A Meta-Analytic Review of Social Identification and Health in Organizational Contexts

Abstract

We provide a meta-analytical review examining two decades of work on the relationship between individuals’ social identifications and health in organizations (102 effect sizes, k=58, N=19,799). Results reveal a mean-weighted positive association between organizational identification and health (r=.21, T=.14). Analysis identified a positive relationship for both workgroup (r=.21) and organizational identification (r=.21), and in studies using longitudinal/experimental (r=.13) and cross-sectional designs (r=.22). The relationship is stronger (a) for indicators of the presence of well-being (r=.27) than absence of stress (r=.18), (b) for psychological (r=.23) than physical health (r=.16), (c) to the extent that identification is shared among group members, and (d) as the proportion of female participants in a sample decreases. Overall, results indicate that social identifications in organizations are positively associated with health but that there is also substantial variation in effect size strength. We discuss implications for theory and practice and outline a roadmap for future research.

Full text

A Meta-Analytic Review of Social Identification and Health in Organizational Contexts
Niklas K. Steffens1,*, S. Alexander Haslam1, Sebastian C. Schuh2, Jolanda Jetten1,
Rolf van Dick3,4
1 School of Psychology, The University of Queensland, Australia
2 Department of Organizational Behavior, China Europe International Business School, China
3 Department of Psychology, Goethe University Frankfurt, Germany
4 Work Research Institute (AFI), Oslo, Norway
* Correspondence concerning this article should be addressed to: Niklas K. Steffens, School of
Psychology, The University of Queensland, Brisbane, St Lucia QLD 4072, Australia. E-mail:
N.Steffens@uq.edu.au; Fax: +61 (0)7 3365 4466; Tel.: +61 (0)7 3346 9555.
To Appear in: Personality and Social Psychology Review
Author Note
This work was supported by two grants from the Australian Research Council awarded to SAH
(FL110100199) and JJ (FT110100238). We thank the researchers whose work is included in this
analysis for their time and effort in helping us to gather all relevant data. We also thank Miriam
Yates, Hannibal Thai, and Nerisa Dozo for help with data analysis and Tom Postmes for helpful
comments on an earlier version of this manuscript.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 2
Abstract
We provide a meta-analytical review examining two decades of work on the relationship between
individuals’ social identifications and health in organizations (102 effect sizes, k=58, N=19,799).
Results reveal a mean-weighted positive association between organizational identification and
health (r=.21, T=.14). Analysis identified a positive relationship for both workgroup (r=.21) and
organizational identification (r=.21), and in studies using longitudinal/experimental (r=.13) and
cross-sectional designs (r=.22). The relationship is stronger (a) for indicators of the presence of
well-being (r=.27) than absence of stress (r=.18), (b) for psychological (r=.23) than physical
health (r=.16), (c) to the extent that identification is shared among group members, and (d) as the
proportion of female participants in a sample decreases. Overall, results indicate that social
identifications in organizations are positively associated with health but that there is also
substantial variation in effect size strength. We discuss implications for theory and practice and
outline a roadmap for future research.
Keywords: social identity; identification; health; well-being; meta-analysis

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 3
The topics of stress and well-being provide a major focus for research in social,
organizational, and health psychology (e.g., Anisman, 2014; Lundberg & Cooper, 2011). This is
for the simple reason that stress, fatigue, and burnout are costly not only for the individual
organizational members who suffer and experience reduced functioning, but also for the
organizations whose effectiveness is compromised by staff underperformance, absence, and
turnover, and for society as a whole that incurs the costs of remediation and treatment. As a
corollary, efforts to promote well-being can have significant benefits at individual,
organizational, and societal levels (e.g., Helliwell, Layard, & Sachs, 2013).
It is noteworthy that, in addition to sparking scholarly interest, health in the workplace has
also become an important issue in society at large as part of efforts to cultivate engaged
workplaces and healthy populations. In this regard, societies across the globe take a keen interest
in issues of health not only as a result of rising health expenditure but also because health metrics
are increasingly seen to be important indicators of national prosperity (e.g., as seen in national
initiatives such as the UK’s Happiness Index and the EU’s Quality of Life Survey and in global
initiatives such as the Social Progress Index and the World Happiness Report; Helliwell et al.,
2013). Moreover, health is an important topic for organizations that care about the well-being of
individuals for either intrinsic or extrinsic reasons (e.g., because they care about creating value
and attracting and retaining talent or because this is associated with enhanced productivity and
profitability). Indeed, speaking to the wider significance of health for organizations, meta-
analytic evidence suggests that individuals’ health often precedes their capacity to perform
(Lyubomirsky, King, & Diener, 2005).
With a view to understanding issues of stress and well-being in organizations, an
emerging body of research suggests that one factor that has important consequences for
psychological health and well-being in the workplace is people’s social group memberships

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 4
(Jetten, Haslam, & Haslam, 2012). More particularly, there is evidence that a person’s
internalization of a relevant group in the work context (a workgroup, an organization) as part of
their self-concept has important consequences for stress and burnout (Horton, McClelland, &
Griffin, 2014; Wegge, van Dick, Fisher, Wecking, & Moltzen, 2006). Here the group furnishes
them with a sense of social identity (as “we” and “us” rather than just “I” and “me”; Ashforth &
Mael, 1989; Tajfel & Turner, 1979) and this has been shown to have profound implications for
the experience of stress, social support, and resilience more generally (e.g., Drury, 2012;
Muldoon & Downes, 2007).
Nevertheless, the precise consequences of workplace social identifications for stress and
health are unclear. On the one hand, research suggests that individuals’ increased social
identification with a workgroup and/or an organization is positively related to their health,
because, among other things, this provides a basis for increased support, control, and resilience
(Greenaway et al., 2015; Haslam, O’Brien, Jetten, Vormedal, & Penna, 2005). On the other hand,
identification may compromise health because, among other things, increased identification is
associated with working long hours (Ng & Feldman, 2008) and this is associated with reduced
health (Golden & Wiens-Tuers, 2006).
As well as establishing whether increased identification enhances or compromises health,
our core aim in the present research is to estimate the nature and precise strength of this
relationship (i.e., with reference to effect-size and confidence intervals). This is important
because such analysis provides a basis (a) for future research to gauge the importance of social
identification for workplace health, (b) for integrating the insights of the present analysis into
such future research, and, ultimately, (c) for building a more accurate and robust quantitative
science (Cumming, 2014). With this goal in mind, we use a range of advanced methods to
estimate and correct for measurement and publication bias so as to gain a more precise estimate

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 5
of the true effect size of this relationship.
In addition to estimating the global summary effect size of the identification–health
relationship, we seek to shed light on theoretically and practically important moderators of the
relationship by examining the extent to which its strength is affected by (a) the focus of
identification (workgroup vs. organization), (b) health valence (positive vs. negative), (c) the
nature of the health index (psychological vs. physical), and (d) the sharedness of identification
(shared vs. non-shared). Furthermore, we perform exploratory analyses of several additional
potential moderators (including the scale used to assess social identification, study methodology,
gender and age of participants, status of profession, and the Inglehart-Welzel cross-cultural
dimensions). Finally, we use the findings of the analysis as a basis for developing an agenda for
future theory, research, and practice with a view to helping the field move forward.
Social Identification and Health in Organizations
A growing body of evidence points to the important role of social factors in people’s
experience of health (Cacioppo, Hawkley, & Bernstein, 2003; Cohen, 2004; Diener & Biswas-
Diener, 2008; Helliwell & Putnam, 2004; Holt-Lunstad, Smith, & Layton, 2010). In recent years,
one significant strand of this work has been informed by social identity theorizing. This starts
from the assertion that people are able to think, feel, and act not just as individuals (i.e., in terms
of a personal identity as “I”) but also as group members (in terms of a social identity as “we”; for
reviews, see Ellemers, 2012; Postmes & Branscombe, 2010; Reicher, Spears, & Haslam, 2010;
Hogg & Abrams, 1988; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; Turner, Oakes,
Haslam, & McGarty, 1994). The approach argues that when people categorize themselves as
members of a group (e.g., “us members of Organization X”), this makes them more likely to see
the world from the perspective of fellow ingroup members, more open to influence from ingroup
members, and more likely to trust and cooperate with ingroup (rather than outgroup) members

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 6
(Turner, 1991). This in turn is the basis for a range of distinctive forms of organizational behavior
— including motivation, communication, collaboration, and leadership (Ashforth & Mael, 1989;
Blader & Tyler, 2009; Ellemers, de Gilder, & Haslam, 2004; Haslam, 2004; Hogg & Terry, 2001;
Tyler & Blader, 2003).
One of the key applications of the social identity approach to the domain of health has
argued that self-categorization in terms of a relevant group membership has significant
implications for the experience of stress and well-being (Branscombe, Schmitt, & Harvey, 1999;
Gleibs et al., 2011; Sani, Herrera, Wakefield, Boroch, & Gulyas, 2012; for recent reviews see
Haslam, Jetten, Postmes, & Haslam, 2009; Jetten et al., 2012). This is for a number of reasons,
many of which speak to insights from the transactional model of stress (Lazarus & Folkman,
1984). First, this approach led researchers to predict that to the extent that people self-categorize
as group members, their primary appraisal of stressors will be affected by the circumstances and
views of their ingroup — a point confirmed in research which has shown that appraisal is
determined both by the significance of a stressor for a particular social identity (Levine &
Reicher, 1996) and by feedback about stressors from ingroup (but not outgroup) members
(Gallagher, Meaney, & Muldoon, 2014). Second, self-categorization in terms of a given group
membership also serves as a basis for active coping processes in the form of secondary appraisal.
In particular, this is because when they are acting in terms of a shared group membership, people
should be more likely both to receive support from fellow ingroup members and to interpret that
support in the spirit in which it is intended (Levine, Cassidy, Brazier, & Reicher, 2002; Levine,
Prosser, Evans, & Reicher, 2005).
As an instantiation of these fundamental points, research has found that people’s
experience of stress and well-being in the workplace is significantly structured by the nature of
their social identifications — that is, their strength of identification with a given group — as

