аЯрЁБс > ўџ ўџџџ џџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџьЅС јП 07 bjbjыШыШ Ђj Ђ Ђ вC r J A џџ џџ џџ З D D Ѓ П П П 4 џџџџ ѓ ѓ ѓ h [ d П d ѓ U[ #! ц ' ( 1' 4 e' e' <- <- <- Y Y Y Y Y Y Y $ ы] Ђ ` ЊY e П <- *, <- <- <- ЊY Л Л e' e' Х [ ? ? ? <- Ъ Л К e' П e' ФT Р ? <- Y ? ? т 0R Є u J T e' џџџџ АOS6д0Ъ ѓ 4 Ф дR АT %[ 0 U[ ъR 6 -a Ъ: H -a , T -a П T <- <- ? <- <- <- <- <- ЊY ЊY ? <- <- <- U[ <- <- <- <- џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ џџџџ -a <- <- <- <- <- <- <- <- <- D M : The cost of status enhancement: Performance effects of individuals status mobility in task groups
Corinne Bendersky
UCLA Anderson School of Management
110 Westwood Plaza, A418
Los Angeles, CA 90095-1481
HYPERLINK "mailto:cbenders@anderson.ucla.edu"cbenders@anderson.ucla.edu
(310) 429-1299
Neha Parikh Shah
UCLA Anderson School of Management
110 Westwood Plaza, A416
Los Angeles, CA 90095-1481
neha.shah.2010@anderson.ucla.edu
(310) 825-2505
Under second review at Organization Science. Please do not cite without authors permission.
The cost of status enhancement: Performance effects of individuals status mobility in task groups
Abstract
Although we know that considerable benefits accrue to individuals with high social status, we do not know the performance effects of gaining or losing status in one's group over time. In two longitudinal studies, we measure the status positions of middle managers currently enrolled in a part-time MBA program at the beginning and end of their study group's life. In both samples, we compare the individual performance (course grades) of the students who gained or lost status to those who maintained high and low stable status positions in their groups. We find that higher status at the end of the group's life is associated with higher performance. We also find, however, that the performance of individuals who gain or lose status over time does not correspond to their final status positions. Instead, those who gain status including those who eventually attain high status perform worse than do those who maintain high status positions for the whole quarter, and they perform no better than did those in stable low status positions throughout. Those who lose status over time actually perform as well as those who maintain high status. We interpret these results to suggest that people might trade off resources they could apply to individual performance for opportunities to enhance their status. After replicating this effect in our second sample, we identify over-investment in increasing dominance and generosity as behavioral mechanisms through which individuals successfully gain status to the detriment of their own performance.
There are two popular images of social status (the amount of respect, influence and prominence one has in the eyes of others ADDIN EN.CITE Anderson200111141114111417Anderson, C.John, O. P.Keltner, D.Kring, A. M.Who attains social status? Effects of personality and physical attractiveness in social groupsJournal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.116-1328112001JulISI:000170456400009Flynn20031123112317Flynn, F. J.How much should I give and how often? The effects of generosity and frequency of favor exchange on social status and productivityAcademy of Management JournalAcademy of Management JournalAcad. Manage. J.539-5534652003Oct0001-4273ISI:000186663200002<Go to ISI>://000186663200002 (Anderson, John, Keltner & Kring 2001; Flynn 2003)) in American society. The first is a narrative about status stability that emphasizes how the rewards for status accumulate and reinforce each other over time. This perspective is empirically supported by the Matthew Effect, for instance, whereby recognition for scientific achievement is disproportionately given to established scientists ADDIN EN.CITE Merton196812871287128717Merton, R. K.The Matthew effect in scienceScienceScience56631591968(Merton 1968). The other narrative focuses on status mobility. Opportunities for status mobility in society are among the most primary human motivators ADDIN EN.CITE Loch200013741374137417Loch, C. H.Huberman, B. A.Stout, S.Status competition and performance in work groupsJournal of Economic Behavior and OrganizationJournal of Economic Behavior and Organization35-554312000Hogan19911191119111915Hogan, R.Hogan, J.Gilbert, D. G.Connolly, J. J.Personality and statusPersonality, social skills and psychopathology: An individual differences approach137 - 1541991New YorkPlenum Press(Hogan & Hogan 1991; Loch, Huberman & Stout 2000). This is because we assume that rewards accrue according to our status positions, whether we are born with a silver spoon in our mouth or we scrape our way to the top.
