Integrating Personality and Social Networks

This article was downloaded by: [2003:ec:cf02:1546:e87f:2700:44d5:699] On: 04 May 2023, At: 15:13
Publisher: Institute for Operations Research and the Management Sciences (INFORMS)
INFORMS is located in Maryland, USA
Organization Science
Publication details, including instructions for authors and subscription information:
http://pubsonline.informs.org
Integrating Personality and Social Networks: A Meta-

Analysis of Personality, Network Position, and Work
Outcomes in Organizations
Ruolian Fang, Blaine Landis, Zhen Zhang, Marc H. Anderson, Jason D. Shaw, Martin Kilduff
To cite this article:

Ruolian Fang, Blaine Landis, Zhen Zhang, Marc H. Anderson, Jason D. Shaw, Martin Kilduff (2015) Integrating Personality and
Social Networks: A Meta-Analysis of Personality, Network Position, and Work Outcomes in Organizations. Organization Science
26(4):1243-1260. https://doi.org/10.1287/orsc.2015.0972
Full terms and conditions of use: https://pubsonline.informs.org/Publications/Librarians-Portal/PubsOnLine-Terms-and-

Conditions
This article may be used only for the purposes of research, teaching, and/or private study. Commercial use
or systematic downloading (by robots or other automatic processes) is prohibited without explicit Publisher
approval, unless otherwise noted. For more information, contact [email protected].
The Publisher does not warrant or guarantee the article’s accuracy, completeness, merchantability, fitness
for a particular purpose, or non-infringement. Descriptions of, or references to, products or publications, or
inclusion of an advertisement in this article, neither constitutes nor implies a guarantee, endorsement, or
support of claims made of that product, publication, or service.
Copyright © 2015, INFORMS
Please scroll down for article—it is on subsequent pages
With 12,500 members from nearly 90 countries, INFORMS is the largest international association of operations research (O.R.)
and analytics professionals and students. INFORMS provides unique networking and learning opportunities for individual
professionals, and organizations of all types and sizes, to better understand and use O.R. and analytics tools and methods to
transform strategic visions and achieve better outcomes.
For more information on INFORMS, its publications, membership, or meetings visit http://www.informs.org
Organization Science
Vol. 26, No. 4, July–August 2015, pp. 1243–1260
ISSN 1047-7039 (print) ISSN 1526-5455 (online) http://dx.doi.org/10.1287/orsc.2015.0972
© 2015 INFORMS
Downloaded from informs.org by [2003:ec:cf02:1546:e87f:2700:44d5:699] on 04 May 2023, at 15:13 . For personal use only, all rights reserved.
Integrating Personality and Social Networks:

A Meta-Analysis of Personality, Network Position,
and Work Outcomes in Organizations
Ruolian Fang
NUS Business School, National University of Singapore, Singapore 119245, [email protected]
Blaine Landis
Department of Management Science and Innovation, University College London, London WCIE 6BT, United Kingdom,
[email protected]
Zhen Zhang
W. P. Carey School of Business, Arizona State University, Tempe, Arizona 85287, [email protected]
Marc H. Anderson
College of Business, Iowa State University, Ames, Iowa 50011, [email protected]
Jason D. Shaw
Faculty of Business, Hong Kong Polytechnic University, Hung Hom, Hong Kong, [email protected]
Martin Kilduff
Department of Management Science and Innovation, University College London, London WCIE 6BT, United Kingdom,
[email protected]
sing data from 138 independent samples, we meta-analytically examined three research questions concerning the
U roles of personality and network position in organizations. First, how do different personality characteristics—self-
monitoring and the Big Five personality traits—relate to indegree centrality and brokerage, the two most studied structurally
advantageous positions in organizational networks? Second, how do indegree centrality and brokerage compare in explaining
job performance and career success? Third, how do these personality variables and network positions relate to work
outcomes? Our results show that self-monitoring predicted indegree centrality (across expressive and instrumental networks)
and brokerage (in expressive networks) after controlling for the Big Five traits. Self-monitoring, therefore, was especially
relevant for understanding why people differ in their acquisition of advantageous positions in social networks. But the
total variance explained by personality ranged between 3% and 5%. Surprisingly, we found that indegree centrality was
more strongly related to job performance and career success than brokerage. We also found that personality predicted
job performance and career success above and beyond network position and that network position partially mediated the
effects of certain personality variables on work outcomes. This paper provides an integrated view of how an individual’s
personality and network position combine to influence job performance and career success.
Keywords: social networks; network position; structural holes; brokerage; indegree centrality; personality;
self-monitoring; Big Five personality traits; meta-analysis
History: Published online in Articles in Advance April 30, 2015.
Introduction have emerged. First, what effects does personality have

