Vous êtes sur la page 1sur 12

Social Networks 35 (2013) 382–393

Contents lists available at SciVerse ScienceDirect

Social Networks
journal homepage: www.elsevier.com/locate/socnet

Pathways to happiness: From personality to social networks


and perceived support
Xiumei Zhu a,∗ , Sang Eun Woo b , Caitlin Porter b , Michael Brzezinski c
a
Brian Lamb School of Communication, Purdue University, Beering Hall of Liberal Arts and Education, Room 2114, 100 North University Street,
West Lafayette, IN 47907, United States
b
Department of Psychological Sciences, Purdue University, 703 Third Street, West Lafayette, IN 47907, United States
c
Office of International Students and Scholars, Purdue University, Young Hall, Room #120, 155 S. Grant Street, West Lafayette, IN 47907-2114, United States

a r t i c l e i n f o a b s t r a c t

Keywords: The present study attempts to incorporate individual personality into a social network account of sub-
Individual difference jective wellbeing (SWB) by proposing and testing a path model in which social relationships (egocentric
Personality social networks and perceived social support, PSS) mediate the relationship between personality and
Social network
SWB. We found that network characteristics (including network size, emotional closeness, and pro-
Social support
portion of new contacts) are related to SWB largely via the mediation of PSS. Additionally, network
Wellbeing
Health size and proportion of new contacts function as important mediating mechanisms linking extraversion,
agreeableness and openness to PSS and SWB.
© 2013 Elsevier B.V. All rights reserved.

1. Introduction (Cohen, 2004; House et al., 1988; Thoits, 1995). The social network
perspective emphasizes the role of structural and compositional
Social networks matter. The social network perspective pro- characteristics of individual ego networks such as the number of
vides a unique approach to understanding how social relationships network contacts, strength of ties, and diversity of network contacts
influence individuals’ instrumental achievements and wellbeing. In (Acock and Hurlbert, 1993; Perry and Pescosolido, 2010; Song and
particular, various characteristics of egocentric social network have Lin, 2009). Researchers argue that characteristics of social networks
been linked to indicators of subjective wellbeing such as depres- determine the availability and adequacy of social support, which
sion and life satisfaction (Acock and Hurlbert, 1993; Burt, 1987; Lin in turn influences wellbeing. However, with few exceptions (Lin
et al., 1999; Perry and Pescosolido, 2010; Totterdell et al., 2008). et al., 1999), social support has rarely been evaluated as mediator
However, little research has fully assessed the mechanisms through of the relationship between network characteristics and wellbeing
which social networks may influence these outcomes (Anderson, (Thoits, 1995).
2008; Reagans and McEvily, 2003). Additionally, the social network In contrast to the sociological literature, psychological research
perspective has not fully acknowledged the role of individual differ- has largely focused on perceived social support, or the perceived
ences, such as personality, in shaping network characteristics (Burt, availability of assistance from others (Bolger and Eckenrode, 1991;
2000; Robins and Kashima, 2008; Stevenson and Greenberg, 2000). Cohen and Wills, 1985). Perceived support is considered a more
In the present study, we seek to advance our knowledge on these proximal measure of the useful features of one’s social networks,
two issues. To do so, we propose and test a path analytic model in and is at the root of cognitive and emotional responses to illness and
which social network characteristics are linked to subjective well- stress (Cohen, 2004). This disjuncture in disciplinary approaches to
being (SWB) via perceived social support (PSS), and social networks the social determinants of wellbeing leaves a number of impor-
and PSS together form an important pathway between personality tant questions unanswered. For example, do social networks and
and wellbeing. perceived social support separately influence individual wellbe-
Research suggests that social networks exert influence on well- ing? Or, is the relationship between social networks and subjective
being primarily through the provision of social support. Social wellbeing mediated by perceived social support?
support is defined as the assistance available via one’s connection to Also, the current literature on social networks largely neglects
others, such as companionship, information and tangible assistance the role of psychological mechanisms. Calling for more attention
to the issue of agency, Emirbayer and Goodwin (1994) pointed out
that social network research fails to show how intentional human
∗ Corresponding author. Tel.: +1 765 494 2951; fax: +1 765 496 1394. action shapes the very social networks that constrain actors in turn.
E-mail address: zhu74@purdue.edu (X. Zhu). Advocating for more communication between network research

0378-8733/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.socnet.2013.04.005
X. Zhu et al. / Social Networks 35 (2013) 382–393 383

and psychological research, Robins and Kashima (2008) stated that the opportunity for regular social interactions and stable social
with little attention paid to the motivation and cognition of indi- roles. When engaging in social interactions and fulfilling social
viduals, social network research risks a seriously undertheorized roles, “. . .individuals gain a sense of identity, predictability and
account of human experience. stability; of purpose; and of meaning, belonging, security, and
Linking personality to social networks is considered as a promis- self-worth” (Cohen, 2004, p. 679). Consequently, social networks
ing way to incorporate individual (or differential) psychology into are likely to promote positive cognitions and emotions (Cassel,
a network account of individual experience and outcomes, such as 1976; Cohen, 1988; Pavot et al., 1990; Thoits, 1983). Furthermore,
wellbeing (Burt et al., 1998; Kalish and Robins, 2006; Totterdell social networks provide multiple sources of information that could
et al., 2008). Personality captures relatively stable patterns of influence health relevant behaviors, lead to more effective use of
thought, emotion, motivation and behavior, and influences per- available health services, and help avoid potentially stressful or
ceptions, attitudes and values, as well as reactions to people and high-risk situations. Network relationships are also critical sources
situations (McAdams, 2009). At the very basic, dyadic level of social of material resources (e.g., money) and tangible help (e.g., a ride to
relationships, personality plays a substantial role in relationship a healthcare provider; Lin et al., 1999; Lin, 2001). The knowledge
development and maintenance (Flynn et al., 2006; McCrae, 1996; that one can obtain necessary assistance from network contacts
Robins and Kashima, 2008). Personality may also shape individuals’ bolsters one’s confidence in the ability to cope with stress and
motivation to leverage advantageous network positions and ability attenuates negative interpretations of stressful situations (Cohen,
to overcome structural constraints (Anderson, 2008; Mehra et al., 2004; Thoits, 1995). Less negative appraisal and greater confidence
2001). in the ability to cope in turn lowers the level of effective stress,
Indeed, several personality traits have been linked to important and promotes more beneficial emotional and cognitive responses
network outcomes, including proactive actions to build network to events. Therefore, social relationships not only routinely pro-
ties (Ashford and Black, 1996; Forret and Dougherty, 2001; Lambert mote wellbeing, but also protect individuals against the potential
et al., 2006; Totterdell et al., 2008; Turban and Dougherty, 1994; negative influence of stressful life events (Cohen and Wills, 1985;
Wanberg and Kammeyer-Mueller, 2000), structural and composi- Thoits, 1995).
tional characteristics of egocentric social networks, and network To capture individuals’ social relationships, the network per-
positions (Burt et al., 1998; Mehra et al., 2001; Oh and Kilduff, 2008; spective of social support has largely relied on an egocentric
Roberts et al., 2008; Totterdell et al., 2008). Further, personality network approach (Wasserman and Faust, 1999), which empha-
dispositions have been linked to PSS (Bowling et al., 2005; Carver sizes that structural characteristics of a network (e.g., the size
and Connor-Smith, 2010; Zellars and Perrewé, 2001), and to well- of one’s network, emotional closeness with contacts, and degree
being (DeNeve and Cooper, 1998; Steel et al., 2008). PSS was also of connectedness among one’s contacts), and the composition of
found to mediate the effect of certain personality variables on cop- contacts in a network (e.g., demographic diversity, and social or
ing behaviors (Von Dras and Siegler, 1997) and depression (Finch formal organizational status) determine the availability of social
and Graziano, 2001). While these studies provide useful informa- support. Early research along this line focused on global measures
tion on the role of personality traits in shaping social relationships of social interaction that assessed the nature of relationship ties,
and wellbeing outcomes, most of them were conducted in a piece- neighborhood interaction, community participation, and group
meal fashion; the entire pathway from personality to SWB via social and organizational membership, and their effects on depres-
relationships (i.e., network structure and PSS) has not been fully sive symptoms, positive feelings about life, and overall wellbeing
explicated. (Cohen and Wills, 1985; Bolger and Eckenrode, 1991; Pinquart
In the present study, we trace the entire pathway from person- and Sorensen, 2000). Recently, social network researchers have
ality to SWB via social networks and PSS. In doing so, we weave applied formal network methodologies (e.g., name generator or
multiple streams of research in the fields of sociology and psychol- position generator) in attempts to link specific characteristics of
ogy into a theoretical model that accounts for both the mediating egocentric networks to wellbeing outcomes such as life satisfac-
processes underlying the relationship between social networks and tion, depressive symptoms, and self reported general health (Acock
wellbeing and the role of personality traits as antecedents of these and Hurlbert, 1993; Lin et al., 1999; Perry and Pescosolido, 2010;
processes. In the following sections, we first provide an integrative Song and Lin, 2009; Totterdell et al., 2008).
critical review of current literatures on: (1) the interconnections Psychological research adopted a different approach to study-
among social networks, perceived social support and wellbeing and ing the connection between social relationships and wellbeing by
(2) the relations between personality characteristics and various focusing largely on perceived availability of social support (Cohen,
social network variables. We then propose and test hypotheses on 2004; Cohen and Wills, 1985; Thoits, 1995). Two mechanisms have
specific associations between these constructs with a sample of 309 been proposed. On one hand, perceived support may reflect the
first-year college students. We feel that a population of incoming actual quantity and quality of support from others, which can help
college students is particularly advantageous for investigating the reduce isolation, avoid stressful situations, develop more positive
relationships among personality, social networks, PSS, and SWB. appraisal of negative events, and cope with difficulties. On the other
First-year college students are in the process of transition into a hand, the perception that others will provide appropriate assis-
new social environment where they most likely do not have well- tance is sufficient for people to acquire a generally more positive
established social networks. Such a situation enables the role of emotional state and develop more positive psychological responses
personality in the development of social networks to be more easily to stress, which may function independently of the reception of
observed. assistance. Empirically, there are a large number of studies docu-
menting the beneficial effect of PSS on subjective wellbeing and
2. Literature review physical health (Cohen, 2004; Cohen and Janicki-Deverts, 2009;
Thoits, 1985; Pinquart and Sorensen, 2000).
2.1. Social networks, perceived social support (PSS), and Both the social network approach and the perception-based
subjective wellbeing (SWB) approach demonstrate the importance of social relationships for
SWB. The emphasis on perceived support reflects the reasonable
Social relationships influence wellbeing via the provision of assumption that people’s perceptions are more direct measures of
social support (Cohen, 2004; Henderson, 1977, 1981; House et al., the helpful features of relationships than the more distal meas-
1988; Thoits, 1995). Social networks represent the structure of and ures of structural features of social relationships (Cohen and Wills,
384 X. Zhu et al. / Social Networks 35 (2013) 382–393

