the rest of the time getting acquainted with all of your group members. Competitive Task Verbal instructions. For this task, you will each be playing the role of a regional representative of the University of Oregon Alumni Association. You have been asked to serve on an awards committee to give out the Jones Foundation Memorial Prizes this year. The Jones Foundation Memorial Prizes are given out each year to a selected group of Oregon alumni. Each prize includes a cash award and corresponding public recognition for the recipient. It is up to you as a committee to decide what criteria to consider while making the decision for this important award. Each regional chapter of the alumni association has selected a nominee for the short list. You must meet as a committee to divide up the $100,000 of prize money among the nominees on the short list. Each of you, as a representative of a different regional chapter, has been assigned to advocate for your chapter’s nominee. Your job is to make sure that your nominee receives full consideration by the committee, and your goal is to get the most money possible for your chapter’s nominee. 66 The information about your nominee is being passed out to you now. Please write your ID, circle group XX, and write the table number on the top. You can use the space on this page to write notes about your nominee before the committee meets. Remember, you want to get as much money as you possibly can for your own nominee. Please take about 2 minutes to review your nominees now. (2 min): Time is up! Now, please meet as a committee to determine the distribution of the prize. There are a few rules regulating how the money can be divided. ➢ You, as the committee, must decide how to divide up $100,000 in prize money among the nominees. ➢ You cannot give all nominees the same amount of money. ➢ You must select one nominee as the First Prize winner who will receive the largest prize and the most recognition. ➢ You must reach a consensus as a group regarding the distribution of prize money among the nominees. When you come to a decision, please write it on the sheets that are coming around now. As long as the rules I just mentioned (which are written on your sheet) are followed, all other decisions about distributing the money should be made by the committee. You have about 10 minutes to make your decisions, starting now. (2 min warning) (10 min): Time is up! Please make sure to write your table number and circle group XX at the top of the award sheet, which we’ll come around to collect now. While we do that, please put the information about your nominee back into your folder and take out a new rating sheet. 67 Group award sheet. One sheet was given to each group. It contained a written version of the instructions, as well as space to allocate the award. 68 Nominee information. Each participant was given a form that had the same written instructions that appeared on the group award form. It also included the following information about the participant’s nominee. Each participant only received information about one nominee. Alexander Chambers. Alexander Chambers is very active in state politics. He has previously served on the board of supervisors of a major city, as well as serving two terms in the state legislature; he now serves as a political consultant. Through connections built up over the years, he has amassed a great deal of influence within the political system, and he has used that influence on behalf of a number of interests, including an anti-tobacco campaign, a petroleum trade group with interests in offshore drilling, and a major retail chain. He has also been active in the debate over funding higher education. People familiar with Alexander say that he is poised to become a major force in national politics during the next Presidential campaign, and many speculate that he could receive a cabinet nomination in the next administration. Barbara Philips. Barbara Philips is a vocal activist in the fight for social justice and environmental protection. She first became involved in social issues several years ago while working for a public policy research institute. The institute issued a number of analyses of the welfare system which were influential on the debate over welfare reform, and Barbara became convinced that she could make a difference in public discourse. She is now working on a number of environmental issues, including preparing a voter ballot initiative that she hopes