According to theories of structural advantage, people on network structure? Second, how do indegree central-
benefit from occupying advantageous network positions ity and brokerage, the two network positions most fre-
that provide access to useful knowledge, career spon- quently associated with structural advantage, compare in
sorship, and psychosocial support (Brass 1984, Burt predicting performance and career success? The third is
1992, Seibert et al. 2001). By contrast, theories of per- a two-part question concerning the integration of per-
sonality suggest that people perform better when they sonality and social network position: (i) Once social
demonstrate specific behavioral tendencies (such as self- network position is taken into account, do personality
monitoring or conscientiousness) valued in work con- characteristics contribute to the explanation of individ-
texts (Barrick et al. 2001, Ng et al. 2005). Recent uals’ performance and career success? And (ii) does
thinking has brought together research on individuals’ network position mediate the relationship between per-
structural positions and personality to forge a more com- sonality and work outcomes?
prehensive approach to performance and career success We address these questions in a meta-analysis
(Kilduff and Tsai 2003). In doing so, three questions of 138 independent samples. Our research examines
1243
Fang et al.: Integrating Personality and Social Networks
1244 Organization Science 26(4), pp. 1243–1260, © 2015 INFORMS
how personality characteristics, as reflected in self- exhibit greater ease and social skills in social interac-
monitoring (Snyder 1974) and the Big Five personal- tions than do low self-monitors (Furnham and Capon
ity traits (extraversion, conscientiousness, openness to 1983). They use humor (Turner 1980), pace conversa-
experience, agreeableness, and neuroticism; McCrae and tions appropriately (Dabbs et al. 1980), reciprocate self-
John 1992), affect the attainment of advantageous net- disclosures when getting to know new people (Shaffer
work positions. Both self-monitoring and the Big Five et al. 1982), seek information and advice from well-
traits have received significant theoretical and empiri- connected coworkers (Fang and Shaw 2009), talk about
cal attention in the network literature (e.g., Klein et al. other people instead of talking about themselves (Ickes
2004, Mehra et al. 2001, Oh and Kilduff 2008, Pollet et al. 1986), and resolve conflict through collaboration
et al. 2011, Sasovova et al. 2010). Integrating and dif- and compromise (Baron 1989). This interpersonal atten-
ferentiating these two approaches to personality enriches tiveness makes high self-monitors desirable targets for
our understanding of how personality relates to the others’ friendship and advice seeking. Research shows
occupation of key network positions. In particular, we that high self-monitors tend to be the recipients of
address the apparent confusion between the traits of self- incoming friendship ties (Sasovova et al. 2010) and that
monitoring and extraversion noted in a recent review employees tend to approach high self-monitoring man-
(Dalal et al. 2015). In the social network domain, agers for help with their emotional problems (Toegel
our research helps clarify whether network advantage et al. 2007). Thus, we predict that self-monitoring will
inheres in indegree centrality (the number of incoming relate positively to indegree centrality in organizational
ties an individual receives from others) or brokerage (the social networks.
extent to which an individual is connected to people or Self-monitoring theory has evolved over a 40-year
clusters of people who are not connected) (Brass 1984, period to include a more recent emphasis on the ways in
Burt 2005, Seibert et al. 2001). Finally, our research which high self-monitors bring others in line with their
addresses a new challenge to the integration of person- own agendas in order to achieve their goals (Fuglestad
ality psychology and social networks in examining the and Snyder 2010). This emphasis on self-monitoring
recent claim (Burt 2012) that, irrespective of the degree agency is strongly suggestive of the qualities that char-
to which personality characteristics influence the attain- acterize social network brokers (Burt et al. 1998).
ment of advantageous network positions, the position a Self-monitoring theory suggests that high self-monitors
person occupies in the social network is the most sub- segregate their contacts from each other, whereas low
stantive predictor of performance. self-monitors bring their contacts together (Snyder 1987,
Oh and Kilduff 2008). This difference between high self-
monitors and low self-monitors is due, in part, to dif-
Integrating Personality and Social Networks ferent preferences regarding choices of activity partners.
Research Question 1: How Does Personality Relate For each specific activity (e.g., chess, tennis), a high self-
to Network Position? monitor will tend to choose a different activity partner,
Self-monitoring has been shown to predict social net- whereas a low self-monitor will tend to choose the same
work positions, particularly brokerage positions (Mehra partner regardless of activity (Snyder et al. 1983). High
et al. 2001, Oh and Kilduff 2008, Sasovova et al. 2010). self-monitors emerge as brokers in friendship networks
The question arises, however, as to whether and how the (Mehra et al. 2001, Sasovova et al. 2010) and acquain-
Big Five traits, which provide a parsimonious approach tanceship networks (Oh and Kilduff 2008). Thus, our
to salient aspects of personality (Digman 1990), relate to prediction is that self-monitoring will relate positively
indegree centrality and brokerage. We focus on indegree to brokerage in organizational networks, particularly in
centrality and brokerage as representative social network expressive networks.
positions because they have long been considered cru- Extraversion. Extraversion refers to the extent to
cial to understanding performance outcomes (Freeman which people are outgoing, active, gregarious, assertive,
et al. 1980). Our predictions apply to both expressive energetic, enthusiastic, and cheerful in their outlook
(e.g., friendship) and instrumental (e.g., advice) net- (McCrae and John 1992). Extraverted individuals are
works unless otherwise indicated. sensitive to reward signals, seek stimulation, and par-
Self-Monitoring. Self-monitoring theory proposes that ticipate in a wide variety of social activities (Ashton
people differ in the extent to which they regulate and et al. 2002, De Pascalis et al. 2005). Extraverts’ social-
control how they present themselves in social settings izing tendencies might suggest that they are likely to
and interpersonal relationships (Snyder 1974). The pro- have larger friendship and advice networks. But there
totypical high self-monitor strives to generate affective is reason to question this line of reasoning, as evidence
states and behaviors appropriate to specific situations, has accumulated showing that extraverts fail to attract
whereas the prototypical low self-monitor generates friends over time (Selfhout et al. 2010) and that the per-
expressive behavior from inner affective states and atti- ceived status of extraverts declines over time in group
tudes (Snyder 1979). Chameleon-like high self-monitors settings (Bendersky and Shah 2013). Thus, research on
Organization Science 26(4), pp. 1243–1260, © 2015 INFORMS 1245
extraversion suggests that extraverts’ tendencies toward (Nettle 2006), and cooperate (Denissen and Penke 2008,
sociability may fail to translate into popularity and Holmes 2002). They help integrate conflicting partners’
may even engender others’ dislike (Klein et al. 2004). views and needs (Jensen-Campbell et al. 2003). These
Theory concerning the relationship between extraver- characteristics make agreeable people attractive friend-
sion and brokerage offers little clear guidance. To the ship partners such that they are more likely to be chosen
extent that extraverts build large networks (Asendorpf as friends over time (Selfhout et al. 2010) and in team
and Wilpers 1998, Pollet et al. 2011), brokerage may friendship networks (Klein et al. 2004). Agreeableness,
increase because the number of missing connections therefore, is likely to relate positively to indegree cen-
among contacts is likely to be high (Bossard 1945). But trality, particularly in expressive networks.
extraverts prefer to bring their different social contacts
Neuroticism. Neuroticism concerns the extent to
together (Kalish and Robins 2006). Therefore, overall,
which a person is anxious, insecure, hostile, and irritable
the relationship of extraversion with indegree central- (McCrae and John 1992). Neuroticism is associated with
ity and brokerage in organizational networks remains the frequency with which people feel uneasy and anx-
unclear (see Stokes 1985). ious throughout the day (Fleeson and Gallagher 2009,
Openness to Experience. Openness to experience cap- Judge et al. 2014). Neurotic people, who often express
tures the extent to which people are imaginative, negative emotions and may be viewed as high-cost inter-
creative, intellectual, open-minded, and have diverse action partners, are likely to be avoided. Neuroticism
interests (McCrae and John 1992). Open people may relates negatively to indegree centrality in team friend-
tire of socializing with others who exhibit conventional ship and advice networks (Klein et al. 2004)—a finding
habits (Cheng et al. 1995, McCrae 1996), thereby poten- we expect to hold in organizational social networks.
tially reducing their popularity as attractive partners for In summary, we anticipate that self-monitoring pre-
interactions (i.e., indegree centrality). Consistent with dicts both indegree centrality and brokerage, whereas
this prediction, research shows that open people not the relation of extraversion to these network outcomes
only have smaller team friendship networks but also remains unclear. Among the other Big Five traits,
are disliked in workgroups (Klein et al. 2004). How- we anticipate that conscientiousness and agreeableness
ever, research has also shown that people who are more relate positively to indegree centrality but that neuroti-
curious are more interesting as conversational partners cism relates negatively to indegree centrality. We also
(Kashdan et al. 2011), which suggests that open peo- anticipate that conscientiousness and openness to expe-
ple are more likely to be sought after for friendship. We rience relate positively to brokerage. We test competing
examine these competing ideas concerning openness in ideas concerning how openness to experience relates to
relation to indegree centrality. With respect to broker- indegree centrality. We do not anticipate how agreeable-
age, we know that open people exhibit diverse interests ness and neuroticism relate to brokerage because their
(McCrae 1996). In their pursuit of contacts from differ- patterns of relationships remain unclear, both theoreti-
ent, unconnected social circles, open people might there- cally and empirically.
fore serve as network brokers.
Research Question 2: How Do Indegree Centrality
Conscientiousness. Conscientiousness concerns the
and Brokerage Compare in Explaining Job
extent to which a person is industrious, organized, duti-
Performance and Career Success?
ful, prepared, persistent, and detail oriented (McCrae and
Theory and research on social networks emphasize
John 1992). Because of people’s preference for work
that central network positions provide access to infor-
partners as competent and hardworking (Hinds et al.
mation and other resources and thereby enhance the
2000), conscientious workers may be selected preferen-
likelihood that individuals will achieve performance
tially in these roles. And despite the finding that consci-
and career success (for reviews, see Burt et al. 2013,
entiousness is unrelated to popularity in team friendship
Kilduff and Brass 2010). There are several different
and advice networks (Klein et al. 2004), conscientious
approaches to understanding network centrality, but two
people might also be selected into brokerage roles if
of the most useful approaches with respect to orga-
colleagues from different organizational areas seek them
nizational behavior are indegree centrality and broker-
out for resolving work-related problems. We propose,
age. Indegree centrality indicates an individual’s level
therefore, that conscientiousness relates positively to
of activity (Freeman 1979), popularity (Wasserman and
both indegree centrality and brokerage, particularly in
Faust 1994), or prominence (Knoke and Burt 1983)
instrumental networks.
and is widely relevant in organizational research (e.g.,
Agreeableness. Agreeableness concerns the extent to Burkhardt and Brass 1990, Kilduff and Krackhardt 1994,
which a person is cooperative, compliant, generous, Klein et al. 2004). Brokerage, measured as betweenness
kind, and trusting (McCrae and John 1992). Agree- centrality (Freeman 1979) or reverse-scored constraint
able people are motivated to develop positive rela- (Burt 1992), indicates the extent to which the individ-
tions with others (Barrick et al. 2002), show empathy ual connects others (and groups of others) who have
no direct connections to each other. Brokerage is a key advantageous positions and that people who occupy
construct in modern organizational network theory and advantageous positions are likely to have better per-
research (e.g., Kleinbaum 2012, Sasovova et al. 2010, formance and career outcomes. But do personality and
Stovel and Shaw 2012). Although indegree centrality network position have independent effects on work
and brokerage have long been recognized as quite dif- outcomes? That is, do personality variables relate to
ferent approaches to network centrality (Freeman et al. work outcomes when social network position is taken
1980), organizational research has overlooked the ques- into account (and vice versa)? Further, does network
tion of how they compare in explaining performance and position mediate the relationship between personality
career outcomes in organizations. and work outcomes? These questions reflect an ongoing
Indegree Centrality. Advantage accrues to people with debate at the intersection of personality psychology and
many connections: they can call on numerous sources of social network research (Balkundi et al. 2011, Kilduff
tangible and intangible resources (Brass and Burkhardt and Tsai 2003).
1993). The availability of these various resources is one Existing research suggests that self-monitoring and
indicator of power (Brass and Burkhardt 1993, Emerson the Big Five traits influence people’s attainment of
1962). In expressive networks, people with many con- indegree centrality and brokerage positions (e.g., Klein
nections are likely to receive social support and other et al. 2004, Mehra et al. 2001, Oh and Kilduff 2008,
psychosocial resources necessary for work performance Sasovova et al. 2010). Evidence also shows that these
and career success (e.g., Baldwin et al. 1997, Gibbons network positions contribute to performance outcomes
2004). In instrumental networks, people with many con- (e.g., Baldwin et al. 1997, Burt 1992, Cross and
nections are likely to receive and accumulate task-related Cummings 2004, Mehra et al. 2001, Seibert et al. 2001).
knowledge, expertise, and information. These resources Furthermore, personality meta-analyses show that self-
are important for enhancing performance and prospects monitoring and the Big Five traits predict job perfor-
of promotion (e.g., Baldwin et al. 1997). People with mance and career success (e.g., Barrick et al. 2001,
higher indegree centrality are likely to achieve higher Day et al. 2002, Ng et al. 2005). Together, findings
performance and greater career success (e.g., Baldwin from these different research streams suggest that peo-
et al. 1997, Sparrowe et al. 2001). ple with certain personality traits tend to attain struc-
Brokerage. Brokerage positions confer different advan- turally advantageous positions and that the occupation
tages than do positions high in indegree centrality. of these advantageous positions, in turn, influences peo-
A brokerage position provides people with access to indi- ple’s performance and career success. The attainment of
viduals (or clusters of individuals) who are disconnected an advantageous network position is one possible mech-
from one another (Burt 1992). Thus, brokerage positions anism by which personality affects work outcomes.
are theorized to offer three advantages: breadth of nonre- Prior work on the relationships among personality,
dundant information from diverse contacts, timeliness of network position, and performance supports this possi-
information passing between disconnected groups, and bility. Mehra et al. (2001) found that self-monitoring
arbitrage in bringing separate groups together (Burt et al.
predicted employees’ occupancy of brokerage posi-
2013). Network brokers in organizations move unknown
tions and that the self-monitoring and network vari-
or misunderstood information to places where it has
ables predicted workplace performance when examined
value. These brokers are rewarded with high performance
separately and simultaneously. These results indicate a
evaluations, generous compensation packages, and early
pattern of partial mediation in which self-monitoring
promotions (Burt et al. 2013).
Overall, indegree centrality and brokerage both pro- affects the attainment of advantageous network position,