Reflecting its importance in society, status is a subject of intense academic research. Scholars have generally treated it as a very stable property of individuals and groups, fitting more within the first narrative described above. Research has emphasized the characteristics of individuals who are most likely to be attributed with high status by their peers ADDIN EN.CITE ADDIN EN.CITE.DATA (e.g., Anderson et al. 2001; Berger, Cohen & Zelditch 1972; Kalkhoff & Thye 2006; Ridgeway & Erickson 2000) and the interpersonal interactions that reinforce and maintain stable status orders in groups and society ADDIN EN.CITE ADDIN EN.CITE.DATA (e.g., Berger, Ridgeway, Fisek & Norman 1998; Bienenstock & Bianchi 2004; Kalkhoff 2005; Magee & Galinsky 2008; Ridgeway, Diekema & Johnson 1995; Zelditch 2001). Furthermore, scholars have documented the immense benefits that accrue to individuals judged as having high status, such as high compensation ADDIN EN.CITE Belliveau199611531153115317Belliveau, M. A.Oreilly, C. A.Wade, J. B.Social capital at the top: Effects of social similarity and status on CEO compensationAcademy of Management JournalAcademy of Management JournalAcad. Manage. J.1568-15933961996Dec0001-4273ISI:A1996WG80700005<Go to ISI>://A1996WG80700005 (Belliveau, Oreilly & Wade 1996), positive performance evaluations ADDIN EN.CITE Berger198010411041104117Berger, J.Rosenholtz, S. J.Zelditch, M.Status Organizing ProcessesAnnual Review of SociologyAnnual Review of SociologyAnnu. Rev. Sociol.479-50861980ISI:A1980KD94900016<Go to ISI>://A1980KD94900016 (Berger, Rosenholtz & Zelditch 1980), valuable exchange partners ADDIN EN.CITE Thye200010621062106217Thye, S. R.A status value theory of power in exchange relationsAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.407-4326532000JunISI:000089045800005<Go to ISI>://000089045800005 Podolny20051145114511456Podolny, J. M.Status signals2005Princeton, N.J.Princeton University Press(Podolny 2005; Thye 2000), opportunities to influence group outcomes, and favorable resource allocations ADDIN EN.CITE Bunderson200310451045104517Bunderson, J. S.Recognizing and utilizing expertise in work groups: A status characteristics perspectiveAdministrative Science QuarterlyAdministrative Science QuarterlyAdm. Sci. Q.557-5914842003DecISI:000225237100001<Go to ISI>://000225237100001 Bales19583773773775Bales, R.Maccoby, E.Newcomb, T.Hartley, E.Task Roles and Social Roles in Problem-solving GroupsSocial Psychology, 3rd edition437-4471958New YorkHolt Rinehart and Winston(Bales 1958; Bunderson 2003). Additionally, having high status leads others to expect high performance ADDIN EN.CITE Berger19741339133913396Berger, J.Conner, T. L.Fisek, M. H.Expectation States Theory: A Theoretical Research Program1974Cambridge, MAWinthrop Publishers(Berger, Conner & Fisek 1974), which may enhance high status individuals actual performance through an expectancy-confirmation process known as the Pygmalion Effect ADDIN EN.CITE Eden198232533253325317Eden, D.Shani, A.B. Pygmalion goes to boot camp: Expectancy, leadership,and trainee performanceJournal of Applied PsychologyJournal of Applied PsychologyJ. Appl. Psychol.1941996721982Rosenthal19683254325432546Rosenthal, R.Jacobson, L. (1968).Pygmalion in the classroom: Teacher expectation and pupils intellectual development1968New YorkHolt, Rinehart, & Winston(Eden & Shani 1982; Rosenthal & Jacobson 1968). The performance of high status group members is, thus, likely to benefit from a positive, reinforcing cycle of high expectations about performance abilities, favorable behavioral exchanges, and positive attributions, all of which provide greater opportunities and access to resources for high status group members than for low status group members ADDIN EN.CITE Magee200815081508150817Magee, J. C. Galinsky, A. D.Social hierarchy: The self-reinforcing nature of power and statusThe Academy of Management AnnalsThe Academy of Management Annals351-398212008Podolny199632453245324517Podolny, J. M.Phillips, D. J.The dynamics of organizational statusIndustrial and Corporate ChangeIndustrial and Corporate Change453-471521996(Magee & Galinsky 2008; Podolny & Phillips 1996).
Despite the enormous attention to status in the literature, little of it has considered status dynamically. There has been some theoretical research on the delegitimation of status orders; that is, how the validity and acceptance of a status order may change ADDIN EN.CITE Berger199810251025102517Berger, J.Ridgeway, C. L.Fisek, M. H.Norman, R. Z.The legitimation and delegitimation of power and prestige orders.American Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.379-405631998(Berger et al. 1998). Status also has been conceptualized as a resource that can be exchanged ADDIN EN.CITE Turner197113301330133017Turner, J. L.Foa, E. B.Foa, U. G.Interpersonal Reinforcers: Classification, Interrelationship, and Some Differential PropertiesJournal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.168 - 1801921971internal-pdf://turner, foa and foa 1971-2653861121/turner, foa and foa 1971.pdf(Turner, Foa & Foa 1971) and auctioned ADDIN EN.CITE Sutton199699499499417Sutton, R. I.Hargadon, A.Brainstorming groups in context: Effectiveness in a product design firmAdministrative Science QuarterlyAdministrative Science QuarterlyAdm. Sci. Q.685-7184141996DecISI:A1996WC64400007<Go to ISI>://A1996WC64400007(Sutton & Hargadon 1996). There has been some empirical work on shifts in status relationships within groups ADDIN EN.CITE ADDIN EN.CITE.DATA (e.g., De Kelaita, Munroe & Tootell 2001; Goar & Sell 2005; Ridgeway 1982), but none has considered the consequences of gaining or losing status.
Given the benefits of high status, we expect that some individuals will strive to be in these coveted positions and resist being in lower ones. However, the process of moving up in the hierarchy may be costly: People may have to expend significant time, energy and attention toward improving their reputation or exerting their influence ADDIN EN.CITE Loch200013741374137417Loch, C. H.Huberman, B. A.Stout, S.Status competition and performance in work groupsJournal of Economic Behavior and OrganizationJournal of Economic Behavior and Organization35-554312000(Loch et al. 2000). In this process, allocating resources toward status-oriented goals and away from task-oriented ones may detract from individual performance ADDIN EN.CITE Kanfer198914751475147517Kanfer, R.Ackerman, P. L.Motivation and cognitive abilities: An integrative/aptitude-treatment interaction approach to skill acquisitionJournal of Applied PsychologyJournal of Applied PsychologyJ. Appl. Psychol.657-6907441989(Kanfer & Ackerman 1989). Other people may place less emphasis on status, and be content with moving down in the hierarchy ADDIN EN.CITE McClelland19873251325132516McClelland, D.Human motivation1987CambridgeCambridge University Press(McClelland 1987). These individuals may allocate their resources toward their task-oriented goals rather than toward status-oriented ones, which could enhance performance. Thus, our research question is: How do people who gain or lose status perform compared to those who maintain stable high and low status positions in task groups over time?