will preserve open space that is home to several endangered species, while at the same time respecting the rights of property owners and business 69 interests. Friends describe her as passionate and intense about the issues that she takes to heart. Catherine Sullivan. Catherine Sullivan is an artist who has been gaining recognition for blazing new trails in the art world. An exhibit of her work recently opened in a major New York gallery, and the critics have been praising it as a fresh and innovative expression of the artist’s vision. Catherine’s body of work is even more remarkable to those who know her personal history. She was raised in a poor and rather conservative home, and explicitly went against her parents’ wishes by pursuing a career as an artist. From the beginning, her work has avoided typical paradigms and the latest trends in the arts world, and instead has seemed to spring from her own individual perspective. Fellow artists admire her for her strong independence and creativity in all aspects of her life. Donald Yee. Donald Yee is the leader of a moderate-sized religious congregation. In a time when many members of his congregation feel that younger people are growing out of touch with their faith, Donald has taken a lead in ensuring that traditional teachings and practices are not lost. He is often asked to preside over marriages and other religious ceremonies, because people value his ability to bring out the meaning behind the rituals and make them fresh to those in attendance. He developed a curriculum for the youth group that was praised by both the children and their parents; so much so, in fact, that it has been picked up by other congregations as a model for inspiring strong moral development among young people. Friends describe Donald as devout and humble, someone who cherishes his faith’s long-held traditions and lives his life accordingly. 70 Erica Dupre. Erica Dupre was recently named as the CEO of a medium-sized technology corporation. Her current position comes as the culmination of years of hard work and advancement in business. She first began gaining attention when she quickly rose through the management ranks of a large investment bank. When she felt like she had started to reach the limits of what she could achieve in banking, though, she switched to what she perceived as the faster-paced field of high tech. Under her leadership, her corporation’s stock has skyrocketed in value, and its products are considered among the best in the industry. Her friends and colleagues suspect that it will not be long before even bigger corporations come knocking at her door, and it is likely that she will achieve great recognition and admiration in the business world. Cooperative Task Verbal instructions. For this part of the study, you’ll be working together to determine the optimal list of items that could save you from a crash landing on the moon. Please take out the sheet in your folder labeled “Lost on the Moon,” and write your ID, table number, and circle Group XX at the top. Now, imagine you all are members of a space crew originally scheduled to rendezvous with a mother ship on the lighted surface of the moon. Due to mechanical difficulties, your ship landed at a spot 200 miles from the final destination point. The rough landing has ruined your ship and damaged much of the equipment aboard. Only the 15 items listed below are available for facilitating your trip to the destination point. Your team’s survival depends on reaching the mother ship, so you should choose the most critical items available for the 200-mile trip. Now, please first individually think about the importance of each item for you as a group collectively reaching the mother 71 ship. You have 2 min to think about the items’ importance for your group’s success, and should write your rankings down on your sheet. (2m): Time is up! Now, work together with your group members to rank the 15 items in terms of their importance for the crew's success. Please note that there exists an optimal rank order, so carefully discuss the pros and cons for each item. Remember, you will work with your group to determine the ranking that will ensure everyone makes it back to the mother ship. Please make your final group ratings on the sheet that I am passing out now. (2 min warning) (8 min): Time is up! Please write your table number and circle group XX on the top of the group rating sheet. I’ll come around to collect the ranking sheets from each group. While I do that, please put your individual rankings back into your folder and take out a new rating sheet. 