vide people with structural advantage. But how do and both self-monitoring and network position influence
these two structural positions compare in facilitat- performance. A recent study on individuals’ multirole
ing job performance and career success? Existing networks (Burt 2012) found that individuals’ personality
research is unclear on this important question. There- characteristics measured as consistent network behaviors
fore, our second contribution to the microfoundations across roles (called “network-related personality”) pre-
of organizational networks is an empirical one: the cur- dicted the occupancy of strategically advantageous posi-
rent meta-analysis examines the relative importance and tions. Although this research asserted that “much of the
predictive power of indegree centrality and brokerage variance in network advantage reflects personality, but
in explaining individuals’ job performance and career that portion of advantage variance has little to do with
success. success” (Burt 2012, p. 586), we suggest that this con-
clusion is premature, given Burt’s (2012) indirect mea-
Research Question 3: How Do Personality and surement of network-related personality and the prior
Network Position Relate to Job Performance and theory and evidence that we have summarized above.
Career Success? Overall, findings from both personality psychology
We have articulated that people with certain personal- and social network research highlight the need to inves-
ity characteristics are more likely to occupy structurally tigate (a) whether personality (self-monitoring and the
Big Five) and advantageous network positions (inde- 2005, Zhang and Peterson 2011). The primary stud-
gree centrality and brokerage) predict job performance ies included in this meta-analysis are provided in the
and career success when taking each other’s effects into appendix, and the full details of our coding decisions are
account and (b) whether the effects of personality on available from the first two authors upon request.
performance outcomes are mediated by the attainment of
structurally advantageous network positions. Our meta- Coded Variables
analysis, which takes into account primary studies across
the personality and network literatures, examines both Personality. Previous meta-analytic research has re-
direct effects and potential mediation relationships. lied on the Big Five (Bono and Judge 2004, Chiaburu
et al. 2011, Judge et al. 2002) as an organizing frame-
work for sorting a number of more specific facets under
Method the five broad factors. The majority of the studies exam-
Literature Search ined the Big Five traits, but in cases where primary
The search for relevant studies began with a keyword studies featured a narrower trait, the first three authors
search of PsycINFO, ABI/Inform, and Web of Sci- independently coded the trait using the Big Five defini-
ence using the terms for personality (e.g., “personal- tions commonly provided in prior research (e.g., Hough
ity,” “traits,” “individual differences,” “self-monitoring,” and Ones 2001, Zhao and Seibert 2006, Zimmerman
“five-factor model traits,” “Big Five,” “extraver- 2008). Disagreements were resolved by discussion. For
sion,” “agreeableness,” “conscientiousness,” “neuroti- example, we categorized trait negative affect as neuroti-
cism,” “openness to experience”), network position cism, achievement as conscientiousness, and tolerance
(e.g., “centrality,” “degree,” “indegree,” “brokerage,” and empathy as agreeableness. These narrower traits
“betweenness,” “constraint”), the term “social networks” have been theoretically argued to be subdimensions of
in general, and terms for network centrality and work their respective Big Five traits (e.g., Hough and Ones
outcomes (e.g., “job performance,” “task performance,” 2001). Both measures of self-monitoring were included
“in-role performance,” “achievement,” “career success,” (Lennox and Wolfe 1984, Snyder 1974).
“promotion,” “compensation,” “bonus,” “salary”). We
Network Position. We included two different types of
also conducted a manual search of journals such
social network positions: indegree centrality and bro-
as Social Networks, Administrative Science Quarterly,
kerage. Indegree centrality was measured as the num-
Organization Science, the Academy of Management
ber of incoming ties received from others (Freeman
Journal, the Journal of Management, the Journal of
1979). Brokerage was measured as betweenness cen-
Personality and Social Psychology, and the Journal of
trality (Freeman 1979) or structural holes (i.e., reverse-
Applied Psychology, which are likely to publish rele-
vant empirical research. To collect unpublished stud- scored constraint; see Burt 1992). The first two authors
ies, we searched Dissertation Abstracts International conducted independent reviews of the studies and then
as well as the conference programs of the Academy compared coding sheets. We resolved disagreements by
of Management and other network conferences (as of discussion. We examined network position separately in
August 2014). We also made announcements on several two types of social networks: expressive (e.g., friendship
electronic mailing lists, soliciting working papers and and social support) versus instrumental (e.g., workflow,
unpublished data. In addition, we contacted researchers information and advice). We also conducted analyses for
directly to collect available but unpublished data. brokerage by analyzing only betweenness centrality. The
results and conclusions remained unchanged.
Inclusion Rules Job Performance and Career Success. We focused
To be included in the meta-analysis, a study had to on two work outcomes: job performance (i.e., success
(a) report an effect size statistic on the relationship of a in completing the tasks and responsibilities required by
personality variable and a measure of network centrality, individuals in a particular role) and career success (i.e.,
or an effect size statistic on the relationship of a measure achievement of objectively observable outcomes such
of network centrality and an outcome, and (b) exam- as promotion and compensation). Job performance mea-
ine one of these relationships in an adult sample. In sures included supervisor or peer ratings, performance
addition, studies had to compute a sociometric measure evaluations based on company records, and objective
of centrality (i.e., indegree or betweenness) from a full measures of job performance. Variables such as actual
network of relations among individuals within organiza- promotion, likelihood of promotion, salary, and bonuses
tions or an egocentric measure of centrality (i.e., struc- were coded as career success.
tural holes or reverse-scored constraint) from an ego or
full network within organizations. We also scrutinized Potential Methodological Moderators. When there
published studies for data that could capture these rela- was evidence showing that a meta-analytic relation-
tionships but were not reported. We obtained these rela- ship exhibited heterogeneity, we examined several
tionships directly from the authors (e.g., Shaw et al. potential methodological moderators concerning study
characteristics: (1) full versus ego network measures could invalidate the meta-analytic procedures. Therefore,
(pertaining to brokerage only), (2) adult student versus before performing the meta-analysis, we conducted a
adult employee samples, (3) concurrent versus predic- Monte Carlo simulation study to examine whether corre-
tive research designs, and (4) published versus unpub- lations derived from such data follow the prescribed dis-
lished studies. The last moderator was examined to tribution of the Hunter and Schmidt (2004) meta-analytic
check for publication bias. Specifically, we compared procedures. We varied the magnitude of the correla-
estimated effect sizes for published versus unpublished tion coefficient, network size, the extent of nestedness
data (the latter group including data from unpublished within the network, and total number of individuals in
dissertations, conference papers, and working papers). the network (N ). For each condition, we generated 5,000
One published social network study examined relation- replications. The empirically derived standard deviation
ships at the organizational level (Shaw et al. 2005). of the sampling error for the correlation (SDr ) was
From the authors, we were able to obtain individual- compared with that based√on the Hunter and Schmidt
level correlations within each social network and treated formula (SDr = 41 − r 2 5/ N − 1). We found that the
these effect sizes as independent (following the guid- bias for SDr was minimal (ranging from −1.5% to
ance from Hunter et al. 1982, pp. 429–443) and com- 1.3%). Thus, we continued with Hunter and Schmidt’s
ing from a published study. Given that the majority of procedures.
the relationships did not show heterogeneity, we provide
Potential Outliers. We used the sample-adjusted meta-
the moderator analysis findings in Online Appendix 2
analytic deviancy statistic to identify potential outliers
(available as supplemental material at http://dx.doi.org/
(Huffcutt and Arthur 1995), with corrections proposed
10.1287/orsc.2015.0972) instead of the Results section.
by Beal et al. (2002). For the relationships between per-
sonality and network position, only one or two outlier
Meta-Analytic Procedures
coefficients were identified for about 10% of these rela-
We followed the random-effects meta-analytic proce-
tionships, and no outliers were found for the rest of the
dures described by Hunter and Schmidt (2004). We cor-
relationships. Including or excluding these potential out-
rected the observed effect sizes for unreliability but not
liers did not change our conclusions regarding the par-
for range restriction. Specifically, personality variables
ticular relationships. For the indegree centrality–career
and job performance were corrected for unreliability
success relationship in instrumental networks, however,
based on reported local reliability estimates (i.e., inter-
we identified one outlier study (Gargiulo et al. 2009)
nal consistency coefficients). For a small proportion of
that had an extremely large sample (N = 21000 invest-
studies that did not report a reliability estimate for the
ment bankers). Including or excluding this study changed
personality variables or for job performance, we used
the true correlation estimate and the path coefficient
the average of available reliabilities for these variables
between centrality and work outcomes in instrumental
(average reliabilities for openness to experience = 0079,
conscientiousness = 0080, extraversion = 0086, agree- networks. Thus, we conducted separate analyses includ-
ableness = 0077, neuroticism = 0082, self-monitoring = ing and excluding this study and report two sets of results
0080, and job performance = 0092). Network centrality for instrumental networks.
variables and career success were not corrected for unre- Meta-Analytic Regression, Relative Weight Analy-
liability because of the lack of information provided in sis, and Path Modeling. Following prior research (e.g.,
primary studies. When the primary studies provided mul- Chiaburu et al. 2011, O’Boyle et al. 2012, Zimmerman
tiple correlations of a relationship based on a single sam- 2008), we used the meta-analyzed true-score correla-
ple, we averaged them into one correlation. This ensured tions matrices to conduct regression and relative weight
the effect sizes in our meta-analysis were independent. analyses (Johnson 2000) as well as meta-analytic path
We report 95% confidence intervals (CIs) around the modeling (Viswesvaran and Ones 1995). Specifically,
estimated true-score correlations. In addition, we report to address Research Questions 1 and 2, we conducted
80% credibility intervals (CVs), which provide infor- regression and relative weight analyses to examine the
mation on the possible range of corrected correlations. relative importance of self-monitoring and the Big Five
If the 80% CV of a particular effect size is wide and traits in predicting indegree centrality and brokerage
includes zero, this suggests the existence of potential and the relative importance of indegree centrality ver-
moderator(s) on the focal effect size. As Whitener (1990, sus brokerage in predicting work outcomes. To address
p. 317) suggested, CVs alone “cannot identify which Research Question 3, we conducted path modeling anal-
moderators are working,” and therefore testing a partic- yses that simultaneously captured (a) the influence of
ular categorical moderator “requires the use of a confi- personality (self-monitoring and the Big Five) on the
dence interval for populations or subpopulations.” two network positions, (b) the influence of personality
Within the same social network, one person’s network and the network positions on work outcomes, and (c) the
position is not independent of a peer’s network posi- potentially mediated effect of personality on work out-
tion. There are potential concerns about whether corre- comes through network position. We ran separate anal-
lations based on nonindependent data in primary studies yses for expressive and instrumental networks. In the
correlation matrices, we supplemented our newly cal- We predicted that self-monitoring would relate pos-
culated true-score correlations with effect sizes from itively to indegree centrality (in both types of orga-
prior meta-analyses. More details are provided in Online nizational networks) and to brokerage (particularly in
Appendix 3. As recommended by Viswesvaran and Ones expressive networks). The results, shown in Table 1, sup-
(1995), we used the harmonic mean of the sample sizes port our predictions. Self-monitoring related positively
in the relevant cells of the correlation matrix to rep- to indegree centrality across expressive ( = 0017, p <
resent the sample size for each path model. Following 00001) and instrumental ( = 0015, p < 00001) networks.
prior meta-analytic research on the Big Five and work Self-monitoring also related positively to brokerage in
outcomes (e.g., Chiaburu et al. 2011), we reported total expressive networks ( = 0016, p < 00001) but not to
explained variance (R2 ) and the change in multiple R brokerage in instrumental networks ( = 0004, n.s.). The
values associated with a predictor or a predictor group. changes in multiple R values (ãRSM_over_Big_Five ), which
The changes in multiple R values show the incremen-
indicate the predictive power of self-monitoring over
tal predicting power of the focal variable (or a variable
the Big Five in explaining the network positions, show
group) above and beyond other predictors in the model.
patterns of results consistent with the regression coeffi-
cients. After controlling for the Big Five, self-monitoring
Results significantly predicted indegree centrality in expressive
Research Question 1: How Does Personality Relate and instrumental networks and brokerage in expressive
to Network Position? networks. The relative weight analysis results, reported
To address this question, we conducted meta-analyses in the “%RW” column in Table 1, also are consistent
on the bivariate correlations between personality (self- with our predictions. Together, our results show that high
monitoring and the Big Five) and network positions self-monitors were sought out for friendship and advice
(indegree centrality and brokerage). We report these and held brokerage positions in expressive networks.
bivariate correlations in Online Appendix 1. These cor- Prior research offered no clear prediction concerning
relations were prerequisites for the matrices used in the relations of extraversion and openness to experience
the multivariate analyses (with details shown in Online to indegree centrality. The results (see Table 1) show
Appendix 3). The meta-analytic regressions and relative that extraversion was not related to indegree central-
weight analyses are shown in Table 1. ity in either expressive ( = 0005, n.s.) or instrumental
Table 1 Meta-Analytic Regression Models for Personality and Network Position with Relative Weight Analysis
Indegree Brokerage
RW %RW RW %RW
Expressive networks
Self-monitoring 0017∗∗∗ 00028 6101 0016∗∗∗ 00024 6200
Extraversion 0005 00006 1208 −0001 00004 1006
Openness to experience −0010∗∗∗ 00005 1107 0009∗∗ 00008 2008
Conscientiousness 0002 00001 204 0006∗ 00002 505
Agreeableness 0004 00003 506 −0001 00000 005
Neuroticism −0005∗ 00003 603 0002 00000 006
Total R (R2 5 0.216 (R2 = 00047∗∗∗ 5 0.198 (R2 = 00039∗∗∗ 5
ãRSM_over_Big_Five 00063∗∗∗ 00063∗∗∗
ãRBig_Five_over_SM 00036∗∗∗ 00028∗∗
Harmonic mean sample size 2,199 1,579
Instrumental networks
Self-monitoring 0015∗∗∗ 00023 4800 0004 00004 1405
Extraversion 0004 00008 1603 0009∗∗ 00009 3208
Openness to experience 0003 00002 501 0003 00003 902
Conscientiousness 0007∗∗ 00006 1203 0013∗∗∗ 00010 3607
Agreeableness −0003 00001 103 −0005∗ 00001 305
Neuroticism −0008∗∗ 00008 1701 0004 00001 302
Total R (R2 5 0.219 (R2 = 00048∗∗∗ 5 0.164 (R2 = 00027∗∗∗ 5
ãRSM_over_Big_Five 00046∗∗∗ 00004
ãRBig_Five_over_SM 00049∗∗∗ 00084∗∗∗
Notes. , standardized regression coefficients; RW, raw relative weights (see Johnson 2000); %RW, relative weights
expressed as a percentage of total explained variance of the dependent variable (R2 ); R, multiple correlation; ãR,
incremental change in multiple R values—for example, ãRSM_over_Big_Five refers to the incremental change in multiple R
values for self-monitoring over the Big Five traits.
∗
p < 0005; ∗∗ p < 0001; ∗∗∗ p < 00001.
Table 2 Meta-Analytic Regression Models for Network Position and Work Outcomes with Relative Weight
Analysis
Job performance Career success