1985). Social network research, on the other hand, posits that per- suggest that the effect of social networks on SWB is better under-
ceptions of social support develop as a result of the existence of stood when we consider the mediating role of PSS. We propose that
social relationships. Specifically, characteristics of social networks PSS is a more proximal predictor of SWB, and that social network
determine the quantity and quality of social support, which in turn characteristics, as indicators of the interpersonal environment and
shapes support perceptions (Lin, 2001; Lin et al., 1999; Thoits, social resources potentially available to an individual, indirectly
1985). These pathways, however, have not been fully tested by relate to SWB via the mediation of PSS. Second, we hypothesize that
social network researchers. A notable exception is a study by Lin as indicators of the quantity and quality of potential and perceived
et al. (1999) which found that the effect of two network characteris- social resources, social networks and PSS together mediate the rela-
tics, the number of weekly contacts and the presence of an intimate tionship between certain personality traits and SWB. We suggest
relationship (e.g., a spouse or partner), on depressed mood was par- that dispositional characteristics facilitate or hinder the devel-
tially mediated by perceived expressive support. More research is opment and maintenance of social relationships that influence
needed to determine the pathway from network characteristics to perceived access to social resources, which then influence SWB.
perceived social support and wellbeing.

3.1. PSS predicts SWB


2.2. Personality and SWB: mediating effects of social networks
and PSS
Three major dimensions of social support have been identified:
Subjective wellbeing (SWB) is a broad psychological phe- emotional, informational, and instrumental support (Cohen, 2004;
nomenon that includes “people’s emotional responses, domain House and Kahn, 1985; Lin et al., 1999). Emotional support refers
satisfactions, and global judgments of life satisfaction” (Diener to the opportunities for trust, companionship, venting, and sup-
et al., 1999, p. 277). Psychological research has documented con- portive communications such as empathy, comfort, reassurance
sistent evidence that personality influences individual wellbeing and encouragement. Informational support refers to the provision
(DeNeve and Cooper, 1998; Steel et al., 2008), via two major path- of relevant information, advice, or guidance that can help the indi-
ways. First, certain personality traits such as positive affectivity (or vidual define, understand, and cope with difficulties and problems.
extraversion) and negative affectivity (or neuroticism) are known Instrumental support refers to tangible help such as lending money,
to be directly related to SWB, as they describe individuals’ predis- providing a ride, or helping with household chores. While differ-
positions to experience positive or negative affect, ceteris paribus ent dimensions of social support can be distinguished conceptually,
(Steel et al., 2008). they are often highly intercorrelated empirically (Cohen and Wills,
Second, as discussed previously, personality may influence 1985).
SWB via the pathway of social relationships and PSS. Personality Researchers have assessed PSS using measures that capture dif-
traits influence how individuals seek out, establish, maintain, and ferent facets of support (e.g., Lin et al., 1999), as well as global
subjectively assess relationships with others (Kalish and Robins, measures that refer to support across dimensions (e.g., Cohen and
2006; McCrae, 1996; Robins and Kashima, 2008), which ulti- Hoberman, 1983). Despite such diverse measures, findings have
mately affects individuals’ SWB. Indeed, studies have shown that converged on the beneficial effect of PSS on SWB. PSS has been
individual differences play a substantial role in explaining individ- positively associated with SWB (Finch et al., 1989; Pinquart and
uals’ proactive social networking behaviors (Lambert et al., 2006; Frohlich, 2009) or psychological wellbeing (Turner, 1981), and neg-
Forret and Dougherty, 2001; Wanberg and Kammeyer-Mueller, atively related to anxiety and depression (Bolger and Eckenrode,
2000). There is some evidence that individual attributes such as 1991; Brissette et al., 2002; Lara et al., 1997; Lin et al., 1999), and has
independence and individualism, self-monitoring, extraversion, been found to mitigate the negative impact of stressful life events
neuroticism, agreeableness, and openness explain significant vari- on psychological and physical functioning (Cohen and Wills, 1985;
ation in the characteristics of individual ego networks (e.g., size, Cohen, 2004). Consistent with existing research, we hypothesize
emotional closeness, diversity, or structural hole) or individual that PSS will have a positive relationship with SWB.
structural positions within a complete network (e.g., centrality)
H1. PSS will be positively related to SWB.
(Burt et al., 1998; Jensen-Campbell et al., 2002; Klein et al., 2004;
Mehra et al., 2001; Oh and Kilduff, 2008; Roberts et al., 2008;
Totterdell et al., 2008). With regard to PSS, existing research has 3.2. Network structure predicts PSS
found that PSS mediates the effect of extraversion on coping behav-
iors (Von Dras and Siegler, 1997) and the effects of extraversion, While various methods are used to assess social networks, the
agreeableness, and neuroticism on depression (Finch and Graziano, egocentric network is the most often studied in the context of social
2001). support (Wasserman and Faust, 1999). In the study of egocentric
Integrating theories and findings dispersed in multiple streams networks, the name generator is the method most commonly used
of research, this study develops and tests a comprehensive model to assess network characteristics and structure (Marsden, 2005).
that traces the full scope of interconnections from personality to In this approach, participants (egos) are given questions that elicit
SWB, via structural features of social networks and subjective per- the name of network contacts (alters) such as people with whom
ception of social support. By doing so, the current study intends to they discuss important matters (Burt, 1984), or people with whom
shed light on the interplay between psychological processes and they chat or visit (Campell and Lee, 1991). Once a list of names
structural factors in shaping SWB. Moreover, linking personality has been produced, participants are presented with a series of
and social networks contributes to the growing effort by social net- follow-up questions (name interpreters) that request information
work researchers to incorporate individual agency and motivation about each alter, the relationship between the ego and the alters,
into a structural account of human experience (Anderson, 2008; and/or the relationships between alters. Data collected through
Burt, 2000; Mehra et al., 2001). name generators and interpreters provide individual profiles of
respondents’ egocentric network members that can be aggre-
3. Theory and hypotheses gated into measures of network composition and structure such
as average tie strength, mean alter characteristics, proportion of
Based on our critical review of the current literature, we pro- contacts with certain characteristics, network range and constraint,
pose an integrative model with two objectives (see Fig. 1). First, we and so on.
X. Zhu et al. / Social Networks 35 (2013) 382–393 385

PERSONALITY SOCIAL NETWORK


STRUCTURE

Extraversion
+ Closeness
+
+ + +
Agreeableness + + +
Size
+
+ +
Conscientiousness Perceived Subjective
social wellbeing
- Upper
+
reachability
- -
Openness
-
+ +
- Proportion of
new contacts
Neuroticism -

Fig. 1. Hypothesized path model (PSS as full mediator). Paths to be focused on are indicated by bold arrows. Control paths included to be consistent with the literature are
in light arrows.