We study these processes in two longitudinal studies of middle managers currently enrolled in a part-time MBA program whose status positions in their study groups are measured at the beginning and end of their study groups life. In both samples, we compare the individual performance (course grades) of the students who gained or lost status with that of students who maintained high and low stable status positions in their study groups. We call these four groups status categories. Controlling for age, work experience and self-reported satisfaction with the group, we find a positive association between ones hierarchical status position at the end of the groups life and his or her performance. We also find, however, controlling for final status positions, that those who gain status even those who attain the highest status positions in their groups perform significantly worse than did those who maintain high status positions for the whole quarter. In fact, the performance of those who gain status is no better than that of stable low status members. People who lose status, on the other hand, perform no worse than did their stable high status counterparts, despite their lower final status positions. Thus, it appears that people who invest in status enhancement do not reap benefits, and people who fall in the hierarchy do not pay a price, in terms of their individual performance.
In our second study we determine that status seekers trade off behavioral resources to enhance their status for those that would improve their performance. Status losers invest fewer resources towards status enhancement, so are able to concentrate on their performance. Specifically, status seekers over-invest in being generous towards their teammates and communicating assertively beyond the point where these activities harm their performance, compared to their teammates. Status losers, however, are less likely to over-invest in assertive communication, relative to their teammates. This different allocation of resources explains their performance outcomes.
This research contributes to the literature in several important ways. First, it highlights the distinction between status as a dynamic versus a static construct. Despite many forces that legitimate and stabilize status orders in groups, individuals do experience shifts in status positions in their groups over time, and the performance implications are substantial. Second, we identify specific behavioral mechanisms that explain the successful attainment of status at the cost of performance. Third, by using actual, rather than perceived, performance in real task groups, our observations of the tradeoff between status enhancement and performance extend recent theoretical and experimental research in this area ADDIN EN.CITE Loch200013741374137417Loch, C. H.Huberman, B. A.Stout, S.Status competition and performance in work groupsJournal of Economic Behavior and OrganizationJournal of Economic Behavior and Organization35-554312000Huberman200413791379137917Huberman, B. A.Loch, C. H.Onculer, A.Status as a valued resourceSocial Psychology QuarterlySocial Psychology QuarterlySoc. Psychol. Q.103-1146712004Mar0190-2725ISI:000227847300008<Go to ISI>://000227847300008 (Huberman, Loch & Onculer 2004; Loch et al. 2000).
STATUS DYNAMICS
Status hierarchies, rank-ordered relationships among actors [that] describe
interactional inequalities, ADDIN EN.CITE Ridgeway19951377: 281137713775Ridgeway, C. L.Walker, H.A.Cook, K.Fine, G.House, J.Status structuresSociological perspectives on social psychology1995Newton, MAAllyn & Bacon(Ridgeway & Walker 1995: 281) inevitably emerge in groups ADDIN EN.CITE Tiedens200713801380138017Tiedens, L. Z.Unzueta, M. M.Young, M. J.An unconscious desire for hierarchy? the motivated perception of dominance complementarity in task partnersJournal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.402-4149332007Sep0022-3514ISI:000248992100006<Go to ISI>://000248992100006 10.1037/0022-3514.93.3.402(Tiedens, Unzueta & Young 2007). Although many social forces serve to stabilize the emergent social orders, not all status orders are legitimated ADDIN EN.CITE Ridgeway198610971097109717Ridgeway, C. L.Berger, J.Expectations, Legitimation, and Dominance Behavior in Task GroupsAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.603-6175151986OctISI:A1986F162600002<Go to ISI>://A1986F162600002 (Ridgeway & Berger 1986). For instance, Anderson, et al. ADDIN EN.CITE Anderson200111141114111417Anderson, C.John, O. P.Keltner, D.Kring, A. M.Who attains social status? Effects of personality and physical attractiveness in social groupsJournal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.116-1328112001JulISI:000170456400009(2001) measured individuals status in a college dormitory at three points in time: Two weeks into the Fall semester (time 1), after four months (time 2) and after nine months (time 3). The overall status correlation from time 1 to time 2 was .61 and from time 2 to time 3 was .86 (p. 124). Thus, although the status positions stabilized considerably over time, there was still notable variation, especially in the first four months.