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 7
members of either a workgroup or the organization as a whole (for a recent review, see van Dick
& Haslam, 2012). Illustrative evidence emerges from a field study conducted by van Dick and
Wagner (2002) which observed that high levels of organizational identification were related not
only to organizational outcomes such as job satisfaction and motivation but also to reduced
experience of strain. Similarly, a study by Bizumic, Reynolds, Turner, Bromhead, and Subasic
(2009) found that a fairer and more inclusive school climate was associated with reduced stress
among teachers partly because it served to enhance teachers’ identification with the school.
It is the case, however, that there is a range of factors associated with increased social
identification may have positive and negative effects on health. Therefore, we can anticipate that
the link between social identification and health in organizations could take one of several
plausible forms: ranging from an overall positive association (an invigoration perspective), to an
overall negative association (an exhaustion perspective), or an overall absence of an association if
these effects were equally strong (a neutralization perspective). Key theoretical backgrounds and
literatures that lend support for these alternative hypotheses concerning the identification–health
relationship are summarized in Table 1. As can be seen from this table, there are at least six
established literatures that support an overall positive association. Briefly, these suggest that
social identifications in organizations will be associated with better health because identification
provides people with (a) a sense of belonging (Baumeister & Leary, 1995; Ryan & Deci, 2000),
(b) a sense of meaning and purpose (van Dick & Wagner, 2002; Wegge et al., 2006), (c) a sense
of control and agency (Greenaway et al., 2015; Reicher & Haslam, 2006), and (d) a source for
self-affirmation and a positive attribution style (Cruwys, South, Greenaway, & Haslam, 2015;
Sherman, Kinias, Major, Kim, & Prenovost, 2007). Furthermore, literatures indicate that social
identification is the basis for people’s experience of (e) social support (Levine et al., 2002;
Haslam et al. 2005) as well as (f) collective self-efficacy (Shamir & Kark, 2004; van Zomeren,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 8
Leach, & Spears, 2012) — experiences that are positively associated with health in the workplace
(Avanzi, Schuh, Fraccaroli, & van Dick, 2015).
At the same time, there are at least four important literatures which suggest that social
identification is associated with factors that have negative effects on people’s health. These imply
that increased social identification will lead to a downward health spiral because increased
identification is associated with (a) excessive involvement as well as heightened demands and
pressure to perform (Herrbach, 2006; Mühlhaus & Bouwmeester, 2016) and (b) working long
hours (Escartiín, Ullrich, Zapf, Schlüter, & van Dick, 2013; Ng & Feldman, 2008), which reduce
people’s capacity to engage with other meaningful groups outside work (Golden & Wiens-Tuers,
2006; Pratt et al., 2000). In addition, lines of research on (c) dirty work and stigma (Ashforth &
Kreiner, 1999; Branscombe et al., 1999) and (d) group norms and identity motivation (Oyserman
& Fryberg, & Yoder, 2007; Perkins & Wechsler, 1996; Tarrant, Hagger, & Farrow, 2012)
indicate that increased identification can lead to increased stress and diminished health to the
extent that groups are stigmatized and endorse norms that prescribe health-damaging habits and
behaviors.
In the present meta-analysis, we review the large body of evidence that speaks to these
issues with a view to establishing whether increases in either workgroup identification or
organizational identification are generally unrelated to health or associated with the experience of
improved or impoverished health. In addition to estimating the global magnitude of the summary
effect of the relationship between social identifications and health in organizations, we also
examine whether the strength of this relationship varies as a function of several key moderating
factors. Specifically, we examine moderating factors that have received relatively little attention
in previous research but that are theoretically important in determining when the positive and
negative effects of identification are stronger or weaker (or absent). In what follows, we will

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 9
briefly introduce these factors. Key literatures that support alternative hypotheses for these a
priori moderators of the positive effects associated with social identification are summarized in
Table 2.
Health as a Function of Workgroup versus Organizational Identification
As outlined earlier, there are reasons to believe that both workgroup and organizational
identification will be positively related to health. However, there are several reasons why health
should be differentially associated with one entity relative to the other. First, one might expect
the identification–health relationship to be stronger for workgroups than for organizations due to
the greater salience of workgroups and their more proximal nature (Ashforth & Johnson, 2001;
Barker & Thompson, 1994; van Knippenberg & van Schie, 2000). Second, one might also expect
stronger effects for workgroups because we know from literatures on group socialization and
cohesion that people are more familiar with members of smaller groups and experience smaller
(work)groups as more cohesive (Anderson & Thomson, 1996; Carron & Spink, 1995; Moreland
& Levine, 2001).
While the previous two arguments suggest that the positive effects of health will be
stronger for workgroup than for organizational identification, other literatures and models suggest
the opposite. First, as workplaces evolve, people are more likely to experience change at the level
of the workgroup (e.g., in terms of roles, tasks, membership) than at the more abstract level of the
entire organization. Building on research on continuity (Sani, 2008; Sani, Bowe, & Herrera,
2008) and enduringness of self (Albert & Whetten, 1985), one might therefore expect that people
are more likely to derive positive effects associated with identification with an organization
because this is more likely to provide them with a sense of stability and continuity. Second,
previous research indicates that public regard is an integral aspect of collective self-esteem
(Dutton, Dukerich, & Harquail, 1994; Luhtanen & Crocker, 1992) and accordingly one might,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 10
expect that positive effects will be stronger for organizational (rather than workgroup)
identification because more encompassing organizations are likely to be recognized and respected
by a wider circle of people in society.
Identification and the Presence of Well-Being versus the Absence of Stress
Following in part from the rise of positive psychology (Diener, 2000; Seligman &
Csikszentmihalyi, 2000), researchers have pointed to the important distinction between negative
aspects of health such as burnout (Lee, & Ashforth, 1996; Maslach, & Jackson, 1981; Maslach,
Schaufeli, & Leiter, 2001) or depression (Cruwys, Haslam, Dingle, Haslam, & Jetten, 2014) and
other more positive aspects such as engagement and salutogenesis (Bakker, Schaufeli, Leiter, &
Taris, 2008; Graeser, 2011; Lindström & Eriksson, 2006). In these terms, health is reflected in
either the absence of negative indicators of health (e.g., the absence of stress such as burnout or
strain) or the presence of positive indicators of health (e.g., the presence of well-being such as
work engagement or general health). This distinction rests on the observation that the absence of
illness is not the same as the presence of well-being and, indeed, that these are qualitatively
distinct states. In short, although these different experiences may often go hand in hand, this is
not inevitably the case. Just because an individual is experiencing stress this does not necessarily
mean that they are unhappy and unfulfilled, and just because an individual feels happy and
fulfilled this does not necessarily mean that they are free of stress.
Theoretical insights from the social identity approach suggest that organizational
identification will be related both to reduced stress and increased well-being. Indeed, this is what
previous studies have tended to report (e.g., Horton et al., 2014). Nevertheless, the literature on
psychological accounts of positive human health (Ryff & Singer, 1998; Thoits, 1994) suggests
that people’s health is influenced by their agency (and not passive surrender to stressors) and a
consequence of individuals’ sense of having meaningful and purposeful lives. Furthermore,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 11
phenomenological analysis indicates that high social identification is generally construed to be a
basis for experiencing positive outcomes (e.g., support, belonging, control, agency; Haslam et al.,
2009) rather than to be a basis for avoiding negative outcomes (Cruwys et al., 2014). This
literature lends support for the notion that stronger positive effects will be found when health is
conceptualized in terms of the presence of well-being rather than the absence of stress. At the
same time, research on the stress-buffering effects of groups indicates that the resources that
groups provide are especially helpful in the context of experiencing stressful events (Branscombe
et al., 1999; Crabtree, Haslam, Postmes, & Haslam, 2010; see also Thoits, 1986), suggesting that
the positive effects associated with identification will be particularly pronounced in helping
people to cope with stressful events and thereby in reducing their negative experience of strain.
Identification and Physical versus Psychological Health
Another common and important distinction that is made in the literature is between
psychological health (Berkman, 2001; Scheier & Carver, 1992; Taylor & Brown, 1988) and
physical health (Cohen, Doyle, Skoner, Rabin, & Gwaltney, 1997; Uchino, 2009; Uchino,
Cacioppo, & Kiecolt-Glaser, 1996). Psychological indicators typically comprise felt and
experienced states that relate to health and well-being (e.g., fatigue, burnout, stress, comfort,
well-being) while physical indicators comprise experiences or biological markers of physical or
physiological functioning (e.g., frequency of physical symptoms, physiological functioning, pain,
cortisol levels). As in the case of positive and negative indicators of health, although
psychological and physical health are likely to be strongly inter-related in many contexts, they
capture distinct aspects of health and, accordingly, can in fact diverge from each other (e.g., so
that reduced physical functioning does not necessarily accompany reduced life satisfaction; see
Howell et al., 2014).
Social identity theorizing suggests that social identification in the workplace is likely to