In the present study, we focus on four structural features of ego- 1984). Thus, we hypothesize that network size has a positive asso-
centric networks that are theoretically relevant to social support: ciation with perceived social support.
emotional closeness, network size (i.e., the number of network
contacts), upper reachability, and proportion of newly developed H2b. Network size will be positively related to PSS.
contacts. Emotional closeness is considered particularly impor-
tant for accessing social assistance (Cohen and Wills, 1985; Lin Upper reachability. Upper reachability refers to the position of
et al., 1999; Thoits, 1995). Network size influences the quantity the network contacts in a hierarchical social structure (Lin, 2001).
of assistance potentially available through one’s network (Acock Researchers have measured it using contacts’ average or maximum
and Hurlbert, 1993; Burt, 1987; Campbell et al., 1986; Lin, 2001). educational level, socioeconomic status, occupational prestige, or
Upper reachability refers to the position of the network contacts position in an organizational hierarchy (Acock and Hurlbert, 1993;
in a hierarchical social structure, and can determine the quality of Morrison, 2002; Seibert et al., 2001; Song and Lin, 2009). It is argued
resources network contacts potentially provide (Lin, 2001). Propor- that contacts occupying higher positions may possess more valu-
tion of newly developed network contacts is particularly important able information and resources, which not only is beneficial for
for individuals making a transition into their first year of college. instrumental actions such as finding a job (Lin, 2001), but also may
Compared with preexisting contacts, newly developed contacts improve life satisfaction (Acock and Hurlbert, 1993), and reduce
have current and firsthand knowledge of the new environment, and depression (Song and Lin, 2009). In our study, we conceptualize
are physically closer, which enables them to more easily provide upper reachability as levels reached within the university struc-
assistance. We propose that these four network characteristics will ture. First-year college students may benefit from social interaction
be related to PSS. Below we describe our theoretical rationales. with more senior students because the latter have more knowledge
Emotional closeness. Emotionally close network contacts are par- about the new environment, and more experience dealing with the
ticularly important for social support (Henderson, 1977). Social various challenges of college life. Further, staff and faculty members
network research suggests that individuals are more likely to seek may possess information about the university that is not readily
emotional, informational, and instrumental support from contacts available to students, and provide more useful guidance on how to
with whom they feel emotionally close, and that closely connected adapt to college life. Therefore, having upperclassmen and/or fac-
network contacts are also more motivated to provide social support ulty in one’s network may lead to more favorable evaluation of the
(Lin, 2001; Granovetter, 1983). Research has consistently linked the availability and adequacy of social support.
presence of confidants or emotionally close network contacts to
perceived availability of social support (Hays and Oxley, 1986; Lin H2c. Upper-reachability will be positively related to PSS.
et al., 1999; Pinquart and Sorensen, 2000; Seeman and Berkman,
1988; Stokes, 1983; Wellman and Wortley, 1990). Therefore, we Proportion of new contacts. Adapting to a new environment, such
hypothesize a positive association between emotional closeness as entering college or a new workplace, requires the acquisition
and perceived support availability. of information through multiple sources (Ostroff and Kozlowski,
1992). Research suggests that network ties within the new envi-
H2a. Emotional closeness will be positively related to PSS.
ronments play a pivotal role in facilitating positive socialization
Network size. The size of one’s network may influence the and adjustment outcomes (Hays and Oxley, 1986; Morrison, 2002).
quantity as well as the variety of resources and assistance poten- Compared with network contacts elsewhere, contacts developed
tially available to the ego. A larger network indicates more people on campus are geographically closer, share common experience,
who can potentially help in different ways, and should therefore and are more knowledgeable about the social and academic
improve one’s opinion about support availability. Although some environment. Therefore, they are more available for emotional,
studies yielded inconclusive findings (Dimond et al., 1987; Vaux informational, and instrumental support than pre-existing con-
and Harrison, 1985), a vast majority of available research has found tacts. As a result, we hypothesize a positive association between
a significant relationship between PSS and network size (Connell proportion of new network contacts and PSS.
and D’Augelli, 1990; Cutrona, 1986; Lin et al., 1999; Norbeck et al.,
1981; Seeman and Berkman, 1988; Turner et al., 1983; Ward et al., H2d. Proportion of new contacts will be positively related to PSS.
386 X. Zhu et al. / Social Networks 35 (2013) 382–393

3.3. Partial versus full mediation of PSS in the relationship compared to disagreeable individuals, agreeable individuals tend
between network characteristics and SWB to have larger friendship networks (Jensen-Campbell et al., 2002),
are more likely to be selected as friends (Klein et al., 2004; Selfhout
The provision of social support has been held as the primary et al., 2010), experience less conflict (Asendorpf and Wilpers, 1998),
mechanism underlying the relation between social networks and and enjoy more intimate relationships (White et al., 2004; Wu et al.,
wellbeing related outcomes. While researchers have considered 2008).
both received social support and PSS as potentially mediating the While existing studies support the beneficial effect of agreeable-
effect of social networks on wellbeing, the literature on received ness on network size and emotional closeness, they provide little
social support has either yielded null findings (e.g., Lin et al., evidence linking agreeableness to other aspects of social networks.
1999) or reported greater depression when received social support Asendorpf and Wilpers (1998) reported an insignificant association
increased (Bolger et al., 2000; Cohen and Wills, 1985). Horenstein between agreeableness and relationship formation. Wanberg and
and Cohen (2009) believe that this finding indicates a causal rela- Kammeyer-Mueller (2000) failed to find any significant relation-
tionship from depression to received social support, suggesting that ship between agreeableness and proactive networking behaviors.
it is those who are most in need (under stress, experiencing neg- Finally, we are not aware of any theory or evidence linking agree-
ative emotions) who actually mobilize their social networks. We ableness to the diversity or status of network contacts. In view of
therefore expect PSS to be the primary mediating mechanism of existing theory and research, we expect agreeable individuals to
the association between social network characteristics and SWB. have larger networks and closer relationships with their contacts.
To assess whether PSS is full or partial mediator, we will evalu-
H3b. Agreeableness will be positively related to network size and
ate an alternative model that includes direct paths from network
emotional closeness.
variables to SWB.
Conscientiousness. This personality dimension refers to “socially
3.4. Personality predicts network structure prescribed impulse control that facilitates task and goal directed
behaviors” (John and Srivastava, 1999, p. 121). Conscientious indi-
Research in personality psychology has converged on a five- viduals tend to be dutiful, organized, disciplined, hard working,
factor model to describe the core dimensions of human personality reliable, and achievement oriented (Costa and McCrae, 1992), and
(the Big Five), namely, Extraversion, Agreeableness, Conscientious- have been found to experience greater achievement such as bet-
ness, Openness to Experience, and Neuroticism (Digman, 1990; ter grades and better job performance (Judge et al., 1999). These
Goldberg, 1990, 1992). Existing research has provided abundant attributes, however, seem to be less relevant in social interac-
theoretical rationales and some empirical evidence linking the tions in which achievement and task related performance are not
Big Five to interpersonal dynamics and characteristics of ego emphasized (Anderson et al., 2001). Indeed, there is less theory or
networks. Below, we present detailed arguments for the hypothe- empirical evidence available to directly connect conscientiousness
sized relationships between the Big Five factors and the network and social network characteristics. Nevertheless, Asendorpf and
characteristics discussed earlier. Wilpers (1998) found that first-year college students high in con-
Extraversion. Extraversion refers to the extent to which an scientiousness were more likely to have frequent contact with their
individual is optimistic, sociable, energetic, enthusiastic, and cheer- family members. Further, Hough (1992) and Doeven-Eggens et al.
ful in outlook (John and Srivastava, 1999; McCrae and John, (2008) noted that conscientious individuals are more motivated to
1992). Extraverted individuals are known to have a preference maintain existing relationships with others. Taken together, these
for social interactions and to be more comfortable and skilled at insights suggest that conscientious individuals are persistent in
handling social interactions than their introverted counterparts various aspects of their lives, and tend to focus on maintaining
(Asendorpf and Wilpers, 1998; Akert and Panter, 1988; Ozer and established relationships rather than establishing new relation-
Benet-Martinez, 2006; Riggio, 1986; White et al., 2004). Empirical ships. Therefore, we expect conscientious individuals to focus on
research has documented consistent positive associations between existing relationships, and to have a smaller proportion of new
extraversion and various aspects of social relationships. Compared contacts.
to introverts, extraverts are more likely to engage in proactive
H3c. Conscientiousness will be negatively related to the propor-
networking behaviors (Forret and Dougherty, 2001; Wanberg and
tion of new contacts.
Kammeyer-Mueller, 2000), are faster at developing relationships
in new environments (Asendorpf and Wilpers, 1998), and tend to Openness. Openness to experience (or Openness) describes
have larger social networks (Jensen-Campbell et al., 2002; Russell individuals who are curious, flexible, receptive to new ideas, and
et al., 1997; Selfhout et al., 2010), closer relationships (Russell et al., motivated to seek novelty and explore new environments (Costa
1997; Wu et al., 2008), and a wider range of contacts (Doeven- and McCrae, 1992; Woo et al., 2013). Openness may facilitate
Eggens et al., 2008). Therefore, we hypothesize that extraversion is the development of new relationships because people who are
positively related to all four network variables, namely, emotional curious and open-minded have an interest in getting to know
closeness, network size, upper reachability, and proportion of new others and seek out interactions with new people (McCrae, 1996).
contacts. On the other hand, focusing on exploring new relationships,
open individuals may be less motivated and have less time to
H3a. Extraversion will be positively associated with emotional
develop close bonds or maintain stable relationships (Wu et al.,
closeness, network size, upper reachability and proportion of new
2008). Further, some attributes displayed by highly open indi-
contacts.
viduals such as unconventional ideas and lack of commitment
Agreeableness. Agreeableness refers to a tendency to exhibit val- to relationships (McCrae, 1996) may be considered undesirable,
ued interpersonal characteristics such as altruism, modesty, trust, and consequently hinder the development of close relationships.
cooperation, empathy and tender concern for others (Digman, Empirically, Selfhout et al. (2010) failed to find any significant
1990; Graziano et al., 1996; Graziano and Eisenberg, 1997). Agree- association between openness and network size. Jensen-Campbell
able individuals are motivated to maintain social relations, and et al. (2002) reported a negative association between openness
to regulate negative affect in interpersonal settings (Ahadi and and the number of close ties. Klein et al. (2004) found that open
Rothbart, 1994). These attributes contribute to interpersonal cohe- individuals were more likely to be included in adversary networks.
sion and harmony, and increase acceptance by others. Indeed, Taken together, the existing theories and findings seem to suggest
X. Zhu et al. / Social Networks 35 (2013) 382–393 387