72 Individual ranking sheet. Each participant received the following ranking sheet. 73 Group ranking sheet. Each group received one copy of the following ranking sheet, on which they were instructed to make their final rankings. 74 Problem Solving Task Verbal instructions. For this part of the study, you’ll be working together to solve 30 math and verbal questions. On the right side of the folder, find the packet that looks like an exam and is labeled “Problem Solving.” Please write your ID and the table number on the top, and circle group XX. Everyone has the same questions in the same order. Please work together as a group to answer as many of the questions as possible. You can use your packet as scrap paper and you can use a calculator if you need it. I’ll pass out the answer sheets now. Every group will get one answer sheet; please write the answers to the questions on this sheet. You have 10 minutes, starting now. (5 min & 1 min warnings) (10 min): Time is up! If you had your phone out to work on the math problems, please put it away now. Please write your table number and circle group XX on the top of the answer sheet. I’ll come around to collect the answer sheets from each group. While I do that, please put your packet back into your folder and take out a new rating sheet. 75 Answer sheet. Each group received one copy of the following answer sheet, on which to write their final answers. 76 Problem solving packet. Each participant received a copy of the following packet of verbal and quantitative problems. 77 78 79 80 81 82 APPENDIX B STUDY 1 POST-TASK QUESTIONNAIRE Verbal Instructions Instructions after first task. Now, please take out one of the rating sheets from your folder. At the top, please write in your ID#, circle Group 1, and write in the number of the table that you’re sitting at. At the bottom of the page, circle where it says “XXX” for the group task. Go ahead and prop up your folders to make a little space for yourself. You’ll now use that 0 – 10 rating scale to report your impressions. The first column, where it says “YOU”, should be used to make ratings about yourself. On one side of the sheet, some of these boxes are crossed off, and you don’t need to make ratings about yourself for those items. Please start your ratings on this side of the sheet. For the rest of the columns, please write in the first names of your group members. Then use those columns to make your ratings about them. Please make sure you write their names on both sides of the sheet so you can keep track of who you’re rating. It’s important that you treat each of your group members as a unique individual. To do so, please make sure that you aren’t giving everyone the same rating for each trait. Please make sure you answer all of the questions for all of your group members and yourself on both sides of the sheet. Any questions? Go ahead and begin; I’ll let you know when time is up. (12 m): Time is up! Please put your sheet back into your folder. 83 Instructions after subsequent tasks. Again, write your ID, the table number, and circle Group 2 on the top. On the bottom, circle “XXX” for the task. Please prop up your folders now and record your impressions of your group members. Remember to write your group members’ first names at the top of each set of ratings so you can keep track of who you’re rating. Please start your ratings on the side that has questions about the task, and be sure to answer all of the questions on both sides of the sheet about all of your group members. I’ll let you know when time is up. 84 Questionnaire 85 86 APPENDIX C STUDY 2 SINGLE-ITEM MEASURES 87 88 89 90 APPENDIX D TABLES 91 Table 1. Self-Report Descriptives Study 1 Study 2 M SD α M SD α BFI Extraversion 3.35 0.79 0.89 3.21 0.86 0.90 Agreeableness 3.78 0.56 0.79 3.79 0.55 0.78 Conscientiousness 3.53 0.53 0.77 3.46 0.61 0.83 Neuroticism 2.95 0.73 0.83 3.07 0.72 0.84 Openness 3.44 0.60 0.84 3.49 0.56 0.78 BFAS [E] Enthusiasm 3.75 0.61 0.72 3.64 0.68 0.74 [E] Assertiveness 3.47 0.68 0.79 3.36 0.73 0.79 [A] Compassionate 3.92 0.57 0.69 3.96 0.65 0.77 [A] Polite 3.64 0.60 0.6 3.67 0.58 0.53 [C] Industriousness 3.56 0.58 0.69 3.43 0.65 0.76 [C] Order 3.59 0.72 0.75 3.55 0.67 0.69 [N] Withdrawal 2.50 0.73 0.7 2.67 0.77 0.73 [N] Volatility 2.74 0.71 0.78 2.76 0.72 0.78 [O] Intellect 3.53 0.59 0.66 3.54 0.61 0.68 [O] Openness 