RW %RW RW %RW
Expressive networks
Indegree 0014∗∗∗ 00019 8103 0011∗∗∗ 00012 8205
Brokerage 0003 00004 1807 0002 00003 1705
Total R (R2 5 0.152 (R2 = 00023∗∗∗ 5 0.121 (R2 = 00015∗∗∗ 5
ãRIndegree_over_Brokerage 00062∗∗∗ 00051∗∗∗
ãRBrokerage_over_Indegree 00002 00001
Indegree 0019∗∗∗ 00034 8408 0019∗∗∗ 00037 7202
Brokerage 0002 00006 1502 0007∗∗∗ 00014 2708
Total R (R2 5 0.200 (R2 = 00040∗∗∗ ) 0.226 (R2 = 00051∗∗∗ )
ãRIndegree_over_Brokerage 00090∗∗∗ 00066∗∗∗
ãRBrokerage_over_Indegree 00000 00006∗∗∗
Notes. , standardized regression coefficients; RW, raw relative weights (Johnson 2000); %RW, relative weights
expressed as a percentage of total explained variance of the dependent variable (R2 ); R, multiple correlation; ãR,
incremental change in multiple R values. When we predict career success with instrumental network indegree and
brokerage centralities, if we add the outlier study (Gargiulo et al. 2009) in the analysis, then harmonic mean sam-
ple size is 5,410, coefficient of indegree = 0.36, p < 00001; ãRIndegree_over_Brokerage = 00190, p < 00001; coefficient of
brokerage = −0001, p < 0030; and ãRBrokerage_over_Indegree = 00000, p > 0030.
∗∗∗
p < 00001.
( = 0004, n.s.) networks. We found a negative rela- of job performance across expressive ( = 0014, p <
tionship between openness to experience and indegree 00001) and instrumental ( = 0019, p < 00001) net-
centrality in expressive networks only ( = −0010, p < works. And, controlling for brokerage, indegree central-
00001). The anticipated positive relationship between ity remained a significant predictor of career success in
conscientiousness and indegree centrality was found for both expressive ( = 0011, p < 00001) and instrumental
instrumental networks only ( = 0007, p < 0001). As ( = 0019, p < 00001) networks. Consistent with these
expected, neuroticism related negatively to indegree cen- regression coefficients, the changes in multiple R values
trality across both expressive ( = −0005, p < 0005) (ãRIndegree_over_Brokerage ) show that indegree centrality was
and instrumental ( = −0008, p < 0001) networks. There predictive of job performance and career success above
are competing processes that may lead extraverts either and beyond brokerage.
to be brokers or to inhabit closed networks. In testing By contrast, controlling for indegree centrality, bro-
these competing theoretical predictions, we found that kerage failed to predict job performance in either expres-
extraversion related positively to brokerage in instrumen- sive ( = 0003, n.s.) or instrumental ( = 0002, n.s.)
tal networks only ( = 0009, p < 0001). As predicted, networks. But controlling for indegree centrality, bro-
openness to experience related positively to brokerage in kerage predicted career success in instrumental ( =
expressive networks ( = 0009, p < 0001), and consci- 0007, p < 00001) but not in expressive ( = 0002, n.s.)
entiousness related positively to brokerage in expressive
networks. Consistent with these regression coefficients,
( = 0006, p < 0005) and instrumental ( = 0013, p <
the changes in multiple R values (ãRBrokerage_over_Indegree )
00001) networks.
show that brokerage only predicted career success in
Research Question 2: How Do Indegree Centrality instrumental networks above and beyond indegree cen-
and Brokerage Compare in Explaining Job trality. Overall, the relative weight analysis results also
Performance and Career Success? supported these patterns of findings. Thus, we found that
The meta-analytic correlations of indegree centrality and indegree centrality was a more potent predictor of job
brokerage with job performance and career success are performance and career success than brokerage in both
shown in Online Appendix 1. These correlations form types of social networks.
the bases of the matrices for the multivariate analy-
ses (with details shown in Online Appendix 3). The Research Question 3: How Do Personality and
meta-analytic regression and relative weight analyses are Network Position Relate to Job Performance and
shown in Table 2. Career Success?
The results in Table 2 show that, controlling for bro- Our third research question has two component parts: (1)
kerage, indegree centrality was a significant predictor To what extent do personality variables (self-monitoring
and the Big Five) and network variables (indegree cen- extraversion ( = 0005, p < 0005), openness to experi-
trality and brokerage) predict job performance and career ence ( = −0011, p < 00001), agreeableness ( = 0005,
success in the presence of each other? (2) Do network p < 0005), and neuroticism ( = −0006, p < 0001) pre-
variables mediate the effects of personality variables dicted indegree centrality. Self-monitoring ( = 0016,
on work outcomes? To address these issues, we con- p < 00001), openness to experience ( = 0008, p <
ducted meta-analytic path modeling analyses for expres- 00001), and conscientiousness ( = 0004, p < 0005)
sive and instrumental networks and report the findings predicted brokerage. Indegree centrality had signifi-
in Figures 1 and 2, respectively. In both figures, the cant effects on job performance ( = 0013, p < 00001)
left-hand panel of the model resembles the results in and career success ( = 0011, p < 00001), but bro-
Table 1, and the right-hand panel of the model resem- kerage failed to predict these two outcomes. Further,
bles the results in Table 2. Going beyond the results direct relations between personality and the outcomes
reported in Tables 1 and 2, which involve one dependent persisted after taking people’s network positions into
variable at a time, Figures 1 and 2 show path modeling account. Job performance was higher for people scoring
results that examine multiple dependent variables simul- higher on conscientiousness ( = 0020, p < 00001), self-
taneously. Thus, we are able to examine various effects monitoring ( = 0005, p < 0005), and openness to expe-
of personality variables on work outcomes (i.e., direct rience ( = 0004, p < 0005). Career success was greater
effects and potentially mediated effects via indegree cen- for people scoring higher on extraversion ( = 0013, p <
trality and brokerage). Table 3 shows these direct and 00001), lower on agreeableness ( = −0019, p < 00001),
mediated effects as well as their 95% CIs. and lower on neuroticism ( = −0016, p < 00001).
The path coefficients presented in Figures 1 and 2 As shown in Figure 2, the pattern of results for
show how personality related to indegree centrality and instrumental networks mirrors those for expressive net-
brokerage and how both personality and network posi- works. Self-monitoring ( = 0015, p < 00001), extraver-
tion related to job performance and career success. The sion ( = 0004, p < 0005), conscientiousness ( = 0005,
models for expressive networks ( 2 = 10087, df = 10, p < 0005), and neuroticism ( = −0010, p < 00001) pre-
p > 0036, comparative fit index (CFI) = 1000, Tucker– dicted indegree centrality. Self-monitoring ( = 0004,
Lewis index (TLI) = 1000, root mean square error of p < 0005), extraversion ( = 0010, p < 00001), con-
approximation (RMSEA) = 00006) and instrumental net- scientiousness ( = 0010, p < 00001), and agreeable-
works ( 2 = 14093, df = 11, p > 0018, CFI = 1000, ness ( = −0004, p < 0005) predicted brokerage. In
TLI = 0099, RMSEA = 00011) both achieved satisfac- turn, indegree centrality predicted job performance and
tory fit. career success (both with = 0017, p < 00001). Broker-
As shown in Figure 1, concerning the link between age predicted career success ( = 0006, p < 0001). Fur-
personality and advantageous positions in expres- ther, direct relations between personality and outcomes
sive networks, self-monitoring ( = 0017, p < 00001), persisted after taking people’s network positions into
Figure 1 Meta-Analytic Path Model for Personality, Expressive Network Position, and Work Outcomes
3ELF
MONITORING