a tradeoff between developing new relationships and deepening opinion of one’s life (respectively) and to experience positive and
existing relationships, and that open people may therefore have a negative affect (respectively), ceteris paribus (Diener et al., 1999;
greater proportion of new contacts in their networks but less close Henderson, 1981; Lutz and Lakey, 2001; Russell et al., 1997; Steel
relationships with their contacts. et al., 2008). Therefore, to be consistent with the current literature,
we include direct paths from extraversion and neuroticism to PSS
H3d. Openness will be negatively related to emotional closeness, and SWB in our theoretical model. Existing theories and empiri-
but positively related to the proportion of new contacts. cal studies are less clear on whether openness, consciousness and
Neuroticism. Neuroticism describes the individual tendency to agreeableness are directly related to PSS or SWB. We will assess
experience substantial and frequent mood swings, to exhibit poor if these personality variables have a direct relationship with PSS
emotional control, and to display negative emotions such as anger, and SWB by evaluating alternative models that include direct links
hostility, impulsiveness, and irritability (Costa and McCrae, 1992). from openness, consciousness and agreeableness to PSS and SWB.
Theories and empirical evidence highlight the disadvantage asso- In summary, we hypothesize a path model that contains six
ciated with high levels of neuroticism in social situations. Neurotic endogenous variables, including four social network characteris-
people are more sensitive to the negative aspects of social relation- tics, one indicator for PSS and one indicator for SWB. In this model,
ships, fearful of rejection, and less likely to initiate relationships PSS has a direct association with SWB, and social network char-
(Turban and Dougherty, 1994) or engage in interpersonal facilita- acteristics are indirectly linked to SWB via the mediation of PSS.
tion (Van Scotter and Motowidlo, 1996; Hurtz and Donovan, 2000). Together, social network characteristics and PSS partially mediate
They are usually less skilled at handling interpersonal differences, the pathway between personality and SWB.
and more likely to adopt an avoidance conflict management strat-
egy characterized by a reluctance to communicate their needs to 4. Methods
others (Antonioni, 1998). Additionally, inclined to see events and
situations in a negative light, they tend to express negative feelings 4.1. Sample and data
toward others, and make vengeful attempts to redress perceived
offenses (Anderson et al., 2001; Barelds, 2005; Donnellan et al., Data were collected with three 20-min long online surveys
2005; McCullough et al., 2001; Robins et al., 2002). In short, neu- administered to first-year students in a public university at three
rotic individuals are less comfortable in social situations and have time points. For each of the three surveys, participants were offered
more problems developing positive relationships than their more the opportunity to enter their name into a drawing. The drawing
emotionally stable counterparts. These insights suggest that neu- randomly selected 30% of the participants to receive a $10 gift card,
roticism may have a negative association with network size and and one participant to receive a $100 gift card.
emotional closeness. Existing empirical studies support the posi- Upon arrival on campus, all freshman students enrolled in the
tive association between neuroticism and poor relationship quality fall of 2010 were invited via email to fill out the first online survey.
(Klein et al., 2004; Russell et al., 1997; Wu et al., 2008), but have They were informed that they would receive subsequent invita-
found little evidence linking neuroticism to network size (Jensen- tions to complete a battery of survey questionnaires again in the
Campbell et al., 2002; Roberts et al., 2008; Sarason et al., 1983; middle of the semester, and at the semester’s end. The first survey
Selfhout et al., 2010). Beyond relationship quality and network included questions about socio-demographic background, prior
size, there is little theory or empirical evidence linking neuroti- ties to the university, and personality. An email reminder was sent
cism to other network characteristics. Based on the above theories out a week later. In total, 1129 students filled out the first survey.
and empirical findings, we expect neurotic people to have smaller Mid-semester, a second email invitation was sent to participants
networks, and to have more distant relationships with their con- who had completed the first survey, followed by a reminder a week
tacts. later. The second survey included questions on structural charac-
teristics of participants’ social networks. Four hundred and twelve
H3e. Neuroticism will be negatively related to emotional close- students participated in the second survey. Finally, a third email
ness and network size. invitation was sent at the end of the semester to participants who
completed the first survey, again followed by a reminder a week
3.5. Direct links from personality to PSS and SWB later. The third survey contained questions on perceived social
support and life satisfaction. Five hundred seventy five students
In addition to the pathways from personality traits to social participated in the third survey.
network characteristics, extraversion and neuroticism may also In total, 309 students completed all three surveys. As presented
be directly related to PSS and SWB, because they describe indi- in Table 1, 58% of the respondents are female, with an average age of
viduals’ predispositions to take on a more positive and negative 18.6. In terms of racial/ethnic background, 78% of the respondents

Table 1
Descriptive statistics and correlations.

Variables Mean SD 1 2 3 4 5 6 7 8 9 10 11

Extraversion 3.24 0.75 (0.88)


Agreeableness 3.79 0.58 0.13* (0.79)
Conscientiousness 3.76 0.55 0.34* 0.28* (0.82)
Openness 3.58 0.51 0.16* 0.15* 0.19* (0.76)
Neuroticism 2.92 0.63 −0.27** −0.33** −0.25** −0.07 (0.81)
Network size 9.26 5.57 0.20** 0.19** 0.05 0.00 −0.07 –
Emotional closeness 0.54 0.29 0.04 0.11* 0.08 0.08 0.01 −0.14* –
Upper-reachability 1.51 1.08 0.15* 0.04 0.04 0.00 −0.04 0.31** −0.20** –
New contacts 0.44 0.34 0.14* −0.04 0.02 0.14* −0.13* −0.02 −0.32** 0.40** –
PSS 3.63 0.74 0.25** 0.17* 0.23** 0.06 −0.18* 0.22** 0.05 0.12* 0.26** (0.88)
SWB 3.39 0.83 0.23** 0.20** 0.21** 0.03 −0.26** 0.19** 0.05 0.05 0.09 0.49** (0.84)

Note: N = 309.
+
p < 0.10, *p < 0.05, and **p < 0.01.
388 X. Zhu et al. / Social Networks 35 (2013) 382–393

are white, 16% are Asian, 2% are African American, 2% are Hispanic, 4 (0 = having no contact affiliated with the university; 1 = having
and 2% belong to the non-specified, “other” category. at least one fellow freshmen student in the network; 2 = having at
least one senior undergraduate student in the network; 3 = having
4.2. Measures at least one graduate student in the network; 4 = having at least one
university staff/faculty in the network). Finally, proportion of new
4.2.1. Subjective wellbeing (SWB) contacts was measured by dividing the number of newly developed
Self-reported life satisfaction was used as the SWB outcome contacts by network size. On average, 44% of the contacts named
variable, which was measured by the 5-item Satisfaction with were developed after the students arrived on campus.
Life Scale developed by Diener et al. (1985). On a 5-point
scale (1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 4.2.4. Personality variables
5 = Strongly Agree), participants rated the extent to which they Big Five personality traits (i.e., Extraversion, Agreeableness,
agreed or disagreed with each of the five statements about their Conscientiousness, Openness, and Neuroticism) were measured
life. Sample items included “In most ways my life is close to my with the forty-four-item Big Five Inventory (BFI) from John
ideal.” “I am satisfied with my life.” “If I could live my life over, I and Srivastava (1999). On a 5-point scale (1 = Strongly Disagree,
would change almost nothing.” The scale has acceptable reliability 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree), participants
(Cronbach’s Alpha = 0.84). indicated the extent to which each of the items was an accurate
description of themselves. The Cronbach’s Alphas for extraversion,
4.2.2. Perceived social support (PSS) agreeableness, conscientiousness, openness, and neuroticism are
PSS was assessed with the short version of the modified 0.88, 0.79, 0.82, 0.76, and 0.81, respectively.
Interpersonal Support Evaluation List (modified ISEL, Cohen and
Hoberman, 1983) adapted to the college context. The ISEL scale 4.3. Analytical strategy
provides a global measure of perceived availability of social
support across all social support dimensions (emotional, informa- To examine the pathways from personality to social networks,
tional, and instrumental). On a 5-point scale (1 = Strongly Disagree, PSS and SWB, we relied upon a structural equation modeling (SEM)
2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree), participants framework, which allows us to simultaneously evaluate all paths by
rated the extent to which they agreed or disagreed with each of 12 fitting a series of equations representing the different paths. All SEM
items about social support. Sample items included “If I were sick, analyses were conducted using Mplus version 6, with maximum
I could easily find someone to help me with my daily chores.” “I likelihood estimate procedure (Muthén and Muthén, 1998–2010).
don’t often get invited to do things with others.” “When I need sug- Mplus reports a number of model fit indices to evaluate how well
gestions on how to deal with a personal problem, I know someone the model fits the data: Comparative Fit Index (CFI), Tucker–Lewis
I can turn to.” The scale has acceptable reliability in this sample index (TLI), Root Mean Square Error of Approximation (RMSEA),
(Cronbach’s Alpha = 0.88). and Standardized Root Mean Square Residual (SRMR). A model is
considered acceptable when CFI and TLI are greater than 0.90 (Hair
4.2.3. Name generator and social network variables et al., 2006), RMSEA is smaller than 0.05, and SRMR is smaller than
The present study focuses on the egocentric network of fresh- 0.08 (Hu and Bentler, 1999). We accounted for measurement errors
man students. Name generator questions were used to elicit the by fixing the residual variance of the personality variables, PSS and
names or initials of up to twenty network contacts. Respondents SWB to (1 − reliability) × sample variance (Bollen, 1989; Muthén,
then assessed their relationship with each contact, and provided 2005).
demographic information for each contact. The notion of functional
specificity suggests that a particular kind of relationship may be 5. Results
effective for one kind of task or problem, but not another (Perry
and Pescosolido, 2010; Sandefur and Laumann, 1998). To assess the 5.1. Correlational analysis
effect of social networks on PSS and SWB, we focus on network ties
to people who had provided participants important social support Table 1 provides the means, standard deviations, scale reliabili-
(e.g., contacts that participants socialize with or go to for help with ties (in parentheses), and correlations of all variables included in the
personal issues), because the skills, information, and resources of analysis. As expected, PSS was significantly correlated with SWB.
these network contacts would be more suitable in the context of PSS had significant positive bivariate correlations with three out of
social support and wellbeing. four network structure variables – network size, upper reachability
Network size was measured with the number of social sup- and proportion of new contacts, but not with emotional closeness.
port contacts named by a respondent. On average, respondents SWB was positively correlated with network size, but not with the
named nine contacts. Percentage measures are commonly used other three network variables.
to describe important network characteristics such as emotional A closer look at the intercorrelations points to the possibility
closeness (e.g., Acock and Hurlbert, 1993; Hays and Oxley, 1986; of correlative suppression effects among the network variables in
Perry and Pescosolido, 2010; Stokes, 1983). In the present study, their correlations with PSS and SWB (Cohen et al., 2003). Notably,
respondents were asked to rate how closely they felt toward emotional closeness was negatively associated with network size
every contact on a 5-point scale (5 = very close, 4 = close, 3 = less (−0.14, p < 0.05), upper reachability (−0.20, p < 0.05), and propor-
than close, 2 = distant, 1 = very distant). The proportion of contacts tion of new contacts (−0.32, p < 0.05). These negative correlations
toward whom the respondents felt very close (i.e., the number make sense theoretically, given that it is harder to maintain close
of very close contacts divided by network size) was computed to relationships with a large number of contacts than with a few, that
measure emotional closeness. The mean of emotional closeness is people tend to develop closer relationships with those who are
0.54, suggesting that, on average, participants had very close rela- more similar in status and life situations, and that it takes time to
tionships with half of their social support contacts. Respondents develop close relationships when one enters a new environment.
were also asked to indicate whether a contact was a fellow fresh- The hypothesized positive relationships of emotional closeness
man, a more senior undergraduate student, a graduate student, or with PSS and SWB could have been suppressed in bivariate situa-
a staff or faculty member. This information was then used to con- tions, and can be expected to surface once the suppressing effect of
struct the measure for upper reachability, which ranged from 0 to network size, upper reachability and proportion of new contacts are
X. Zhu et al. / Social Networks 35 (2013) 382–393 389

