Beyond Status: Relating Status Inequality to Performance and Health in Teams

Abstract

Status structures in organizations are ubiquitous yet largely ignored in organizational research. We offer a conceptualization of team status inequality, or the extent to which status positions on a team are dispersed. Status inequality is hypothesized to be negatively related to individual performance and physical health for low-status individuals when uncooperative behavior is high. Trajectories of the outcomes across time are also explored. Analyses using multilevel modeling largely support our hypotheses in a sample of National Basketball Association players across six time points from 2000 to 2005.

Full text

Beyond Status: Relating Status Inequality to Performance and
Health in Teams
Amy M. Christie
Wilfrid Laurier University
Julian Barling
Queen’s University
Status structures in organizations are ubiquitous yet largely ignored in organizational research. We offer
a conceptualization of team status inequality, or the extent to which status positions on a team are
dispersed. Status inequality is hypothesized to be negatively related to individual performance and
physical health for low-status individuals when uncooperative behavior is high. Trajectories of the
outcomes across time are also explored. Analyses using multilevel modeling largely support our
hypotheses in a sample of National Basketball Association players across six time points from 2000 to
2005.
Keywords: status, inequality, teams
Team composition has increasingly become the object of orga-
nizational research. The main focus of this research has been on
demographic composition factors, such as age, gender, ethnicity,
functional background, and education (Jackson, Joshi, & Erhardt,
2003), and deep-level composition factors, such as personality,
values, and abilities (e.g., Bell, 2007). Largely escaping this work
has been an emphasis on the status composition of teams, despite
a vast sociological literature demonstrating the inevitable emer-
gence of status hierarchies in small groups (e.g., Berger, Ridge-
way, Fisek, & Norman, 1998). These hierarchies can vary widely
in form and become legitimized in the group through social inter-
action (Ridgeway & Walker, 1995). Given that the study of status
hierarchies in small groups has an entrenched history in the socio-
logical literature, its near omission from research on teams is
intriguing (DiTomaso, Post, & Parks-Yancy, 2007), especially
because organizations are thought to have a clear role in creating
and maintaining status distinctions (Pfeffer, 1998).
Responding to recent calls (Pearce, 2001; Ravlin & Thomas,
2005), our purpose here is to examine how team status hierarchies
influence individual outcomes. Evidence suggests that status hier-
archies are related to basal properties of successful teamwork, such
as processes of social influence and interaction, and can become
organizers of behavior within groups (Berger, Rosenholtz, &
Zelditch, 1980; Ridgeway & Correll, 2004). Our interest is on one
feature of status hierarchies, namely status inequality, which has
been a focus of scholarship on status for decades but has rarely
been conceptualized as a defining feature of team structures on
which to make between-group comparisons (e.g., DiTomaso et al.,
2007). Instead, inequality is typically treated as the background for
studying the enduring nature of stratification and the experiences
of individuals occupying various social positions (e.g., Blackburn
& Prandy, 1997). We explored the effects of variation in the level
of status inequality between teams.
We provide a definition of status inequality and explore the way
in which it shapes individual performance and physical health
under two contingencies: an individual’s status position and un-
cooperative behavior. Our hypotheses also outline the predicted
relationship between status inequality and these outcome trajecto-
ries over time and were tested in a study of National Basketball
Association (NBA) players across a 6-year period.
Conceptual Background and Hypotheses
Status is a primary motivation (Barrick, Stewart, & Piotrowski,
2002), core to a person’s self-worth and social esteem (Berdahl,
2007). Common to all status-based approaches is the implicit inter-
pretation of status as a relationship between individuals in a social
structure. Whether characterized by performance expectations (e.g.,
Berger et al., 1980), attributions of prestige (Perretti & Negro, 2006),
or unearned privileges (Washington & Zajac, 2005), status is foremost
a relative construct. Unlike an individual’s reputation, for example,
status cannot be defined as an isolated, individual attribute. Instead,
status is a “positional or relational element of a social structure”
(Washington & Zajac, 2005, p. 282). More specifically, status repre-
sents an individual’s social standing or rank order among others
within a social system, which is based on prestige, prominence, and
respect (e.g., Anderson, John, Keltner, & Kring, 2001; Berger, Cohen,
& Zelditch, 1972; Huberman, Loch, O
¨
nc¸u¨ler, 2004; Perretti & Negro,
2008; Ridgeway & Walker, 1995; Washington & Zajac, 2005).
This article was published Online First August 16, 2010.
Amy M. Christie, School of Business and Economics, Wilfrid Laurier
University, Waterloo, Ontario, Canada; Julian Barling, Queen’s School of
Business, Queen’s University, Kingston, Ontario, Canada.
An earlier version of this article was presented at the Annual Meeting of
the Academy of Management 2008. Financial support from the Social
Sciences and Humanities Research Council of Canada to both authors is
acknowledged gratefully. The data source for the study is STATS LLC,
copyright 2007. We thank Nick Turner, Jana Raver, Natalie Allen, Karl
Keane, Susan Brodt, William Cooper, Tina Dacin, Kelly Packalen, Rod-
erick Iverson, Colette Hoption, Sean Tucker, Tony Carroll, Katherine
Alexander, Alyson Byrne, and Stacie Byrne for constructive comments on
this article.
Correspondence concerning this article should be addressed to Amy M.
Christie, School of Business and Economics, Wilfrid Laurier University,
Waterloo, Ontario, Canada N2L 3C5. E-mail: achristie@wlu.ca
Journal of Applied Psychology © 2010 American Psychological Association
2010, Vol. 95, No. 5, 920–934 0021-9010/10/$12.00 DOI: 10.1037/a0019856
920

Therefore, status relates each individual to all others within the social
system. Recognizing status as an individual’s place in an ordered
distribution is fundamental to understanding how status influences
team outcomes. Specifically, because status can be defined by the
relative space between group members, we suggest that status differ-
entials, and therefore the status distribution overall, provide a mean-
ingful approach to understanding how status operates in team settings.
Thus, we extend most previous research, which accounts for effects of
an individual’s status without considering the hierarchy from which it
is drawn.
We define status inequality as the extent to which status posi-
tions in a hierarchy are dispersed. A team’s degree of status
inequality captures the overall pattern of status differentials among
members. The study of status distributions and inequality is not
new. Sociologists characterize societies by the extent to which they
are stratified, where stratification is defined as hierarchically or-
ganized social inequality (Morris & Scott, 1996). Our approach
differs from studies of stratification, which focus on the emergence
of social injustice within a society, by conceptualizing status
inequality as a distributional property that varies between teams
and reflects status differences between those belonging to the same
social class or strata of society. Variation in status inequality
across teams is hypothesized to relate to individual outcomes.
Status Inequality and Performance
Status inequality is foremost a measure of social distance. Sociol-
ogists have long since recognized that inequality represents not only
relative advantage and prestige but a structure of relationships (e.g.,
Laumann & Guttman, 1966). In fact, recent approaches have concep-
tualized status inequalities by the social space that distances individ-
uals from one another (Bottero & Prandy, 2003; Prandy, 1999). The
resulting tendency is for individuals to feel detached from more
distant others (Chattopadhyay, 1999; Tsui & O’Reilly 1989), partic-
ularly on comparisons of status, which are core to self-perceptions of
inferiority and superiority (Locke, 2003, 2005). Individuals feel alien-
ated from a target when they find their traits to be less desirable than
those of the target, and they feel less similar and connected to targets
with less desirable traits than themselves. Locke (2003) concluded
that “target characteristics that maximize feelings of status (namely,
undesirable characteristics) also tend to undermine feelings of soli-
darity” (p. 629). In teams, lack of unity can be problematic because
performance is often maximized though collaboration.
As status inequality within a team increases, these status differ-
ences achieve greater relevance and importance, as evidenced by
the disparity in resources accorded across positions. This is why
personal characteristics are less likely to derive status value when
members hold them equally (e.g., Bunderson, 2003). The motiva-
tion to achieve higher status is then stronger with greater status
inequality because status is more salient and valuable. However,
status striving is a self-focused pursuit, where emphasis is placed
on individual advancement, often irrespective of collective inter-
ests (Huberman et al., 2004; Loch, Huberman, & Stout, 2000;
Loch, Yaziji, & Langen, 2001), especially when individuals feel
socially distanced from their teammates.
This logic is consistent with tournament theory, which suggests
that tournaments can evoke self-focused motivation and uncoop-
erative behavior. In tournament structures, pay across organiza-
tional positions is dispersed, so that as movement toward higher
level positions occurs, pay spreads widen and the positions become
scarce (e.g., Lazear & Rosen, 1981; Rosen, 1986). In the same way
that status becomes a prominent source of motivation as status
inequality increases, an incentive to achieve higher compensation
arises with greater pay dispersion. Support has been found for this
basic tenet of tournament theory—that widening pay spreads mo-
tivates individual effort (see Devaro, 2006). However, because
individuals advance according to relative performance (Lazear &
Rosen, 1981), like status inequalities, tournaments can prompt
proself and not prosocial motivation, deterring cooperation (De
Dreu, 2007; Pfeffer, 1998). Lazear and Rosen (1981) referred to
the tournament as a competition between rivals, such that prizes
can be won on the basis of personal merit and the downfall of
others. In some cases, individuals sabotage others to ensure that
they are not eliminated from the tournament (Lazear, 1989).
Yet, this dismal perspective of inequality is contingent on team
members working selfishly or uncooperatively. Researchers have
suggested that tournaments are more likely to evoke such behaviors
under various conditions, such as the perceived legitimacy or fairness
of the social structure (e.g., Shaw, Gupta, & Delery, 2002), the nature
of the task (Beersma et al., 2003), or an individual’s position in the
hierarchy (Lazear, 1989). Individual differences should also influence
how people interpret and respond to competitive pressures and in-
equality (Beersma et al., 2003; Trevor & Wazeter, 2006).
Accordingly, we suggest that individual performance will be
negatively related to status inequality only when selfish or unco-
operative behaviors are adopted. Although perhaps not related to
performance in all settings, in a team context requiring interde-
pendence, such uncooperative behavior is detrimental (e.g., De
Dreu, 2007; Shaw et al., 2002). Performance is jeopardized if
members do not combine their efforts and capitalize on the nec-
essary inputs of others or the combined potential of the team. For
this reason, compensation researchers have argued that if dishar-
mony is created by tournaments, individual performance suffers in
settings where work is interdependent (e.g., Bloom, 1999; Cow-
herd & Levine, 1992; Shaw et al., 2002). If individuals work
cooperatively on teams with greater status inequality, they can
draw on the combined talents of their teammates, and any adverse
consequences should be negated.
The relationship between status inequality and performance
should be contingent not only on uncooperative behavior but also
on an individual’s status. Status theories recognize the benefits of
high status. Status characteristics theory argues that high status is
attributed to those group members who are expected to perform
well; thus, high-status individuals receive more opportunities to
contribute to task deliberation and decision making (e.g., Belli-
veau, O’Reilly, & Wade, 1996; Driskell & Mullen, 1990; Weis-
band, Schneider, & Connolly, 1995). Further, the performance of
low-status individuals is often undervalued. Weisband et al. (1995)
found that when high-status members were mislabeled to the group
as low-status members, they received worse evaluations even
when they gave equal input. Therefore, Weisband et al. concluded
that status differences are related to people’s expectations and how
they interact with and evaluate others.
To the extent that status inequality is greater and the distinctions
between positions are accentuated, the opportunities and favorable
evaluations afforded (or denied) to those with high (or low) status
should also be heightened (e.g., Berger, Fisek, Norman, &
Zelditch, 1977). Kirchler and Davis (1986) manipulated status
921
TEAM STATUS INEQUALITY, PERFORMANCE, AND HEALTH