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 12
be associated with health as captured not only by psychological markers (e.g., Bizumic et al.,
2009; Harris & Cameron, 2005) but also physical ones (e.g., Häusser, Kattenstroth, van Dick, &
Mojzisch, 2012; Wegge et al., 2006). Nevertheless, there are at least two important reasons why
social identification might be associated more strongly with psychological than physical health.
First, in line with the literature on attitude–behavior compatibility in levels-of-analysis (Fishbein
& Aijzen, 1974; see also Riketta & van Dick, 2005), as a psychological construct (grounded in,
and experienced through, cognitions, emotions and behaviors; Ashforth, Harrison, & Corley,
2008; Ellemers, Kortekaas, & Ouwerkerk, 1999; Leach et al., 2008; Turner, 1982), social
identification is likely to be linked particularly strongly to experiences at this psychological level.
Second, a body of work indicates that individuals often connect socially with others by engaging
in risky behavior and by jeopardizing their physical health (Howell et al., 2014; Hopkins &
Reicher, 2016), suggesting that increases in identification will be related more strongly to
increases in psychological health than increases in physical health.
Shared Social Identification
Research informed by the social identity approach asserts that social identity is the
“cognitive mechanism that makes group behavior possible” but that it is nevertheless an aspect of
the self-concept that is represented and realized in the mind of the individual (Turner, 1982, p.
21). Along these lines, research conceptualizes and captures social identification at the level of
the individual, as reflected in standard identification measures that assess the extent to which an
individual identifies with, or self-categorizes in terms of, a group as an individual group member
(e.g., assayed via items such as “I identify with [this group]”; I see myself as a member of [this
group]”; e.g., Abrams, Ando, & Hinkle, 1998; Brown, Condor, Mathews, Wade, & Williams,
1986; Cameron, 2004; Doosje, Ellemers, & Spears, 1995; Ellemers et al., 1999; Jackson, 2002;
Leach et al., 2008; Mael & Ashforth, 1992; Phinney, 1992; Postmes, Haslam, & Jans, 2013;

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 13
Roccas, Sagiv, Schwartz, Halevy, & Eidelson, 2008; Sellers, Smith, Shelton, Rowley, &
Chavous, 1998; van Dick, Wagner, Stellmacher, & Christ, 2004). Significantly, identification as
assessed by identification scales can be more or less shared among group members. This degree
of sharedness in social identification may be important to consider for theoretical reasons in
allowing for a fuller understanding of what social identification is and how it plays out in the
psychology and behavior of group members (see also Haslam, 1997; Tajfel, 1981).
Here it is important to note that the issue of sharing attributes more generally has
previously been proposed to be an important feature of group attitudes and behavior. For
instance, in the multicomponent model of group identification, Leach and colleagues (2008)
identify individual self-stereotyping (sharing attributes with other members) and perceived
homogeneity (being similar to other group members) as distinct components of social
identification. There are some interesting parallels between perceived homogeneity and
individual self-stereotyping on the one hand and sharedness in identification on the other. In
particular, perceived homogeneity and individual self-stereotyping are similar to sharedness in
identification in that both refer to shared attributes among group members. At the same time, they
also differ from each other in two ways: (a) the former refer at an abstract level to any shared
attributes that are seen to define a group (which could relate to identification but need not) while
the latter refers specifically to identification as the attribute that is shared, and (b) the former refer
to perceptions of shared attributes, while the latter refers to actual sharedness in identification.
Sharedness in identification is likely to be important for group members’ psychology
because an individual whose level of social identification is similar to that of fellow group
members is likely to have a different orientation to, and experience of, the group than an
individual whose level of social identification is not shared in this way. This is for two reasons.
First, because shared social identity is likely to provide a particularly strong basis for trust and

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 14
mutual influence (e.g., Haslam, Reicher, & Platow, 2011; Tanis & Postmes, 2005; Steffens et al.,
2014; Turner & Reynolds, 2001). Second, because research indicates that group entitativity is
important in itself (e.g., Campbell, 1958; Hamilton & Sherman, 1996; Lickel et al., 2000), and in
newly forming groups, entitativity has been shown to strengthen both people’s perceptions of
groups and their feeling of belongingness in them (Jans, Postmes, & van der Zee, 2011).
Accordingly, and as shown by Jans, Leach, Garcia, and Postmes (2015), while it is useful and
meaningful to treat identification as an individual difference measure on its own, it is likely that
this fails to capture the full range of identity-informed psychology and behavior. Specifically,
Jans and colleagues (2015) found that identification at the collective level had an additional
impact (above and beyond an individual’s own level of identification) on a group member’s
subsequent development of identification with the group. Surprisingly, prior social identity
research has not examined this aspect in relation to health and our meta-analysis provides a
unique opportunity to shed light on the effects of sharedness on the identification–health
relationship.
One reason why the positive identification–health relationship is likely to become more
pronounced as social identity is more shared among group members relates to the idea that shared
social identity is the basis for solidarity (Drury, Cocking, & Reicher, 2009; Reicher, Cassidy,
Wolpert, Hopkins, & Levine, 2006) that may promote their experience of well-being.
Furthermore, the positive effects are likely to become amplified when identification is shared
because shared social identity is likely to lead to more cohesive groups that in turn are likely to
amplify the positive health benefits associated with identification (Blanchard, Amiot, Perreault,
Vallerand, & Provencher, 2009; Griffith, 2002). At the same time, however, the literature on the
benefits of individual distinctiveness (Cantor, Kemmelmeier, Basten, & Prentice, 2002; Hornsey
& Jetten, 2004; Markus & Kitiyama, 1994; Sheldon & Bettencourt, 2002) might lead one to

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 15
expect that sharedness in identification is likely to weaken the positive effects associated with
identification due to greater obstacles to harness the benefits of individual distinctiveness. Again,
then, this is an important theoretical and empirical question upon which our analysis can shed
powerful light.
The Present Research
In the present meta-analysis we aim to examine the above hypotheses by taking stock of
the extensive body of research that pertains to the relationship between social identification and
health in organizations in order to provide a more robust and precise estimation of the size of this
relationship (Cumming, 2014). In order to assess the extent to which the magnitude of the effect
size estimates based on the present set of studies are influenced by various biases that hinder the
advancement of science (e.g., selective publishing, small study effects, p-hacking) and to yield
estimates of the true unbiased effect size, we also employ a number of state-of-the-art techniques
to estimate and correct for measurement and publication bias. In addition, we examine the
capacity for a variety of novel and theoretically important variables — specifically, focus of
social identification (workgroup vs. organization), health valence (negative vs. positive
indicators), nature of health index (physical vs. psychological health), and sharedness in
identification — to moderate the strength of this relationship. Moreover, for exploratory
purposes, we also examine several additional variables (scale of identification, study
methodology, gender an age of participants, status of profession, and the Inglehart-Welzel cross-
cultural dimensions) as potential moderators of the identification–health relationship. In pursuing
these aims, we follow the meta-analysis reporting standards specified by the American
Psychological Association (APA Publications and Communications Board Working Group on
Journal Article Reporting Standards, 2008) concerning both our meta-analytic procedure and the
reporting of relevant information.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 16
Method
Study Search
Two independent coders searched for studies in the major databases Web of Science,
PsychINFO, PsyArticles, and MEDline by using the keywords “social identity/social
identification/team identification/workgroup identification/organizational identification and
health/stress/burnout/exhaustion/fatigue/sickness/physical symptoms/well-being”. The coders
also searched for further studies on GoogleScholar by screening all papers that were indicated to
have cited the papers that the primary search identified. Moreover, to obtain studies that were not
easily accessible or unpublished, we sent out calls for data (via email lists and newsletters)
through major associations in social and organizational psychology and management including
the Society for Personality and Social Psychology (SPSP), the Society for Industrial and
Organizational Psychology (SIOP), the European Association of Social Psychology (EASP), the
Social and Organizational Sections of the German Psychological Society (DGPs), and the Society
of Australasian Social Psychologists (SASP).
Study Inclusion Criteria
We included all published and unpublished studies reported in English if they reported
any quantitative assessments concerning the relationship between individuals’ social
identification with either workgroup or organization and at least one health indicator. We
included studies if the report was in English. In an attempt to include reports of studies in
languages other than English, in our calls for data, we asked for results from any additional
(published or unpublished) studies, which could be based on reports in other languages, as long
as the authors sent us reports concerning the identification–health relationship in English.
To be included, studies had to fulfill the criterion of being based on explicit measures of
social identification with either workgroup or organization (including standard identification