Table 2
Nested model comparison (N = 309).

Model description 2 (df) Model 2 (df) CFI TLI RMSEA SRMR
compared with

1 Hypothesized model 25.691 (19) Model 3 28.2 (4), p = 0.000 0.989 0.963 0.027 0.020
2 Model 1 + direct paths from social network 21.599 (15) Model 1 4.1 (4), p = 0.393 0.988 0.956 0.032 0.027
variables to SWB
3 Network variables and PSS as full mediators 53.888 (23) 0.932 0.890 0.066 0.057
between personality and SWB; PSS as full
mediator between network variables and SWB
4 Model 2 + direct paths from openness, 16.746 (16) Model 1 8.9 (3), p = 0.031 0.998 0.967 0.012 0.010
conscientiousness, agreeableness to PSS
5 Model 4 + direct paths from openness, 15.216 (13) Model 1 10.5 (6), p = 0.106 0.996 0.954 0.022 0.019
conscientiousness, agreeableness to SWB

Table 3
Path coefficients for the hypothesized model on personality, social networks, PSS, and SWB as shown in Fig. 1.

Independent variables Dependent variables

Emotional closeness Network size Upper reachability Proportion of new contacts PSS SWB

1. Extraversion 0.026 0.194** 0.155** 0.156* 0.161* 0.058


2. Agreeableness 0.113 0.176** – – – –
3. Conscientiousness – – – −0.076 – –
4. Openness 0.055 – – 0.143* – –
5. Neuroticism −0.039 −0.035 – – −0.114+ −0.197**
6. Emotional closeness – – – – 0.151** –
7. Network size – – – – 0.240** –
8. Upper reachability – – – – −0.075 –
9. Proportion of new contacts – – – – 0.306** –
10. PSS – – – – – 0.516**

R2 0.015 0.081* 0.024 0.051* 0.200** 0.373**


CFI 0.989
TLI 0.963
RMSEA 0.027
SRMR 0.020

Note: N = 309. CFI, comparative fit index; TLI, Tucker–Lewis index; RMSEA, root mean square error of approximation; SRMR, standardized root mean square residual.
+
p < 0.10, *p < 0.05, and **p < 0.01.

controlled for. Similarly, the hypothesized positive effects of upper test of the two models showed that including the direct paths
reachability and proportion of new contacts are also expected to from social network variables to SWB did not improve model fit
become visible when emotional closeness is taken into account. (2 /df = 4.1/4, p = 0.393). Therefore, we retained the more par-
The pattern of correlations between personality variables and simonious Model 1 in which PSS serves as a full mediator between
network variables was largely consistent with our expectations social network variables and SWB.
with a few exceptions. Extraversion was positively correlated with We also evaluated a few neighboring alternatives.1 Model 3 in
all network characteristics except for emotional closeness. Agree- Table 2 specifies social network variables and PSS as full mediators
ableness was positively correlated with network size, but not with between personality variables and SWB. As the Chi-square differ-
emotional closeness. Conscientiousness was not correlated with ence test suggested, Model 1, the hypothesized model with direct
the proportion of new contacts. Openness was positively corre- paths from neuroticism and extraversion to PSS and SWB fitted
lated with the proportion of new contacts. Neuroticism was not the data significantly better than Model 3. The direct paths from
correlated with emotional closeness or with network size. Finally, neuroticism to PSS and SWB, and from extraversion to PSS were
all personality variables were significantly correlated with PSS and statistically significant. Model 4 in Table 2 further includes direct
SWB in the expected direction with one exception: Openness was paths from agreeableness, openness and conscientiousness to PSS,
not significantly correlated with PSS or with SWB. and Model 5 includes direct paths from these variables to both PSS
and SWB. The Chi-square difference test suggested that Model 4
5.2. Path model fitted the data slightly better than Model 1. A closer look at the
results revealed a significant direct path from conscientiousness
To examine the relationships among personality, social to PSS (0.178, p = 0.014). The direct paths from agreeableness and
networks, PSS and SWB, we first evaluated two variations of the openness to PSS, however, were not significant. Model 5, however,
hypothesized model. One model represents our theoretical model did not significantly improve model fit, and none of the direct paths
as shown in Fig. 1, which specifies PSS as a full mediator between from agreeableness, openness and conscientiousness to SWB were
social network variables and SWB, and includes direct links from significant.
extraversion and neuroticism to PSS and SWB as control variables Table 3 presents model fit indices and standardized coefficients
(Model 1 in Table 2). The alternative model (Model 2 in Table 2) for all hypothesized paths as shown in Fig. 1 (see Model 1 in
includes both direct and indirect paths from social network vari- Table 2). The fit indices for our hypothesized model suggest that the
ables to SWB. Fit indices suggested that both models fit the data model fitted the data well (CFI = 0.989; TLI = 0.963; RMSEA = 0.027;
well, with both CFI and TLI greater than 0.95, and RMSEA and SRMR
smaller than 0.05 for both models (Hair et al., 2006; Hu and Bentler,
1999). However, none of the direct paths from social network char-
acteristics to SWB were significant, and the Chi-square difference 1
Details on the models are available from the authors upon request.
390 X. Zhu et al. / Social Networks 35 (2013) 382–393

Extraversion Closeness

0.161*
0.151*
Agreeableness 0.194**
Size

0.176**
0.240* * 0.516**
Conscientiousness Perceived Subjective
social wellbeing
support
0.155*
Upper
reachability
Openness

-0.114+ -0.197**
0.156* 0.306**
Proportion of
Neuroticism new contacts
0.143*

Fig. 2. Significant standardized path coefficients, significant at + p < 0.10, *p < 0.05, and **p < 0.01. Hypothesized paths that were significant are represented by solid bold
arrows. Hypothesized paths that were not significant are represented by dashed arrows. Control paths that were significant are represented by solid light arrows.