inequality in experimental groups and found that groups with
greater status inequality were more likely to use a power-wins
approach to decision making, in which high-status individuals
controlled decision outcomes. Conversely, equal status groups
were more likely to make their decisions on the basis of the quality
of input regardless of the contributor’s status. As a result, under
greater status inequality, lower status individuals should have
relatively fewer developmental opportunities to enhance their self-
efficacy and performance. Thus, we hypothesize the following:
Hypothesis 1: Status inequality is negatively associated with
individual performance for individuals who behave uncoop-
eratively and are in lower status positions on the team (i.e., for
individual performance, there is a three-way Status Inequal-
ity ⫻ Uncooperative Behavior ⫻ Status interaction).
Status Inequality and Absences
We propose that when team members act uncooperatively,
greater status inequality also influences absences from work be-
cause of physical ill health for those with lower status. Greater
status inequality may influence health by altering perceptions of
social support that help alleviate stress (e.g., Karasek & Theorell,
1990) and through the onset of negative, stress-producing interac-
tions (Cacioppo et al., 2002; S. Cohen, 2004; Rook, 1984). Com-
petitive strategies, such as politicking and undermining, have been
shown to be destructive to health (e.g., Cropanzano, Howes,
Grandey, & Toth, 1997; Duffy, Ganster, & Pagon, 2002; Harris &
Kacmar, 2005). When members of a collective do not share
congruent goals (i.e., between the self and the collective), they do
not develop the social support systems that occur naturally in
cohesive collectives and help to protect health. Socially integrated
individuals are healthier than their disintegrated counterparts who
feel a sense of detachment and distress (Seeman, 1996, 2000;
Uchino, 2004), which can interfere with neuroendocrine, cardio-
vascular, and immune bodily functions (S. Cohen, 2004).
However, as with performance, such negative effects of greater
status inequality are contingent on team members behaving self-
ishly or uncooperatively. If such strategies are not adopted and
individuals connect with their teammates despite inequalities, then
any adverse effects of social isolation and stressful social interac-
tions on health would be alleviated. As a second contingency,
higher status individuals are also more likely to be protected from
the potential social harms of greater status inequality. Low-status
individuals tend to experience less favorable health (see Marmot,
2004); even in prestigious social classes, relative status is related
to health. Redelmeier and Singh (2001) found that Academy
Award winners live approximately 4 years longer than those who
are nominated but never win the award.
Marmot (2004) argued that health disparities are, in part, ex-
plained by the inability of lower status individuals to fully connect
and interact with others and to have autonomy over life events.
First, in groups, high-status group members are accorded more
attention (Weisband, Schneider, & Connolly, 1995). The fact that
attention is not only derived from others with high status is
noteworthy, given that both high- and low-status group members
prefer to interact with those of higher status (Perretti & Negro,
2006). Not surprisingly, to protect these advantages, high-status
members tend to avoid close association with lower status mem-
bers, resulting in downward discrimination and augmented social
disintegration (Wilkinson, 2005).
Second, as discussed, lower status individuals are given fewer
opportunities to control group decision making and are expected to
defer to those with higher status. Research has linked health problems
to a lack of autonomy at work and to lower personal control beliefs
(e.g., Bailis, Segall, Mahon, Chipperfield, & Dunn, 2001; Karasek &
Theorell, 1990). Control helps individuals to cope with stressors and
influences health through psychobiological pathways (e.g., Karasek,
1979; Penninx et al., 1997). Accordingly, the physical health of
low-status individuals should be more vulnerable to widening status
inequality. One way that physical ill health manifests itself is in
withdrawal behaviors, such as absenteeism (Johns, 2008).
Hypothesis 2: Status inequality is positively associated with
absences due to physical ill health for individuals who behave
uncooperatively and are in lower status positions on the team
(i.e., for individual absences because of physical ill health,
there is a three-way Status Inequality ⫻ Uncooperative Be-
havior ⫻ Status interaction).
The Poor Get Poorer?
We also consider the temporal nature of individuals’ experi-
ences in teams. We propose that a Status Inequality ⫻ Status
interaction influences individuals’ performance and absence tra-
jectories across time. First, because low-status individuals receive
fewer developmental opportunities and are evaluated more harshly
under conditions of high-status inequality, their performance
should suffer increasingly over time as these disadvantages com-
pound. Those individuals consistently occupying low-status posi-
tions may eventually internalize the lowered expectations afforded
to them and approach their work with less confidence and moti-
vation. Wrzesniewski, Dutton, and Debebe (2003) suggested that
“the experience of receiving evaluative information about the
worth of one’s job, role, or self is powerful. Its impact strikes at the
core of the self and its worth in the organization” (p. 113).
Likewise, over time the perceived legitimacy of these status posi-
tions should be reinforced or strengthened (Ridgeway & Walker,
1995), further perpetuating the biased performance opportunities
against individuals who consistently occupy low-status positions.
Hypothesis 3A: Status inequality is associated with increas-
ingly lower individual performance over time for individuals
in lower status positions on a team.
Second, lower status individuals on teams with greater status
inequality not only are disadvantaged in terms of performance
opportunities and evaluations, but also face more social isolation
and reduced control, the health effects of which can intensify
across time. Low-status individuals who initially lack these social
and personal resources should have more difficulty coping with
stressors, enhancing the negative consequences of stress and im-
peding a full recovery (Gallo & Mathews, 2003; Hobfoll, 1989,
2001). Once weakened and with diminished resources, individuals
risk increased reactivity to stressors (e.g., Christie & Barling,
2009; Gallo & Mathews, 2003; Holahan, Moos, Holahan, & Cron-
kite, 1999). Thus, we hypothesize that the positive relationship
between status inequality and absences for low-status individuals
922
CHRISTIE AND BARLING

grows stronger over time as they consistently experience the
disadvantages of their positions.
Hypothesis 3B: Status inequality is associated with increas-
ingly more individual absences due to physical ill health over
time for individuals in lower status positions on a team.
Study Context
We used data from the NBA to test our hypotheses. Organiza-
tional research advocates the use of sport contexts to study orga-
nizational phenomena (Wolfe et al., 2005), and in the present
study, professional basketball provided a rich empirical context.
First, our hypothesized relationships are salient in sports; Wolfe et
al. (2005) stated explicitly that
the pervasive competitiveness in sport may have relevance for a
concept such as status contests. Team sports are eternally beset by the
tension of team cooperation that is impeded by individuals who are
more concerned with their own statistics, visibility, and heroics. The
reverse of that situation also is noteworthy as when potential star
individuals resist the lure of heroic individualistic visibility and facil-
itate team functioning. (p. 201)
Basketball is particularly conducive to studying cooperative be-
havior, which is “voluntary and discretionary. Players repeatedly
face situations in which they can elect whether or not to cooperate”
(Keidel, 1987, p. 593).
A second advantage is that performance is characterized by
individual and team components. Third, boundaries exist that
separate individuals into teams, which are differentiated from other
functional areas in the organization. Thus, the team represents a
logical unit in which to define social structure, avoiding cross-
memberships that exist in some organizations. Fourth, although
players by no means constitute the entire organization, studying
sports teams does allow for a type of cross-organization analysis,
allowing for greater generalizability. Fifth, because professional
sports take place in the public realm, a wealth of data are publicly
recorded. Finally, performance is (to some extent) objectively
indicated, making the data amenable to research.
Method
Data Source
The primary data were acquired from STATS, a leading source
of statistical information and analysis of sports leagues in the
United States. We supplemented this data set with award records
from the Official NBA Guide (Anderson & Reheuser, 2004, 2005,
2006; Carter & Hareas, 2001; Carter & Reheuser, 2002; Reheuser
& Smith, 2003). We collected data across the six consecutive
basketball seasons from 2000/2001 to 2005/2006 that occurred
after a major NBA lockout that ended in 1999. The sample
included all 30 NBA teams.
1
Following past research (Trevor,
Gerhart, & Reilly, 2006), we enhanced the reliability of our mea-
sures by including only those players who played in at least 20 of
the 82 games in a given season, which provided 2,280 individual-
level data points from 635 players across the 6 sample years.
Measures
Status and status inequality. To develop a measure of status
inequality, we first needed to measure the status of each player on
a team. Status is contextually embedded, meaning that understand-
ing its constitution requires knowledge of the research setting and
what might be pertinent indicators of status in that setting (Bunder-
son, 2003; B. P. Cohen & Zhou, 1991). Working from our original
definition of status as a ranking based on prominence, prestige, and
respect, we identified and measured five indicators of status for
NBA players in a given season: salary, games started, tenure,
awards/recognitions, and celebrity status.
First, the extravagance of player salaries in professional sports
receives public attention and scrutiny. In fact, some have criticized
players for becoming too focused on money and abandoning their
intrinsic motivations (e.g., Pierce, Thomsen, & McCallum, 2005;
Rosenberg, 2007). Facilitating the salience of salaries is their
public disclosure, allowing comparisons within and across teams.
Accordingly, salary is a source of prominence, and because final
salaries are negotiated individually, they also indicate respect from
those who make payroll decisions.
The second indicator used was the number of games that a
player started. During the course of play, five players participate in
a game at any one time; however, there are a total of 12 active
players on a team. Players are distinguished as starters (designated
to start a game) versus nonstarters (playing with greater irregular-
ity). Starters are afforded higher performance expectations and are
likely to receive more recognition by constituents, including own-
ers, coaches, teammates, rival teams, the media, and fans. Cer-
tainly, the way in which starting players are introduced and an-
nounced to the public prior to each game reinforces this
prominence. Likewise, an NBA player’s designation as a starter is
an indicator of his talent as perceived by organizational decision
makers and thus signals that he is in a respected position.
Third, organizational tenure is often considered a marker of status,
where individuals with longer tenure are more likely to have social
power (Mehra, Kilduff, & Brass, 2001) and carry valuable informal
knowledge (Rollag, 2004). A player’s tenure is consequential in
professional sports because of high turnaround rates; after spending a
few seasons in the league, players may be forced to retire (Hoffer,
2006; Witnauer, Rogers, & Saint Onge, 2007). Similarly, players’
abilities decline with age, forcing the exit of older players (Groothuis
& Hill, 2004). However, although older players typically exhibit
performance declines (Berri, Schmidt, & Brook, 2006), they bring a
wisdom to the game (e.g., Sabino, 2005) and are credited with a
mental toughness underdeveloped in less tenured players (Wolff,
2001). Longer tenured players are those who have been successful in
the sport. NBA salary schedules, which award longer tenured players
with higher minimum salaries, attest to the respect given to these
players. Thus, we used tenure in the league (rather than tenure on a
specific team) as an indicator of status.
Next, the awards and recognitions that players receive denote
prestige (Perretti & Negro, 2006). External judges of the game,
such as sports writers, broadcasters, and fans are responsible for
choosing the winners of awards presented to players throughout
the course of the season (i.e., all-star game selections and awards,
player of the week awards, and player of the month awards) and at
the season’s end (i.e., most valuable player, rookie of the year,
defensive player of the year, sixth-man award, most improved
player, sportsmanship award, citizenship award, all NBA team
1
The total number of teams in the league varied slightly across years.
923
TEAM STATUS INEQUALITY, PERFORMANCE, AND HEALTH