%XTRAVERSION
*OB
n )NDEGREE
/PENNESS PERFORMANCE

#ONSCIEN
TIOUSNESS #AREER
"ROKERAGE
SUCCESS

!GREEABLE
NESS Expressive networks
n
n
.EUROTICISM
n
Notes. Harmonic mean sample size = 21199. Path coefficients that were nonsignificant were fixed to zero, and these paths are omitted
from the figure. The model fit after fixing these paths to zero is 2 = 10087, df = 10, p > 0036, CFI = 1000, TLI = 1000, and RMSEA = 00006.
∗
p < 0005; ∗∗ p < 0001; ∗∗∗ p < 00001.
Figure 2 Meta-Analytic Path Model for Personality, Instrumental Network Position, and Work Outcomes

3ELF
MONITORING

%XTRAVERSION
*OB
)NDEGREE
/PENNESS PERFORMANCE

#ONSCIEN
TIOUSNESS #AREER
"ROKERAGE
SUCCESS
n
!GREEABLE
NESS
n n
n
.EUROTICISM
n
Notes. Harmonic mean sample size = 21720. Path coefficients that were nonsignificant were fixed to zero, and these paths are omitted
from the figure. The model fit after fixing these paths to zero is 2 = 14093, df = 11, p > 0018, CFI = 1000, TLI = 0099, and RMSEA = 00011.
∗
p < 0005; ∗∗ p < 0001; ∗∗∗ p < 00001.
account. Specifically, job performance was higher for study, changes in Figure 2 are that (1) the indegree–
people scoring higher on conscientiousness ( = 0019, career success path coefficient becomes 0.34 (p < 00001),
p < 00001) and self-monitoring ( = 0005, p < 0001); and (2) the brokerage–career success path coefficient
career success was greater for people scoring higher on becomes nonsignificant ( = −0002, p > 0030). All other
extraversion ( = 0013, p < 00001), lower on openness to coefficients in Figure 2 remained unchanged. Thus, our
experience ( = −0005, p < 0005), lower on agreeable- overall conclusion of the partially mediated relationships
ness ( = −0018, p < 00001), and lower on neuroticism remains the same for instrumental networks.
( = −0015, p < 00001). We also checked for evidence of an alternative to
To examine whether network position mediated the the partial mediation model: perhaps personality had no
relationship between personality and work outcomes, direct effects on outcomes but only affected performance
we conducted a series of mediation tests (MacKinnon and career success by facilitating the movement of peo-
2008; MacKinnon et al. 2002). Table 3 shows evidence ple into advantageous social network positions (see Burt
for partial mediation for several personality variables 2012). In testing this alternative approach (i.e., full
in expressive and instrumental networks. In terms of mediation model), we found that constraining the paths
job performance, personality characteristics (i.e., self- between personality and outcomes to zero resulted in
monitoring, openness to experience, and conscientious- models with significantly poorer fit to the data than the
ness) had both direct and mediated effects (via indegree partial mediation models (ã 2 = 210073, ãdf = 6, p <
centrality) on work outcomes. For example, the effect of 00001 for expressive networks; ã 2 = 891065, ãdf = 11,
self-monitoring on job performance was partially medi- p < 00001 for instrumental networks).
ated by indegree centrality in both expressive and instru- In summary, our overall findings across expressive
mental networks, as the first column in Table 3 indicates. and instrumental networks support a partial mediation
In terms of career success, the pattern of results also model among personality, network position, and work
showed evidence of partial mediation such that some outcomes. Our findings suggest that personality indi-
personality characteristics (i.e., neuroticism, agreeable- rectly influences performance and career success by facil-
ness, and extraversion) had both direct and mediated itating individuals’ occupation of advantageous positions
effects via network positions. For example, the nega- in organizational networks but that there is also a direct
tive effect of neuroticism on career success was partially effect of personality on the performance outcomes above
mediated by indegree centrality in expressive and instru- and beyond the mediated effects via advantageous net-
mental networks. work positions.
If the outlier study that concerned only instrumental
networks (Gargiulo et al. 2009) was included in the anal- Discussion
ysis presented in Figure 2, then the harmonic mean sam- There has been a structural hole between network schol-
ple size becomes 2,738. With the inclusion of this outlier ars who focus on social ties and researchers who focus
Table 3 Mediated and Direct Effects of Personality on Work Outcomes
Job performance Career success