SRMR = 0.020; Hair et al., 2006; Hu and Bentler, 1999). For a clearer size, and openness was not related to PSS except via its positive
view of the results, Fig. 2 presents standardized path coefficients relationship with the proportion of new contacts. Extraversion
for all hypothesized significant paths along with paths that were was linked to PSS via three network paths (network size, upper
controlled for. In support of H1, PSS had a significant positive reachability, and proportion of new contacts) as well as directly.
relationship (0.516, p < 0.01) with SWB. The paths from emotional It seems reasonable to infer that both objective and subjective
closeness (0.151, p < 0.01), network size (0.240, p < 0.01), and pro- aspects of social relationships (i.e., social networks and PSS, respec-
portion of newly developed network contacts (0.306, p < 0.01) to tively) constitute an important path linking extraversion to SWB.
PSS were all positive and significant, supporting H2a, H2b, and Social network variables played no role in linking conscientious-
H2d. Upper reachability, however, was not significantly related ness to PSS and SWB, as conscientiousness had only a direct path
to PSS. Thus H2c was not supported. Largely consistent with H3a, to PSS. Finally, the negative relationship between neuroticism and
extraversion was directly related to network size (0.194, p < 0.01), SWB was not mediated by any social network variable, and only
upper reachability (0.155, p < 0.05), and proportion of new contacts marginally mediated by PSS.
(0.156, p < 0.05). Extraversion also had a significant direct path to
PSS (0.161, p < 0.05). The direct path from extraversion to SWB, 6. Discussion and conclusion
however, was not significant. Inconsistent with H3a, extraver-
sion was not related to emotional closeness. In partial support of The findings reported in this study contribute to the current
H3b, agreeableness was positively linked to network size (0.176, social network theory in two important ways. First, the results
p < 0.01), but was not related to emotional closeness. Openness was corroborate the often posited, but rarely tested, psychological
positively related to proportion of new contacts (0.143, p < 0.05), mechanisms underlying the beneficial effects of social networks
but not with emotional closeness, partially supporting H3d. Con- on individual wellbeing (Acock and Hurlbert, 1993; Lin et al., 1999;
scientiousness and neuroticism were not significantly related to Perry and Pescosolido, 2010; Thoits, 1995). The results are consis-
any network variables, which was inconsistent with H3c and H3e. tent with the proposition that social networks may influence SWB
Consistent with our expectations, however, neuroticism had a sig- by promoting positive perceptions of social support. One impor-
nificant direct negative relationship with SWB (−0.197, p < 0.01), tant question for future research is whether different types of social
and a marginally significant direct link to PSS (−0.114, p < 0.1). support mediate relationships between social network character-
A number of results are noteworthy. First, PSS fully mediated istics and SWB. In addition, by addressing how social networks are
the positive associations that emotional closeness, network size related to SWB, the present study opens up new avenues for investi-
and proportion of new contacts had with SWB. Second, results gating the question “When do social networks matter?” Given that
regarding the mediating role of social network characteristics var- PSS mediates the relationship between social networks and SWB,
ied. Network size mediated the relationships that extraversion and we can infer that factors that affect the correspondence between
agreeableness had with PSS, and proportion of new contacts medi- social network characteristics and PSS can potentially influence the
ated the relationships between extraversion and PSS, and between strength of association that social networks have with SWB.
openness and PSS. Although positively related to PSS, emotional Second, the study shows that social networks and PSS together
closeness did not mediate relationships between any personality constitute important pathways through which some personality
variable and PSS. Likewise, upper reachability did not mediate the characteristics are linked to SWB, suggesting that individual dif-
relationships between any personality variable and PSS or SWB, as ferences play important roles in shaping the social environment
it was not associated with either variable. that subsequently promotes or impedes positive life experience.
Third, results regarding the paths linking personality variables Although it would require a true longitudinal study to better under-
to SWB varied a great deal. Social network variables appeared to be stand the causality and interplay of these factors, the research
the primary path linking agreeableness and openness to PSS and validates the need to consider the role of individuals’ personality
SWB, as none of the direct paths from agreeableness and openness traits in a network account of individual wellbeing and other out-
to PSS and SWB were significant. More specifically, agreeableness comes (Mehra et al., 2001; Robins and Kashima, 2008; Totterdell
was related to PSS, and subsequently to SWB, largely via network et al., 2008). Taken together, the study suggests a more complete
X. Zhu et al. / Social Networks 35 (2013) 382–393 391

story of individual experience in which relatively stable individual conscientious people would be more persistent and more moti-
characteristics shape the social structure that consequently influ- vated to maintain existing relationships. However, conscientious
ences subjective perceptions of social resources, the more proximal people are also more oriented toward achievement (Costa and
cause of SWB. McCrae, 1992), which might motivate them to actively develop
While we focused on social support networks and SWB, it new relationships on campus so as to achieve better performance
is also important to investigate the possibility that instrumental (Hough, 1992). As such, the net association between conscien-
networks (such as advice networks) may mediate the association tiousness and proportion of newly developed network contacts
between personality and instrumental outcomes such as career may well depend on how the two opposing forces play out against
advancement. Mehra et al. (2001) pinpointed two theoretical issues each other. Future research is needed to probe these possibilities.
relevant to personality, “. . .the network position occupied by indi- We note a few limitations with the present study. First, although
viduals might be influenced by their psychology. . . personality and the study collected data at three time points, with the assess-
social network position might combine to influence important out- ment of personality variables, egocentric network data, PSS and
comes such as work performance.” The present study focused on SWB arranged sequentially in time, the study is not truly longi-
the first one, that is, the role of personality as predictors of social tudinal and therefore does not lend itself to causal inferences. Our
networks. Future research might examine the moderating role of findings can only be considered as suggestive of the causal relation-
personality on the relationships that social networks have with PSS ships between the variables. Further, we are aware of the potential
and SWB. Research suggests that personality may combine with threat of unobserved contemporaneous variables that may cause
social networks to influence instrumental outcomes (Anderson, personality and social relationships to covary. For example, cer-
2008; Mehra et al., 2001). We have little theory or empirical tain environmental aspects about the university may have lead to
research on whether and which personality variables interact with the covariation between personality and social relationships. Even
which network characteristics in shaping wellbeing-related out- with this limitation noted, this study provides evidence of the role
comes. Our earlier discussion suggests one possibility: personality individual differences play in the development of social networks,
attributes that capture the level of sensitivity to the external envi- and highlights the need to take into consideration both personality
ronment and social relationships may influence how individuals and social structure in shaping individual life experience.
respond to their social environment and subjectively assess support Second, given the data collection methods, we were not able to
availability. Consequently, such personality attributes may mod- compare the respondents and nonrespondents, or to determine the
erate the strength and nature of social network effects on SWB. extent of nonresponse bias. Studies have suggested that no matter
Continued research along this direction can greatly advance our how small the nonresponse rate may be, bias is of concern for the
knowledge of social networks and wellbeing. researcher (Sheikh and Mattingly, 1981). However, our study was
A few hypotheses were not supported in our study. Notably, conducted on a fairly homogenous group of individuals, namely,
emotional closeness did not mediate the association between per- freshmen college students.
sonality and PSS, as none of the five personality facets predicted Finally, as we have pointed out earlier, we focused on positive
emotional closeness, which is inconsistent with existing research social relationships only. Although there are various other types of
evidence (Russell et al., 1997; Wu et al., 2008). Neuroticism had a social networks that may be of interest for researchers (e.g., adver-
direct path to SWB that was mediated by neither PSS nor any of sarial network), we chose to focus on positive social relationships
the social network variables. Upper reachability did not mediate because our hypotheses were concerned with the effects this type
the relationship between personality and PSS or SWB, as it was not of relationship when transitioning into a new environment. As such,
associated with PSS or SWB. Finally, conscientious was not related the results should not be generalized beyond that relationship and
to the proportion of newly developed network contacts. personality aspects that were assessed in the present study.
One possible explanation of these insignificant results has to In conclusion, the present study brings together research from
do with the fact that the network data were collected when the different academic disciplines to develop an integrated theoretical
participants were just a little over two months into their first year model that traces the pathway from personality to social networks,
in college, which may not have provided sufficient time to fully to PSS and SWB. Results support PSS as the mediating mechanism
develop some social network characteristics. In the case of emo- underlying the beneficial effects of social networks on SWB, and
tional closeness, respondents reported contacts that fall on the highlight the potential role of social and personality psychology in
“closer” end of the continuum, as the average parentage of very shaping the development of social networks.
close contacts is 0.54. This is probably due to the fact that we
focused on social support network contacts, which tend to draw
from the closer end of social relationships. The variability in emo- References
tional closeness in this study was therefore less than ideal, which
may partially explain the insignificant results. Acock, A.C., Hurlbert, J.S., 1993. Social networks, marital-status, and wellbeing.
Social Networks 15, 309–334.
The intercorrelations of the personality variables may be Ahadi, S.A., Rothbart, M.K., 1994. Temperament, development, and the Big Five. In:
another factor. For example, neuroticism may be linked to the posi- Halverson Jr., C.F., Kohnstamm, G.A., Martin, R.P. (Eds.), The Developing Structure
tive network variables primarily because of its negative association of Temperament and Personality from Infancy to Adulthood. Erlbaum, Hillsdale,
NJ, pp. 189–207.
with extraversion (r = −0.27, p < 0.05) and agreeableness (r = −0.33, Akert, R., Panter, A., 1988. Extraversion and the ability to decode nonverbal com-
p < 0.05). Once the effects of extraversion and agreeableness were munication. Personality and Individual Differences 9, 965–972.
controlled for, neuroticism did not have any unique association Anderson, C., John, O.P., Keltner, D., Kring, A.M., 2001. Who attains social status?
Effects of personality and physical attractiveness in social groups. Journal of
with network variables.
Personality and Social Psychology 81, 116–132.
The relationships between some personality facets and network Anderson, M.H., 2008. Social networks and the cognitive motivation to realize
characteristics may require more nuanced theoretical account. network opportunities: a study of managers’ information gathering behaviors.
Journal of Organizational Behavior 29, 51–78.
For example, as pointed out earlier, neurotic people are more
Antonioni, D., 1998. Relationship between the Big Five personality factors and
sensitive to the negative aspects of social relationships than the conflict management styles. International Journal of Conflict Management 9,
positive aspects (Turban and Dougherty, 1994), which was not 336–355.
fully considered in our study. In addition, we hypothesized a Asendorpf, J.B., Wilpers, S., 1998. Personality effects on social relationships. Journal
of Personality and Social Psychology 74, 1531–1544.
negative association between conscientiousness and the propor- Ashford, S.J., Black, J.S., 1996. Proactivity during organizational entry: the role of
tion of newly developed network contacts on the ground that desire for control. Journal of Applied Psychology 81, 199–214.
392 X. Zhu et al. / Social Networks 35 (2013) 382–393