selections, selection for the all NBA rookie team, and selection for
the all NBA defensive team). Award prestige differs (i.e., being
deemed the league’s most valuable player is more prestigious than
being player of the week); thus, to create a measure of total player
awards, we attributed greater weight to major awards (as defined
by the Official NBA Guide) and distinguished between medium
and minor awards (i.e., those awarded to players at an end of the
season ceremony vs. awards for player of the week and month). In
a given season, players were given one point for minor awards
(e.g., player of the week award), two points for medium level
awards (e.g., an all-NBA team selection), and three points for
major awards (e.g., most valuable player).
Last, professional basketball is a public occupation, which can
result in fame for some. Critics have suggested that the quest for
superstardom has become an overly dominant concern. Although
celebrity status is unlikely equally desired, those garnering media
attention are likely to be attended to in the organization, particu-
larly because popularity or star power is significantly related to fan
attendance and gate revenues (Berri et al., 2006). Therefore, an
indicator of prominence is celebrity. To proxy celebrity, we tabulated
the number of articles mentioning the player’s name in Sports Illus-
trated magazine for a given season. Sports Illustrated is a weekly
American sports magazine. A similar procedure for ascribing atten-
tion from media reports to public figures has been used in past studies
of organizations (e.g., Hayward & Hambrick, 1997).
We verified the relevance of these status indicators by surveying
11 prominent North American sportswriters for the NBA. Of the
journalists, 10 were male and one was female, and their experience
as sportswriters for the NBA ranged from 1 to 35 years. Partici-
pants were given the definition of status and asked to rate on a
scale from 1 (not well at all)to5(extremely well) how well each
of the five status indicators discussed represents status in the NBA
(participants were also given the option to select don’t know). The
mean ratings for each of the indicators are as follows: salary (M ⫽
4.55, SD ⫽ 0.82), games started (M ⫽ 3.82, SD ⫽ 0.60), tenure
(M ⫽ 3.40, SD ⫽ 0.97), awards (M ⫽ 4.64, SD ⫽ 0.51), and
celebrity (M ⫽ 4.36, SD ⫽ 0.67); for the indicators amalgamated,
the mean was as follows: M ⫽ 4.13, SD ⫽ 0.38. The participants
strongly agreed on their overall status ratings (rwg ⫽ .93). The
correlations between status indicators appear in Table 1.
The five indicators were used to create a status measure for each
player observation. We followed past field research in adapting
Berger et al.’s (1977) methods for creating status composites in
simulation studies (e.g., Berger & Fisek, 2006) in a field setting
(Bunderson, 2003). It is noteworthy that this method is suitable for
“multicharacteristic status situations . . . 关where兴 the characteristics
may be either consistent or inconsistent” (Berger et al., 1977, p.
61); thus, we do not assume that the status indicators are neces-
sarily correlated.
A player’s score for each indicator was divided by the team
maximum value (e.g., SALARY/SALARY
team_max
). This deter-
mined the indicator strength for each player observation, rang-
ing from 0 to 1. The indicators were then equally weighted and
combined into a single status score with the formula that
follows, which reflects the notion that as the number of status
indicators accounted for increases, each indicator provides little
additional unique information about a player’s status. This
attenuation principle is an underlying tenet of status character-
istics theory (Berger et al., 1977). Last, consistent with the
definition of status as a position in an ordered distribution, we
placed the scores on a scale from 0 to 1 by dividing a player’s
status score by the maximum score for his team (Bunderson,
2003). See the equation at the bottom of the page, where SS
i,k
is
the status score for player observation i, s共SALARY
i,k
兲 is the
weighted status indicator strength of player observation i’s
salary, s共STARTS
i,k
兲 is the weighted status indicator strength for
the number of games started by player observation i,
s共TENURE
i,k
兲 is the weighted status indicator strength of player
observation i’s tenure, s 共AWARDS
i,k
兲 is the weighted status
indicator strength for player observation i’s total awards,
s共CELEB
i,k
兲 is the weighted status indicator strength for the
celebrity status of player observation i, and SS
team_max,k
is the
maximum status score on team k.
We operationalized status inequality, using the Gini coefficient
(formula next), a common metric of inequality used in organiza-
tional studies (Harrison & Klein, 2007).
G共SS兲
k
⫽
¥
i⫽1
n
兩 SS
i
⫺ SS
j
兩
2*n
2
* SS
mean
,
where G共SS兲
k
is the Gini coefficient of status scores for team k,
SS
mean
is the mean status score on team k, and n is the total number
of players on team k. The Gini coefficient is valued between 0 and
1, where 0 represents perfect status equality, and 1 represents
perfect status inequality; teams with higher Gini coefficients have
greater inequality.
Uncooperative behavior. We used player transgressions to
measure uncooperative behavior. Transgressions included suspen-
sions from play and ejections from a game (given for breaking a
number of rules, including physical contact restrictions, fighting,
assaulting officials, and general unsportsmanlike behavior). Trans-
gressions reflect unproductive, uncooperative, and noninstrumental
team behavior. Unlike most personal fouls, which also penalize play-
ers for breaking rules, transgressions are not the result of strategic play
or a focus on team-oriented outcomes; they are inconsistent with team
goals. Moreover, Kendall (2008) argued that players may perceive
transgressions to be a source of publicity and external popularity,
ultimately heightening their status in the league. This is likely because
transgressions elicit media and fan attention. Therefore, transgressions
are a proxy for a self-orientated or uncooperative playing style; they
represent a tendency for players to focus on themselves to the detri-
ment of their team. A recent public apology made by a suspended
NBA player illustrates these effects; he stated, “I apologize to my
teammates, our fans, our ownership and the N.B.A. for the negativity
this has created and the poor example that I set” (Associated Press,
2007, p. SP5). Transgressions were adjusted for the number of games
played.
SS
i,k
⫽
关1 ⫺ 共1 ⫺ s共SALARY
i,k
兲兲 ⴱ 共1 ⫺ s共STARTS
i,k
兲兲 ⴱ 共1 ⫺ s共TENURE
i,k
兲兲 ⴱ 共1 ⫺ s共AWARDS
i,k
兲兲 ⴱ 共1 ⫺ s共CELEB
i,k
兲兲兴
SS
team_max,k
,
924
CHRISTIE AND BARLING

Table 1
Means, Standard Deviations, and Correlations
Variable MSD123456789101112131415
Status indicators
1. Awards 0.461 1.64 —
2. Celebrity status 1.400 3.25 .56 —
3. Games started 34.670 30.66 .35 .34 —
4. Salary 4,444,226 4,182,800 .39 .47 .45 —
5. Tenure 6.647 3.70 .04 .13 .02 .36 —
Study variables
1. Ethnicity (AA ⫽ 1) 0.769 0.42 —
2. Center (Center ⫽ 1) 0.138 0.35 ⫺.22 —
3. Guard (Guard ⫽ 1) 0.469 0.50 .16 ⫺.38 —
4. Prior absences 0.083 0.15 ⫺.06 .08 ⫺.07 —
5. Prior performance (Eff) 0.424 0.10 .06 .01 ⫺.01 ⫺.21 —
6. Prior performance (Win) 0.160 0.07 .03 .02 ⫺.02 ⫺.22 .85 —
7. Pay dispersion 29.891 17.24 .02 ⫺.02 .00 ⫺.02 .04 .03 —
8. Mean status 0.599 0.06 .04 .05 .00 ⫺.01 .01 .00 ⫺.48 —
9. Winning percentage 0.016 0.28 ⫺.02 .00 .02 ⫺.08 .14 .15 .02 ⫺.02 —
10. Ejections 0.003 0.01 .07 .02 ⫺.05 ⫺.02 .14 .09 ⫺.01 .01 ⫺.01 —
11. Suspensions 0.002 0.01 .09 .01 ⫺.10 ⫺.01 .08 .04 ⫺.02 ⫺.01 .02 .36 —
12. Status 0.605 0.26 .16 .02 .01 ⫺.25 .54 .41 ⫺.08 .19 ⫺.02 .15 .09 —
13. Status inequality 0.281 0.05 ⫺.02 ⫺.04 .00 .01 .01 .02 .40 ⫺.77 ⫺.01 ⫺.01 .02 ⫺.13 —
14. Absences 0.086 0.15 ⫺.07 .09 ⫺.08 .13 ⫺.14 ⫺.12 ⫺.03 .02 ⫺.01 .02 .08 ⫺.24 .01 —
15. Performance (Eff) 0.420 0.10 .04 .00 .01 ⫺.15 .76 .61 .00 .04 .14 .12 .03 .51 ⫺.04 ⫺.21 —
16. Performance (Win) 0.158 0.07 .01 .02 .00 ⫺.15 .62 .71 .00 .04 .17 .07 ⫺.02 .41 ⫺.03 ⫺.24 .85
Note. N ⫽ 1,574 observations (listwise deletion). Correlations of .06 are significant at the p ⬍ .05 level. AA ⫽ African American; Eff ⫽ player efficiency; Win ⫽ player win score.
925
TEAM STATUS INEQUALITY, PERFORMANCE, AND HEALTH