3.60 0.68 0.74 3.69 0.64 0.66 PSPS Sense of Power 4.87 0.80 0.83 4.75 0.84 0.82 Willingness to Use Power 4.18 0.86 0.77 4.09 0.86 0.77 Dominance-Prestige Dominance 3.29 1.13 0.84 3.20 1.13 0.84 Prestige 5.23 0.71 0.72 5.13 0.87 0.82 Need to Belong Need to belong 3.39 0.66 0.81 3.64 0.99 NPI Narcissism 4.82 3.05 0.71 4.58 3.05 0.72 92 Table 1. (continued). Study 1 Study 2 M SD α M SD α Subjective SES Growing Up 6.45 1.76 6.42 1.79 Now 6.29 1.69 6.23 1.65 Aspire 8.07 1.52 8.00 1.50 Social Class Growing Up 3.35 0.88 3.38 0.98 Now 3.25 0.88 3.24 0.92 Single-Item Measures Self-Esteem 3.33 1.09 Intelligence 3.94 0.71 Status 3.09 0.95 Trustworthy 4.41 0.66 Funny 3.91 0.76 Arrogant 2.34 0.90 Influence over Others 3.54 0.84 Life Satisfaction 3.58 0.95 Speaks Mind 3.49 1.03 Loyal 4.47 0.78 Compassionate 4.38 0.86 Gossips 2.72 1.11 Self-discloses 3.34 1.20 Compliments Others 4.06 0.88 Knows Impression 3.36 1.08 Has Trouble getting Along 2.15 0.90 Tells Truth 3.57 1.07 Patient 3.58 1.06 Good Friend 4.47 0.79 Note: BFI, BFAS, PSPS, Need to Belong, Social Class, and all single-item measures were measured on a 5-point scale. In Study 1, Need to Belong was measured with the full 10-item scale; in Study 2, it was measured with a single item. Dominance-Prestige was measured on a 7-point scale. NPI was scored by summing the number of narcissistic responses selected by the participant, which ranged from 0 - 16. Subjective SES was measured on a 10-point scale. Single-item measures were only asked of participants in Study 2. 93 94 Table 3. Participants and Groups by Activity n k Social 236 56 LGD 278 66 LOM 219 52 Problem Solving 279 66 Note: Participants in every session completed the Problem Solving task and the LGD. The LOM and Social tasks were each completed by half of the sessions that could not complete all four activities. Table 4. Target Effect Descriptives M SD Extraversion 6.69 1.61 Agreeableness 7.20 1.11 Conscientiousness 7.12 1.02 Neuroticism 4.31 1.11 Openness 6.21 0.87 Trustworthy 6.87 1.11 Intelligent 7.52 0.95 Assertive 6.07 1.73 Had the right skills 7.26 1.49 Cooperated 8.37 1.20 Difficult to work with 1.47 1.45 Self-Confident 7.06 1.57 Complemented others 5.39 1.72 Pointed out others' weaknesses 1.25 1.26 Liking 7.59 1.25 Respect 8.03 1.11 Influence 6.70 1.67 Status 6.82 1.53 Note: All responses made on a scale from 0 - 10. Averages are not group-mean centered, and were computed across all four tasks. 95 Table 5. Post-Task Manipulation Checks Whole Sample Social LGD LOM Problem Solving M SD M SD M SD M SD M SD This task was difficult for me. 3.32 3.00 1.57 2.36 2.64 2.62 3.10 2.78 5.64 2.60 I enjoyed the task. 5.75 2.92 7.66 2.34 5.60 2.69 6.08 2.43 4.04 2.92 I would not want to do this task again. 4.73 3.28 2.61 3.00 4.86 2.87 4.62 2.91 6.48 3.11 During the task, my group members and I cooperated with each other. 8.54 1.79 8.88 1.63 8.59 1.64 8.55 1.83 8.20 1.96 During the task, my group members and I competed with each other. 2.76 2.89 1.35 2.03 4.18 3.27 2.59 2.43 2.67 2.81 During the task, my group members and I got to know each other well. 4.64 2.86 7.38 1.98 4.56 2.50 3.99 2.43 2.92 2.45 This task required special knowledge (e.g., about math, English, or space) to complete. 4.33 3.98 0.72 1.84 2.12 2.50 5.27 3.19 8.85 1.83 I liked working with my group. 7.55 2.16 8.27 1.79 7.48 2.13 7.30 2.18 7.19 2.31 Note: All questions measured on a 0 - 10 scale. 96 Table 6. Variance Decomposition Estimate Std. Estimate SE % of Total Whole Sample Status Perceiver 1.85 0.36 0.19 36.41 Target 1.48 0.29 0.14 29.14 Residual 1.75 0.35 0.12 Liking Perceiver 1.67 0.47 0.12 46.87 Target 0.50 0.14 0.09 14.11 Residual 1.39 0.39 0.08 Respect Perceiver 2.12 0.62 0.15 62.10 Target 0.21 0.06 0.07 6.27 Residual 1.08 0.32 0.10 Influence Perceiver 2.04 0.34 0.24 33.66 Target 1.78 0.30 0.18 29.47 Residual 2.23 0.37 0.13 Social Status Perceiver 2.49 0.47 0.54 46.90 Target 1.10 0.21 0.18 20.79 Residual 1.72 0.32 0.23 Liking Perceiver 1.40 0.48 0.17 47.56 Target 0.33 0.11 0.11 11.08 Residual 1.22 0.41 0.14 Respect Perceiver 2.08 0.69 0.35 69.02 Target 0.09 0.03 0.07 2.85 Residual 0.85 0.28 0.13 Influence Perceiver 4.29 0.59 0.73 58.78 Target 1.02 0.14 0.27 13.92 Residual 1.99 0.27 0.27 97 Table 6. (continued). Estimate Std. Estimate SE % of Total LGD Status Perceiver 1.85 0.38 0.40 37.50 Target 1.29 0.26 0.27 26.19 Residual 1.79 0.36 0.33 Liking Perceiver 1.64 0.50 0.21 50.32 Target 0.39 0.12 0.13 11.82 Residual 1.23 0.38 0.16 Respect