Barelds, D.P.H., 2005. Self and partner personality in intimate relationships. Euro- Graziano, W.G., Jensen-Campbell, L.A., Hair, E.C., 1996. Perceiving interpersonal con-
pean Journal of Personality 19, 501–518. flict and reacting to it: the case for agreeableness. Journal of Personality and
Bolger, N., Eckenrode, J., 1991. Social relationships, personality, and anxiety during a Social Psychology 70, 820–835.
major stressful event. Journal of Personality and Social Psychology 61, 440–449. Graziano, W.G., Eisenberg, N., 1997. Agreeableness: a dimension of personality.
Bolger, N., Zuckerman, A., Kessler, R.C., 2000. Invisible support and adjustment to In: Briggs, S., Hogan, R., Jones, W. (Eds.), Handbook of Personality Psychology.
stress. Journal of Personality and Social Psychology 79, 953–961. Academic Press, San Diego, pp. 795–824.
Bollen, K.A., 1989. Structural Equations with Latent Variables. John Wiley & Sons, Hair, J.F., Black, W., Babin, B., Anderson, R.E., Tatham, R., 2006. Multivariate Data
Inc., New York. Analysis, 6th ed. Prentice Hall, Upper Saddle River, NJ.
Bowling, N.A., Beehr, T.A., Swader, W.M., 2005. Giving and receiving social support Hays, R.B., Oxley, D., 1986. Social network development and functioning during a
at work: the roles of personality and reciprocity. Journal of Vocational Behavior life transition. Journal of Personality and Social Psychology 50, 305–313.
67, 476–489. Henderson, S., 1977. The social network, support and neurosis: the function of
Brissette, I., Scheier, M.F., Carver, C.S., 2002. The role of optimism in social network attachment in adult life. British Journal of Psychiatry 131, 185–191.
development, coping, and psychological adjustment during a life transition. Henderson, S., 1981. Neurosis and the Social Environment. Academic Press, New
Journal of Personality and Social Psychology 82, 102–111. York.
Burt, R.S., 1984. Network items and the general social survey. Connections 8, Horenstein, J., Cohen, S., 2009. Social support. In: Ingram, R.E. (Ed.), The Interna-
119–122. tional Encyclopedia of Depression. Springer Publishing Company, New York, pp.
Burt, R.S., 1987. A note on strangers, friends, and happiness. Social Networks 9, 514–517.
311–331. Hough, L.M., 1992. The “Big Five” personality variables-construct confusion:
Burt, R.S., 2000. The network structure of social capital. In: Staw, B.M., Sutton, R.I. description versus prediction. Human Performance 5, 139–156.
(Eds.), Research in Organizational Behavior, vol. 22. JAI Press, Greenwich, CT. House, J.S., Landis, K.R., Umberson, D., 1988. Social relationships and health. Science
Burt, R.S., Jannotta, J.E., Mahoney, J.T., 1998. Personality correlates of structural 241, 540–545.
holes. Social Networks 20, 63–87. House, J.S., Kahn, R.L., 1985. Measures and concepts of social support. In: Cohen,
Campell, K.E., Lee, B.A., 1991. Name generators in surveys of personal networks. S., Syme, S.L. (Eds.), Social Support and Health. Academic Press, New York, pp.
Social Networks 13, 203–221. 83–108.
Campbell, K.E., Marsden, P.V., Hurlbert, J.S., 1986. Social resources and socioeco- Hu, L.-T., Bentler, P.M., 1999. Cutoff criteria for fit indexes in covariance struc-
nomic status. Social Networks 8, 97–117. ture analysis: conventional criteria versus new alternatives. Structural Equation
Cassel, J.C., 1976. The contribution of the social environment to host resistance. Modeling 6, 1–55.
American Journal of Epidemiology 104, 107–123. Hurtz, G.M., Donovan, J.J., 2000. Personality and job performance: the big five revis-
Carver, C.S., Connor-Smith, J., 2010. Personality and coping. Annual Review of Psy- ited. Journal of Applied Psychology 85, 869–879.
chology 61, 679–704. Jensen-Campbell, L.A., Adams, R., Perry, D.G., Workman, K.A., Furdella, J.Q., Egan,
Cohen, S., 1988. Psychosocial models of social support in the etiology of physical S.K., 2002. Agreeableness, extraversion, and peer relations in early adolescence:
disease. Health Psychology 7, 269–297. winning friends and deflecting aggression. Journal of Research in Personality 36,
Cohen, J., Cohen, P., West, S.G., Aiken, L.S., 2003. Applied multiple regres- 224–251.
sion/correlation analysis for the behavioral sciences, 3rd ed. Erlbaum, Mahwah, John, O.P., Srivastava, S., 1999. The big-five trait taxonomy: history, measurement
NJ. and theoretical perspectives. In: Pervin, L., John, O.P. (Eds.), Handbook of Per-
Cohen, S., Hoberman, H., 1983. Positive events and social supports as buffers of life sonality: Theory and Research. , 2nd ed. Guilford Press, New York, pp. 102–138.
change stress. Journal of Applied Social Psychology 13, 99–125. Judge, T.A., Higgins, C.A., Thoresen, C.J., Barrick, M.R., 1999. The Big Five personality
Cohen, S., Wills, T.A., 1985. Stress, social support, and the buffering hypothesis. traits, general mental ability, and career success across the life span. Personnel
Psychological Bulletin 98, 310–357. Psychology 52, 621–652.
Cohen, S., 2004. Social relationships and health. American Psychologist 59, 676–684. Kalish, Y., Robins, G., 2006. Psychological predispositions and network structure: the
Cohen, S., Janicki-Deverts, D., 2009. Can we improve our physical health by altering relationship between individual predispositions, structural holes and network
our social networks? Perspectives on Psychological Science 4, 375–378. closure. Social Networks 28, 56–84.
Connell, C.M., D’Augelli, A.R., 1990. The contribution of personality characteristics to Klein, K.J., Lim, B.C., Saltz, J.L., Mayer, D.M., 2004. How do they get there? An examina-
the relationship between social support and perceived physical health. Health tion of the antecedents of centrality in team networks. Academy of Management
Psychology 9, 192–206. Journal 47, 952–963.
Costa Jr., P.T., McCrae, R.R., 1992. Revised NEO personality inventory NEO PI-R Lambert, T.A., Eby, L.T., Reeves, M.P., 2006. Predictors of networking intensity and
and NEO five five-factor inventory NEO-FFI: professional manual. Psychological network quality among white collar job seekers. Journal of Career Development
Assessment Resources, Odessa, FL. 32, 351–365.
Cutrona, C.E., 1986. Objective determinants of perceived social support. Journal of Lara, M.E., Leader, J., Klein, D.N., 1997. The association between social support and
Personality and Social Psychology 50, 349–355. course of depression: is it confounded with personality? Journal of Abnormal
DeNeve, K.M., Cooper, H., 1998. The happy personality: a meta-analysis of 137 Psychology 106, 478–482.
personality traits and subjective wellbeing. Psychological Bulletin 124, 197–229. Lin, N., 2001. Social Capital: A Theory of Social Structure and Action. Cambridge
Dimond, M.F., Lund, D.A., Caserta, M.S., 1987. The role of social support in the first University Press, Cambridge.
two years of bereavement in an elderly sample. Gerontologist 27, 599–604. Lin, N., Ye, X.L., Ensel, W.M., 1999. Social support and depressed mood: a structural
Diener, E., Emmons, R., Larsen, J., Griffin, S., 1985. The satisfaction with life scale. analysis. Journal of Health and Social Behavior 40, 344–359.
Journal of Personality Assessment 49, 71–75. Lutz, C.J., Lakey, B., 2001. How people make support judgments: individual differ-
Digman, J.M., 1990. Personality structure – emergence of the 5-factor model. Annual ences in the traits used to infer supportiveness in others. Journal of Personality
Review of Psychology 45, 417–440. and Social Psychology 81, 1070–1079.
Diener, E., Suh, E.M., Lucas, R.E., Smith, H.L., 1999. Subjective wellbeing: three Marsden, P., 2005. Recent developments in network measurement. In: Carring-
decades of progress. Psychological Bulletin 125, 276–302. ton, P.J., Scott, J., Wasserman, S. (Eds.), Models and Methods in Social Network
Doeven-Eggens, L., Fruyt, F.D., Hendriks, A.A.J., Bosker, R.J., Werf, V.D., 2008. Per- Analysis. Cambridge University Press, Cambridge, UK, pp. 8–30.
sonality and personal network type. Personality and Individual Differences 45, McAdams, D.P., 2009. The Person: An Introduction to the Science of Personality.
689–693. Psychology, 5th ed. Wiley, New York.
Donnellan, M.B., Larsen-Rife, D., Conger, R., 2005. Personality, family history, and McCrae, R.R., 1996. Social consequences of experiential openness. Psychological
competence in early adult romantic relationships. Journal of Personality and Bulletin 120, 323–337.
Social Psychology 88, 562–576. McCrae, R.R., John, O.P., 1992. An introduction to the five-factor model and its
Emirbayer, M., Goodwin, J., 1994. Network analysis, culture, and the problem of applications. Journal of Personality 60, 175–215.
agency. American Journal of Sociology 99, 1411–1454. McCullough, M.E., Bellah, C.G., Kilpatrick, S.D., Johnson, J.L., 2001. Vengefulness:
Finch, J.F., Graziano, W.G., 2001. Predicting depression from temperament, person- relationships with forgiveness, rumination, wellbeing, and the Big Five. Person-
ality, and patterns of social relations. Journal of Personality 69, 27–55. ality and Social Psychology Bulletin 27, 601–610.
Finch, J.F., Okun, M.A., Barrera Jr., M., Zautra, A.J., Reich, J.W., 1989. Positive and Mehra, A., Kilduff, M., Brass, D.J., 2001. The social networks of high and low self moni-
negative social ties among older adults: measurement models and the predic- tors: implications for workplace performance. Administrative Science Quarterly
tion of psychological distress and well-being. American Journal of Community 46, 121–146.
Psychology 17, 585–604. Morrison, E.W., 2002. Newcomers’ relationships: the role of social network ties
Flynn, F.J., Reagans, R.E., Amanatullah, E.T., Ames, D.R., 2006. Helping one’s way to during socialization. Academy of Management Journal 45, 1149–1160.
the top: self monitors achieve status by helping others and knowing who helps Muthén, L.K., Muthén, B.O., 1998. 2010 Mplus User’s Guide, 5th ed. Muthén &
whom. Journal of Personality and Social Psychology 91, 1123–1137. Muthén, Los Angeles, CA.
Forret, M.L., Dougherty, T.W., 2001. Correlates of networking behavior for man- Muthén, B.O., 2005. Correcting for measurement error, April 28, 6:30 pm. Mes-
agerial and professional employees. Group & Organization Management 26, sage posted to http://www.statmodel.com/discussion/messages/11/645.html?
283–311. 1332900000
Goldberg, L.R., 1990. An alternative “description of personality”: the Big-Five factor Norbeck, J.S., Lindsey, A.M., Carrieri, V.L., 1981. The development of an instru-
structure. Journal of Personality and Social Psychology 59, 1216–1229. ment to measure social support. Nursing Research 30 (September–October (5)),
Goldberg, L.R., 1992. The development of markers for the Big-Five factor structure. 264–269.
Psychological Assessment 4, 26–42. Oh, H., Kilduff, M., 2008. The ripple effect of personality on social structure: self
Granovetter, M., 1983. The strength of weak ties: a network theory revisited. Socio- monitoring origins of network brokerage. Journal of Applied Psychology 93,
logical Theory 1, 201–233. 1155–1164.
X. Zhu et al. / Social Networks 35 (2013) 382–393 393