Dependent variables. Hypotheses were tested with two per-
formance measures. First, we used a metric of player efficiency
advocated by the NBA (Berri et al., 2006). It is calculated as
follows:
PP
i
⫽ 关POINTS
i
⫹ REB
i
⫹ ASST
i
⫹ BLK
i
⫹ STL
i
⫺ 共FGA
i
⫺ FGM
i
兲 ⫺ 共FTA
i
⫺ FTM
i
兲 ⫺ TURN
i
兴,
where PP
i
is player observation i’s performance, POINTS
i
is player
observation i’s total points scored, REB
i
is number of rebounds
made by player observation i, ASST
i
is the number of assists for
player observation i, BLK
i
is the number of blocked shots by player
observation i, STL
i
is the number of steals made by player obser-
vation i, FGA
i
is the number of field goals attempted by player
observation i, FGM
i
is the number of field goals made by player
observation i, FTA
i
is player observation i’s total free throw
attempts, FTM
i
is player observation i’s total free throws made,
and TURN
i
is the total turnovers made by player observation i.
Second, economists have developed and validated a measure
(win score) that captures a player’s overall contribution to team
wins:
PW
i
⫽
冋
POINTS
i
⫹ REB
i
⫹
1
2
ASST
i
⫹
1
2
BLK
i
⫹ STL
i
⫺ FGA
i
⫺
1
2
FTA
i
⫺ TURN
i
⫺
1
2
PF
i
册
,
where PW
i
is player observation i’s win score, and PF
i
is player
observation i’s personal fouls.
Because playing time and position provide players with non-
equivalent performance opportunities, we also adjusted both per-
formance measures for the number of minutes played and the
player’s position on the team (Berri et al., 2006).
We used player absences due to injury or illness to measure
physical health. Absenteeism is a common metric of health and
well-being. However, employees may miss time for reasons other
than injury or illness. This is a small threat in the present research
because teams have private doctors who assess a player’s physical
condition, and teams release the nature of the absence. We ex-
cluded cases due to reasons other than injury or illness. The
demanding nature of the profession makes injuries commonplace.
Although occurring for accidental reasons, evidence shows that
psychosocial factors, such as life stress, and relational factors,
including social support, predict athletic injury as well (see Wil-
liams, 2001, for a review). Likewise, absences due to illnesses are
a direct measure of physical health. Absences were measured as
the number of occurrences of games missed. Because players who
receive more playing time are at greater risk for injury and because
those who are not expected to play need not be officially absent
when unable to play, we divided absences by the total number of
games in which a player appeared.
Control variables. The mean of an attribute is incorporated in
the calculation of inequality. Therefore, without controlling for the
Table 2
Parameter Estimates for Performance Analyses
Variable
Performance (efficiency) Performance (win score)
Model 1 Model 2 Model 3 Model 1 Model 2 Model 3
␥ SE ␥ SE ␥ SE ␥ SE ␥ SE ␥ SE
Intercept .05
ⴱⴱ
.02 .05
ⴱⴱ
.02 .05
ⴱⴱ
.02 .02 .03 .03 .03 .03 .03
Ethnicity ⫺.03
⬃
.02 ⫺.03 .02 ⫺.03
⬃
.02 ⫺.02 .03 ⫺.02 .03 ⫺.02 .03
Absences ⫺.10
ⴱⴱ
.03 ⫺.10
ⴱⴱ
.03 ⫺.10
ⴱⴱ
.03 ⫺.16
ⴱⴱ
.04 ⫺.15
ⴱⴱ
.04 ⫺.16
ⴱⴱ
.04
Prior performance .63
ⴱⴱ
.03 .62
ⴱⴱ
.07 .62
ⴱⴱ
.07 .29
ⴱⴱ
.04 .29
ⴱⴱ
.04 .29
ⴱⴱ
.04
Winning percentage .06
ⴱⴱ
.02 .06
ⴱⴱ
.02 .06
ⴱⴱ
.02 .12
ⴱⴱ
.02 .12
ⴱⴱ
.02 .11
ⴱⴱ
.02
Pay dispersion ⫺.02 .02 ⫺.02 .02 ⫺.02 .02 ⫺.02 .02 ⫺.02 .02 ⫺.02 .02
Status .17
ⴱⴱ
.03 .17
ⴱⴱ
.03 .17
ⴱⴱ
.03 .22
ⴱⴱ
.03 .21
ⴱⴱ
.03 .21
ⴱⴱ
.03
Mean status ⫺.06
⬃
.03 ⫺.06
⬃
.03 ⫺.06
⬃
.03 ⫺.05 .03 ⫺.05 .03 ⫺.04 .03
Status inequality ⫺.07
ⴱ
.03 ⫺.07
ⴱ
.03 ⫺.07
ⴱ
.03 ⫺.05 .03 ⫺.05 .03 ⫺.06
⬃
.03
Status ⫻ Status Inequality .01 .02 .02 .02 .04
⬃
.02 .04
ⴱ
.02
Suspensions ⫺.01 .02 ⫺.03 .02
Suspensions ⫻ Status ⫺.02 .02 ⫺.03 .02
Suspensions ⫻ Status Inequality ⫺.02 .02 ⫺.01 .02
Suspensions ⫻ Status ⫻ Status Inequality (Hypothesis 1) .05
ⴱ
.02 .04
ⴱ
.02
Ejections .03 .02 .03 .02
Ejections ⫻ Status ⫺.01 .02 ⫺.03 .03
Ejections ⫻ Status Inequality ⫺.01 .03 ⫺.02 .02
Ejections ⫻ Status ⫻ Status Inequality (Hypothesis 1) .04
⬃
.02 .06
ⴱⴱ
.02
Random components
Level 1 residual .37
ⴱⴱ
.03 .36
ⴱⴱ
.01 .36
ⴱⴱ
.03 .32
ⴱⴱ
.02 .31
ⴱⴱ
.02 .31
ⴱⴱ
.02
Level 2 intercept .03 .03 .03 .03 .03 .03 .29
ⴱⴱ
.04 .29
ⴱⴱ
.04 .29
ⴱⴱ
.04
Random slope (status inequality) .00 .01 .00 .01 .01 .01 .00 .01 .01 .01 .00 .01
Covariance (intercept and slope) .00 .01 .00 .01 .00 .01 .01 .01 .01 .01 .01 .01
LR␹
2
1,674.62
ⴱⴱ
10.03
⬃
4.36 1,498.87
ⴱⴱ
13.75
ⴱ
10.83
⬃
Note. LR␹
2
is the change in the ⫺2 log likelihood statistic across models. Model 1 was compared with the null model; Models 2 and 3 were compared
with Model 1.
⬃
p ⬍ .10.
ⴱ
p ⬍ .05.
ⴱⴱ
p ⬍ .01.
926
CHRISTIE AND BARLING

mean level of status on a team, we would be unable to separate the
effects of status inequality from those of mean team status. Fol-
lowing recommendation, we controlled for a team’s mean status in
all analyses (Harrison & Klein, 2007). In addition, given past
research showing a relationship between pay dispersion and per-
formance (Bloom, 1999), we considered the relationship between
status inequality and outcomes after controlling for its effects. We
measured pay dispersion as the salary of the highest paid player
divided by the salary of the lowest paid player. Ethnicity is related
to health and well-being (see Stanton, Revenson, & Tennen, 2007),
and thus was controlled. In analyses of player absences, we con-
trolled for dominant position played. Injuries may interfere with
performance (Berri & Krautmann, 2006), and thus absences were
included as a covariate in the performance models. Likewise, team
winning percentage (i.e., the number of team wins divided by the
number of games in the season) controlled for the possibility that
being on a winning or losing team may affect performance
(Bloom, 1999). Finally, past performance and absences are prob-
able predictors of future performance and absenteeism; thus, we
controlled for performance and absences in the previous season.
Results
Analytic Strategy
The temporal nesting of our data implies that ordinary least
squares regression’s assumption of independent observations may
be violated. We conducted our analyses using multilevel modeling
in Mplus 5.1, which provides parameter estimates analogous to
regression coefficients, but is appropriate for nested datasets. We
first calculated interclass correlations (ICC) to determine the ex-
tent to which our dependent variables were clustered by team
membership and across time. The ICC for performance nested
within teams was .01 (for efficiency and win score), suggesting
that team membership accounts for a very small portion of the
variance in performance. Accounting for performance across sam-
ple years revealed that performance was correlated across time
periods; ICC ⫽ .70 (efficiency) and ICC ⫽ .67 (win score),
necessitating the use of multilevel modeling. A similar pattern
emerged for player absences, where observations were largely
independent across team membership; ICC ⫽ .01, yet, signifi-
cantly correlated across league seasons; ICC ⫽ .28.
We used a series of multilevel equations that accounted for the
nested observations of time periods (Level 1) within individuals
(Level 2). Comparisons between models were made using the
change in the ⫺2 log likelihood statistic (LR␹
2
), which follows a
␹
2
distribution and indicates the relative fit of nested models
(Singer & Willett, 2003). To help avoid problems of multicol-
linearity, we followed others in standardizing all measures, which
simultaneously centered them on their sample means (e.g., Chen,
Kirkman, Kanfer, Allen, & Rosen, 2007; Ployhart, Weekley, &
Baughman, 2006). Descriptive statistics are presented in Table 1.
2
Hypothesis Tests
Hypothesis 1: Status inequality—Performance analyses.
We expected an interaction between status inequality, uncoopera-
tive behavior, and status on performance. The null model indicated
significant random variation across the sample years in the
performance intercept and status inequality-performance slope
components, and thus these parameters were retained in further
analyses (see Table 2). Hypothesis 1 was tested by entering a
three-way interaction term into the multilevel analyses after con-
trolling for main effects, two-way interactions, and controls. Sep-
arate models were estimated for both uncooperative behavior and
performance measures. Supporting Hypothesis 1, all interactions
were significant. Probing the interaction at one standard deviation
above and below the mean of the variables (Figure 1) showed a
negative relationship between status inequality and performance
for low-status players who played uncooperatively (efficiency and
ejections b ⫽⫺.14, p ⬍ .01, efficiency and suspensions b ⫽⫺.15,
p ⬍ .01, win score and ejections b ⫽⫺.18, p ⬍ .01, and win score
and suspensions b ⫽⫺.15, p ⬍ .01). By contrast, no relationship
emerged between status inequality and performance for low-status
players when uncooperative behavior was low (efficiency and
ejections b ⫽⫺.04, ns, efficiency and suspensions b ⫽⫺.01, ns,
win score and ejections b ⫽⫺.02, ns, and win score and suspen-
2
Although correlations of .9 (Tabachnick & Fidell, 2001) between
predictor variables are often used as a benchmark to judge whether mul-
ticollinearity may be a relevant statistical concern, because of the relatively
high correlation between status inequality and mean status, we conducted
additional tests to determine whether multicollinearity may be present in
our models. We found that the variance inflation factors were well below
conventional standards (i.e., 10) for detecting multicollinearity, with the
most extreme case in any of the analyses being 3.20. Similarly the toler-
ance levels for all analyses were above those that would prove worrisome
(i.e., .10), with .31 being the lowest in any of the analyses.
-0.4
-0.2
0
0.2
0.4
Performance
Status Inequality
High Uncooperative Behavior
Low status
High status
hgiHwoL
-0.4
-0.2
0
0.2
0.4
Performance
Status Inequality
Low Uncooperative Behavior
Low status
Hig h status
hgiHwoL
Figure 1. For performance, there is a Status ⫻ Status Inequality ⫻
Uncooperative Behavior interaction. Low status inequality, status, and
uncooperative behavior (high status inequality, status, and uncooperative
behavior) refer to status inequality, status, and uncooperative behavior one
standard deviation below (or above) the mean. Plotted interactions are
shown for ejection and player efficiency measures of uncooperative be-
havior and performance, respectively.
927
TEAM STATUS INEQUALITY, PERFORMANCE, AND HEALTH