Perceiver 2.21 0.66 0.32 65.82 Target 0.14 0.04 0.08 4.15 Residual 1.01 0.30 0.27 Influence Perceiver 1.37 0.26 0.33 26.14 Target 1.57 0.30 0.27 30.01 Residual 2.29 0.44 0.23 LOM Status Perceiver 1.56 0.30 0.30 29.48 Target 1.87 0.35 0.37 35.23 Residual 1.87 0.35 0.19 Liking Perceiver 1.70 0.44 0.26 44.17 Target 0.54 0.14 0.21 14.04 Residual 1.61 0.42 0.19 Respect Perceiver 2.02 0.59 0.25 59.42 Target 0.13 0.04 0.07 3.67 Residual 1.26 0.37 0.17 Influence Perceiver 0.83 0.15 0.25 15.10 Target 2.12 0.39 0.40 38.58 Residual 2.55 0.46 0.30 98 Table 7. Correlations among Status and Related Constructs across Tasks Liking Respect Status Respect 0.71 Status 0.38 0.30 Influence 0.30 0.26 0.85 Note: All correlations are significant, p < .05. Table 6. (continued). Estimate Std. Estimate SE % of Total Problem Solving Status Perceiver 1.53 0.32 0.20 31.62 Target 1.67 0.35 0.26 34.53 Residual 1.64 0.34 0.18 Liking Perceiver 1.90 0.46 0.28 45.73 Target 0.74 0.18 0.25 17.84 Residual 1.52 0.36 0.16 Respect Perceiver 2.15 0.56 0.25 56.05 Target 0.47 0.12 0.24 12.35 Residual 1.21 0.32 0.17 Influence Perceiver 1.81 0.29 0.38 28.84 Target 2.37 0.38 0.43 37.77 Residual 2.09 0.33 0.22 Note: Percentage of perceiver variance indicates degree of variance attributable to perceivers; higher values indicate more idiosyncrasies in perceivers' perceptions of targets. Percentage of target variance indicates variance attributable to targets; higher values indicate more agreement among perceivers about targets' traits. 99 Table 8. Consistency of Status across Tasks Status Influence Liking Respect Comparison raw disattenuated raw disattenuated raw disattenuated raw disattenuated LGD - SOC 0.21 0.32 0.28 0.45 0.14 0.31 0.04 0.16 LGD - PS 0.24 0.34 0.21 0.30 0.08 0.15 0.03 0.08 LGD - LOM 0.33 0.47 0.31 0.45 0.29 0.61 0.27 1.07 SOC - PS 0.20 0.29 0.17 0.25 0.19 0.38 0.10 0.29 SOC - LOM 0.42 0.60 0.35 0.54 0.30 0.66 0.06 0.26 PS - LOM 0.36 0.48 0.31 0.46 0.15 0.28 0.08 0.23 Average 0.30 0.42 0.27 0.42 0.19 0.42 0.10 0.55 Note: Correlations among centered target effects; dissattenuated correlations account for variances in the reliability of each target effect. Values in bold are significant, p < .05. Table 9. Reliability of Status Target Effects Status Liking Respect Influence Social 0.65 0.44 0.23 0.59 LGD 0.68 0.47 0.28 0.66 LOM 0.74 0.48 0.23 0.71 Problem Solving 0.75 0.58 0.53 0.77 Note: Reliability was computed across perceivers in the SRM analyses in TripleR. 100 Table 10. Correlations between Status and Personality by Task Whole Sample Social LGD LOM Problem Solving BFI Extraversion 0.20 0.27 0.23 0.17 0.16 Agreeableness 0.03 0.13 0.04 0.03 -0.06 Conscientiousness 0.07 0.02 0.02 0.11 0.13 Neuroticism -0.02 -0.01 -0.06 -0.04 0.01 Openness 0.05 0.11 -0.01 0.05 0.06 BFAS [E] Enthusiasm 0.17 0.19 0.21 0.23 0.07 [E] Assertiveness 0.22 0.20 0.23 0.21 0.24 [A] Compassionate 0.14 0.16 0.13 0.17 0.11 [A] Polite 0.00 0.05 -0.03 0.05 -0.06 [C] Industriousness 0.11 0.12 0.11 0.19 0.04 [C] Order -0.03 -0.01 -0.11 0.00 0.00 [N] Withdrawal -0.07 -0.02 -0.12 -0.08 -0.06 [N] Volatility 0.02 0.00 -0.02 0.02 0.06 [O] Intellect 0.13 0.02 0.15 0.17 0.17 [O] Openness 0.02 0.10 -0.01 0.00 0.00 PSPS Sense of Power 0.12 0.10 0.10 0.16 0.13 Willingness to Use Power 0.03 0.04 0.04 -0.03 0.05 Dominance - Prestige Dominance 0.02 -0.09 0.02 -0.03 0.14 Prestige 0.08 0.10 0.09 0.10 0.05 Note: Status is the group-mean centered target effect. Personality variables were self-reported prior to the interaction. Values in bold are significant, p < .05 101 102 103 104 105 106 107 108 109 110 111 112 113 114 Table 23. Participants' Perceptions of Informants Participant of Informant Meta- Perception M SD M SD Liking 8.45 0.98 8.23 1.10 Extraversion 3.53 1.02 3.46 1.00 Agreeableness 3.75 0.81 3.76 0.76 Conscientiousness 3.93 0.80 3.47 0.75 Neuroticism 2.80 0.97 2.98 1.07 Openness 3.69 0.81 3.75 0.81 Self-Esteem 3.56 1.09 3.55 1.10 Intelligence 4.39 0.64 4.33 0.58 Status 3.88 0.90 3.75 0.83 Trustworthy 4.36 0.65 Funny 4.19 0.71 Arrogant 2.22 0.96 Has influence over others 3.61 0.90 Need to belong 3.44 1.05 Life satisfaction 3.78 0.83 Speaks mind 4.00 0.87 Loyal 4.39 0.57 Compassionate 4.29 0.68 Gossips 2.82 1.08 Self-discloses 3.08 1.22 Compliments others 3.79 0.78 Has trouble getting along with others 2.19 0.95 Tells truth 3.55 0.91 Patient 3.52 1.02 Good Friend 4.40 0.56 Note: Liking measured on a 1 - 9 scale; all other variables measured on a 1 - 5 scale. Data are only available from participants recruited for Study 2. 