Some research suggests that status is continuously re-negotiated as people assert expertise and claim legitimacy ADDIN EN.CITE ADDIN EN.CITE.DATA (Barley 1996; Barley 1986; Bechky 2003; Owens & Sutton 2001; Strauss 1978; Zhou 2005). Research from markets and social networks indicates that status orders may change as a function of the status of ones affiliations ADDIN EN.CITE Lin199911491149114917Lin, N.Social networks and status attainmentAnnual Review of SociologyAnnual Review of SociologyAnnu. Rev. Sociol.467-4872519990360-0572ISI:000082825900019<Go to ISI>://000082825900019 Podolny199632453245324517Podolny, J. M.Phillips, D. J.The dynamics of organizational statusIndustrial and Corporate ChangeIndustrial and Corporate Change453-471521996(Lin 1999; Podolny & Phillips 1996) or exchange partners ADDIN EN.CITE Thye200010621062106217Thye, S. R.A status value theory of power in exchange relationsAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.407-4326532000JunISI:000089045800005<Go to ISI>://000089045800005 (Thye 2000). In task groups, status orders may be delegitimated if task performance requirements change such that expectations of competence become inconsistent with an established hierarchy ADDIN EN.CITE Berger199810251025102517Berger, J.Ridgeway, C. L.Fisek, M. H.Norman, R. Z.The legitimation and delegitimation of power and prestige orders.American Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.379-405631998Goar200511351135113517Goar, C.Sell, J.Using task definition to modify racial inequality within task groupsSociological QuarterlySociological QuarterlySociol. Q.525-5434632005Sum0038-0253ISI:000230417300006<Go to ISI>://000230417300006 (Berger et al. 1998; Goar & Sell 2005). Interventions that make individuals task competencies appear inconsistent with their status characteristics can equalize the status and influence differentials in task interactions ADDIN EN.CITE Markovsky19841189e.g.`, 1189118917Markovsky, B.Smith, L. R. F.Berger, J.Do Status Interventions PersistAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.373-3824931984WOS:A1984ANQ8300008<Go to ISI>://WOS:A1984ANQ8300008 Pugh198311901190119017Pugh, M. D.Wahrman, R.Neutralizing Sexism in Mixed-Sex Groups - Do Women Have to Be Better Than MenAmerican Journal of SociologyAmerican Journal of SociologyAm. J. Sociol.746-7628841983WOS:A1983QE30800004<Go to ISI>://WOS:A1983QE30800004 (e.g., Markovsky, Smith & Berger 1984; Pugh & Wahrman 1983). This may be particularly effective when low status individuals influence attempts are perceived to be motivated by advancing the groups interests rather than their self-interests ADDIN EN.CITE Ridgeway198210541054105417Ridgeway, C. L.Status in Groups - the Importance of MotivationAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.76-884711982ISI:A1982ND54300006<Go to ISI>://A1982ND54300006 (Ridgeway 1982).
In addition to the ways in which status orders may be destabilized due to changes in the task environment, certain types of people are more likely to seek status than others, independent of their demographic characteristics or expertise on a particular task ADDIN EN.CITE Overbeck20051023102310235Overbeck, J. R.Correll, J. Park, B.Thomas-Hunt, M.C. Mannix, E. A.Neale, M.A.Internal status sorting in groups: The problem of too many stars.Research on Managing Groups & Teams169-19972005Oxford, UKElsevierElangovan199912891289128917Elangovan, A. R.Xie, J. L.Effects of perceived power of supervisor on subordinate stress and motivation: The moderating role of subordinate characteristicsJournal of Organizational BehaviorJournal of Organizational Behavior359 - 373201999(Elangovan & Xie 1999; Overbeck, Correll & Park 2005). Some individuals may be fundamentally motivated by a need for power ADDIN EN.CITE McClelland19873251325132516McClelland, D.Human motivation1987CambridgeCambridge University PressWinter19733252325232526Winter, D.The power motive1973New YorkThe Free Press(McClelland 1987; Winter 1973) or status ADDIN EN.CITE Flynn200611551155115517Flynn, F. J.Reagans, R. E.Amanatullah, E. T.Ames, D. R.Helping one's way to the top: Self-monitors achieve status by helping others and knowing who helps whomJournal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.1123-11379162006Dec0022-3514ISI:000242400000010<Go to ISI>://000242400000010 (Flynn, Reagans, Amanatullah & Ames 2006). In their pursuit of prestige, such individuals may challenge an emergent hierarchy in which they are attributed with lower status than they think they deserve ADDIN EN.CITE Porath200812521252125217Porath, C. L.Overbeck, J. R.Pearson, C. M.Picking up the gauntlet: How individuals respond to status challengesJournal of Applied Social PsychologyJournal of Applied Social PsychologyJ. Appl. Soc. Psychol.1945-19803872008Jul0021-9029ISI:000256854700011<Go to ISI>://000256854700011 Polzer20081167116711675Polzer, J., Caruso, H. Brief, A.P. Identity negotiation processes amidst diversityDiversity at Work89 - 1262008Cambridge, UKCambridge University Press(Polzer & Caruso 2008; Porath, Overbeck & Pearson 2008). Thus, individuals status positions in groups may change over time.
THEORETICAL MODEL: RESOURCE TRADEOFFS
Our theoretical model is predicated on the assumption that some individuals actively seek to enhance their status in their groups. Doing so is costly, and potentially more costly than is maintaining a high status position that is conferred at the beginning of a groups life. The costs associated with attaining higher status may undermine the benefits that accrue from ones enhanced structural position in the group. We want to understand this potential tradeoff between status enhancement and performance, but we are not trying to explain the individual or contextual characteristics that lead to status seeking behavior in this study.
Some research indicates that as individuals expend relatively more effort to enhance their status, their performance suffers. Loch, et al. ADDIN EN.CITE Loch200081818117Loch, C. H.Huberman, B. A.Stout, S.Status competition and performance in work groupsJournal of Economic Behavior and OrganizationJournal of Economic Behavior and Organization35-554312000(2000) suggest that individuals have a finite amount of effort that they can allocate between work (enhancing performance) and politics (enhancing status). Whereas Loch, et al. (2000) examine the effects of status seeking behavior on group performance, we consider its effects on individual performance. We assume that the deleterious effects of status seeking behavior may be even more pronounced than in their models because there is no opportunity to free-ride on the work effort of other group members. Thus, as individuals choose between exerting efforts to attain status or working hard at their individual tasks, their status seeking behavior may offset whatever positional benefits their attained status rank would afford.