sions b ⫽⫺.03, ns). Likewise, the simple slopes for high-status
players did not differ from zero.
Hypothesis 2: Status inequality—Absences analyses. Hypoth-
esis 2 related status inequality to absences due to physical ill
health. Estimating the null model for absences showed that the
absences intercept varied significantly across sample years; yet,
the slope components (between status inequality and absences) did
not, and thus were fixed in further analyses (see Table 3 for
models). As can be seen in Table 3, the interaction between status
inequality, ejections, and status was significantly associated with
absences. Figure 2 shows that status inequality was positively
related to the absences of low-status players under conditions of
high ejections (b ⫽ .31, p ⬍ .01), but not under conditions of low
ejections (b ⫽⫺.13, ns). For high-status players, the simple slopes
relating status inequality to absences did not differ significantly
from zero at either high or low levels of ejections. A marginally
significant relationship emerged for the interaction between status
inequality, suspensions, and status. However, although for low-
status players the positive relationship between status inequality
and absences was greater when suspensions were higher, it did not
differ significantly from zero when suspensions were low (b ⫽ .01,
p ⬍ ns) or when they were high (b ⫽ .10 p ⬍ ns). Evaluating the
simple slopes for high-status players showed that the relationship
between status inequality and absences was positive when suspen-
sions were low (b ⫽ .15, p ⬍ .05) and not significantly different
from zero when suspensions were high (b ⫽ .03, ns). Thus, the
data provide only partial support for Hypothesis 2.
Hypothesis 3: Longitudinal analyses. Our final hypotheses
linked status inequality and status to individuals’ performance and
absences trajectories over time. To test these hypotheses, we used
a multilevel model for individual growth with our six NBA sea-
sons as time points (centered on time 1). This approach allowed us
to model within-individual change over time (Level 1) and be-
tween individual differences in change over time (Level 2) simul-
taneously (e.g., Singer & Willet, 2003). The independent variable
slope (coded as 2000 ⫽ 0, 2001 ⫽ 1, 2002 ⫽ 2, 2003 ⫽ 3, 2004 ⫽
4, and 2005 ⫽ 5) denotes linear time. Thus, the estimated intercept
of the model represents the sample average starting point of the
dependent variable in 2000, whereas the estimates for slope rep-
resent the sample average rate of change in the dependent variable
from 2000 to 2005 (conditional on the remaining variables). Mod-
eling linear time as a variable allows us not only to consider the
relationship between time and performance and absences (i.e., the
sample average rate of change in performance/absences across
time), but also how other variables moderate this relationship. In
other words, we can test the effects of interactions between the
slope variable and other variables on the dependent variables. To
investigate our hypotheses about the relationship between status
inequality, status, and the rate of change in performance and
absences, we created interaction terms between these variables and
slope.
Status inequality and status were modeled as time-varying pre-
dictors, meaning that a player was permitted to have different
scores on these variables for each season (Singer & Willet, 2003).
Table 3
Parameter Estimates for Absences Due to Physical Ill-Health Analyses
Variable
Absences
Model 1 Model 2 Model 3
␥ SE ␥ SE ␥ SE
Intercept ⫺.01 .03 .00 .03 ⫺.01 .03
Guard ⫺.05 .03 ⫺.04 .03 ⫺.05
⬃
.03
Center .07
ⴱ
.03 .07
ⴱ
.03 .07
⬃
.04
Ethnicity .00 .03 ⫺.01 .03 ⫺.01 .03
Prior absences ⫺.04 .03 ⫺.04 .03 ⫺.04 .03
Pay dispersion ⫺.02 .02 ⫺.01 .02 ⫺.01 .02
Status ⫺.25
ⴱⴱ
.03 ⫺.26
ⴱⴱ
.03 ⫺.26
ⴱⴱ
.04
Mean status .10
ⴱ
.04 .11
ⴱ
.04 .11
ⴱⴱ
.04
Status inequality .07
⬃
.04 .07
⬃
.04 .09
ⴱ
.04
Status ⫻ Status Inequality .02 .03 ⫺.01 .03
Suspensions .09
ⴱⴱ
.02
Suspensions ⫻ Status ⫺.05
ⴱ
.03
Suspensions ⫻ Status Inequality ⫺.01 .03
Suspensions ⫻ Status ⫻ Status Inequality (H2) ⫺.05
⬃
.03
Ejections .02 .03
Ejections ⫻ Status .01 .02
Ejections ⫻ Status Inequality .10
ⴱ
.04
Ejections ⫻ Status ⫻ Status Inequality (H2) ⫺.12
ⴱⴱ
.04
Random components
Level 1 residual .57
ⴱⴱ
.03 .56
ⴱⴱ
.03 .56
ⴱⴱ
.12
Level 2 intercept .14
ⴱⴱ
.04 .14
ⴱⴱ
.04 .14
ⴱ
.07
Random slope (status inequality) — — — — — —
Covariance (intercept and slope) — — — — — —
LR␹
2
2074.47
ⴱⴱ
13.92
ⴱ
14.21
ⴱ
Note. Model 1 was compared to the null model; Models 2 and 3 were compared to Model 1. Dashes reflect
fixed parameters based on results of null models. H2 ⫽ Hypothesis 2.
⬃
p ⬍ .10.
ⴱ
p ⬍ .05.
ⴱⴱ
p ⬍ .01.
928
CHRISTIE AND BARLING

The results were examined by plotting the significant interactions
and comparing prototypical trajectories of players under four con-
ditions: when players continuously occupy (a) high-status posi-
tions on teams with high status inequality, (b) high-status positions
on teams with low status inequality, (c) low-status positions on
teams with high status inequality, and (d) low-status positions on
teams with low status inequality. When probing an interaction with
time-varying predictors, many other conditions implied by the
models could also be examined (Singer & Willet, 2003). For
example, one may wish to consider the average trajectory of a
high-status player whose team changes from having high to low
status inequality in 2002; however, to illustrate the findings in
relation to our hypotheses, we present only these four contrasts.
The results for performance appear in Table 4. Examining the
coefficient for the slope term after accounting for control variables
revealed that on average player performance declined over time
(conditional on the remaining independent variables). An interac-
tion emerged between the slope term, status inequality, and status
(marginally significant for win score). This fitted trajectory (at one
standard deviation above and below the means; Figure 3), shows
that, on average, consistently low-status players on teams with
consistently high status inequality experienced performance de-
clines across time (efficiency b ⫽⫺.41, p ⬍ .01, and win score
b ⫽⫺.25, p ⬍ .01), whereas performance did not change across
time for players who were consistently in low-status positions on
teams with low status inequality (efficiency b ⫽ .05, ns, and win
score b ⫽ .06, ns). Under consistently high status inequality, the
slope of the declining trajectory was steeper for players in consis-
tently low compared to high-status positions (efficiency b ⫽⫺.41,
p ⬍ .01, vs. b ⫽⫺.30, p ⬍ .01, and win score b ⫽⫺.25, p ⬍ .01
vs. b ⫽⫺.17, p ⬍ .01).
Table 5 displays the findings for Hypothesis 3B. We found no
effect of the slope term; on average, player absences were static
over time (at centered values of the independent variables). How-
ever, the results showed a significant relationship between status
inequality, status, and the slope of absences across time (see Figure 4).
Contrary to our hypothesis, players consistently occupying low-
status positions on teams with higher status inequality missed
increasingly fewer games (b ⫽⫺.02, p ⬍ .01), whereas those on
teams with lower status inequality did not (b ⫽ 0, p ⬍ ns). Further,
although players consistently occupying high-status positions on
teams with greater status inequality were absent more often than
those on teams with lesser status inequality in each season, the rate
of change in the absences across time did not differ between these
two groups (for high status inequality b ⫽ .01, p ⬍ ns, and for low
status inequality b ⫽ .01, p ⬍ ns).
Discussion
Despite calls to bring status to organizational research (Pearce,
2001; Ravlin & Thomas, 2005), the applicability of status theories
to past studies (e.g., Berdahl, 2007; Bunderson, 2003), and the role
of status in organizations (Pfeffer, 1998), it is surprising that with
few exceptions (e.g., Bacharach, Bamberger, & Mundell, 1993;
Washington & Zajac, 2005), very little research has explicitly
studied status in teams. Our study addresses this void. The results
suggest that the relationship between status inequality and out-
comes within the team is complex. When low-status individuals
exhibited more uncooperative behavior, greater status inequality
was associated with weaker performance and, in some cases, more
absences. Yet, status inequality was not associated with adverse
outcomes when uncooperative behavior was low and players had
high status.
Mixed evidence emerged for the dynamic effects of status
inequality on performance and absences trajectories. As hypothe-
sized, status inequality thwarted performance across time for lower
status players in particular. The trajectory of absences was less
consistent with our hypotheses: for higher status players, absences
were more frequent among those on teams with greater status
inequality. By contrast, low-status players on teams with greater
status inequality saw their absences decline, becoming more sim-
ilar to low-status players on teams with lesser status inequality.
Together these results suggest that low-status players played more
games across time, however, they performed worse in those
games. Although unexpected, this pattern may have resulted be-
cause these players reside in the most vulnerable positions—they
begin with the lowest performance records and most missed
games, and thus feel pressure to play whether ill or injured to
secure their jobs. This would be consistent with research that
shows that sickness presenteeism is correlated with job insecurity
(Caverley, Cunningham, & MacGregor, 2007).
Strengths, Limitations, and Future Research
Directions
Inherent in this study are numerous strengths. Objective mea-
sures of the outcome variables were used. The dataset allowed for
the control of potentially biasing variables (e.g., past performance,
past absences, team performance), and for examining the effects of
-0.5
-0.3
-0.1
0.1
0.3
0.5
0.7
Absences
Status Inequality
High Uncooperative Behavior
Low status
Hig h status
hgiHwoL
-0.5
-0.3
-0.1
0.1
0.3
0.5
0.7
Absences
Status Inequality
Low Uncooperative Behavior
Low status
Hig h status
hgiHwoL
Figure 2. For absences, there is a Status ⫻ Status Inequality ⫻ Unco-
operative Behavior interaction. Low status inequality, status, and uncoop-
erative behavior (high status inequality, status, and uncooperative behav-
ior) refer to status inequality, status, and uncooperative behavior one
standard deviation below (or above) the mean. Plotted interactions are
shown for the ejection measure of uncooperative behavior.
929
TEAM STATUS INEQUALITY, PERFORMANCE, AND HEALTH