Ostroff, C., Kozlowski, S.W.J., 1992. Organizational socialization as a learning process Stevenson, W.B., Greenberg, D., 2000. Agency and social networks: strategies of
– the role of information acquisition. Personnel Psychology 45, 849–874. action in a social structure of position, opposition, and opportunity. Adminis-
Ozer, D.J., Benet-Martinez, V., 2006. Personality and the prediction of consequential trative Science Quarterly 45, 651–678.
outcomes. Annual Review of Psychology 57, 401–421. Stokes, J.P., 1983. Predicting satisfaction with social support from social network
Pavot, W., Diener, E., Fujita, F., 1990. Extroversion and happiness. Personality and structure. American Journal of Community Psychology 11, 141–152.
Individual Differences 11, 1299–1306. Thoits, P.A., 1983. Multiple identities and psychological well-being: A reformation
Perry, B.L., Pescosolido, B.A., 2010. Functional specificity in discussion networks: the and test of the social isolation hypothesis. American Sociological Review 48,
influence of general and problem specific networks on health outcomes. Social 174–187.
Networks 32, 345–357. Thoits, P.A., 1995. Stress, coping, and social support processes: where are we? What
Pinquart, M., Frohlich, C., 2009. Psychosocial resources and subjective wellbeing of next? Journal of Health and Social Behavior 35, 53–79.
cancer patients. Psychology and Health 24, 407–421. Totterdell, P., Holman, D., Hukin, A., 2008. Social networkers: measuring and
Pinquart, M., Sorensen, S., 2000. Influences of socioeconomic status, social net- examining individual differences in propensity to connect with others. Social
work, and competence on subjective wellbeing in later life: a meta-analysis. Networks 30, 283–296.
Psychology and Aging 15, 187–224. Turban, D.B., Dougherty, T.W., 1994. Role of protege personality in receipt of men-
Reagans, R., McEvily, B., 2003. Network structure and knowledge transfer: the effects toring and career success. Academy of Management Journal 37, 688–702.
of cohesion and range. Administrative Science Quarterly 48, 240–267. Turner, R.J., 1981. Social support as a contingency in psychological well-being.
Riggio, R.E., 1986. Assessment of basic social skills. Journal of Personality and Social Journal of Health and Social Behavior 22, 357–367.
Psychology 51, 649–660. Turner, R.J., Frankel, G., Levin, D.M., 1983. Social support: conceptualization, mea-
Robins, G., Kashima, Y., 2008. Social psychology and social networks: individuals surement and implications for mental health. In: Greenley, J.R., Simmons, R.G.
and social systems. Asian Journal of Social Psychology 11, 1–12. (Eds.), Research in Community and Mental Health. JAI Press, Greenwich, CT, pp.
Robins, R., Caspi, A., Moffitt, T., 2002. It’s not just who you’re with, it’s who you are: 67–111.
personality and relationship experiences across multiple relationships. Journal Van Scotter, J.R., Motowidlo, S.J., 1996. Interpersonal facilitation and job dedication
of Personality 70, 925–964. as separate facets of contextual performance. Journal of Applied Psychology 81,
Roberts, S.G.B, Wilson, R., Fedurek, P., Dunbar, R.I.M., 2008. Individual differences 525–531.
and personal social network size and structure. Personality and Individual Dif- Vaux, A., Harrison, D., 1985. Support network characteristics associated with
ferences 44, 954–964. support satisfaction and perceived support. American Journal of Community
Russell, D.W., Booth, B., Reed, D., Laughlin, P.R., 1997. Personality, social networks, Psychology 13, 245–268.
and perceived social support among alcoholics: a structural equation analysis. Von Dras, D.D., Siegler, I.C., 1997. Stability in extraversion and aspects of social
Journal of Personality 65, 649–692. support at midlife. Journal of Personality and Social Psychology 72, 233–241.
Sandefur, R.L., Laumann, E.O., 1998. A paradigm for social capital. Rationality and Wellman, B., Wortley, S., 1990. Different strokes from different folks: community
Society 10, 481–501. ties and social support. The American Journal of Sociology 96, 558–588.
Sarason, I.G., Levine, H.M., Basham, R.B., Sarason, B.R., 1983. Assessing social Wanberg, C.R., Kammeyer-Mueller, J.D., 2000. Predictors and outcomes of proactiv-
support: the social support questionnaire. Journal of Personality and Social Psy- ity in the socialization process. Journal of Applied Psychology 85, 373–385.
chology 44, 127–139. Ward, R., Sherman, S., LaGory, M., 1984. Subjective network assessments and sub-
Seibert, S.E., Kraimer, M.L., Liden, R.C., 2001. A social capital theory of career success. jective wellbeing. Journal of Gerontology 39, 93–101.
Academy of Management Journal 44, 219–237. Wasserman, S., Faust, K., 1999. Social Network Analysis. Cambridge University Press,
Seeman, T.E., Berkman, L.F., 1988. Structural characteristics of social networks and Cambridge, UK.
their relationship with social support in the elderly: who provides support. White, J.K., Hendrick, S.S., Hendrick, C., 2004. Big Five personality variables and
Social Science and Medicine 26, 737–749. relationship constructs. Personality and Individual Differences 37, 1519–1530.
Selfhout, M., Burk, W., Branhje, S., Denissen, J., van Aken, M., Meeus, W., 2010. Woo, S.E., Chernyshenko, O.S., Longley, A., Zhang, Z., Chiu, C-y., Stark, S.E., 2013.
Emerging late adolescent friendship networks and big five personality traits: a Openness to experience: its lower-level structure, measurement, and cross-
social network approach. Journal of Personality 72, 509–538. cultural equivalence. Journal of Personality Assessment, in press.
Sheikh, K., Mattingly, S., 1981. Investigating non-response bias in mail surveys. Wu, P.-C., Foo, M.-D., Turban, D.B., 2008. The role of personality in relationship close-
Journal of Epidemiology and Community Health 35, 293–296. ness, developer assistance, and career success. Journal of Vocational Behavior 73,
Song, L.J., Lin, N., 2009. Social capital and health inequality: evidence from Taiwan. 440–448.
Journal of Health and Social Behavior 50, 149–163. Zellars, K.L., Perrewé, P.L., 2001. Affective personality and the content of emo-
Steel, P., Schmidt, J., Shultz, J., 2008. Refining the relationship between personality tional social support: coping in organizations. Journal of Applied Psychology
and subjective wellbeing. Psychological Bulletin 134, 138–161. 86, 459–467.

Vous aimerez peut-être aussi