Empirical evidence supports this resource tradeoff proposition. Huberman, Loch and Onculer ADDIN EN.CITE Huberman200413791379137917Huberman, B. A.Loch, C. H.Onculer, A.Status as a valued resourceSocial Psychology QuarterlySocial Psychology QuarterlySoc. Psychol. Q.103-1146712004Mar0190-2725ISI:000227847300008<Go to ISI>://000227847300008 (2004), for instance, determined that individuals gave up material gain for opportunities to enhance their status. In a two-stage experiment, they required participants to either use their budget for increasing the chances of winning stage 1, in which there was an experimental condition where those winners status would be enhanced through public recognition, or increase their chances of winning stage 2, worth $20. The researchers found that participants in the condition where the winner of stage 1 was publicly acknowledged invested 14% more in stage 1, on average, than did those in the condition without this status-enhancing opportunity. In an organizational example, Flynn ADDIN EN.CITE Flynn200311591159115917Flynn, F. J.How much should I give and how often? The effects of generosity and frequency of favor exchange on social status and productivityAcademy of Management JournalAcademy of Management JournalAcad. Manage. J.539-5534652003Oct0001-4273ISI:000186663200002<Go to ISI>://000186663200002 (2003) indicates that one way individuals gain status in their groups is by exchanging favors; in particular, by being especially generous. He finds that this generosity is costly, however, because overly generous individuals gain no practical advantage from the favors they are owed. Analysis of a curvilinear effect of imbalanced favor exchange on performance suggests that the level of generosity needed to affect status hurts individual performance.
Thus, we posit that the behavioral resources that individuals direct towards status enhancing activities deplete the effects of those resources on their individual performance. We summarize our theoretical model in Figure 1.
Insert Figure 1 about Here
OVERVIEW OF THE STUDIES
We conduct two studies with separate samples of working MBA students at a large, public university in the United States. These are ideal samples for this kind of research for a number of reasons. First, these student groups had no existing formal or prescribed overarching hierarchy. Therefore, we can study the inevitable emergent status sorting processes without confounding the influence of an organizational reporting structure. Second, the groups tasks and consequences are real, objectively measurable and meaningful for the members in the form of a class grade on which their employers tuition reimbursement depends so the setting is more realistic than it would be for a short-term group created in a laboratory to work on an experimental task. Last, the groups all began and ended at the same time and individuals worked on the same tasks, enabling direct performance comparisons that are often difficult in the field.
Although high status may follow performance abilities, having high status in these student groups also may improve performance on individual assignments for several reasons. Higher status individuals may have their queries to the group answered more quickly and more comprehensively than do low status group members. As part-time students, this would be particularly important since the groups meet infrequently. In addition, the high status group members may be able to set the discussion agenda for the group in ways that serve their interests, like working first on assignments with which the person needs the most input from others. The Pygmalion Effect where peoples performance conforms to others expectations ADDIN EN.CITE Rosenthal19683254325432546Rosenthal, R.Jacobson, L. (1968).Pygmalion in the classroom: Teacher expectation and pupils intellectual development1968New YorkHolt, Rinehart, & Winston(Rosenthal & Jacobson 1968) may also benefit high status group members performance in this context.
The research procedure was essentially the same in both studies, although different non-status questions were asked in the two studies. Data were collected from two sections in Study 1, and four sections in Study 2, of 65 to 72 students who were enrolled in a required Organizational Behavior course during the first 10-week quarter of the MBA program. In both studies, students were surveyed during the first week of the course (time 1) and again after the last class and right before taking their final exam (time 2), 10 weeks later. Following protocols used by Anderson, et al. ADDIN EN.CITE Anderson200611461146114617Anderson, C.Srivastava, S.Beer, J.Spataro, S.Chatman, J. A.Knowing your place: Self-perceptions of status in face-to-face groupsJournal of Personality & Social PsychologyJournal of Personality & Social Psychology1094 - 11109162006(2006) and Flynn ADDIN EN.CITE Flynn200311591159115917Flynn, F. J.How much should I give and how often? The effects of generosity and frequency of favor exchange on social status and productivityAcademy of Management JournalAcademy of Management JournalAcad. Manage. J.539-5534652003Oct0001-4273ISI:000186663200002<Go to ISI>://000186663200002 (2003), both surveys asked students to rate each member of their study group, including themselves, on a scale from 1 = very little to 7 = very much, about the following characteristics: 1) To what extent does each individual make valuable contributions to the group, 2) To what extent does each individual influence the groups decisions, and 3) How much status (social respectability) does each individual have in the group?
The time 2 survey in both studies also asked students to rate their satisfaction with the group. In Study 2 only, students additionally rated each others assertive communication and generosity behaviors in both surveys. Demographic data and work histories were collected from the MBA program office. Assignment grades were generated by two teaching assistants (TAs), each of whom graded all assignments for students from both sections on alternating assignments, thereby ensuring within-assignment grading consistency across sections and reducing the potential bias that might be introduced by the TAs familiarity with the students. On a subset of the assignments that both TAs independently graded, they achieved a Cohens Kappa measure of inter-rater reliability equal to .86. Although it is plausible that the TAs sensed the relative status of students in the entire section, they had no knowledge about the status dynamics within each study group, which are the focus of this research. Their grades, therefore, should not have been influenced by the students within-group status ranks. Final course grades were determined based on weighted points from all individual assignments throughout the quarter prior to the authors examining the survey data. Although students were obligated to complete the surveys for course credit, they voluntarily agreed to let the authors use the data for research.