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 17
scales and other scales that were constructed for the purpose of assessing identification; for a
review, see Postmes et al., 2013). Studies were excluded if they measured other organizational
constructs (such as commitment; for detailed discussions of conceptual differences, see Meyer,
Becker, & van Dick, 2006; Ng, 2015; van Knippenberg & Sleebos, 2006). Moreover, studies also
had to fulfill the inclusion criterion of assessing an explicit measure of psychological or
physiological health and well-being including burnout (including its sub-dimensions emotional
exhaustion, cynicism, and reduced accomplishment), subjective stress or strain, physical stress or
symptoms including health complaints or symptoms, physiological functioning, general health,
subjective well-being, work engagement, and positive and negative emotions at work.
Studies were excluded if they assessed other organizational outcomes such as job
satisfaction, turnover, or performance (Riketta & van Dick, 2005). Moreover, studies had to
fulfill the inclusion criterion of being conducted and retrieved up to April 2014, while we did not
set inclusion criteria for starting dates (i.e., studies could have been conducted any time prior to
2014). Moreover, studies that used any methodology (e.g., correlational, longitudinal, and
experimental designs) were included but they had to provide statistical information about the
quantitative assessment of the identification–health relationship. Studies were excluded if they
used other assessments of this relationship (e.g., qualitative studies).
The search strategy based on the above inclusion and exclusion criteria yielded a total of
102 effect sizes based on 58 independent samples (N = 19,799) concerning the (workgroup and
organizational) identification–health relationship. To provide a comprehensive and transparent
description of the studies including the coding of variables (as described in more detail below),
an overview of the studies is presented in Table 3.
Coding Procedure
We coded all variables that were relevant for the present analysis including author names,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 18
year of publication, publication status, sample size, effect size, and the type of sample. Moreover,
we coded all potential moderators as discussed previously as well as several further variables for
exploratory purposes including the particular identification scale that was used to assess
identification, the study’s methodology, the gender and age of participants, the status of their
profession, and Inglehart-Welzel cross-cultural dimensions of the country in which study was
conducted. All coded variables are described in more detail in what follows.
A team of four coders, all of whom had completed a university degree in psychology and
received an induction about the meaning of the variables of interest, coded all variables. At least
two coders coded any single variable. Discrepancies in coding were discussed between the coders
and resolved by the author team. In cases where it was not possible to extract relevant data
(because data were missing or the report was ambiguous), we contacted the authors via email and
requested either the data or an indication of the nature of missing data (and sent out two
additional reminders if authors did not respond within two weeks). We contacted the authors of
18 primary studies, all of whom responded.
To provide a more detailed description of the included studies, overall, 31 of the 102
effect sizes in the final sample (30.4%) are based on items adapted from Mael and Ashforth’s
(1992) identification scale, 27 effect sizes (26.5%) are based on items adapted from Doosje et
al.’s (1995) scale, and 9 effect sizes (889%) are based on items adapted from van Dick et al.’s
(2004) scale. Following the use of these three scales, the use of other scales varied significantly.
Another 4 effect sizes (3.9%) are based on items adapted from Ellemers’ (1993) scale, 3 effect
sizes (2.9%) each are based on items adapted from the scales by Haslam (2004), Postmes et al.
(2013), and Sellers et al. (1998), 2 effect sizes (2.0%) each are based on items adapted from the
scales by Brown et al. (1986) and Leach et al. (2008), while the remainder are either based on
items adapted from various other scales (e.g., Cameron, 2004) or were created for the purpose of

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 19
the research with no scale used for more than one effect size.
Moreover, studies used a variety of different measures to assess health constructs.
Overall, 24 of the 102 effect sizes in the final sample (23.5%) are based on scales that assess
subjective well-being and general health (e.g., Goldberg, 1972; Warr, 1990), 22 (21.6%) are
based on scales and biological measures that assess physical stress or symptoms (e.g., Brähler,
Hinz, & Scheer, 2008; Patchen, 1970), 20 (19.6%) are based on scales that assess burnout or sub-
dimensions of it (e.g., Maslach & Jackson, 1981; Singh, Goolsby, & Rhoads, 1994), 17 (16.7%)
are based on scales that assess subjective stress (e.g., Cohen, Kamarck, & Mermelstein, 1983;
Parker & DeCotiis, 1983), and 15 (14.7%) are based on scales that assess positive or negative
emotions at work. In line with the present theoretical focus and as discussed previously, we
conduct moderation analysis by health valence and index and refrain from analyzing effect size
strength as a function of particular health constructs — not least because of the variability in the
assessed health constructs (and the variability in the scales within constructs) limits the
meaningfulness of results. Nevertheless, we conducted an exploratory moderator analysis, which
revealed that the relationship between identification and these different outcomes did not differ in
strength. In what follows, we provide a detailed description of the coding of variables.
Country. We coded studies based on the country that a study was conducted in (i.e., the
country that participants were residing in).
Sample. We coded the sample of participants as a function of participants’ job or
profession as described in the studies (e.g., whether the sample consisted of teachers, bank
assistants, call center employees).
Publication status. We coded effect sizes as a function of publication status — recording
whether an effect size was published or unpublished.
Identification focus. We coded effect sizes as a function of whether participants

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 20
indicated their identification with a workgroup (or team, department) or their organization as a
whole.
Health valence. To capture health valence, we coded whether health was measured in
positive or negative terms — that is, whether an effect size referred to the presence of well-being
(e.g., subjective health, psychological well-being, work engagement, positive emotions at work)
or the absence of stress (e.g., strain, burnout, physical health symptoms, negative emotions at
work). To render effect sizes comparable, for the purpose of the analyses we reverse-coded the
sign of the effect size in the case of indicators of the absence of stress to denote health.
Health index. To denote health index, we coded whether the health indicator represented
psychological (e.g., reported psychological well-being, stress, burnout) or physical health (e.g.,
reported or assessed physiological functioning, physical health, cortisol levels).
Identification scale. We coded the identification scale as a function of the inventory that
was used to assess social identification. As described above, by far the most commonly used
scales were Mael and Ashforth’s (1992) scale (31 effect sizes, k = 19) and Doosje et al.’s (1995)
scale (26 effect sizes, k = 12). We also coded effect sizes that used van Dick et al.’s (2004) scale
(9 effect sizes, k = 4). However, due to their low frequency of use, we coded all remaining scales
into a fourth residual category.
Study methodology. We coded studies based on whether they employed a cross-sectional
design, or a longitudinal or experimental design that manipulated social identification to address
issues of causality. As the sample included only three studies that used experimental designs
(designs identified as Research Design 1 by Cooper, Robinson, & Patall, 2006) and only five
studies that used longitudinal designs (Research Design 2), we combined studies that used
experimental and longitudinal designs (k = 8) and compared them to studies that used
correlational designs (k = 50; Research Design 3). Experimental studies that manipulated other

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 21
variables but assessed social identification were coded as cross-sectional designs for the present
purposes and effect sizes were based on the zero-order correlations between identification and
health measures.
Gender of participants. To denote participants’ gender, we coded a study based on the
proportion of female participants in the sample (as a percentage).
Age of participants. To denote participants’ age, we coded the mean age of participants
in a sample.
Status of profession. We coded studies based on the status of participants’ profession by
means of the Cambridge Social Class and Stratification Scale (CAMSIS; Lambert & Prandy,
2014; Prandy, 1990; Sacker, Firth, Fitzpatrick, Lynch, & Bartley, 2000; for a review on status
and health, see Gallo & Matthews, 2003). We used country-specific indices where they existed
and if indices differed by gender, we weighted the sample score by the gender of participants in a
sample. For studies conducted in countries for which there did not exist country-specific indices,
we used the international CAMSIS scale. In samples that included distinct sets of professions, we
weighted the score by the number of participants in each set and in samples that included
participants from different countries, we weighted the score by country. For 11 studies, we were
not able to code status because participants worked in a range of different jobs in various
industries.
Inglehart-Welzel cross-culture dimensions. We coded studies as a function of the
country’s Inglehart-Welzel cultural value dimensions — the traditional/secular-rational
dimension (reflecting a country’s secular values) and the survival/self-expression dimension
(reflecting a country’s emancipative values; Inglehart & Baker, 2000) — using the most recently
updated data from Wave 6 (World Values Survey, 2014). If data for a given country was not
available in the most recent wave, we used the data from the preceding Wave 5. Because there

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 22
are no data available for Belgium and Greece in any of the waves, the sample size was reduced
for this analysis (k = 54).
Shared social identification. We also coded the standard deviation (SD) in social
identification in samples in which participants shared membership, and reported identification
with, the same single group as indicators of sharedness in social identification. These criteria
were fulfilled in 21 effect sizes based on 16 independent samples (N = 8,179) of the identified
studies assessing the social identification–health relationship. Fifteen samples assessed
organizational identification and one sample assessed workgroup identification (all remaining
studies that assessed workgroup identification provided indicators of identification from
participants who were part of different workgroups and pooled data across participants in
multiple workgroups). Scores were standardized across samples to form a 1 to 7-point Likert
scale.
Analytic Procedure
Consistent with the majority of primary research that has reported correlations between
social identification and health, we used mean-weighted r as the effect size of choice. We
analyzed effect sizes using Fisher’s Z-transformation and for ease of interpretation we then
converted all effect sizes into an r-statistic (see also Lipsey & Wilson, 2001). We estimated the
amount of variation in the effect size distribution by inspecting indicators of homogeneity of
variance (Q) and heterogeneity of variance (T and I2). We also estimated the 95% prediction
interval following recommendations and using the formula provided by Higgins, Thompson, and
Spiegelhalter (2009). To perform the analyses, we used comprehensive meta-analysis version 3
from Biostat Inc. (Engelwood, New Jersey, USA).
Some studies reported multiple outcomes. Because effect sizes were not independent in
such cases, we used the approach suggested by Cooper (1998) to account for non-independent