status inequality across multiple levels of analysis, including time.
In fact, gaining an understanding of the role of time may be one of
the greatest oversights in organizational research (Ancona, Good-
man, Lawrence, & Tushman, 2001). Nevertheless, a number of
limitations should be noted. First, sample characteristics may limit
generalizability. NBA players are elite athletes in highly visible
occupations. Although we identified status markers relevant in the
NBA, their salience may be enhanced due to public visibility. The
extent to which status cues are less explicit in other organizational
settings may limit generalizability. Furthermore, the NBA employs
only males. Given that men and women may approach relation-
ships differently (e.g., Connell, 2002), future research should con-
sider whether these findings can be extended to females.
Perhaps the most productive way to interpret the findings is to
consider the types of organizational teams most similar to sports
teams. Sports teams may be best classified as performance or
action teams, which are often characterized by concentrated bursts
of effort, complex tasks, visible output or audiences, expert or
specialized members, and/or working in challenging circumstances
(e.g., Sundstrom, McIntyre, Halfhill, & Richards, 2000). Wolfe et
al. (2005) suggest that “generalizing from a sport team to other
types of teams must consider the extent to which the teams in
either setting are characterized by innate ability, insulation from
the rest of the organization, and visible production” (p. 202).
Basketball teams specifically resemble teams in organizations that
rely on reciprocal interdependence and require cooperation (Kei-
del, 1987). Accordingly, our results may be most reflective of
interdependent teams of individuals who are hired for their exper-
tise or ability, and whose combined performance is the organiza-
tion’s product (e.g., airplane cockpit teams, surgery teams, military
teams, firefighting teams, creative teams) as opposed to those with
more independent or physically distant members.
Second, the results of the study should also be interpreted
mindful of unmeasured intervening variables. We drew on multi-
ple theories that suggest that status structures are related to moti-
vation and behavior under various conditions; however, given the
constraints of archival data, we could not measure individual
cognitions, motivation, or social interactions directly. For exam-
ple, we argue that the longitudinal interaction between status and
status inequality emerges in part because low-status individuals
internalize the expectations of others, lowering their self-efficacy
16
17
18
19
20
21
2000 2001 2002 2003 2004 2005
Performance
Season
Low status inequality and high status
High status inequality and high status
Low status inequality and low status
High status inequality and low status
Figure 3. Individual growth trajectory for performance. For all time
points, low status inequality (high status inequality) refers to status in-
equality one standard deviation below (or above) the centered mean, and
low status (or high status) refers to status at one standard deviation below
(or above) the centered mean. Plot is shown for the player efficiency
measure of performance.
Table 4
Parameter Estimates for Performance Trajectory
a
Variable
Performance (efficiency) Performance (win score)
Model 1 Model 2 Model 1 Model 2
␥ SE ␥ SE ␥ SE ␥ SE
Intercept 19.90
ⴱⴱ
0.19 19.89
ⴱⴱ
0.19 7.31
ⴱⴱ
0.14 7.31
ⴱⴱ
0.14
Slope ⫺0.22
ⴱⴱ
0.05 ⫺0.20
ⴱⴱ
0.05 ⫺0.11
ⴱⴱ
0.03 ⫺0.11
ⴱⴱ
0.03
Ethnicity ⫺0.21 0.35 ⫺0.21 0.35 ⫺0.07 0.25 ⫺0.07 0.25
Absences ⫺2.51
ⴱⴱ
0.71 ⫺2.63
ⴱⴱ
0.73 ⫺2.45
ⴱⴱ
0.57 ⫺2.53
ⴱⴱ
0.59
Winning percentage 0.90
⬃
0.50 0.79 0.49 0.67
⬃
0.37 0.60
⬃
0.36
Pay dispersion 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Status 5.04
ⴱⴱ
0.41 5.29
ⴱⴱ
0.58 3.05
ⴱⴱ
0.30 3.14
ⴱⴱ
0.43
Mean status ⫺1.85 1.80 ⫺1.88 1.80 ⫺1.17 1.33 ⫺1.16 1.33
Status inequality 0.45 2.30 5.41
ⴱ
2.67 0.18 1.54 4.43
ⴱ
1.90
Status Inequality ⫻ Slope ⫺2.76
ⴱⴱ
0.71 ⫺2.03
ⴱⴱ
0.54
Status ⫻ Slope ⫺0.10 0.17 ⫺0.05 0.12
Status Inequality ⫻ Status ⫺9.45 7.16 ⫺3.38 5.31
Status Inequality ⫻ Status ⫻ Slope (Hypothesis 3A) 5.50
ⴱ
2.65 3.28
⬃
1.90
Random components
Level 1 residual 5.30
ⴱⴱ
0.25 5.28
ⴱⴱ
0.25 2.96
ⴱⴱ
0.15 2.95
ⴱⴱ
0.15
Level 2 intercept 11.74
ⴱⴱ
1.12 11.48
ⴱⴱ
1.09 6.53
ⴱⴱ
0.60 6.40
ⴱⴱ
0.60
Random slope 0.38
ⴱⴱ
0.07 0.37
ⴱⴱ
0.07 0.17
ⴱⴱ
0.04 0.16
ⴱⴱ
0.04
Covariance (intercept and slope) ⫺0.67
ⴱⴱ
0.25 ⫺0.64
ⴱⴱ
0.24 ⫺0.40
ⴱⴱ
0.13 ⫺0.39
ⴱⴱ
0.13
LR␹
2
799.53
ⴱⴱ
16.10
ⴱⴱ
655.62
ⴱⴱ
15.66
ⴱⴱ
Note. Model 1 was compared to the null model; Model 2 was compared to Model 1.
a
Because standardizing dependent measures in growth modeling is inappropriate, the dependent measure represents performance per game.
⬃
p ⬍ .10.
ⴱ
p ⬍ .05.
ⴱⴱ
p ⬍ .01.
930
CHRISTIE AND BARLING

and/or motivation to perform well. Alternatively, feelings of in-
justice or relative deprivation could hold greater explanatory
power. Justice perceptions have been related to both performance
(Ambrose & Schminke, 2009) and health outcomes (Kivima¨ki et
al., 2004) in previous studies. Emotional responses to status hier-
archies and status positions may also be important mechanisms to
explore in future research, given that emotional pathways relate
socioeconomic status to health (Gallo & Matthews, 2003).
Third, future research that accounts for various patterns of
interaction and status compositions in teams would be beneficial.
In large teams, faultlines may divide the team members into
subgroups of similarly ranked status positions, creating an in-
group—out-group effect. Thus, our model may not generalize
easily to teams that are not comparable in size to those of the NBA.
Larger teams may be more likely to have greater status inequality,
which makes extending our results to teams of varying sizes an
important avenue for future research.
Fourth, status was operationalized using objective indicators.
Although status hierarchies are thought to originate from objective
or visible characteristics (Berger et al., 1980; Sauder, 2005), sub-
jective evaluations of status may be equally important. For exam-
ple, individuals’ perceived level of socioeconomic status is asso-
ciated with health after accounting for objective socioeconomic
status (e.g., S. Cohen et al., 2008). Such perceptual measures could
be created by asking participants how they rank themselves on
various objective status indicators, such as salary. Alternatively,
future research could measure status by asking each member of a
team to rate or rank one another’s status irrespective of any
objective status indicators (e.g., Anderson et al., 2001). These
approaches could capture the subtleties of status distinctions that
emerge in groups, which may escape objective status measures.
Last, future replications of our results with alternate health
measures are warranted. The data did not account for incidences
where individuals continued to play basketball while ill or injured,
or for minor health problems not requiring absences. Measures of
sickness absences are among the most appropriate global health
indicators, particularly if based on medical certification (e.g., Kivi-
ma¨ki et al., 2003). Va¨a¨na¨nen, Buunk, Kivima¨ki, Pentti, and
Vahtera (2005) explained that “in the Whitehall II study, sickness
absence was found to be a more powerful predictor of all-cause
mortality than were established self-reported health measures or
available objective measures of specific physical illnesses and
medical conditions” (p. 188). However, undoubtedly ill and in-
jured individuals show up for work (e.g., Aronsson, Gustafsson, &
Dallner, 2000). These health problems are not captured in ab-
sences, and have been found to be negatively correlated with
income, job security, and job satisfaction (Aronsson et al., 2000;
Caverley et al., 2007). Low-status players may be more likely to
face job insecurity and dissatisfaction. Our results should be in-
terpreted cautiously, accounting for the potential that health prob-
lems were underestimated more so for this specific subsample.
Theoretical Contributions and Practical Implications
We add to the growing presence of status theories that account
for behavior in organizations. Extending previous conceptualiza-
tions of status at the individual-level by exploring status hierar-
chies in teams, we show that status inequality explains unique
variance beyond individual status in performance and absences
due to physical ill health. Our results suggest that any negative
effects of status inequality may be mitigated by promoting team
cooperation. These results warrant further investigation of status
hierarchies in teams. Of importance to managers may be recog-
nizing the status characteristics that are valued within the organi-
zation and making strategic composition decisions when designing
teams, particularly if interdependence is integral to team success.
0
0.05
0.1
0.15
0.2
0.25
2000 2001 2002 2003 2004 2005
Absences
Seasons
High status inequality and high status
Low status inequality and low status
High status inequality and low status
Low status inequality and high status
Figure 4. Individual growth trajectory for absences due to physical
ill-health. For all time points, low status inequality (high status inequality)
refers to status inequality one standard deviation below (or above) the
centered mean, and low status (or high status) refers to status at one
standard deviation below (or above) the centered mean.
Table 5
Parameter Estimates for Absences Due to Physical
Ill-Health Trajectory
a
Variable
Absences
Model 1 Model 2
␥ SE ␥ SE
Intercept .12
ⴱⴱ
.02 .12
ⴱⴱ
.01
Slope ⫺.01 .01 .00 .00
Guard ⫺.01 .01 ⫺.01 .01
Center .04
ⴱⴱ
.02 .04
ⴱ
.02
Ethnicity ⫺.01 .01 ⫺.01 .01
Pay dispersion .00 .00 .00 .00
Status ⫺.17
ⴱⴱ
.03 ⫺.26
ⴱⴱ
.04
Mean status .39
ⴱⴱ
.11 .37
ⴱⴱ
.11
Status inequality .43
ⴱⴱ
.13 .60
ⴱⴱ
.15
Status Inequality ⫻ Slope ⫺.09
ⴱⴱ
.03
Status ⫻ Slope .03
ⴱⴱ
.01
Status Inequality ⫻ Status ⫺.93
ⴱ
.41
Status Inequality ⫻ Status ⫻
Slope (Hypothesis 3B) .26
ⴱ
.13
Random components
Level 1 residual .02
ⴱⴱ
.01 .02
ⴱⴱ
.00
Level 2 intercept .01 .01 .01 .01
Random slope .00 .00 .00 .00
Covariance (intercept and slope) .00 .00 .00 .00
LR␹
2
49.33
ⴱⴱ
20.56
ⴱⴱ
Note. Model 1 was compared with the null model; Model 2 was com-
pared with Model 1.
a
Because standardizing dependent measures in growth modeling is inap
-
propriate, the dependent measure represents absences per game.
ⴱ
p ⬍ .05.
ⴱⴱ
p ⬍ .01.
931
TEAM STATUS INEQUALITY, PERFORMANCE, AND HEALTH