115 Table 24. Informant-Report Descriptives Study 1 Study 2 M SD a M SD a BFI Extraversion 3.64 0.77 0.88 3.41 0.78 0.87 Agreeableness 3.87 0.68 0.88 3.92 0.60 0.85 Conscientiousness 3.72 0.60 0.83 3.75 0.66 0.87 Neuroticism 2.83 0.73 0.85 2.78 0.77 0.87 Openness 3.62 0.54 0.82 3.70 0.57 0.85 BFAS [E] Enthusiasm 3.95 0.65 0.77 3.81 0.67 0.77 [E] Assertiveness 3.65 0.68 0.79 3.59 0.64 0.72 [A] Compassionate 3.98 0.65 0.80 3.95 0.67 0.82 [A] Polite 3.67 0.68 0.71 3.79 0.61 0.65 [C] Industriousness 3.90 0.61 0.75 3.89 0.64 0.80 [C] Order 3.37 0.72 0.73 3.34 0.71 0.71 [N] Withdrawal 2.42 0.69 0.73 2.39 0.68 0.71 [N] Volatility 2.72 0.81 0.85 2.65 0.72 0.83 [O] Intellect 3.87 0.60 0.73 4.08 0.51 0.71 [O] Openness 3.60 0.60 0.70 3.64 0.63 0.74 PSPS Sense of Power 5.30 0.68 0.71 5.31 0.58 0.65 Willingness to Use Power 4.30 1.01 0.84 4.15 0.98 0.84 Dominance- Prestige Dominance 3.25 1.18 0.85 3.07 1.08 0.82 Prestige 5.66 0.85 0.82 5.73 0.72 0.75 Single-Item Measures Closeness 7.94 1.40 7.96 1.38 Liking 8.56 0.88 8.60 0.82 Relationship Quality 8.10 1.22 8.10 1.17 Status in Relationship 7.86 1.51 7.88 1.49 Self Esteem 3.62 0.98 3.52 1.01 Intelligence 4.59 0.54 4.64 0.50 Respect 4.73 0.53 4.75 0.50 Status 3.88 0.73 3.82 0.71 Trustworthy 4.53 0.63 4.60 0.61 Funny 4.48 0.58 4.48 0.62 Arrogant 1.92 0.96 1.84 0.90 Influence 3.91 0.81 4.03 0.77 Need to belong 3.50 0.99 3.37 1.00 Note: BFI, BFAS, PSPS, and self-esteem through respect measured on a 5-point scale. Closeness, liking, relationship quality, and status in relationship measured on a 9-point scale; Dominance-Prestige measured on a 7-point scale. 116 Table 26. Informant Relationship Type and Length Years Known Relationship Study 1 Study 2 Total % of Total M SD Aunt 5 2 7 1.08% 18.14 0.38 Brother 13 7 20 3.07% 19.4 3.86 Cousin 4 1 5 0.77% 13.4 7.89 Coworker 11 4 15 2.30% 4.45 3.29 Father 40 23 63 9.68% 18.73 3.05 Friend 123 52 175 26.88% 6.75 3.33 Friend from College 62 30 92 14.13% 2.19 2.09 Mother 87 41 128 19.66% 18.91 1.87 Romantic Partner 32 7 39 5.99% 3.3 2.26 Roommate 37 23 60 9.22% 3.49 3.98 Sister 26 18 44 6.76% 18.51 2.54 Teammate 3 0 3 0.46% 5 1.41 Total 443 208 651 Table 27. Quality of Relationship with Informants Participant Report Informant Report M SD M SD r Closeness 7.67 1.58 7.96 1.38 0.53 Liking 8.39 1.16 8.60 0.82 0.25 Relationship Quality 7.90 1.39 8.10 1.17 0.26 Status in Relationship 7.81 1.97 7.88 1.49 0.25 Note: All variables measured on a 1 - 9 scale; combines reports from Study 1 and Study 2. Correlations are between informant- and participant-reported variables. All correlations are significant, p < .05. Table 25. Distribution of Informants by Participant N Informants N Participants % Participants 1 98 36% 2 71 26% 3 52 19% 4 34 12% 5 10 4% 6 8 3% 7 3 1% 117 Table 28. Correlations among Informant Reports Status in Relationship Status Influence Respect Status 0.11 Influence 0.40 0.24 Respect 0.31 0.22 0.32 Liking 0.51 0.21 0.32 0.52 Table 29. Informant-Report Variance Decomposition Variance ICC Status 17% Person 0.09 Residual 0.43 Status in Relationship 5% Person 0.12 Residual 2.11 Liking 6% Person 0.04 Residual 0.64 Respect 7% Person 0.02 Residual 0.24 Influence 8% Person 0.05 Residual 0.59 Relationship Quality 5% Person 0.07 Residual 1.31 Note: ICC reflects the proportion of variance attributable to the target (i.e., it is a measure of consensus); higher values indicate more agreement. 118 Table 30. Status in Different Relationships Variance Estimates b SE df t p Person Residual Intercept 7.89 0.07 178.70 108.50 <.001 0.12 2.11 Intercept 7.73 0.12 395.50 63.42 <.001 0.19 1.62 Parent 0.99 0.16 438.80 6.13 <.001 Sibling 0.40 0.23 446.30 1.76 0.08 Other -0.41 0.30 448.50 -1.36 0.18 College Friend -1.06 0.22 454.10 -4.74 <.001 Roommate -0.56 0.23 455.90 -2.41 0.02 Romantic Partner 0.06 0.28 450.90 0.23 0.82 Note: Relationship was dummy coded with Friend as the referent. Status was the informant-report of the participant's status in the relationship, measured on a 1 – 9 scale. Degrees of freedom estimated using the Satterwaithe approximation. 