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 23
effect sizes and combined non-independent effect sizes by averaging across them. In line with
Hedges and Olkin’s (1985) method, we correct for measurement bias due to sampling error by
weighting effect sizes by sample size using a random-effects model (Borenstein, Hedges,
Higgins, & Rothstein, 2009; Cumming, 2014; Hunter & Schmidt, 2000).
Furthermore, following recommendations by Ferguson and Brannick (2012) and Kepes,
Banks, McDaniel, and Whetzel (2012) to use triangulation of methods to address publication
bias, we used a number of different traditional and advanced techniques to estimate and correct
for measurement and publication bias (for an overview, see also Borenstein et al., 2009).
Specifically, we inspected (a) Orwin’s (1983) fail-safe N for effect size analysis to determine the
number of studies with a correlation of zero that would be needed in file drawers to reduce any
potentially significant effects to a small correlation of .05, (b) a funnel plot asymmetry test
(Egger, Smith, Schneider, & Minder, 1997) to assess potential asymmetry in effect sizes and
standard error (asymmetry would be indicative of publication bias), (c) trim and fill analysis
(Duval & Tweedie, 2000) to assess the unbiased point estimate of the effect size after adjusting
for missing studies, (d) subgroup analysis by publication status to assess the extent to which the
effect size in unpublished studies is significant and differs from that in published studies (Egger,
Juni, Bartlett, Holenstein, & Sterne, 2003; McAuley, Tugwell, & Moher, 2000), (e) p-curve
analysis to assess whether there the set of studies have evidential value (whether there is evidence
of p-hacking) and whether the evidential value is inadequate (Simonsohn, Nelson, & Simmons,
2014), and (f) cumulative meta-analysis that involves sorting studies in the sequence from largest
to smallest before performing cumulative meta-analysis with the addition of each study to assess
the extent to which the effect size changes as smaller studies are added (if effect size does not
shift, then there is no reason to suspect bias; Borenstein et al., 2009).
To assess the impact of the present categorical moderators, we again used a mixed-effects,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 24
rather than a fixed-effects, model. For the analyses of categorical moderators, the variance across
the subgroups was pooled (i.e., we do not assume that the variance within each level of a
categorical moderator should differ from each other). We used the Q statistic to determine
whether the variation in effect sizes was greater than expected due to sampling error. To examine
whether a categorical moderator accounted for variability in effect sizes, we conducted subgroup
analyses (Borenstein et al., 2009).
The data structure was somewhat complex in that for three moderators (identification
focus, health valance, and health index), some studies included dependent subgroups (i.e., they
included measures of both values of a moderator and thus observations within each level of the
moderator are not independent in these studies). Dependent subgroups are particularly
informative in assessing moderation to the extent that the two levels of a moderator are correlated
with each other. To account for non-independence of observations, we followed procedures
outlined by Borenstein and colleagues (2009) by calculating a difference score in the
identification–health relationship as a function of the level of a moderator while taking into
consideration the strength of the correlation between the two levels of a moderator as found in a
particular study. We combined the resulting difference score of dependent samples with the
difference score of independent samples to calculate overall variability in effect size as explained
by a moderator.
To assess whether the continuous moderators affected the identification–health
relationship, we performed univariate meta-regression by means of method of moments to
estimate parameters (Borenstein et al., 2009). To analyze the impact of social status and
sharedness in identification, we performed meta-regression on the sub-sample in which data for
these variables were available.
In addition to univariate moderation analyses, we conducted multiple regression analysis

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 25
by means of meta-regression to estimate the combined variance of the distribution of true effect
sizes that can be explained by the moderators that the univariate analysis revealed as significant
(Borenstein et al., 2009). Because the data had a complex structure (such that categorical
moderators included a set of dependent subgroups that furthermore differed between the
moderators) that does not allow us to estimate explained variance at a sample level, we specified
a meta-regression model that assumed independence between outcomes within a sample for this
purpose. This model provides a conservative test of the strength of a moderator to the extent that
the categories of a moderator are in fact correlated (Borenstein et al., 2009), and we thus
recommend interpreting these results with caution. To account for the fact that only a sub-sample
of studies (k = 16) provided data concerning shared identification, we ran a separate a meta-
regression model that included SD in social identification only to estimate explained variance in
effect size distribution in this sample.
Results
Summary Effect Analyses
We estimated the overall effect size of the relationship across all 102 effect sizes based on
the 58 independent samples (N = 19,799) that examined social identifications in organizations
and health outcomes (as presented in Table 3). Results of a random-effects model analysis
indicated that the mean-weighted relationship between social identification and health was
positive with an effect size of r = .21, 95%CIs: [.17, .25]; Z = 10.11. Figure 1 presents a forest
plot displaying effect sizes, confidence intervals, and weights of independent samples and the
effect size distribution around the summary effect. Analysis indicated that the overall
heterogeneity in effect sizes across samples was larger than would be expected due to sampling
error, Q(57) = 406.61, p < . 001. Moreover, a substantial amount (I2 = 85.98%) of the observed
variation was due to real differences in true effect sizes between studies (i.e., 14.02% of the

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 26
observed variation was due to sampling error). Furthermore, there was significant absolute
dispersion in true effect sizes (T = .138).1 Estimation of the approximate 95% prediction interval
(Higgins et al., 2009) indicates that there is a probability of 95% that the true effect in a new
study would fall within a range of –.07 to .46.
Measurement Correction and Publication Bias Analyses
In addition to using random-effects model analysis to estimate a weighted-effect size, we
used several advanced techniques to address publication bias. An overview of the key questions
about the robustness of the evidence and answers provided by the results of the various analyses
that address measurement and publication bias is presented in Table 4.
Fail-safe N effect size analysis. Fail-safe N for effect size analysis (Orwin, 1983)
indicated that 217 studies in file drawers with a mean correlation of zero would be needed to
reduce the present effect size to a small correlation of .05. Given that 58 studies were identified in
the present meta-analysis, it is unlikely that 217 studies (more than three times as many identified
studies) were missed.
Funnel plot asymmetry test. Visual inspection of the funnel plot indicated that there
were no particular causes for concern about asymmetry as a function of effect size and standard
error, and the funnel plot asymmetry regression test (Egger et al., 1997) indicated that there was
no evidence that the funnel plot was asymmetrical (intercept, b = –1.02, 95%CIs: [–2.37, .33],
t(56) = 1.51, p = .14). Overall, these analyses suggest that there is no evidence of publication bias
due to small study effects.
Trim and fill method. The trim and fill approach (Duval & Tweedie, 2000) estimates the
number and magnitude of effect sizes missing due to publication bias before then correcting the
estimate of the summary effect size by imputing any missing effect sizes. This approach
identified some asymmetry in effect sizes to the left side of the mean due to potential

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 27
underreporting of effect sizes below the point estimate. However, an adjustment for missing
studies by imputing six missing studies indicated an unbiased estimation of the social
identification–health relationship of a similar effect size, r = .18, 95%CIs: [.14, .22]. This
unbiased effect size has the same substantial implications as the effect size estimated by the
random-effects model, thereby providing no evidence that the effect size was impacted by bias.
Analysis by publication status. We conducted subgroup analysis as a function of
publication status by estimating effects sizes in published and unpublished studies. Results
indicated a significant positive identification–health relationship on the basis of both published
effect sizes, r = .23, 95%CIs: [.18, .28]; Z = 9.15, unpublished effect sizes, r = .16, 95%CIs: [.07,
.24]; Z = 3.70. The effect size was larger in published effect sizes but the between-group
difference was not greater than expected due to sampling error, Q(1) = 2.30, p = .317, indicating
no evidence that the strengths of the effect sizes differed as a function of publication status.
Further inspection of the data indicated that the overall lower estimate of the effect size was
largely due to one study by Galang and Jones (2014) that found a contrasting moderate-to-strong
negative identification–health relationship of r = –.41, 95%CIs: [–.59, –.19], which given the
small sample of unpublished effect sizes (k = 14), has a large influence on the estimate of the
overall effect size. Analysis without this study leads to a significant increase in the estimated
effect size based on unpublished effect sizes, r = .19, 95%CIs: [.11, .27]; Z = 4.45, and again a
non-significant difference between the effect size based on publication status, Q(1) = .83, p =
.659.
P-curve analysis. We p-analyzed the set of findings (Simonsohn et al., 2014) to examine
whether or not the set of findings have evidential value as a result of selective reporting (in
particular, p-hacking). Because the p-values in studies that report multiple dependent variables
are not independent of each other, we followed Simonsohn et al.’s recommendations to focus