Effect mediated Effect mediated Direct Effect mediated Effect mediated Direct
by indegree by brokerage effect by indegree by brokerage effect
Expressive networks
Self-monitoring 0002∗∗∗ — 0005∗ 0002∗∗∗ — —
(0.01, 0.03) (0.01, 0.09) (0.01, 0.03)
Extraversion — — — — — 0.13∗∗∗
(0.09, 0.17)
Openness to experience −0002∗∗∗ — 0004∗ −0001∗∗ — —
(−0.02, −0.01) (0.003, 0.09) (−0.02, −0.01)
Conscientiousness — — 0020∗∗∗ — — —
(0.16, 0.24)
Agreeableness 0001∗ — — 0001∗ — −0019∗∗∗
(0.00, 0.01) (0.00, 0.01) 4−00231 −00145
Neuroticism −0001∗∗ — — −0001∗ — −0016∗∗∗
4−00011 −000025 (−0.01, −0.002) (−0.21, −0.12)
Self-monitoring 0003∗∗∗ — 0005∗∗ 0003∗∗∗ — —
(0.02, 0.04) (0.02, 0.09) (0.02, 0.03)
Extraversion 0.01 — — 0.01 0001∗ 0013∗∗∗
(0.00, 0.02) (0.001, 0.01) (0.00, 0.01) (0.09, 0.17)
Openness to experience — — — — — −0005∗
4−00091 −00015
Conscientiousness 0001∗ — 0019∗∗∗ 0001∗ 0001∗∗ —
(0.001, 0.02) (0.16, 0.23) (0.001, 0.02) (0.002, 0.01)
Agreeableness — — — — −00003 −0018∗∗∗
4−00011 00005 4−00221 −00145
Neuroticism −0002∗∗∗ — — −0002∗∗∗ — −0015∗∗∗
4−00021 −00015 4−00021 −00015 4−00191 −00115
Notes. Reported in parentheses are the 95% CIs. After adding the outlier study (Gargiulo et al. 2009) in the analysis, brokerage no longer
served as a mediator in the instrumental networks path model.
∗
p < 0005; ∗∗ p < 0001; ∗∗∗ p < 00001.
on the personalities of individuals (Kilduff and Tsai the leading personality variables of relevance to social
2003). Our meta-analytic approach integrates these two network research—self-monitoring and the Big Five
research streams and provides a bridge across this dis- traits—in predicting indegree centrality and brokerage.
connect. One of the main findings to emerge from The overall pattern of results depicts high self-monitors
our analyses is that individuals’ personality characteris- (relative to low self-monitors) as garnering more inde-
tics and the positions they occupy within organizational gree connections such that people approach them for
social networks both matter for job performance and both expressive resources (e.g., friendship) and instru-
career success. In linking personality to network struc- mental resources (e.g., information and advice). High
ture, our inclusion of the Big Five traits beyond self- self-monitors are also more likely to hold broker-
monitoring enriches our understanding of who occupies age positions bridging disconnected friends. Given the
structurally advantageous positions in organizational net- importance of indegree centrality and brokerage for per-
works. Of relevance to theories of structural advantage, formance and career success, self-monitoring emerges as
perhaps the most surprising finding is that indegree an especially relevant personality variable in the predic-
centrality in instrumental networks such as advice and tion of networking behavior and individual success.
workflow is a stronger predictor of performance and With respect to the Big Five traits, our results provide
career success than is brokerage. new evidence concerning the relationship between per-
sonality and social network outcomes. First, we found
Contributions to Research and Theory that extraverted people are more likely to hold brokerage
Our meta-analysis has important theoretical insights for positions in instrumental networks, but not in expres-
research on social networks and personality psychology. sive networks. These differences suggest that aspects of
Our first contribution is to the literature on personality extraversion, such as assertiveness and ambition, may
psychology relevant to social networks. We examined be especially important for brokerage in instrumental
networks but inconsequential for brokerage in expres- workflow constraints, and job design restrictions (Brass
sive networks. Second, in clarifying the link between 1981, 1995). Within these parameters, even modest
openness to experience and friendship, we found that amounts of explained variance can represent significant
people higher in openness to experience tend to have structural advantages (see Prentice and Miller 1992).
smaller friendship networks, but they are also more Our second contribution to the microfoundations of
likely to act as go-betweens connecting disconnected organizational networks is an empirical one: our meta-
friends, suggesting that having diverse interests drives analysis indicates that indegree centrality emerges as a
them to interact with people from different social cir- stronger predictor of job performance and career success
cles for friendship. Third, we found that conscientious than does brokerage. Our analyses take into account a
individuals are more likely to be approached for work- wide literature on network position and individual suc-
related advice and information and to bridge the flow of cess, reflecting a diverse pool of jobs, work settings, and
work-related information between people disconnected demographics. By contrast, structural hole theory (Burt
at work. This finding helps explain the well-established 1992, 2005), and the evidence upon which it is based, is
association between conscientiousness and job perfor- often focused on managers whose careers benefit from
mance across occupations (see Barrick et al. 2001). timely access and movement of information and knowl-
Fourth, we found that neurotic people, who may be con- edge from one place to another. Considering the work-
sidered high-cost interaction partners, are less likely to force more broadly, rather than focusing just on managers
be approached for friendship and advice. in the knowledge economy, it appears that people benefit
Our overall findings highlight that self-monitoring is more from having larger networks or being in receipt of
especially relevant for understanding why people differ resources from many others, rather than being in posi-
in their acquisition of advantageous positions in social tions that span otherwise unconnected network clusters.
networks. The finding that self-monitoring played a sig- Overall, theories of structural advantage are likely to be
nificant role in predicting indegree centrality and bro- informed by our finding that the number of incoming ties
kerage after controlling for the Big Five traits extends in organizational social networks is a strong correlate of
prior studies that neglected to take into account the individual success in organizations and a stronger overall
Big Five personality variables (e.g., Mehra et al. 2001, predictor than is brokerage.
Sasovova et al. 2010). As such, our meta-analysis pro- Our third major contribution is to the integration of
vides further insights into self-monitoring theory by personality and network variables. We tested two models
establishing this personality variable as a distinct con- concerning how personality and network position relate
struct from the Big Five traits in the field of social net- to individual success. The first model posits that person-
work research, thereby responding to a recent call in the ality is unrelated to both performance and career success
management literature for self-monitoring to be differ- once individuals attain advantageous network positions.
entiated from extraversion (Dalal et al. 2015). Further, The second model incorporates two predictions: (a) that
our meta-analysis also advances knowledge concerning the individual’s personality is related, in part, to the like-
how the Big Five personality factors, which still receive lihood that the individual attains an advantageous net-
limited scholarly attention in social network research, work position and (b) that the individual’s personality
affect people’s attainment of structurally advantageous and the individual’s network position directly affect both
network positions. Our findings extend prior work (Klein the individual’s chances of performance and career suc-
et al. 2004) that focused on indegree centrality within cess. This second model implies a partial mediation rela-
team friendship and advice networks. We found that tionship. In this meta-analysis, we ask, which model best
three personality traits (openness to experience, extraver- fits the data?
sion, and conscientiousness) related positively to broker- Interpreting the overall pattern of findings, we gen-
age in expressive or instrumental networks. Overall, our erally found support for the second model—personality
meta-analysis on the personality–network position rela- helps individuals move into advantageous network posi-
tionships suggests that it is important for researchers to tions, and both personality tendencies and network
broaden their examination of other personality variables, positions relate to performance and career success. The
beyond self-monitoring, in linking to structurally advan- overall tests of relative model fit showed that the data
tageous network positions such as indegree centrality represented a better fit with the second model. We
and brokerage. also found that both personality and network position
One caveat, however, is that personality, as reflected were each incremental predictors of performance and
in self-monitoring and the Big Five traits, contributed career success, as shown by the significant paths from
modestly to the prediction of indegree centrality and bro- these predictors to the outcome variables (see Figures 1
kerage. The total amount of variance explained in these and 2). Personality explained unique variance in perfor-
network outcomes ranged between 3% and 5%. The evi- mance and career success not captured by network posi-
dence suggests that organizational networks represent tion, and vice versa. Further, the results of the specific
relatively strong situations with hierarchical structures, mediation tests showed that, in approximately half of
the cases, network position partially mediated the rela- each Big Five trait (Dudley et al. 2006) and “contextual-
tionship between personality and work outcomes. For ized” measures of personality (Shaffer and Postlethwaite
example, high self-monitors and conscientious people 2012) could have incremental predictive power over
performed better at work not only because of the direct and above self-monitoring and the broad Big Five
influence of their personality characteristics on perfor- traits examined here. Future research should also focus
mance but also because of these tendencies helping them on other individual difference variables (e.g., positive
move into advantageous network positions (e.g., inde- affectivity, negative affectivity, general cognitive abil-
gree centrality), which in turn enhanced their perfor- ity, emotional intelligence, core self-evaluations, proac-
mance. The overall pattern of results supports a partially tive personality) that may deserve empirical attention.
mediated model for the relationships among personality, For instance, in ancillary meta-analyses we found that,
network position, and work outcomes. across expressive and instrumental networks, positive
The meta-analysis emphasizes the importance of inte- affectivity had significant and modest positive correla-
grating network structure and personality into exam- tions with indegree centrality and brokerage ( ranged
inations of performance outcomes. Network scholars, from 0.09 to 0.14), whereas negative affectivity had
who work within the structural legacy, have largely smaller, although significant, negative correlations with
neglected the possibility that, beyond network positions, indegree centrality and brokerage ( ranged from −0.04
individual differences relate to performance outcomes. to −0.08).
For example, Burt (2012) measured personality char- Third, our research focused on indegree centrality and
acteristics in terms of consistency of network behav- brokerage, two of the most studied network positions, for
ior across roles that individuals play in online games. answering the three research questions. Future research
His results suggested that personality had no relation to could explore whether our findings hold for other struc-
work outcomes in the presence of network position. Our turally advantageous positions. For example, consider-
findings challenge this conclusion. Drawing on specific ing their differential effects on indegree centrality, we
measures of individuals’ personalities in work organiza- also examined whether self-monitoring and extraversion
tions, our meta-analysis shows evidence that personality relate differently to the number of outgoing ties to oth-
(self-monitoring and the Big Five) and network posi- ers (i.e., outdegree centrality) in our ancillary meta-
tion (indegree centrality and brokerage) both relate to analytic regression analyses. We found that extraversion
job performance and career success and, further, that
related positively to outdegree centrality across expres-
network position partially mediates certain relationships
sive and instrumental networks, whereas self-monitoring
between personality and the performance outcomes. As
was unrelated to outdegree centrality in either type of
such, our integration of personality and networks, built
social networks. This evidence suggests a key difference
on 138 independent samples examining these links, pro-
between self-monitoring and extraversion—high self-
vides additional insight concerning the pathways toward
monitors tend to be sought out for advice and friendship,
performance and career success in organizations.
whereas extraverts tend to seek out others for advice and
Limitations and Future Research friendship.
Our research has several limitations that point to direc- Fourth, future research should consider the bound-
tions for future research. First, given that many of the ary conditions of the relationships studied here. We
studies involve nonexperimental data, we are not able focused on expressive and instrumental networks as
to offer strong inferences concerning causality among one potential moderator. But (as noted by an anony-
personality, network position, and work outcomes. Prior mous reviewer) other moderators, such as work char-
work on personality and social ties (e.g., Asendorpf and acteristics (the extent of social interaction and task
Wilpers 1998) and the genetic basis of personality (e.g., interdependence) and employees’ hierarchical level in
Yamagata et al. 2006) suggests that personality is likely organizations, are also likely to be. In post hoc analy-
to precede network position because of the former’s rel- ses, we did not find any consistent pattern of moderation
ative stability. However, we recognize that specific facets for social interaction, task interdependence, and hierar-
of personality are known to change in response to rela- chical level for either the personality–network position
tionship factors (Mund and Neyer 2014). Furthermore, or the network–position–outcome relationships. Future
it is also possible that individuals with higher perfor- research could explicitly measure work characteristics
mance and greater career success achieve more advanta- and hierarchical level for each individual and investigate
geous network positions. Future research is needed to better their moderating effects across the personality–network
understand the causal direction of the relationships among position or the network–position–outcome relationships.
network position, performance, and career success. Fifth, we focused on a partial mediation model in
Second, we focused on self-monitoring and the most answering the broad question on how personality and
popular personality framework (i.e., the Big Five). How- network position work together in relating to work out-
ever, other approaches may improve the predictive valid- comes. An alternative model (that we were unable to test
ity of personality. For example, lower-level “facets” of because of the absence of information on correlations
with interaction terms) captures the interaction of per- Bono JE, Anderson MH (2003) Advice and influence networks
sonality and network position. This model has been sug- of transformational leaders. Working paper, University of Minnesota,
gested by several authors (e.g., Burt 2007, Kilduff and Minneapolis.
Brass 2010, Mehra et al. 2001) and has been exam- Brands RA, Kilduff M (2014) Just like a woman? Effects of
gender-biased perceptions of friendship network brokerage on attri-
ined in empirical work (e.g., Anderson 2008, Fang and
butions and performance. Organ. Sci. 25:1530–1548.
Shaw 2009, Fang et al. 2015). We recommend that future
*Brass DJ (1984) Being in the right place: A structural analysis
research report zero-order correlations with the moder- of individual influence in an organization. Admin. Sci. Quart. 29:
ated regression terms to facilitate meta-analytic tests. 518–539.
In addition, in instrumental networks, our path model- *Burt RS (2007) Secondhand brokerage: Evidence on the impor-
ing results showed that the coefficient of the path from tance of local structure for managers, bankers, and analysts. Acad.
brokerage to career success became nonsignificant after Management J. 50:119–148.
including the outlier study (i.e., Gargiulo et al. 2009), Burton PE (2007) Dimensions of social networks as predictors
which greatly increased the harmonic mean size as well of employee performance. Doctoral dissertation, University of North
as the coefficient of the path from indegree centrality Texas, Denton.
to career success. Thus, we should be cautious in inter- Carboni I, Casciaro T (2013) Love me or hate me: Performance
preting the mediating role played by brokerage in instru- effects of sociometric status in organizational networks. Acad. Man-
agement Proc. 2013(Meeting abstract supplement):Abstract 17033.
mental networks in the relationship of personality and
Carboni I, Ehrlich K (2013) The effect of relational and team char-
work outcomes.
acteristics on individual performance: A social network perspective.
In conclusion, the findings of our meta-analysis show Human Resource Management 52:511–535.
that it is pivotal to integrate both characteristics of indi- Casciaro T (1998) Seeing things clearly: Social structure, person-
viduals (e.g., self-monitoring and the Big Five) and their ality, and accuracy in social network perception. Soc. Networks 20:
social network positions (e.g., indegree centrality and 331–351.
brokerage) into theory and research for a better under- Choi K, Kim MJ (2014) Energizing networks and public sector
standing of how the personalities of people and their employees’ attitudes. Acad. Management Proc. 2014(Meeting abstract
social networks matter for important work and career supplement):Abstract 13241.
outcomes. Clement J, Shipilov AV, Galunic CD (2014) The making
of brokers: When second-order brokerage leads to formation of
neighborhood-spanning ties. Acad. Management Proc. 2014(Meeting
Supplemental Material abstract supplement):Abstract 11505.
Supplemental material to this paper is available at http://dx.doi
*Cross R, Cummings JN (2004) Tie and network correlates of indi-
.org/10.1287/orsc.2015.0972.
vidual performance in knowledge-intensive work. Acad. Management
J. 47:928–937.
Acknowledgments Dolgova E (2013) Fake it till you make it? The effect of self-
The first two authors contributed equally to the paper. monitoring and competence on friendship dynamics. Acad. Manage-
The authors thank Tiziana Casciaro, Sigal Barsade, David ment Proc. 2013(Meeting abstract supplement):Abstract 13888.
Krackhardt, and three anonymous reviewers for their helpful
Erdogan B, Bauer TN, Walter J (2015) Deeds that help and
comments and suggestions on previous drafts. The authors words that hurt: Helping and gossip as moderators of the relation-
also thank In-Sue Oh for his suggestions on meta-analytic ship between leader–member exchange and advice network centrality.
procedures. Personnel Psych. 68:185–214.
Appendix. Primary Studies Included in Meta-Analysis Fang R (2010) Peer influence on undermining behaviors in the
Below we list the studies included in our meta-analysis. Those workplace: A social network perspective. Doctoral dissertation, Uni-
with asterisks are also cited in the reference list. versity of Minnesota, Minneapolis.
*Anderson MH (2008) Social networks and the cognitive motiva- *Fang R, Shaw JD (2009) Self-monitoring, status, and justice-
tion to realize network opportunities: A study of managers’ informa- related information flow. J. Occupational Organ. Psych. 82:405–430.
tion gathering behaviors. J. Organ. Behav. 29:51–78. Fang R, Duffy MK, Shaw JD (2011) A multi-wave examination of
Anderson MH (2012) Team member traits and capabilities, social social capital, core self-evaluations, and newcomer adjustment. Work-
capital, and transactive memory development. Working paper, Iowa ing paper, National University of Singapore, Singapore.
State University, Ames. Flynn FJ, Reagans RE, Amanatullah ET, Ames DR (2006) Help-
Anderson MH, Haar J (2012) Individual differences, social capi- ing one’s way to the top: Self-monitors achieve status by helping
tal, and organizational behavior outcomes. Working paper, Iowa State others and knowing who helps whom. J. Personality Soc. Psych. 91:
University, Ames. 1123–1137.
*Asendorpf JB, Wilpers S (1998) Personality effects in social rela- *Gargiulo M, Ertug G, Galunic C (2009) Two faces of control:
tionships. J. Personality Soc. Psych. 74:1531–1544. Network closure and individual performance among knowledge work-
Baer M (2010) The strength-of-weak-ties perspective on creativity. ers. Admin. Sci. Quart. 54:299–333.
J. Appl. Psych. 95:592–601. Geletkanycz MA, Boyd BK, Finkelstein S (2001) The strategic
Bizzi L (2013a) The dark side of structural holes: A multilevel value of CEO external directorate networks: Implications for CEO
investigation. J. Management 39:1554–1578. compensation. Strategic Management J. 22:889–898.
Bizzi L (2013b) The social structure of job crafting: Exploring the Grosser TJ (2014) The effect of political skill and social network
ambivalent nature of networks. Acad. Management Proc. 2013(Meet- structure on innovation and career success. Acad. Management Proc.
ing abstract supplement):Abstract 13159. 2014(Meeting abstract supplement):Abstract 10650.
Ho VT, Pollack JM (2014) Passion isn’t always a good thing: ties. Acad. Management Proc. 2013(Meeting abstract supplement):
Examining entrepreneurs’ network centrality and financial perfor- Abstract 15003.
mance with a dualistic model of passion. J. Management Stud. *Oh H, Kilduff M (2008) The ripple effect of personality on social
51:433–459.
structure: Self-monitoring origins of network brokerage. J. Appl.