119 120 121 122 123 124 125 126 127 128 129 REFERENCES CITED Adler, N. E., Epel, E. S., Castellazzo, G., & Ickovics, J. R. (2000). Relationship of subjective and objective social status with psychological and physiological functioning: Preliminary data in healthy, White women. Health Psychology, 19(6), 586-592. Ames, D. R., Rose, P., & Anderson, C. P. (2006). The NPI-16 as a short measure of narcissism. Journal of Research in Personality, 40(4), 440-450. Anderson, C., Hildreth, J. A. D., & Howland, L. (2015). Is the desire for status a fundamental human motive? A review of the empirical literature. Psychological Bulletin, 141(3), 574–601. doi:10.1037/a0038781 Anderson, C., John, O. P., & Keltner, D. (2012). The personal sense of power. Journal of Personality, 80(2), 313-344. Anderson, C., John, O. P., Keltner, D., & Kring, A. M. (2001). Who attains social status? Effects of personality and physical attractiveness in social groups. Journal of Personality and Social Psychology, 81, 116-132. doi: 10.1037//0022- 3514.81.1.116 Anderson, C., Brion, S., Moore, D. A., & Kennedy, J. A. (2012). A status-enhancement account of overconfidence. Journal of Personality and Social Psychology, 103(4), 718-735. doi: http://dx.doi.org/10.1037/a0029395 Anderson, C., & Kilduff, G. J. (2009a). The pursuit of status in social groups. Current Directions in Psychological Science, 18, 295–298. doi:10.1111/j.1467- 8721.2009.01655.x Anderson, C., & Kilduff, G. J. (2009b). Why do dominant personalities attain influence in face-to-face groups? The competence-signaling effects of trait dominance. Journal of Personality and Social Psychology, 96, 491–503. doi:10.1037/a0014201 Anderson, C., Spataro, S.E. & Flynn, F.J. (2008). Personality and organizational culture as determinants of influence. Journal of Applied Psychology, 93, 702-710. doi: 10.1037/0021-9010.93.3.702 Anderson, C., Srivastava, S., Beer, J. S., Spataro, S. E., & Chatman, J. A. (2006). Knowing your place: Self-perceptions of status in face-to-face groups. Journal of Personality and Social Psychology, 91, 1094–1110. doi:10.1037/0022- 3514.91.6.1094 130 Anderson, C., Willer, R., Kilduff, G. J., & Brown, C. E. (2012). The origins of deference: when do people prefer lower status?. Journal of Personality and Social Psychology, 102(5), 1077-1088. doi: http://dx.doi.org/10.1037/a0027409 Barnlund, D.C. (1962). Consistency of emergent leadership in groups with changing tasks and members. Speech Monographs, 29, 45-52. Berger, J., Rosenholtz, S., & Zelditch, M. (1980). Status organizing processes. Annual Review of Sociology, 6, 479–508. Bolger, N. & Schilling, E. A. (1991). Personality and the problems of everyday life: The role of neuroticism in exposure and reactivity to daily stressors. Journal of Personality, 59(3), 355-386. Carlson, E. N., & Lawless DesJardins, N. M. (2015). Do mean guys always finish first or just say that they do? Narcissists’ awareness of their social status and popularity over time. Personality and Social Psychology Bulletin, 41(7). doi: 10.1177/0146167215581712. Charles, E.P. (2005). The correction for attenuation due to measurement error: Clarifying concepts and creating confidence sets. Psychological Methods, 10(2), p. 206 – 226. DOI: 10.1037/1082-989X.10.2.206 Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., & Henrich, J. (2012). Two ways to the top: Evidence that dominance and prestige are distinct yet viable avenues to social rank and influence. Journal of Personality and Social Psychology, 104(1), 103–125. doi:10.1037/a0030398 Cheng, J. T., Tracy, J. L., & Henrich, J. (2010). Pride, personality, and the evolutionary foundations of human social status. Evolution and Human Behavior, 31, 334-347. Costa, P. T., & McCrae, R. R. (1995). Domains and facets: Hierarchical personality assessment using the Revised NEO Personality Inventory. Journal of Personality Assessment, 64, 21-50. Denissen, J. J., & Penke, L. (2008). Neuroticism predicts reactions to cues of social inclusion. European Journal of Personality, 22(6), 497-517. DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10 aspects of the Big Five. Journal of Personality and Social Psychology, 93(5), 880-896. DeYoung, C. G., Weisberg, Y. J., Quilty, L. C., & Peterson, J. B. (2013). Unifying the aspects of the Big Five, the interpersonal circumplex, and trait affiliation. Journal of Personality, 81, 465-475. 