STUDY 1
Before presenting our hypotheses, we ask an exploratory research question about the degree of stability of status orders within groups. As we identified earlier, there has been little empirical research on the evolution of status orders over time to determine the extent to which status positions are immutable. The research by Anderson, et al (2001) is an exception, and as we already reported, they find a high degree of correlation between status positions in a college dormitory over a nine-month period, but considerably lower correlations over the first four-months of the study, especially among women. Their samples were large, non-task oriented social groups, which they studied for an entire academic year. It is unclear to what extent their observations would generalize to smaller, task oriented groups with shorter life cycles, which are more consistent with the majority of status research. We, therefore, first ask how stable are task-oriented peer group status orders over a complete group life cycle?
Hypotheses
Next, we develop hypotheses regarding the effects of status positions over time on individual performance. The extant status literature indicates that the higher ones position is in the groups initial, emergent status order (hereafter called time 1), the higher is ones performance ADDIN EN.CITE Bales19583773773775Bales, R.Maccoby, E.Newcomb, T.Hartley, E.Task Roles and Social Roles in Problem-solving GroupsSocial Psychology, 3rd edition437-4471958New YorkHolt Rinehart and Winston(Bales 1958). This research suggests that, since the group hierarchy is very stable ADDIN EN.CITE Anderson200111141114111417Anderson, C.John, O. P.Keltner, D.Kring, A. M.Who attains social status? Effects of personality and physical attractiveness in social groupsJournal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.116-1328112001JulISI:000170456400009Chase198010461046104617Chase, I. D.Social-Process and Hierarchy Formation in Small-Groups - a Comparative PerspectiveAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.905-9244561980ISI:A1980KV94000002<Go to ISI>://A1980KV94000002 (Anderson et al. 2001; Chase 1980), and benefits accrue to high status individuals, which reinforce their status positions and performance outcomes ADDIN EN.CITE ADDIN EN.CITE.DATA (Berger et al. 1974; Podolny 1993; Ridgeway & Berger 1986), the groups final status order (hereafter, time 2) should be positively associated with performance. In other words, high status should be sustained and lead to positive performance outcomes, as those individuals accrue benefits over time from their structural positions.
Status characteristics theory ADDIN EN.CITE Ridgeway199110941094109417Ridgeway, C. L.The Social Construction of Status Value - Gender and Other Nominal CharacteristicsSocial ForcesSocial Forces367-3867021991DecISI:A1991HB00900004<Go to ISI>://A1991HB00900004 (Ridgeway 1991), however, suggests that the status valuation of personal attributes may unfold over time. She observes that resources are inequitably distributed in society based on nominal demographic factors (e.g. gender) due to various historical and structural forces. Doubly dissimilar encounters interactions where the actors differ both on levels of resources and nominal categories lead the actors to perceive an association between level of resources and nominal category because one nominal group tends to have more resources than the other. Thus, interactions between resource rich members of the advantaged nominal group and resource poor members of the disadvantaged nominal group enhance the associations between status characteristics and performance expectations, legitimating the status order over time ADDIN EN.CITE Ridgeway199810551055105517Ridgeway, C. L.Boyle, E. H.Kuipers, K. J.Robinson, D. T.How do status beliefs develop? The role of resources and interactional experienceAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.331-3506331998JunISI:000074355100002<Go to ISI>://000074355100002 (Ridgeway, Boyle, Kuipers & Robinson 1998). Status characteristics theory, therefore, would predict that individuals status positions in their groups may change over time as a function of the status construction process, such that only the time 2 status order represents the legitimated hierarchy, and is related to individuals performance.
Although there is much evidence that ones status position enhances ones performance opportunities ADDIN EN.CITE ADDIN EN.CITE.DATA (e.g., Bales 1958; Berger et al. 1980; Bunderson 2003), it is also plausible that ones final status position is a function of ones performance. To the extent that there is error in the initially formed performance expectations, or that the task requirements change over time, individuals may gain or lose status in their group as their true performance abilities are revealed. If this were the case, we would still expect a positive relationship between ones time 2 status position in the group and ones individual performance. Thus, whether due to the stability and reinforcement of the initial status order, the social construction of legitimated status orders, or the evolution of performance expectations, the first hypothesis we derive from the literature is:
Hypothesis 1: At time 2, status rank will be positively associated with performance.
Importantly, this hypothesis implies that ones time 2 status rank position will be associated with individual performance regardless of when the status position was attained. In other words, there should be no difference in the relationship between time 2 status position and individual performance for those whose status was conferred at the outset and they maintained it, and those who gained or lost status positions in their groups over time. If, however, individuals incur costs associated with seeking status, as we propose in our theoretical model, then the performance of those who gain status in their groups may not correspond to their time 2 status positions. The costs of status enhancing activities may undermine the positional benefits that are enjoyed by those who achieve high status at the outset and maintain it without expending as many resources. We, therefore, hypothesize that controlling for time 2 status positions:
Hypothesis 2: Those individuals who gain status over time do not perform as well as do those who maintain high status positions in their groups from the beginning.