Team research has focused on process and performance-based
outcomes. Our inclusion of physical health extends this collection
of outcomes. Practically, identifying the potential causes of missed
time from work is of significance to organizations, with health care
costs a growing concern. Further, the results suggest that, at least
in some organizations, absences can impact performance, intensi-
fying their effects. Theoretically, our conceptualization of status at
both the position and distribution levels contributes to the growing
study of status and health in occupational health psychology. To
understand the relation between status and health, this research
must explore both an individual’s status and the distribution from
which it is drawn.
Conclusion
The goals of this study were to extend individual-level perspec-
tives of status and to demonstrate the importance of status hierar-
chies to team research. More specifically, our goal was to explore
the role of status inequality in teams. To that end, we provided a
conceptualization of status inequality and showed the conditions
under which status inequality was most relevant to performance
and health and across time. Thus, we provided a platform for
future research into the study of status structures and status in-
equality in organizations.
References
Ambrose, M. L., & Schminke, M. (2009). The role of overall justice
judgments in organizational justice research: A test of mediation. Jour-
nal of Applied Psychology, 94, 491–500.
Ancona, D. G., Goodman, P. S., Lawrence, B. S., & Tushman, M. L.
(2001). Time: A new research lens. Academy of Management Review,
26, 645– 663.
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.
Anderson, C., & Reheuser, R. (Eds.). (2004). Official NBA guide (2004 –
2005 ed.). St. Louis, MO: Sporting News.
Anderson, C., & Reheuser, R. (Eds.). (2005). Official NBA guide (2005–
2006 ed.). St. Louis, MO: Sporting News.
Anderson, C., & Reheuser, R. (Eds.). (2006). Official NBA guide (2006 –
2007 ed.). St. Louis, MO: Sporting News.
Aronsson, G., Gustafsson, K., & Dallner, M. (2000). Sick but yet at work.
An empirical study of sickness presenteeism. Journal of Epidemiology
and Community Health, 54, 502–509.
Associated Press. (2007, July 15). N.B.A. suspends Artest and Stephen
Jackson. New York Times, p. SP5.
Bacharach, S., Bamberger, P., & Mundell, B. (1993). Status inconsistency
in organizations: From social hierarchy to stress. Journal of Organiza-
tional Behavior, 14, 21–36.
Bailis, D. S., Segall, A., Mahon, M. J., Chipperfield, J. G., & Dunn, E. M.
(2001). Perceived control in relation to socioeconomic and behavioral
resources for health. Social Science & Medicine, 52, 1661–1676.
Barrick, M. R., Stewart, G. L., & Piotrowski, M. (2002). Personality and
job performance: Test of the mediating effects of motivation among
sales representatives. Journal of Applied Psychology, 87, 43–51.
Beersma, B., Hollenbeck, J. R., Humphrey, S. E., Moon, H., Conlon, D. E.,
& Ilgen, D. R. (2003). Cooperation, competition, and team performance:
Toward a contingency approach. Academy of Management Journal, 46,
572–590.
Bell, S. T. (2007). Deep-level composition variables as predictors of team
performance: A meta-analysis. Journal of Applied Psychology, 92, 595–
615.
Belliveau, M. A., O’Reilly, C. A. I., & Wade, J. B. (1996). Social capital
at the top: Effects of social similarity and status on CEO compensation.
Academy of Management Journal, 39, 1568 –1593.
Berdahl, J. L. (2007). Harassment based on sex: Protecting social status in
the context of gender hierarchy. Academy of Management Review, 32,
641–658.
Berger, J., Cohen, B. P., & Zelditch, M. (1972). Status characteristics and
social interaction. American Sociological Review, 37, 241–255.
Berger, J., & Fisek, M. H. (2006). Diffuse status characteristics and the
spread of status value: A formal theory. American Journal of Sociology,
111, 1038 –1079.
Berger, J., Fisek, M. H., Norman, R. Z., & Zelditch, M. (1977). Status
characteristics and social interaction: An expectation-states approach.
New York, NY: Elsevier.
Berger, J., Ridgeway, C. L., Fisek, M. H., & Norman, R. Z. (1998). The
legitimation and delegitimation of power and prestige orders. American
Sociological Review, 63, 379–405.
Berger, J., Rosenholtz, S. J., & Zelditch, M., Jr. (1980). Status organizing
processes. Annual Review of Sociology, 6, 479 –508.
Berri, D. J., & Krautmann, A. C. (2006). Shirking on the court: Testing for
the incentive effects of guaranteed pay. Economic Inquiry, 44, 536 –546.
Berri, D. J., Schmidt, M. B., & Brook, S. L. (2006). The wages of wins:
Taking measure of the many myths in modern sport. Stanford, CA:
Stanford University Press.
Blackburn, R. M., & Prandy, K. (1997). The reproduction of social in-
equality. Sociology, 31, 491–509.
Bloom, M. (1999). The performance effects of pay dispersion on individ-
uals and organizations. Academy of Management Journal, 42, 25–40.
Bottero, W., & Prandy, K. (2003). Social interaction distance and stratifi-
cation. British Journal of Sociology, 54, 177–197.
Bunderson, J. S. (2003). Recognizing and utilizing expertise in work
groups: A status characteristics perspective. Administrative Science
Quarterly, 48, 557–591.
Cacioppo, J. T., Hawkley, L. C., Crawford, E., Ernst, J. M., Burleson,
M. H., Kowalewski, R. B., . . . Berntson, G. G. (2002). Loneliness and
health: Potential mechanisms. Psychosomatic Medicine, 64, 407– 417.
Carter, C., & Hareas, J. (Eds.). (2001). Official NBA guide (2001–2002
ed.). St. Louis, MO: Sporting News.
Carter, C., & Reheuser, R. (Eds.). (2002). Official NBA guide (2002–2003
ed.). St. Louis, MO: Sporting News.
Caverley, N., Cunningham, J. B., & MacGregor, J. N. (2007). Sickness
presenteeism, sickness absenteeism, and health following restructuring
in a public service organization. Journal of Management Studies, 44,
304–319.
Chattopadhyay, P. (1999). Beyond direct and symmetrical effects: The
influence of demographic dissimilarity on organizational citizenship
behavior. Academy of Management Journal, 42, 273–287.
Chen, G., Kirkman, B. L., Kanfer, R., Allen, D., & Rosen, B. (2007). A
multilevel study of leadership, empowerment, and performance in teams.
Journal of Applied Psychology, 92, 331–346.
Christie, A. M., & Barling, J. (2009). Disentangling the indirect links
between SES and health: The dynamic roles of work stressors and
personal control. Journal of Applied Psychology, 94, 1466–1478.
Cohen, B. P., & Zhou, X. (1991). Status processes in enduring work
groups. American Sociological Review, 56, 179 –188.
Cohen, S. (2004). Social relationships and health. American Psychologist,
59, 676 –684.
Cohen, S., Alper, C. M., Doyle, W. J., Adler, N., Treanor, J. J., & Turner,
R. B. (2008). Objective and subjective socioeconomic status and sus-
ceptibility to the common cold. Health Psychology, 27, 268–274.
Connell, R. W. (2002). Gender. Malden, MA: Blackwell.
Cowherd, D. M., & Levine, D. I. (1992). Product quality and pay equity
932
CHRISTIE AND BARLING