131 Downey, G., & Feldman, S. I. (1996). Implications of rejection sensitivity for intimate relationships. Journal of Personality and Social Psychology, 70(6), 1327-1343. Fiske, S. T., & Berdahl, J. (2007). Social power. In A.W. Kruglanski & T.E. Higgins (Eds), Social psychology: Handbook of basic principles (2nd ed.), 678-692. New York: Guilford Press. Galinsky, A. D., Gruenfeld, D. H., & Magee, J. C. (2003). From power to action. Journal of Personality and Social Psychology, 85(3), 453–466. doi:10.1037/0022- 3514.85.3.453 Galinsky, A. D., Magee, J. C., Inesi, M. E., & Gruenfeld, D. H. (2006). Power and perspectives not taken. Psychological Science, 17(12), 1068–1074. Horowitz, L. M., Wilson, K. R., Turan, B., Zolotsev, P., Constantino, M. J., & Henderson, L. (2006). How interpersonal motives clarify the meaning of interpersonal behavior: A revised circumplex model. Personality and Social Psychology Review, 10, 67–86. doi:10.1207/s15327957pspr1001_4 John, O.P. & Robins, R.W. (1994). Accuracy and bias in self-perception: Individual differences in self-enhancement and the role of narcissism. Journal of Personality and Social Psychology, 66, 206-219. John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed., pp. 102-138). New York: Guilford. Judge, T.A., Bono, J.E., Ilies, R. & Gerhardt, M.W. (2002). Personality and leadership: A qualitative and quantitative review. Journal of Applied Psychology, 87, 765-780. doi: 10.1037//0021-9010.87.4.765 Kahn Academy (2015). SAT. [Online Practice Test] Retrieved from https://www.khanacademy.org/test-prep/sat. Keltner, D., Van Kleef, G. A., Chen, S., & Kraus, M. (2008). A reciprocal influence model of social power: Emerging principles and lines of inquiry. Advances in Experimental SocialPsychology, 40, 151-192. Kenny, D. A. (1994). Interpersonal perception: A social relations analysis. New York, NY: Guilford Press. Kenny, D.A. & Zaccaro, S.J. (1983). An estimate of variance due to traits in leadership. Journal of Applied Psychology, 68, 678-685. Lashley, B. R., & Kenny, D. A. (1998). Power estimation in social relations analyses. Psychological Methods, 3(3), 328-338. 132 Lawless DesJardins, N. M., Srivastava, S., Küfner, A. C., & Back, M. D. (2015). Who attains status? Similarities and differences across social contexts. Social Psychological and Personality Science, 6(6), 692-700. doi: 10.1177/1948550615580171. Lawless DesJardins, N.M. & Srivastava, S. (unpublished data). Power, stability, and emotion. Leary, M. R., Kelly, K. M., Cottrell, C. A., & Schreindorfer, L. S. (2013). Construct validity of the need to belong scale: Mapping the nomological network. Journal of Personality Assessment, 95(6), 610-624. Lord, R. G., De Vader, C. L., & Alliger, G. M. (1986). A meta-analysis of the relation between personality traits and leadership perceptions: An application of validity generalization procedures. Journal of Applied Psychology, 71, 402-410. doi: 10.1037/0021-9010.71.3.402 Magee, J. C., & Galinsky, A. D. (2008). Social hierarchy: The self-reinforcing nature of power and status. Academy of Management Annals, 2, 351–398. doi:10.1080/19416520802211628 Markey, P. M., Funder, D. C., & Ozer, D. J. (2003). Complementarity of interpersonal behaviors in dyadic interactions. Personality and Social Psychology Bulletin, 29, 1082–1090. doi:10.1177/0146167203253474 Milgram, S. (1974/2009). Obedience to Authority. New York: Harper Collins. Rammstedt, B., & John, O. P. (2007). Measuring personality in one minute or less: A 10- item short version of the Big Five Inventory in English and German. Journal of Research in Personality, 41(1), 203-212. Robins, R. W. & Beer, J. S. (2001). Positive illusions about the self: Short-term benefits and long-term costs. Journal of Personality and Social Psychology, 80, 340-352. Schönbrodt, F. D., Back, M. D., & Schmukle, S. C. (2012). TripleR: An R package for social relations analyses based on round-robin designs. Behavior Research Methods, 44, 455-470. doi:10.3758/s13428-011-0150-4 Staw, B.M. & Barsade, S.G. (1993). Affect and managerial performance: A test of the sadder-but-wiser vs. happier-and-smarter hypotheses. Administrative Science Quarterly, 38, 304-331. Stogdill, R.M. (1948). Personal factors associated with leadership: A survey of the literature. Journal of Personality, 25, 35-71. 133 Tiedens, L. Z., & Fragale, A. R. (2003). Power moves: Complementarity in dominant and submissive nonverbal behavior. Journal of Personality and Social Psychology, 84, 558–568. doi:10.1037/0022-3514.84.3.558 Vazire, S. (2010). Who knows what about a person? The self-other knowledge asymmetry (SOKA) model. Journal of Personality and Social Psychology, 98, 291-300. Yukl, G., & Van Fleet, D. D. (1992). Theory and research on leadership in organizations. In M.D. Dunnette & L.M. Hough (Eds). Handbook of Industrial and Organizational Psychology, 147 – 197. Palo Alto, CA: Consulting Psychologists Press.