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 28
only on the p-value that was reported first (in a correlation table and if there was none, in the
text) and then ran another robustness analysis using the p-value that was reported last. One effect
size from Frisch et al. (2014) could not be used in p-curve analysis because the study design was
an experimental attenuation interaction design and inclusion of the simple effect for identification
on health would bias p-curve analysis (Simonsohn et al., 2014). Results are presented in Figure 2.
The figure shows no uptick in the p-curve at the level of .05 (which would indicate some level of
p-hacking). Both the full p-curve (testing p-curve against threshold p < .05) and the half p-curves
(testing p-curve against threshold p < .025) are significantly right-skewed at less than p < .10: full
p-curve: Z = 21.24, p < .0001; half p-curve: Z = 22.09, p < .0001. This indicates that the set of
studies contains evidential value and that there is no evidence of p-hacking or ambitious p-
hacking (up to p < .025; Simonsohn, Simmons, & Nelson, 2015). Furthermore, the p-curve is not
flatter than would be expected if the studies had a power of 33%, full p-curve: Z = 15.18, p >
.9999; half p-curve: Z = 19.23, p > .9999, thereby providing no evidence that the existing
evidential value is inadequate (i.e., that effects are too small given the size of the samples). A
separate robustness test using the p-value associated with the effect size that was reported last in
studies that reported multiple effect sizes indicated virtually identical results. Both the full p-
curve and the half-curve are significantly right-skewed, full p-curve: Z = 21.70, p < .0001; half p-
curve: Z = 22.09, p < .0001, indicating that the set of studies contain evidential value. Results
also indicate that the p-curve is not flatter than a curve that would be expected if the studies were
powered at 33%, full p-curve: Z = 15.62, p > .9999; half p-curve: Z = 19.23, p > .9999, again
providing no evidence to suggest that the evidential value of the suite of studies under
consideration is inadequate.
Cumulative meta-analysis. We conducted cumulative meta-analysis to inspect the
studies’ effect sizes as a function of their sample size (Borenstein et al., 2009). We first inspected

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 29
the data visually as shown by the cumulative forest plot in which studies are sorted by sample
size and this is presented in Figure 3. As can be seen from this figure there was no evidence of a
drift of the effect size towards the right as smaller studies are added (towards the bottom of the
plot), thereby suggesting no evidence of bias due to small-study effects. To investigate this
further, we conducted meta-regression with sample size as a predictor of the effect size. Results
provided no evidence that studies’ estimated effect size was related to sample size, b < .0001,
95%CIs [–.0001, .0001], Z = .79, Q(1) = .63, p = .429. Estimation of the effect size based on the
largest samples (k =29) yielded an effect size, r = .20, 95%CIs: [.15, .25]; Z = 7.85, that has the
same substantive implications as the summary effect size generated by the random-effects model
based on the total sample. Overall, then, estimation and correction for measurement and
publication bias indicates that publication bias has negligible impact on the core findings we
report.
Categorical Moderation Analyses
Identification focus: Workgroup versus organizational identification. Results
concerning all categorical moderators are presented in Table 5. First, we examined the extent to
which the effect sizes between organizational identification and health and between workgroup
identification and health differed in strength. Analysis revealed that organizational identification
was positively related to health, r = .21, 95%CIs: [.17, .26]; Z = 9.20, and that workgroup
identification was also positively related to health, r = .21, 95%CIs: [.14, .28]; Z = 5.90. The
variance between the groups of studies was not greater than could be expected due to sampling
error and did not differ statistically from zero, Q(1) = .14, p = .706, suggesting that there was no
evidence that the strengths of the effect sizes differed from each other.
Health valence: Absence of stress versus presence of well-being. Analysis indicated
that the relationship between social identification and the absence of stress was positive, r = .18,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 30
95%CIs: [.12, .23]; Z = 6.43. The association with the presence of well-being was also positive, r
= .27, 95%CIs: [.22, .33]; Z = 9.37. Overall, there was evidence that the variance between the
groups was greater than would be expected on the basis of sampling error, Q(1) = 7.71, p = .005,
suggesting that the relationship between social identification and the presence of well-being was
stronger than the relationship between social identification and the absence of stress.
Health index: Physical versus psychological health. Analysis indicated that the
association between social identification and psychological health was positive, r = .23, 95%CIs:
[.18, .28]; Z = 9.44. The same was true for the association between social identification and
physical health, r = .16, 95%CIs: [.10, .22]; Z = 5.39. The between-group variance was
significantly greater than expected due to sampling error, Q(1) = 4.87, p = .027, indicating that
social identification was more strongly associated with psychological than physical health
indices.
Identification scale. The social identification–health relationship varied as a function of
(dummy coded) identification scale, Q(3) = 9.80, p = .020. Analysis revealed that the relationship
was strongest in samples that employed the Doosje et al. (1995) scale, r = .34, 95%CIs: [.25, .43];
Z = 6.98, and stronger than in samples that used the Mael and Ashforth (1992) scale, r = .18,
95%CIs: [.11, .24]; Z = 5.05, the van Dick et al. (2004) scale, r = .17, 95%CIs: [.02, .32]; Z =
2.22, or other identification scales, r = .18, 95%CIs: [.12, .25]; Z = 5.54. To examine the
robustness of the identification–health relationship, we conducted additional sensitivity analysis
excluding effect sizes using the Doosje et al. (1995) scale (k = 46). This revealed an overall effect
size of, r = .18, 95%CIs: [.14, .22]; Z = 7.99 — an effect size that has the same substantive
implications that the effect size revealed by the summary random-effects model.
Study methodology. The social identification–health relationship was positive for studies
that relied both on experimental and longitudinal designs, r = .13, 95%CIs: [.01, .24]; Z = 2.11, as

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 31
well as those that relied on cross-sectional designs, r = .22, 95%CIs: [.18, .26]; Z = 10.04. The
effect size was larger for correlational studies but the between-group difference was not greater
than expected due to sampling error, Q(1) = 2.37, p = .124.
Continuous Moderation Analyses
Gender. Analysis revealed that the social identification–health link was weaker to the
extent that the proportion of female participants in the sample increased, b = –.0028, 95%CIs [–
.0047, –.0010], Z = 3.05, Q(1) = 9.28, p < .001.
Age. Analysis of the slope of the mean age did not differ statistically from zero, b =
.0019, 95%CIs [–.0033, .0071], Z = .72, Q(1) = .52, p = .510, providing no evidence that the
social identification–health link was moderated by age.
Status. To examine the extent to which social status of a job moderated the social
identification–health relationship, we analyzed all samples that afforded the possibility of coding
for this variable (k = 47, N = 17,977). The point estimate and confidence intervals of the
estimation of the slope of social class of the job was not statistically significant, b = –.0022,
95%CIs [–.0051, .0006], Z = 1.52, Q(1) = 2.31, p = .128, indicating that there was no evidence
that the social identification–health relationship was moderated by the social status of
participants’ job.
Inglehart-Welzel culture dimensions. We ran separate analysis to examine the extent to
which the social identification–health relationship was moderated by a country’s
traditional/secular values and survival/emancipative values (k = 54). There was no evidence that
the relationship was moderated by traditional/secular values, b = –.7982, 95%CIs [–1.8445,
.2481], Z = 1.50; or survival/emancipative values, b = –.0070, 95%CIs [–.7167, .7027], Z = .02.2
Shared social identification. To assess the extent to which sharedness in identification
moderated the social identification–health relationship, we analyzed all samples where

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 32
participants shared membership in, and indicated their identification with, the same group (k =
16, N = 8,179). The point estimate and confidence intervals of the estimation of the slope of SD
in identification was negative, b = –.424, 95%CIs [–.65, –.22], Z = 3.97, Q(1) = 15.74, p < .001,
indicating that the social identification–health link was moderated by SD in identification such
that the positive relationship between social identification and health became weaker to the extent
that SD in social identification decreased. Reversing the signs, this suggests that the positive
association between social identification and health became stronger as sharedness in
identification among group members increased.
We conducted further sensitivity analysis by controlling for the sample’s mean level of
identification (standardized to a 1 to 7-point scale) in order to examine whether effects of
sharedness are in fact driven by the mean level in identification (Harrison & Klein, 2007; see also
Liao, Liu, & Loi, 2010; Steffens, Shemla, Diestel, & Wegge, 2014). Analysis indicated that mean
social identification did not significantly moderate the social identification–health relationship, b
= .02, 95%CIs [–.10, .15], Z = .37, while the effect of SD in social identification remained largely
unchanged, b = –.39, 95%CIs [–.72, –.07], Z = 2.37. Overall, these results suggest that there is
limited evidence that the moderation of the social identification–health relationship by SD in
identification was accounted for by mean level of identification.
Multiple Regression Analyses
We conducted multiple moderator analysis to estimate the combined variance that can be
explained by the moderators that the previous analysis revealed to be significant (Borenstein et
al., 2009). The regression model that included all reliable moderators health valence, health
index, identification scale (as dummy coded variables), and gender, was significant, Q(6) =
35.25, p < .001, R2 = .115, indicating that overall the moderators accounted for 12% of the
between–study variance in effect sizes (i.e., of the distribution of true effect sizes across studies).