between lower level employees and top management: An investigation
of distributive justice theory. Administrative Science Quarterly, 37,
302–320.
Cropanzano, R., Howes, J. C., Grandey, A. A., & Toth, P. (1997). The
relationship of organizational politics and support to work behaviors,
attitudes, and stress. Journal of Organizational Behavior, 18, 159 –180.
De Dreu, C. K. W. (2007). Cooperative outcome interdependence, task
reflexivity, and team effectiveness: A motivated information processing
perspective. Journal of Applied Psychology, 92, 628 – 638.
Devaro, J. (2006). Strategic promotion tournaments and worker perfor-
mance. Strategic Management Journal, 27, 721–740.
DiTomaso, N., Post, C., & Parks-Yancy, R. (2007). Workforce diversity
and inequality: Power, status, and numbers. Annual Review of Sociology,
33, 473–501.
Driskell, J. E., & Mullen, B. (1990). Status, expectations, and behavior: A
meta-analytic review and test of the theory. Personality and Social
Psychology Bulletin, 16, 541–553.
Duffy, M. K., Ganster, D. C., & Pagon, M. (2002). Social undermining in
the workplace. Academy of Management Journal, 45, 331–351.
Gallo, L. C., & Matthews, K. A. (2003). Understanding the association
between socioeconomic status and physical health: Do negative emo-
tions play a role. Psychological Bulletin, 129, 10–51.
Groothuis, P. A., & Hill, J. R. (2004). Exit discrimination in the NBA: A
duration analysis of career length. Economic Inquiry, 42, 341–349.
Harris, K. J., & Kacmar, K. M. (2005). Easing the strain: The buffering role
of supervisors in the perceptions of politics–strain relationship. Journal
of Occupation and Organizational Psychology, 78, 337–354.
Harrison, D. A., & Klein, K. (2007). What’s the difference? Diversity
constructs as separation, variety, or disparity in organizations. Academy
of Management Review, 32, 1199–1228.
Hayward, M. L., & Hambrick, D. C. (1997). Explaining the premium paid
for large acquisitions: Evidence of CEO hubris. Administrative Science
Quarterly, 42, 103–128.
Hobfoll, S. E. (1989). Conservation of resources: A new attempt at con-
ceptualizing stress. American Psychologist, 44, 513–524.
Hobfoll, S. E. (2001). The influence of culture, community, and the
nested-self in the stress process: Advancing conservation of resources
theory. Applied Psychology: An International Review, 50, 337–421.
Hoffer, R. (2006, July 31). It’s great to be average. Sports Illustrated,
105(4), 56.
Holahan, C. J., Moos, R. H., Holahan, C. K., & Cronkite, R. C. (1999).
Resource loss, resource gain, and depressive symptoms: A 10-year
model. Journal of Personality and Social Psychology, 77, 620– 629.
Huberman, B. A., Loch, C. H., & Önc¸u¨ler, A. (2004). Status as a valued
resource. Social Psychology Quarterly, 67, 103–114.
Jackson, S., Joshi, A., & Erhardt, N. L. (2003). Recent research on team
and organizational diversity: SWOT analysis and implications. Journal
of Management, 29, 801–830.
Johns, G. (2008). Absenteeism and presenteeism: Not at work or not
working well. In C. L. Cooper & J. Barling (Eds.), The Sage handbook
of organizational behavior (Vol. 1, pp. 160 –177). London, England:
Sage.
Karasek, R. A. (1979). Job demands, decision latitude, and mental strain:
Implications for job design. Ithaca, NY: Cornell University Press.
Karasek, R. A., & Theorell, T. (1990). Healthy work: Stress, health, and
the reconstruction of working life. New York, NY: Basic Books.
Keidel, R. W. (1987). Team sports models as a generic organizational
framework. Human Relations, 40, 591–612.
Kendall, T. D. (2008). Celebrity misbehavior in the NBA. Journal of
Sports Economics, 9, 231–249.
Kirchler, E., & Davis, J. H. (1986). The influence of member status
differences and task type on group consensus and member position
change. Journal of Personality and Social Psychology, 51, 83–91.
Kivima¨ki, M., Ferrie, J. E., Head, J., Shipley, M. J., Vahtera, J., & Marmot,
M. G. (2004). Organisational justice and change in justice as predictors
of employee health: The Whitehall II study. Journal of Epidemiology &
Community Health, 58, 931–937.
Kivima¨ki, M., Head, J., Ferrie, J. E., Shipley, M. J., Vahtera, J., & Marmot,
M. G. (2003). Sickness absences as a global measure of health: Evidence
from mortality in the Whitehall II prospective cohort study. British
Medical Journal, 327, 364–372.
Laumann, E. O., & Guttman, L. (1966). The relative associational conti-
guity of occupations in an urban setting. American Sociological Review,
31, 169 –178.
Lazear, E. P. (1989). Pay equality and industrial politics. Journal of
Political Economy, 97, 561–581.
Lazear, E. P., & Rosen, S. (1981). Rank-order tournaments as optimum
labor contracts. Journal of Political Economy, 89, 841–864.
Loch, C., Huberman, B. A., & Stout, S. (2000). Status competition and
performance in work groups. Journal of Economic Behavior & Orga-
nization, 43, 35–55.
Loch, C., Yaziji, M., & Langen, C. (2001). The fight for alpha position:
Channelling status competition in organizations. European Management
Journal, 19, 16 –25.
Locke, K. (2003). Status and solidarity in social comparison: Agentic and
communal values and vertical and horizontal directions. Journal of
Personality and Social Psychology, 84, 619 –631.
Locke, K. (2005). Connecting the horizontal dimension of social compar-
ison with self-worth and self-confidence. Personality and Social Psy-
chology Bulletin, 31, 795–803.
Marmot, M. G. (2004). The status syndrome: How social standing affects
our health and longevity. New York, NY: Times Books.
Mehra, A., Kilduff, M., & Brass, D. J. (2001). The social networks of high
and low self-monitors: Implications for workplace performance. Admin-
istrative Science Quarterly, 46, 121–146.
Morris, L., & Scott, J. (1996). The attenuation of class analysis: Some
comments on G. Marshall, S. Roberts and C. Burgoyne, “Social Class
and the Underclass in Britain in the USA.” British Journal of Sociology,
47, 45–55.
Pearce, J. L. (2001). Less epistemology; more government and social
status. Human Relations, 54, 85–89.
Penninx, B. W. J. H., van Tilburg, T., Kriegsman, D. M. W., Deeg, D. J. H.,
Boeke, A. J. P., & van Eijk, J. T. M. (1997). Effects of social support and
personal coping resources on mortality in older age: The Longitudinal
Aging Study Amsterdam. American Journal of Epidemiology, 146,
510–519.
Perretti, F., & Negro, G. (2006). Filling empty seats: How status and
organizational hierarchies affect exploration versus exploitation in team
design. Academy of Management Journal, 49, 759 –777.
Pfeffer, J. (1998). The human equation: Building profits by putting people
first. Boston, MA: Harvard Business School Press.
Pierce, C. P., Thomsen, I., & McCallum, J. (2005, February 21). Why fans
are tuning out the NBA. Sports Illustrated, 102(8), 64–73.
Ployhart, R. E., Weekley, J. A., & Baughman, K. (2006). The structure and
function of human capital emergence: A multilevel examination of the
attraction–selection–attrition model. Academy of Management Journal,
49, 661– 677.
Prandy, K. (1999). Class, stratification and inequalities in health: A com-
parison of the registrar-general’s social class and the Cambridge Scale.
Sociology of Health & Illness, 21, 466 – 484.
Ravlin, E. C., & Thomas, D. C. (2005). Status and stratification processes
in organizational life. Journal of Management, 31, 966–987.
Redelmeier, D. A., & Singh, S. M. (2001). Survival in academy award-
winning actors and actresses. Annals of Internal Medicine, 134, 955–
962.
Reheuser, R., & Smith, R. (Eds.). (2003). Official NBA guide (2003–2004
ed.). St. Louis, MO: Sporting News.
Ridgeway, C. L., & Correll, S. J. (2004). Unpacking the gender system: A
933
TEAM STATUS INEQUALITY, PERFORMANCE, AND HEALTH

theoretical perspective on gender beliefs and social relations. Gender &
Society, 18, 510 –531.
Ridgeway, C. L., & Walker, H. A. (1995). Status structures. In K. S. Cook,
G. A. Fine, & J. S. House (Eds.), Sociological perspectives on social
psychology (pp. 281–310). Boston, MA: Allyn & Bacon.
Rollag, K. (2004). The impact of relative tenure on newcomer socialization
dynamics. Journal of Organizational Behavior, 25, 853– 872.
Rook, K. S. (1984). The negative side of social interaction: Impact on
psychological well-being. Journal of Personality and Social Psychology,
46, 1097–1108.
Rosen, S. (1986). Prizes and incentives in elimination tournaments. Amer-
ican Economic Review, 76, 701–716.
Rosenberg, S. S. (2007, April 8). NBA can’t match college for fans.
Greeley Tribune. Retrieved from http://www.greeleytribune.com
Sabino, D. (2005, October 24). MEMPHIS Grizzlies. Sports Illustrated,
103(1), 129.
Sauder, M. (2005). Symbols and contexts: An interactionist approach to the
study of social study. Social Psychology Quarterly, 46, 279–298.
Seeman, T. E. (1996). Social ties and health: The benefits of social
integration. Annals of Epidemiology, 6, 442– 451.
Seeman, T. E. (2000). Health promoting effects of friends and family on
health outcomes in older adults. American Journal of Health Promotion,
14, 362–370.
Shaw, J. D., Gupta, N., & Delery, J. E. (2002). Pay dispersion and
workforce performance: Moderating effects of incentives and interde-
pendence. Strategic Management Journal, 23, 491–512.
Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis:
Modeling change and event occurrence. New York, NY: Oxford Uni-
versity Press.
Stanton, A. L., Revenson, T. A., & Tennen, H. (2007). Health psychology:
Psychological adjustment to chronic disease. Annual Review of Psychol-
ogy, 58, 565–592.
Sundstrom, E., McIntyre, M., Halfhill, T., & Richards, H. (2000). Work
groups: From the Hawthorne studies to work teams of the 1990s and
beyond. Group Dynamics: Theory, Research, and Practice, 4, 44– 67.
Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th
ed.). Needham Heights, MA: Allyn & Bacon.
Trevor, C., Gerhart, B., & Reilly, G. (2006, August). Reconciling compet-
ing perspectives on pay dispersion and organizational performance: The
criticality of pay strategies and employee inputs. Paper presented at the
annual meeting of the Academy of Management, Atlanta, Georgia.
Trevor, C. O., & Wazeter, D. L. (2006). A contingent view of reactions to
objective pay conditions: Interdependence among pay structure charac-
teristics and pay relative to internal and external referents. Journal of
Applied Psychology, 91, 1260–1275.
Tsui, A. S., & O’Reilly, C. A. I. (1989). Beyond simple demographic
effects: The importance of relational demography in superior–
subordinate dyads. Academy of Management Journal, 32, 402–423.
Uchino, B. N. (2004). Social support and physical health: Understanding
the health consequences of relationships. New Haven, CT: Yale Uni-
versity Press.
Va¨a¨na¨nen, A., Buunk, B. P., Kivima¨ki, M., Pentti, J., & Vahtera, J. (2005).
When it is better to given than to receive: Long-term health effects of
perceived reciprocity in support exchange. Journal of Personality and
Social Psychology, 89, 176–193.
Washington, M., & Zajac, E. J. (2005). Status evolution and competition:
Theory and evidence. Academy of Management Journal, 48, 281–296.
Weisband, S. P., Schneider, S. K., & Connolly, T. (1995). Computer-
mediated communication and social information: Status salience and
status differences. Academy of Management Journal, 38, 1124–1151.
Wilkinson, R. G. (2005). The impact of inequality: How to make sick
societies healthier. New York, NY: New Press.
Williams, J. M. (2001). Psychology of injury risk and prevention. In R. N.
Singer, H. A. Hausenblas, & C. Janelle (Eds.), Handbook of research in
sport psychology (2nd ed., pp. 766 –786). New York, NY: Macmillan.
Witnauer, W. D., Rogers, R. G., & Saint Onge, J. M. (2007). Major league
baseball career length in the 20th century. Populational Research and
Policy Review, 26, 371–386.
Wolfe, R. A., Weick, K. E., Usher, J. M., Terborg, J. R., Poppo, L.,
Murrell, A. J., . . . Jourdan, J. S. (2005). Sport and organizational studies:
Exploring synergy. Journal of Management Inquiry, 14, 182–210.
Wolff, A. (2001, May 21). The old guard. Sports Illustrated, 94(21),
50–51.
Wrzesniewski, A., Dutton, J. E., & Debebe, G. (2003). Interpersonal
sensemaking and the meaning of work. In R. M. Kramer & B. Staw
(Eds.), Research in organizational behavior (pp. 93–135). San Fran-
cisco, CA: Interscience.
Received January 5, 2009
Revision received February 1, 2010
Accepted February 9, 2010 䡲
934
CHRISTIE AND BARLING