An additional implication of this logic is that the less people invest in status enhancement or maintenance efforts, the better they should perform because their resources can be focused more on their own performance. Given the structural benefits to being in a high status position at time 2, however, it is unclear a priori whether people who lose status in their groups over time would perform better than or equal to stable high status group members. Thus, we will conduct an exploratory comparison, assuming that if status losers performance is either equal to or higher than that of stable high status group members it is further evidence of our resource tradeoff proposition.
Sample
99 individuals (66 men) from 24 groups of 4 to 6 people (M = 4.35, s d. = .92) agreed to participate in the research project for a 78 percent response rate. Three people were dropped from the analyses due to incomplete data for a final sample size of 96. They are 30.67 years old, on average (s.d. = 3.65), 62.5 percent are Asian or South Asian and 28 percent are Caucasian. On average they have 81.63 months of post-graduate work experience (s.d. = 39.99).
Variables
Dependent variable. We measured objective individual performance (as opposed to peer-rated perceived performance) as the overall course grade based on two individual assignments (case write-ups) worth 40 percent of the final grade, a final exam comprised of a case write-up and short essay questions worth 30 percent of the final grade and in-class participation from each class cumulatively worth 15 percent of the final grade. A group assignment that was worth 15 percent of the final grade is excluded from this variable. We used their aggregated grade rather than the grade from their final exam only (which would measure their performance closest to the time 2 status assessment) because it is a more accurate indication of their actual performance than is any single assignment ADDIN EN.CITE ADDIN EN.CITE.DATA (Polzer, Milton & Swann 2002; Rothstein 2004; Slack & Porter 1980). This variable does not differ across class sections (t = 1.15, n.s.), so we pooled the data.
Independent variables. We use the rating data of each persons contribution, influence and status items to determine the appropriateness of aggregating these ratings to the group level and combining them into a scale. We calculated the intra-class coefficient, or ICC(1) (an index of inter-rater reliability that takes into account group size), for the peer ratings of status, contribution and influence within groups in both time 1 and time 2. An ICC(1) value greater than .12 is generally considered acceptable ADDIN EN.CITE James198213871387138717James, L.Aggregation bias in estimates of perceptual agreementJournal of Applied PsychologyJournal of Applied PsychologyJ. Appl. Psychol.219-2296721982(James 1982). All ICC(1) are greater than .28, indicating substantial within group reliability in the peer ratings. The scale alphas (at time 1 = .90 at time 2 = .94) suggest that combining the three items into a single scale is appropriate.
The raw rating data is confounded by between-group differences in mean rating levels, which makes them inappropriate for testing our hypotheses regarding positions within a groups status hierarchy. Thus, we constructed within-group status hierarchy variables. We considered two options: Using group mean-centered rating variables or constructing within-group rank-ordered variables based on the relative ratings of group members. The groups mean-centered rating variables create different scales in each group and time period, which make it virtually impossible to define stability in the groups over time. In other words, although a person may not have shifted their rank position, their mean-centered rating values may, nonetheless, be different due to the incompatible scales. In contrast, since the ranking data are all on the same scale, and our theoretical interest is in stability of hierarchical positions versus movement up and down the rank order, we opted to use rank-ordered variables. We, thus, constructed rankings in each time period based on mean teammates ratings of each person on the three items in our status scale, allowing for ties. We use the time 2 rank order scale to test our first hypothesis (note that since a top-ranked individual has a value = 1, a positive effect of status is indicated by a negative regression coefficient).
We also considered several options for how best to examine the dynamic effects of status over time. The most straightforward approach seemed to be testing the main effects of time 1 status rank, time 2 status rank, and their interaction. The problem with this approach is that it constrains incremental changes in ranking over time to yield equivalent changes in performance. This means that the performance differences comparing someone who stays in a stable position to someone who moves by one unit is the same as comparing the performance of two mobile people, one of whom increases or decreases status by one position more than the other does. For example, the marginal performance difference between not losing status at all compared to losing one unit of status (e.g., time 1 status rank equal one and time 2 status rank equal one, versus time 1 status rank equal one and time 2 status rank equal two) would yield the same marginal performance effect as two people who both lost status to different incremental degrees (e.g., time 1 status rank equal one and time 2 status rank equal four versus time 1 status rank equal 1 and time 2 status rank equal five). Since we expect that the performance gap between the second pair would be greater than the performance gap between the first pair because the performance of those who maintain high status over time may be disproportionately high, this model is not an appropriate way to test our hypotheses.
Instead, in line with our theoretical questions, we created a status category variable, a four-part measure indicating if a subject maintains stable high or stable low status, gains status or loses status in his or her group between time 1 and time 2. We operationalize high status as being in the top ranked position in ones group. This is because individuals in the highest status positions in their groups are the unambiguously dominant members of their groups, and are most likely to reap the benefits of being high status ADDIN EN.CITE Chase198010461046104617Chase, I. D.Social-Process and Hierarchy Formation in Small-Groups - a Comparative PerspectiveAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.905-9244561980ISI:A1980KV94000002<Go to ISI>://A1980KV94000002 (Chase 1980). Supporting this assertion, we find that the status ratings of the top ranked individuals are significantly higher than are the ratings of the second ranked group member in both time periods (Mrank = 1 = 5.88, s.d. = .54; Mrank = 2 = 5.26, s.d. = .82; t = -3.11, p < .001; at time 2, Mrank = 1 = 6.02, s.d. = .49; Mrank = 2 = 5.53, s.d. = .82; t = -2.65, p < .01). In addition, the performance of those who maintain the highest status positions (M = 86.46, s. d = 5.18) is significantly higher than the performance of those who maintain the second highest status positions (M = 82.00, s.d. = 6.58), t = -1.78, p < .05). In contrast, the performance of those who stay in the second-ranked positions is not significantly higher than that of people who maintain the third-ranked positions (M = 88.13, s. d. = 8.61) t = .11, p = .45.