Hopp C, Zenk L (2012) Collaborative team networks and implica- Psych. 93:1155–1164.
tions for strategic human resource management. Internat. J. Human Oldroyd JB (2007) The burdens and benefits of information flow:
Resource Management 23:2975–2994.
Social structure, interdependence, information flow and performance.
*Kalish Y, Robins G (2006) Psychological predispositions and net- Doctoral dissertation, Northwestern University, Evanston, IL.
work structure: The relationship between individual predispositions,
*Pollet TV, Roberts SGB, Dunbar RIM (2011) Extraverts have
structural holes and network closure. Soc. Networks 28:56–84.
larger social network layers: But do not feel emotionally closer to
Kilduff M (1988) Decision making in context: Social and per- individuals at any layer. J. Individual Differences 32:161–169.
sonality correlates of choices in organizations. Doctoral dissertation,
Cornell University, Ithaca, NY. *Sasovova Z, Mehra A, Borgatti SP, Schippers MC (2010) Net-
work churn: The effects of self-monitoring personality on brokerage
Kim E, Bhave D (2012) Too smart for your own good? Cognitive
dynamics. Admin. Sci. Quart. 55:639–670.
ability, personality, and network centrality. Working paper, Georgia
Institute of Technology, Atlanta. Schulte, M, Cohen NA, Klein KJ (2012) The coevolution of net-
work ties and perceptions of team psychological safety. Organ. Sci.
*Klein KJ, Lim B, Saltz JL, Mayer DM (2004) How do they get
23:564–581.
there? An examination of the antecedents of centrality in team net-
work. Acad. Management J. 47:952–963. Scott BA (2007) Employee popularity: Its nature, measurement,
Kleinbaum AM, Jordan A, Audia PG (2013) Do you read me? and organizational relevance. Doctoral dissertation, University of
How perceptions of empathy shape self-monitors’ brokerage in social Florida, Gainesville.
networks. Acad. Management Proc. 2013(Meeting abstract supple- *Seibert SE, Kraimer ML, Liden RC (2001) A social capital theory
ment):Abstract 10950. of career success. Acad. Management J. 44:219–237.
Lau DC (2001) Job consequences of trustworthy employees: *Selfhout M, Burk W, Branje S, Denissen J, van Aken M, Meeus
A social network analysis. Doctoral dissertation, University of Hong W (2010) Emerging late adolescent friendship networks and Big
Kong, Hong Kong. Five personality traits: A social network approach. J. Personality 78:
Lee YH, Yang LS, Wan KM, Chen GH (2010) Interactive effects 509–538.
of personality and friendship networks on contextual performance. Setton RP, Mossholder KW (2002) Relationship quality and rela-
Soc. Behav. Personality 38:197–208. tionship context as antecedents of person- and task-focused interper-
Liu Y, Ipe M (2010) How do they become nodes? Revisiting team sonal citizenship behavior. J. Appl. Psych. 87:255–267.
member network centrality. J. Psych. 144:243–258. *Shaw JD, Duffy MK, Johnson JL, Lockhart DE (2005) Turnover,
Luria G, Kalish Y (2013) A social network approach to peer social capital losses, and performance. Acad. Management J. 48:
assessment: Improving predictive validity. Human Resource Manage- 594–606.
ment 52:537–560. Shea C, Fitzsimons G (2012) A goal driven view of social network
Marrone JA (2004) Cutting across team boundaries: Antecedents structure. Working paper, Duke University, Durham, NC.
and implications of individual boundary spanning behavior within Shum C, Zhou L, Carter DR, Mo S, Chen H (2014) Influence
consulting teams. Doctoral dissertation, University of Maryland, Col- of subordinates’ and supervisors’ network positions on the effect of
lege Park.
abusive supervision. Acad. Management Proc. 2014(Meeting abstract
McFadyen MA, Cannella AA Jr (2004) Social capital and knowl- supplement):Abstract 15652.
edge creation: Diminishing returns of the number and strength of
*Sparrowe RT, Liden RC, Wayne SJ, Kraimer ML (2001) Social
exchange relationships. Acad. Management J. 47:735–746.
networks and the performance of individuals and groups. Acad. Man-
*Mehra A, Kilduff M, Brass DJ (2001) The social networks of agement J. 44:316–325.
high and low self-monitors: Implications for workplace performance.
Admin. Sci. Quart. 46:121–146. Tasselli S (2013) Network structure, individual agency and out-
comes: Power dynamics following a change in technology. Acad.
Merluzzi JM (2014) The myth of the queen bee: A study and
Management Proc. 2013(Meeting abstract supplement):Abstract
gender and work conflict at two firms. Acad. Management Proc.
13039.
2014(Meeting abstract supplement):Abstract 11654.
*Toegel G, Anand N, Kilduff M (2007) Emotion helpers: The role
Milton LP, Westphal JD (2005) Identity confirmation networks and
of high positive affectivity and high self-monitoring managers. Per-
cooperation in work groups. Acad. Management J. 48:191–212.
sonnel Psych. 60:337–365.
Miners CTH (2008) It is who you know and what you know:
An examination of the relations among emotional intelligence, social Totterdell P, Holman D, Hukin A (2008) Social networks: Mea-
network centrality, and performance. Doctoral dissertation, University suring and examining individual differences in propensity to connect
of Toronto, Toronto. with others. Soc. Networks 30:283–296.
Moore GJ (2006) The longitudinal effects of self-monitoring and Totterdell P, Wall T, Holman D, Diamond H, Epitropaki O (2004)
locus of control on social network position in friendship networks. Affect networks: A structural analysis of the relationship between
Master’s thesis, Air Force Institute of Technology, Wright-Patterson work ties and job-related affect. J. Appl. Psych. 89:845– 867.
Air Force Base, OH. Venkataramani V, Green SG, Schleicher DJ (2010) Well-connected
Neubert MJ, Taggar S (2004) Pathways to informal leadership: leaders: The impact of leaders’ social network ties on LMX and mem-
The moderating role of gender on the relationship of individual dif- bers’ work attitudes. J. Appl. Psych. 95:1071–1084.
ferences and team member network centrality to informal leadership Wang L, Kishore R, Balkundi P (2012) Knowledge network evo-
emergence. Leadership Quart. 15:175–194. lution: A cross-level study of individual antecedents and team learn-
Neuman EJ, Avgar AC (2013) A bridge over troubled water? ing effectiveness. Working paper, State University of New York at
The push and pull of team conflict on between-team network Buffalo, Buffalo.
Wei J, Zheng W (2012) Linking motivation to social capital: The Brass DJ (1984) Being in the right place: A structural analysis of
roles of regulatory focus and self-monitoring. Acad. Management individual influence in an organization. Admin. Sci. Quart. 29:
Proc. 2012(Meeting abstract supplement):Abstract 15619. 518–539.
Wisker ZL (2011) The effect of personality, emotional intelligence, Brass DJ (1995) A social network perspective on human resources
and social network characteristics on sales performance: The mediat- management. Res. Personnel and Human Resources Management
ing roles of market intelligence use, adaptive selling behaviour, and 31:39–79.
improvisation. Doctoral dissertation, University of Waikato, Hamil- Brass DJ, Burkhardt ME (1993) Potential power and power use: An
ton, New Zealand. investigation of structure and behavior. Acad. Management J.
Xiao Z, Tsui AS (2007) When brokers may not work: The cultural 36:441–470.
contingency of social capital in Chinese high-tech firms. Admin. Sci. Burkhardt ME, Brass DJ (1990) Changing patterns or patterns of
Quart. 52:1–31. change: The effect of a change in technology on social network
Yuan YC, Gay G (2006) Homophily and network ties and bonding structure and power. Admin. Sci. Quart. 35:104–127.
and bridging social capital in computer-mediated distributed teams. Burt RS (1992) Structural Holes: The Social Structure of Competition
J. Comput.-Mediated Comm. 11:1062–1084. (Harvard University Press, Cambridge, MA).
*Zhang Z, Peterson SJ (2011) Advice networks in teams: The role Burt RS (2005) Brokerage and Closure (Oxford University Press,
of transformational leadership and members’ core self-evaluations. New York).
J. Appl. Psych. 96:1004–1017. Burt RS (2007) Secondhand brokerage: Evidence on the importance
Zhang M, Wei J (2014) The contingent value of centrality on indi- of local structure for managers, bankers, and analysts. Acad.
vidual performance: A multi-level analysis. Acad. Management Proc. Management J. 50:119–148.
2014(Meeting abstract supplement):Abstract 14465. Burt RS (2012) Network-related personality and the agency ques-
Zou X (2009) Social networks and subjective well-being: Regu- tion: Multi-role evidence from a virtual world. Amer. J. Sociol.
latory focus between self-regulation and network structure. Doctoral 118:543–591.
dissertation, Columbia University, New York. Burt RS, Jannotta JE, Mahoney JT (1998) Personality correlates of
structural holes. Soc. Networks 20:63–87.
Burt RS, Kilduff M, Tasselli S (2013) Social network analysis:
References Foundations and Frontiers on advantage. Annual Rev. Psych.
64:527–547.
Anderson MH (2008) Social networks and the cognitive motivation to Cheng C, Bond MH, Chan SC (1995) The perception of ideal best
realize network opportunities: A study of managers’ information friends by Chinese adolescents. Internat. J. Psych. 30:91–108.
gathering behaviors. J. Organ. Behav. 29:51–78. Chiaburu DS, Oh IS, Berry CM, Li N, Gardner RG (2011) The five-
Asendorpf JB, Wilpers S (1998) Personality effects in social relation- factor model of personality traits and organizational citizenship
ships. J. Personality Soc. Psych. 74:1531–1544. behaviors: A meta-analysis. J. Appl. Psych. 96:1140–1166.
Ashton MC, Lee K, Paunonen SV (2002) What is the central fea- Cross R, Cummings JN (2004) Tie and network correlates of individ-
ture of extraversion? Social attention versus reward sensitivity. ual performance in knowledge-intensive work. Acad. Manage-
J. Personality Soc. Psych. 83:245–252. ment J. 47:928–937.
Baldwin TT, Bedell MD, Johnson JL (1997) The social fabric of a Dabbs JM, Evans MS, Hopper CH, Purvis JA (1980) Self-monitors in
team-based M.B.A. program: Network effects on student satis- conversation: What do they monitor? J. Personality Soc. Psych.
faction and performance. Acad. Management J. 40:1369–1397. 39:278–284.
Balkundi P, Kilduff M, Harrison DA (2011) Centrality and charisma: Dalal RS, Meyer RD, Bradshaw RP, Green JP, Kelly ED, Zhu M
comparing how leader networks and attributions affect team per- (2015) Personality strength and situational influences on behav-
formance. J. Appl. Psych. 96:1209–1222. ior: A conceptual review and research agenda. J. Management
Barrick MR, Mount MK, Judge TA (2001) Personality and perfor- 41:261–287.
mance at the beginning of the new millennium: What do we Day DV, Schleicher DJ, Unckless AL, Hiller NJ (2002) Self-
know and where do we go next? Internat. J. Selection Assess. monitoring personality at work: A meta-analytic investigation of
9:9–30. construct validity. J. Appl. Psych. 87:390–401.
Barrick MR, Stewart GL, Piotrowski M (2002) Personality and job Denissen JJA, Penke L (2008) Neuroticism predicts reactions to cues
performance: Test of the mediating effects of motivation among of social inclusion. Eur. J. Personality 22:497–517.
sales representatives. J. Appl. Psych. 87:43–51. De Pascalis V, Arwari B, Matteucci M, Mazzocco A (2005) Effects
Baron RA (1989) Personality and organizational conflict: Effects of of emotional visual stimuli on auditory information processing:
the type A behavior pattern and self-monitoring. Organ. Behav. A test of J.A. Gray’s reinforcement sensitivity theory. Personality
Human Decision Processes 44:281–296. Individual Differences 38:163–176.
Beal DJ, Corey DM, Dunlap WP (2002) Identifying outlier stud- Digman JM (1990) Personality structure: Emergence of the five-factor
ies in meta-analysis: Alternatives to Huffcutt and Arthur’s 1995 model. Annual Rev. Psych. 41:417–440.
approach. J. Appl. Psych. 87:583–589. Dudley NM, Orvis KA, Lebiecki JE, Cortina JM (2006) A meta-
Bendersky C, Shah NP (2013) The downfall of extraverts and the rise analytic investigation of conscientiousness in the prediction of
of neurotics: The dynamic process of status allocation in task job performance: Examining the intercorrelations and the incre-
groups. Acad. Management J. 56:387–406. mental validity of narrow traits. J. Appl. Psych. 91:40–57.
Bono JE, Judge TA (2004) Personality and transformational and trans- Emerson RM (1962) Power-dependence relations. Amer. Sociol. Rev.