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 33
When controlling for each other, the regression weights were overall very similar to those
revealed by univariate analyses: health valence (b = .059, 95%CIs [–.012, .130], Z = 1.63), health
index (b = .075, 95%CIs [–.008, .158], Z = 1.77), identification scale (Q[3] = 15.24, p =.002), and
gender (b = –.0026, 95%CIs [–.0041, –.0012], Z = 3.64).
The overall regression model on the sub-sample of studies concerning SD in identification
was significant, Q(1) = 15.74, p < .001, R2 = .565, explaining 56.5% of the between–study
variance in effect sizes. When mean identification was added, the overall model was also
significant, Q(2) = 14.87, p = .001, R2 = .520, accounting for 52.0% of the between–study
variance in effect sizes (parameters of the explained variance can be lower in models with a
greater number of moderators due to imprecision in estimating parameters; Borenstein et al.,
2009).3
Discussion
The present research provides a quantitative review of an increasingly large field of
research that has interrogated the relationship between social identifications in organizations on
the one hand and individuals’ health on the other. The present study provides meta-analytic
results from primary published and unpublished studies that indicate that social identifications in
organizations are overall positively related to individuals’ health. Moreover, results reveal an
instructive estimation of the precise summary magnitude of the social identification–health
relationship of r = .21, which is of small-to-moderate size (Cohen, 1992). On the whole, results
suggest that rather than being related to employees’ exhaustion, social identification with (a) a
workgroup and (b) an organization is instead positively related to employees’ invigoration
(thereby providing support for an invigoration hypothesis).
We used several methods to estimate and correct for measurement and publication bias in
the identified set of studies. Fail-safe N for effect size analysis (Orwin, 1983) indicates that 217

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 34
unidentified samples (more than three times the number of identified samples) from the file
drawer with a correlation of zero would be needed to reduce the effect size to a small size of r =
.05. The funnel plot asymmetry test (Egger et al., 1997) revealed no evidence of publication bias
as shown by a relationship between a study’s effect size and standard error, while the trim and fill
method (Duval & Tweedie, 2000) revealed an unbiased effect size of a similar size of r = .18
after imputing potentially missing studies. Analysis of the set of studies by publication bias
(Egger et al., 2003) indicated a significant positive effect size in both unpublished and published
studies and no statistical evidence that the effect size differed as a function of publication status.
Furthermore, p-curve analysis (Simonsohn et al., 2014) revealed that the set of samples had
evidential value (showing no evidence of p-hacking across the samples), while yielding no
evidence that the evidential value of the set of studies was inadequate. Finally, cumulative meta-
analysis (Borenstein et al., 2009) furnished no evidence of publication bias due to small-study
effects (as shown by no evidence of a drift of the effect size as smaller studies are added to the
analysis), while analysis based on studies with larger sample sizes yielded an effect size of r =
.20 — a magnitude that is similar to that of the summary effect size and that has the same
substantive implications. In sum, results across the various techniques that estimate and correct
for measurement and publication bias reveal negligible impact of publication bias on the
estimated summary effect size in the present set of samples.
In addition to estimating the magnitude of the global summary effect and correcting for
measurement and publication bias, the present meta-analysis also revealed substantial variation in
effect size distribution. Accordingly, we examined several novel and theoretically relevant
variables as potential boundary and qualifying conditions of the observed relationship between
social identification and health. Here there was no evidence that the strength of this relationship
varied depending on the focus of identification (with a workgroup versus an organization).

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 35
However, results indicate (a) that social identification was associated more strongly with
(positive) experiences of health than with the absence of (negative) experiences of ill-health, and
that the identification–health relationship was (b) stronger for psychological than physical health,
(c) stronger in samples that used the Doosje et al. (1995) social identification scale (rather than
other identification scales such as the Mael & Ashforth, 1992, or the van Dick et al., 2004, scale),
and (d) stronger to the extent that the proportion of female participants in the sample decreased.
Finally, there was evidence that the social identification–health relationship became (e) more
pronounced with increasing levels of sharedness in identification.
Social Identification in Organizations as a Backbone to Individuals’ Health
The present research substantiates assertions that individuals’ self-categorization in terms
of a workgroup and an organization has implications for stress and well-being in the workplace
(van Dick & Haslam, 2012). At the same time, it extends our understanding of previous empirical
studies by providing a more comprehensive, precise, and quantitative assessment of the
identification–health relationship. Here the evidence reveals that by and large individuals’
strength of internalization of a workgroup and organization-based membership relates to their
experience of diminished discomfort and stress as well as enhanced comfort and well-being. In
this way, the present research expands upon a wider literature that has asserted that health and
well-being are influenced by social factors in general and by issues of group membership and
identity in particular (Adler, Lodi-Smith, Philippe, & Houle, 2016; Cruwys et al., 2014; Diener &
Biswas-Diener, 2008; Drury, 2012; Helliwell & Putnam, 2004; Jetten et al., 2012; Jetten, Haslam,
Haslam, Dingle, & Jones, 2014; Sani et al., 2012). In particular, in providing meta-analytic
support for an invigoration hypothesis it lends substance to the claim that overall social
identification is best conceptualized as a resource rather than a liability in relation to people’s
health and well-being at work.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 36
Previous research on social identification in organizational contexts has focused largely
on workgroup and organizational identification as contributors to classic work-related outcomes
such as motivation, in-role and extra-role performance, or turnover intentions (Ng, 2015; Riketta,
2005; Riketta & van Dick, 2005; van Dick, 2004; van Knippenberg & van Schie, 2000) but had
paid far less attention to the role these play to the domain of health (Jetten et al., 2012). As well
as complementing research that has advanced our understanding of motivation and performance
in organizational contexts, these findings are particularly timely as research has started to point to
ways in which people’s health is a building block for subsequent performance and success
(Lyubomirsky et al., 2005).
Speaking to the literature on health in organizations (Lundberg & Cooper, 2011), findings
indicate that social identification is an energizing source, whose significance aligns with (and is
at least as marked as in the case of) resources that other meta-analyses have pointed to, such as
social support (Thoits, 1995; Viswesvaran, Sanchez, & Fisher, 1999), hardiness (Eschleman,
Bowling, & Alarcon, 2010), humor in the workplace (Mesmer-Magnus, Glew, & Viswesvaran,
2012), or the promotion of individuals’ mindfulness, mediation, or relaxation (Grossman
Niemann, Schmidt, & Walach, 2004; Richardson & Rothstein, 2008). At the same time, the
present findings have unique theoretical and practical value because they suggest that
individuals’ stress and health at work are grounded not just in individuals’ personal capabilities
and shortcomings but also in groups and organizations as a whole. In particular, it would appear
that organizations that seek to foster the health of their workforce are likely to be more successful
if they (a) allow members to contribute to and shape organizational identity so that people are
able to feel ‘at home’ in the workplace (Knight & Haslam, 2010b) or (b) create structures that
facilitate the development of identities associated with lower-level workgroups (Hornsey &
Hogg, 2000).

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 37
Boundary Conditions of the Identification–Health Relationship
It is worth noting that even though the observed relationship between social identification
and health was positive, there was significant variation in its strength. To help understand the
nature and source of this variability, we examined the role of a range of theoretically important
factors. In what follows, we focus on those factors that have greatest theoretical relevance and on
those that exploratory analyses revealed as significant boundary and qualifying conditions of the
identification–health relationship.
Identification focus: Workgroup and organizational identification. Analysis revealed
that individuals’ health was positively related to their social identification with both their
workgroup and their organization. In this regard, literatures on salience of more proximal groups
(Ashforth & Johnson, 2001; van Knippenberg & van Schie, 2000) and group socialization
(Carron & Spink, 1995; Moreland & Levine, 2001) might have led one to expect that
identification with a workgroup would be associated with stronger health effects; while other
work on stability and continuity (Albert & Whetten, 1985; Sani, 2008) as well as public regard of
collectives (Dutton et al., 1994; Luhtanen & Crocker, 1992) might have led one to expect the
opposite. Overall, however — perhaps because these competing influences are equally important
— the present findings reveal that the strength of the identification–health relationship did not
differ as a function of identification focus but was of comparable magnitude for organizational
identification (r = .21) and for workgroup identification (r = .21).
Health valence: Presence of well-being and absence of stress. Our analysis provides
clear evidence that social identification is related to the absence of (negative) forms of ill-health
as well as the presence of (positive) forms of well-being. Nevertheless, social identification had
somewhat stronger benefits for the enhancement of salutogenic aspects of health than for the
prevention of pathogenic ones. More broadly, in indicating that social identification feeds more

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 38
strongly into the promotion of what is ‘good for us’ than into the prevention of what is ‘bad for
us’, these findings support previous work which has made the point that the absence of stress is
not equivalent to the presence of well-being (Diener, 2000; Schaufeli, Salanova, González-Romá,
& Bakker, 2002; Seligman & Csikszentmihalyi, 2000). More specifically, findings are consistent
with a psychological conception of positive human health (Ryff & Singer, 1998; Thoits, 1994)
and phenomenological accounts of social identification which suggest that increases in
identification capture increases in positive experiences (Haslam et al., 2009) and are therefore
related particularly strongly to positive forms of well-being. As a corollary, although increases in
identification also involve the amelioration of suffering (Branscombe et al., 2009), such
consequences are less pronounced (in part too because the absence of identification is more likely
to relate to a lack of positive experiences rather than to the presence of negative ones).
Indeed, it is possible that compared to the promotion of well-being that relates more
strongly to social identifications in organizations, the prevention of ill-health may depend to a
larger degree on factors other than identification with a single work-related entity (e.g.,
identification with other non work-related groups as well as multiple group memberships; see
Cruwys et al., 2013; Iyer, Jetten, Tsivrikos, Postmes, & Haslam, 2009; van Steenbergen,
Ellemers, Haslam, & Urlings, 2008). Similarly, it is conceivable that increasing levels of
disidentification (i.e., actively distancing self from the group; Kreiner & Ashforth, 2004; Pratt,
2000) are more strongly related to the presence of unease, discomfort, and stress than to the
absence of ease, comfort, and well-being. These are interesting lines of inquiry that future
research should address more programmatically.
Health index: Physical and psychological health. Consistent with the notion of
compatibility in levels-of-analysis (Fishbein & Ajzen, 1974; Riketta & van Dick, 2005), the
present findings indicate that identification–health relationships are particularly strong to the