Thus, we categorize a person as stable high status when he or she is in the top-ranked position in both time 1 and time 2. Stable low status is operationalized as being in the same rank position other than the highest-ranked position at time 1 and time 2. People gained status (called status seekers to reflect our assumption of pro-activity) if they moved up the rank order between time 1 and time 2 and lost status (called status losers) if they moved down the rank order from time 1 to time 2.
Control variables. We control for the diffuse status characteristic of age. Age is particularly salient in this sample of working MBA students and has been associated with individuals influence within groups ADDIN EN.CITE Ziller195832773277327717Ziller, R. C.Exline, R. V.Some consequences of age heterogeneity in decision-making groupsSociometrySociometry198-2111958(Ziller & Exline 1958), even when groups have interacted for a short time ADDIN EN.CITE Berger197222372237223717Berger, JosephCohen, Bernard P.Zelditch, Morris Jr.Status characteristics and social interactionAmerican Sociological ReviewAmerican Sociological ReviewAm. Sociol. Rev.241-2553731972June 1972(Berger et al. 1972). We also included a specific status characteristic, months of post-graduate work experience, because those with more experience have been exposed to more organizational behavior issues at work and may have had more managerial experience, which could help their performance on their class assignments. In addition, we included subjects satisfaction with their group (I was satisfied working on this team (1= to no extent, 7= to a great extent)) because people are more satisfied with interpersonal interactions when they experience complementary dominance and deference behaviors than when they encounter incongruent behaviors ADDIN EN.CITE Dryer199710271027102717Dryer, D. C.Horowitz, L. M.When do opposites attract? Interpersonal complementarity versus similarity.Journal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.592-603721997Tiedens200310311031103117Tiedens, L. Z.Fragale, A. R.Power moves: Complementarity in dominant and submissive nonverbal behavior.Journal of Personality and Social PsychologyJournal of Personality and Social PsychologyJ. Pers. Soc. Psychol.558-568842003(Dryer & Horowitz 1997; Tiedens & Fragale 2003). Thus, satisfaction with the group likely reflects how content one is with his or her position in the status hierarchy, and thus may negatively correlate with status seeking behavior. Furthermore, job satisfaction is positively associated with job performance ADDIN EN.CITE Judge200132493249324917Judge, T. A.Thoresen, C. J.Bono, J. E.Patton, G. K.The job satisfaction-job performance relationship: A qualitative and quantitative reviewPsychological BulletinPsychological BulletinPsychol. Bull.376-40712732001(Judge, Thoresen, Bono & Patton 2001). People also might be more satisfied with higher quality groups, either in terms of the quality of the individuals or the quality of the group processes ADDIN EN.CITE Hackman20023258325832586Hackman, J. R.Leading teams: Setting the stage for great performances2002Harvard Business School PressGladstein198421172117211717Gladstein, Deborah L.Groups in context: A model of task group effectivenessAdministrative Science QuarterlyAdministrative Science QuarterlyAdm. Sci. Q.499-517294group effectiveness, team effectiveness1984(Gladstein 1984; Hackman 2002). Having high status in a high quality group, therefore, may mean that the resources one can access through that position are especially beneficial to ones performance, which would confound our results if it were not taken into account.
Results
Descriptive statistics and correlations of the variables in this study are presented in Table 1a. In Table 1b, we report mean levels of performance across the status categories as well as the numbers of individuals in each status category.
Insert Tables 1a and 1b about Here
To test our exploratory research question about how stable the status orders are, we preformed a Kendalls tau-b correlation, which adjusts for tied rank positions ADDIN EN.CITE Abdi20071256125612565Abdi, H.Salkind N.J. Kendall rank correlationEncyclopedia of Measurement and Statistics87 - 892007Thousand Oaks, C.A.Sage(Abdi 2007) between the time 1 and time 2 status rank variables. Although significant (p < .01), the correlation was .47, indicating substantial instability in the status orders of these groups. We also examined the mobility of individuals within their groups. Of the status seekers, 61.2 percent started at time 1 in the second or third highest-ranked positions (M = 3.32, s. d. = 1.04). Of those who moved up in their status ranking, only 41.9 percent moved all the way to the top ranked position by time two (M = 1.87, s. d. = .92). Among the status losers, 39 percent started in the top-ranked positions at time 1 (M = 2.03, s. d. = 1.03). At time 2, 53.8 percent were in the second- or third-highest positions (M = 3.50, s. d. = .95). Thus, there was considerable mobility in this sample, but most of it was within a few rank positions.
In Hypothesis 1, we proposed that status rank at time 2 would be positively associated with performance. We test this hypoth e s i s b y c o n d u c t i n g a n O L S r e g r e s s i o n o n i n d i v i d u a l p e r f o r m a n c e w i t h o u r t i m e 2 s t a t u s r a n k v a r i a b l e a l o n g w i t h t h e c o n t r o l s ( T a b l e 2 , C o l u m n 1 ) . T h e r e s u l t s i n d i c a t e t h a t p e r f o r m a n c e i s p o s i t i v e l y a s s o c i a t e d w i t h t i m e 2 s t a t u s r a n k ( В = - 1 . 7 7 , p <