actional leadership: A meta-analysis. J. Appl. Psych. 89:901–910. 27:31–41.
Bossard JH (1945) The law of family interaction. Amer. J. Sociol Fang R, Shaw JD (2009) Self-monitoring, status, and justice-related
50:292–294. information flow. J. Occupational Organ. Psych. 82:405–430.
Brass DJ (1981) Structural relationships, job characteristics, and Fang R, Chi L, Chen M, Baron RA (2015) Bringing political skill into
worker satisfaction and performance. Admin. Sci. Quart. 26: social networks: Findings from a field study of entrepreneurs.
331–348. J. Management Stud. 52:175–212.
Fleeson W, Gallagher P (2009) The implications of big five standing Kilduff M, Tsai W (2003) Social Networks and Organizations (Sage,
for the distribution of trait manifestation in behavior: Fifteen London).
experience-sampling studies and a meta-analysis. J. Personality Klein KJ, Lim B, Saltz JL, Mayer DM (2004) How do they get there?
Soc. Psych. 97:1097–1114. An examination of the antecedents of centrality in team network.
Freeman LC (1979) Centrality in social networks: Conceptual clarifi- Acad. Management J. 47:952–963.
cation. Soc. Networks 1:215–239. Kleinbaum AM (2012) Organizational misfits and the origins of bro-
Freeman LC, Roeder D, Mulholland RR (1980) Centrality in social kerage in intrafirm networks. Admin. Sci. Quart. 57:407–452.
networks, II: Experimental results. Soc. Networks 2:119–141.
Knoke D, Burt RS (1983) Prominence. Burt RS, Miner MJ, eds.
Fuglestad PT, Snyder M (2010) Status and motivational founda- Applied Network Analysis: A Methodological Introduction (Sage,
tions of self-monitoring. Soc. Personality Psych. Compass 4: Beverly Hills, CA), 195–222.
1031–1041.
Lennox RD, Wolfe RN (1984) Revision of the self-monitoring scale.
Furnham A, Capon M (1983) Social skills and self-monitoring pro-
J. Personality Soc. Psych. 46:1349–1364.
cesses. Personality Individual Differences 4:171–178.
Gargiulo M, Ertug G, Galunic C (2009) Two faces of control: MacKinnon DP (2008) Introduction to Statistical Mediation Analysis
Network closure and individual performance among knowledge (Routledge, New York).
workers. Admin. Sci. Quart. 54:299–333. MacKinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V
Gibbons DE (2004) Friendship and advice networks in the context of (2002) A comparison of methods to test mediation and other
changing professional values. Admin. Sci. Quart. 49:238–262. intervening variable effects. Psych. Methods 7:83–104.
Hinds PJ, Carley KM, Krackhardt D, Wholey D (2000) Choos- McCrae RR (1996) Social consequences of experiential openness.
ing work group members: Balancing similarity, competence, Psych. Bull. 120:323–337.
and familiarity. Organ. Behav. Human Decision Processes 81: McCrae RR, John OP (1992) An introduction to the five-factor model
226–251. and its applications. J. Personality 60:175–215.
Holmes JG (2002) Interpersonal expectations as the building blocks Mehra A, Kilduff M, Brass DJ (2001) The social networks of high
of social cognition: An interdependence theory perspective. Per- and low self-monitors: Implications for workplace performance.
sonal Relationships 9:1–26. Admin. Sci. Quart. 46:121–146.
Hough LM, Ones DS (2001) The structure, measurement, validity, Mund M, Neyer FJ (2014) Treating personality-relationship trans-
and use of personality variables in industrial, work, and organi- actions with respect: Narrow facets, advanced models, and
zational psychology. Anderson N, Ones DS, Sinagil HK, Viswes- extended time frames. J. Personality Soc. Psych. 107:352–368.
varan C, eds. Handbook of Industrial, Work and Organizational
Psychology, Vol. 1 (Sage, London), 233–277. Nettle D (2006) The evolution of personality variation in humans and
other animals. Amer. Psychologist 61:622–631.
Huffcutt AI, Arthur W (1995) Development of a new outlier statistic
for meta-analytic data. J. Appl. Psych. 80:327–334. Ng TWH, Eby LT, Sorensen KL, Feldman DC (2005) Predictors of
Hunter JE, Schmidt FL (2004) Methods of Meta-Analysis: Correcting objective and subjective career success: A meta-analysis. Person-
Error and Bias in Research Findings, 2nd ed. (Sage, Thousand nel Psych. 58:367–408.
Oaks, CA). O’Boyle EH Jr, Forsyth DR, Banks GC, McDaniel MA (2012)
Hunter JE, Schmidt FL, Jackson GB (1982) Meta-Analysis: Cumulat- A meta-analysis of the dark triad and work behavior: A social
ing Research Findings Across Studies (Sage, Beverly Hills, CA). exchange perspective. J. Appl. Psych. 97:557–579.
Ickes WJ, Reidhead S, Patterson M (1986) Machiavellianism and Oh H, Kilduff M (2008) The ripple effect of personality on social
self-monitoring: As different as “me” and “you.” Soc. Cognition structure: Self-monitoring origins of network brokerage. J. Appl.
4:58–74. Psych. 93:1155–1164.
Jensen-Campbell LA, Gleason KA, Adams R, Malcolm KT (2003) Pollet TV, Roberts SGB, Dunbar RIM (2011) Extraverts have larger
Interpersonal conflict, agreeableness, and personality develop- social network layers: But do not feel emotionally closer to indi-
ment. J. Personality 71:1059–1086. viduals at any layer. J. Individual Differences 32:161–169.
Johnson JW (2000) A heuristic method for estimating the relative Prentice DA, Miller DT (1992) When small effects are impressive.
weight of predictor variables in multiple regression. Multivariate Psych. Bull. 112:160–164.
Behav. Res. 35:1–19. Sasovova Z, Mehra A, Borgatti SP, Schippers MC (2010) Network
Judge TA, Heller D, Mount MK (2002) Five-factor model of per- churn: The Effects of self-monitoring personality on brokerage
sonality and job satisfaction: A meta-analysis. J. Appl. Psych. dynamics. Admin. Sci. Quart. 55:639–670.
87:530–541.
Seibert SE, Kraimer ML, Liden RC (2001) A social capital theory of
Judge TA, Simon LS, Hurst C, Kelley K (2014) What I experienced career success. Acad. Management J. 44:219–237.
yesterday is who I am today: Relationship of work motivations
and behaviors to within-individual variation in the five-factor Selfhout M, Burk W, Branje S, Denissen J, van Aken M, Meeus W
model of personality. J. Appl. Psych. 99:199–221. (2010) Emerging late adolescent friendship networks and Big
Five personality traits: A social network approach. J. Personality
Kalish Y, Robins G (2006) Psychological predispositions and network
structure: The relationship between individual predispositions, 78:509–538.
structural holes and network closure. Soc. Networks 28:56–84. Shaffer JA, Postlethwaite BE (2012) A matter of context: A meta-
Kashdan TB, McKnight PE, Fincham FD, Rose P (2011) When analytic investigation of the relative validity of contextualized
curiosity breeds intimacy: Taking advantage of intimacy oppor- and noncontextualized personality measures. Personnel Psych.
tunities and transforming boring conversations. J. Personality 65:445–494.
79:1369–1402. Shaffer DR, Smith JE, Tomarelli M (1982) Self-monitoring as a deter-
Kilduff M, Brass DJ (2010) Organizational social network research: minant of self-disclosure reciprocity during the acquaintance pro-
Core ideas and key debates. Acad. Management Ann. 4:317–357. cess. J. Personality Soc. Psych. 43:163–175.
Kilduff M, Krackhardt D (1994) Bringing the individual back in: Shaw JD, Duffy MK, Johnson JL, Lockhart DE (2005) Turnover,
A structural analysis of the internal market for reputation in orga- social capital losses, and performance. Acad. Management J.
nizations. Acad. Management J. 37:87–108. 48:594–606.
Snyder M (1974) The self-monitoring of expressive behavior. J. Per- Zimmerman RD (2008) Understanding the impact of personality traits
sonality Soc. Psych. 30:526–537. on individuals’ turnover decisions: A meta-analytic path model.
Snyder M (1979) Self-monitoring processes. Berkowitz L, ed. Personnel Psych. 61:309–348.
Advances in Experimental Social Psychology, Vol. 12 (Academic
Press, New York), 85–128. Ruolian Fang is assistant professor in management and
Snyder M (1987) Public Appearances, Private Realities: The Psy- organization at NUS Business School, National University
chology of Self-Monitoring (W. H. Freeman & Company, of Singapore. She received her Ph.D. from the Univer-
San Francisco). sity of Minnesota. Her research interests include social net-
Snyder M, Gangestad S, Simpson JA (1983) Choosing friends as works, newcomer socialization, workplace deviant behavior,
activity partners: The role of self-monitoring. J. Personality Soc. and entrepreneurship.
Psych. 51:181–190. Blaine Landis is postdoctoral research fellow in organi-
Sparrowe RT, Liden RC, Wayne SJ, Kraimer ML (2001) Social net- zational behavior at University College London. He received
works and the performance of individuals and groups. Acad. his Ph.D. from the University of Cambridge. His research
Management J. 44:316–325. focuses on the psychology of individuals’ social networks in
Stokes RC (1985) The effects of price, package design, and brand organizations.
familiarity on perceived quality. Jacoby J, Olson J, eds. Perceived Zhen Zhang is associate professor of management at the
Quality: How Consumers View Stores and Merchandise (Lexing-
W. P. Carey School of Business, Arizona State University.
ton Books, Lexington, MA), 233–246.
He received his Ph.D. from the University of Minnesota.
Stovel K, Shaw L (2012) Brokerage. Annual Rev. Sociol. 38:139–158.
His research concentrates on leadership processes and leader-
Toegel G, Anand N, Kilduff M (2007) Emotion helpers: The role
ship development, the biological basis of work behavior, and
of high positive affectivity and high self-monitoring managers.
Personnel Psych. 60:337–365.
research methods.
Marc H. Anderson is associate professor in management at
Turner RG (1980) Self-monitoring and humor production. J. Person-
ality 48:163–172.
the College of Business at Iowa State University. He received
his Ph.D. from the University of Minnesota. His research inter-
Viswesvaran C, Ones DS (1995) Theory testing: Combining psycho-
metric meta-analysis and structural equations modelling. Person- ests include the effects of personality, social networks, and
nel Psych. 48:865–885. their interaction on organizational behavior, leadership, man-
Wasserman S, Faust K (1994) Social Network Analysis: Methods and agerial and organizational cognition, sensemaking, and timing.
Applications (Cambridge University Press, Cambridge, UK). Jason D. Shaw is chair professor and codirector of the
Whitener EM (1990) Confusion of confidence intervals and credibility Centre for Leadership and Innovation at the Hong Kong Poly-
intervals in meta-analysis. J. Appl. Psych. 75:315–321. technic University. He received his Ph.D. from the University
Yamagata S, Suzuki A, Ando J, Ono Y, Kijima N, Yoshimura K, of Arkansas. His research interests include team effective-
Ostendorf F, et al. (2006) Is the genetic structure of human per- ness, employment relationships, envy, financial incentives, and
sonality universal? A cross-cultural twin study from North Amer- turnover.
ica, Europe, and Asia. J. Personality Soc. Psych. 90:987–998. Martin Kilduff is professor of organizational behavior in
Zhang Z, Peterson SJ (2011) Advice networks in teams: The role of the Management Science and Innovation department, Univer-
transformational leadership and members’ core self-evaluations. sity College London. He received his Ph.D. from Cornell
J. Appl. Psych. 96:1004–1017. University. His research focuses on the microfoundations and
Zhao H, Seibert SE (2006) The big five personality dimensions and consequences of individuals’ social networks, with particular
entrepreneurial status: A meta-analytical review. J. Appl. Psych. emphasis on the role of personality, cognition, and emotion in
91:259–271. these processes.
This work is licensed under a Creative Commons Attribution 4.0 International

License. You are free to copy, distribute, transmit and adapt this work, but you
must attribute this work as “Organization Science. Copyright 2015 INFORMS.
http://dx.doi.org/10.1287/orsc.2015.0972, used under a Creative Commons
Attribution License: http://creativecommons.org/licenses/by/4.0/.”

Integrating Personality and Social Networks

Uploaded by

Copyright:

Available Formats

Integrating Personality and Social Networks

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Integrating Personality and Social Networks

Uploaded by

Copyright:

Available Formats

This article was downloaded by: [2003:ec:cf02:1546:e87f:2700:44d5:699] On: 04 May 2023, At: 15:13