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 39
extent that social identification relates to the psychological experience of health. Nevertheless, it
is noteworthy that the identification–health link was also positive (albeit weaker) in relation to
physical health (e.g., as assessed by physiological indicators). In the first instance, we would note
that common-method variance may have contributed to this differential association (Podsakoff,
MacKenzie, Lee, & Podsakoff, 2003). However, it seems unlikely that common-method variance
can account fully for this divergence not least because some studies relied on experimental
designs and some of the physical health indicators relied on self-report (e.g., symptom
checklists). Taken as a whole, rather than providing support for a bio-medical model of stress and
well-being, findings thus demonstrate the usefulness of bio-psycho-social models which suggest
that physical and physiological markers of health are determined in no small part by the broader
social-psychological environment in which people find themselves (Cacioppo, Berntson,
Sheridan, & McClintock, 2000; Jetten et al., 2012; Helliwell & Putnam, 2004).
Shared social identification. Finally, the present research breaks new ground in being
the first to review a large empirical domain in a way that allowed us to investigate the extent to
which shared social identification has particular significance for stress and well-being. Here
findings indicate that people’s experience of better health was particularly strongly associated
with increased social identification when that social identification was shared among group
members (accounting for more than 52% of the between–study variance in this sample, while
also indicating that this finding was not driven by the mean level of identification). In this way, it
appears that in addition to being influenced by personal experiences of the relationship with a
collective entity, individuals’ health and well-being are also shaped by other group members’
experiences of their relationship with that entity.
Speaking to the construct of social identification itself, these findings also underscore the
point that sharedness infuses this with a qualitatively different meaning, and hence that this is an

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 40
important (but typically overlooked) aspect of social identification (Jans et al., 2015; Tajfel,
1981; see also Alnabulsi & Drury, 2014; Pandey, Stevenson, Shankar, Hopkins, & Reicher, 2014;
Stott & Drury, 2000). Indeed, the social component of identification matters not only because
individuals internalize a (social) group as part of their self-concept but also because, and to the
extent that, identification shifts from being a merely personally referenced experience to one that
is collectively shared.
Additional findings. The vast majority of studies employed cross-sectional designs.
Nevertheless, it is important to note that the social identification–health relationship was
consistently positive across different methodologies such that the relationship was found both in
studies that employed cross-sectional designs and in those that employed experimental or
longitudinal designs, providing some evidence of directionality. Moreover, the relationship was
stronger the greater the proportion of male participants in a sample. We cannot know with any
certainty why this was the case, but only speculate. Nevertheless, in the first instance it is worth
noting this finding could be seen as surprising in light of evidence that women tend to have larger
social support networks than men (Kendler, Myers, & Prescott, 2005). At the same time, this
finding is in line with claims and evidence that men tend to place particular emphasis on their
collective self and that, compared to female workers, males tend to show stronger collective
identification and greater satisfaction with group tasks (Zhang, Chen, Chen, Liu, & Johnson,
2014). It is also entirely possible that these gender differences reflect the fact that many
workplaces are stereotypically masculine (e.g., in terms of their culture) and that women are often
a minority within organizational contexts. These two factors mean that female workers routinely
encounter more obstacles to identification and experience greater difficulty ‘fitting in’ than their
male counterparts (Peters, Ryan, Haslam, & Fernandes, 2012).
Finally, the identification–health link was strongest in studies that used the Doosje et al.’s

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 41
(1995) identification scale. This may in part be due to the fact that one of the four items in this
scale taps into satisfaction with membership in a group (“I am glad to be a member of the
group”), which may have a particularly strong bearing on people’s experiences of general health
and well-being than other identification items. At the same time, and perhaps less evident at first
sight, this stronger association may also in part be due the fact that two of the items refer to group
members rather than the group as a whole (“I identify with other group members”; “I feel strong
ties with other group members”; see also Karasawa, 1991), while most other scales refer only to
the group as a whole (including the most widely used scale by Mael & Ashforth, 1992, and the
scale by van Dick et al., 2004). It may be the case that items referring to group members are
related more strongly to health and well-being because they are more likely to tap into intragroup
attraction than intergroup relations. Indeed, the literature on social exclusion demonstrates that
when individuals feel that they are socially excluded by other close individuals this can have
devastating consequences on health and well-being (Leary, 1990; MacDonald & Leary, 2005). In
any event, these findings suggest that this issue of the relative importance of intragroup attraction
and intergroup relations for health is an important issue for future research to address and resolve.
Limitations and Roadmap for Future Research
Elaborating on aspects of the forgoing discussion, in what follows, we highlight five
themes that emerged from the present findings and that will be important for future research to
address. First, the global effect sizes revealed in the present analysis support the notion that social
identification is associated with individuals’ feelings of vitalization rather than exhaustion. While
this appears to be true in general terms, it would nevertheless be interesting to understand the
precise processes that underpin this relationship by disentangling the various elements that are
bound up with social identification and that may have differential bearing on people’s health and
well-being. Specifically, the relationship between increasing social identification and health is

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 42
likely to be complex such that it might be determined not only by the specific beliefs, norms,
values, and ideals that a workgroup or organization promotes (which may be more or less
conducive to a healthy habits; Oyserman et al., 2007; Tarrant et al., 2012) but also by the
(negative or positive) treatment that it receives from other groups (Amiot, Terry, & Callan, 2007;
Jetten, Branscombe, Schmitt, & Spears, 2001; Pascoe & Smart Richman, 2009; Postmes &
Branscombe, 2002; for a recent review see Schmitt, Branscombe, Postmes, & Garcia, 2014).
More generally, it is likely that increasing identification feeds into a health-promoting
pathway by enhancing an individual's sense of purpose, belonging, and collective self-efficacy
(Cruwys et al., 2014), as well as by laying the foundations for the receipt and positive construal
of social support (Levine et al., 2005; Platow et al., 2007). On the other hand, elevated social
identification in the workplace may also feed into a health-diminishing pathway to the extent that
it compromises individuals’ capacity to build and maintain memberships in other relevant groups
(van Steenbergen et al., 2008). To provide us with a richer understanding of the various
implications of identification for health, there would clearly be value in research that disentangles
these processes as well as investigates their simultaneous effects (e.g., by means of elaborate
experimental and longitudinal designs that control for the influences of each other).
Second, the notion of sharedness as a consequential part of people’s identification with a
group (in addition to personally referenced identification) strikes us as particularly interesting and
hence worth exploring in more detail. Nevertheless, one of the obvious reasons why this aspect of
social identification has tended to be overlooked in previous research is that most standard
identification scales assess identification in terms of individual group members’ self-
categorization as group members without explicit reference to their sense of shared identification
(e.g., Abrams et al., 1998; Brown et al., 1986; Cameron, 2004; Doosje et al., 1995; Ellemers et
al., 1999; Jackson, 2002; Leach et al., 2008; Mael & Ashforth, 1992; Mael & Tetrick, 1992;

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 43
Roccas et al., 2008; for a review, see Postmes et al., 2013). Thus while identification measures
are highly reliable and clearly provide extremely valuable information in their standard form
(Ashforth et al., 2008), the present findings suggest that important information might be lost to
the extent that these fail to assess actual sharedness in identification. From a methodological
point of view, then, to provide a more in-depth understanding of the importance of sharedness, it
would be fruitful to employ multi-level approaches as well as to develop new forms of
assessment that tap into sharedness of social identification. It is also important to note that in the
present analysis we investigated actual sharedness in identification. In this regard, there would be
value in future work that examines when and why actual sharedness goes hand in hand with (or
diverges from) an individual’s sense of sharedness as well as the unique and overlapping
consequences of both for their experience of health.
Third, there is a need for more studies that employ experimental and intervention as well
as longitudinal designs to examine the impact of organizational identifications on individuals’
health. In this regard, future research should expand our understanding of the various factors that
can be harnessed (e.g., collective empowerment, collective decision making) with a view to
fostering social identification and health (e.g., Gleibs et al., 2011; Haslam, 2014). Indeed, given
that by and large organizational identification can be seen as a health resource at work, it appears
that organizations also have some responsibility for creating workspaces that are not only
motivating but also healthy. In this regard, future research is needed to shed light on the capacity
for social and organizational policies as well as job and work design (Parker, 2014) to build
social identification (e.g., in ways formulated by the ASPIRe model; Haslam, Eggins, &
Reynolds, 2003).
Fourth, the majority of studies that have investigated the relationship between social
identification in the workplace and health have focused on social identification as a uni-

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 44
dimensional construct. Very few studies, and too few for a systematic analysis, assessed sub-
dimensions of identification in relation to people’s health. However, there is evidence that social
identification comes in different forms (such as ambivalent identification; Ashforth, Rogers,
Pratt, & Pradies, 2014; Kreiner & Ashforth, 2004; Pratt, 2000) and can be conceptualized as
comprising different qualities and sub-dimensions (e.g., Ashmore, Deaux, & McLaughlin-Volpe,
2004; Bergami & Bagozzi, 2000; Cameron, 2004; Jackson, 2002; Leach et al., 2008). In future
work, examination of various sub-components and forms of identification may be interesting
because it is conceivable that different components have a differentiated impact on particular
health outcomes (cf. Harris & Cameron, 2005). To provide a more complete understanding of the
identification–health relationship, future research should therefore attempt to disentangle the
ways in which health is influenced by the multi-faceted forms that social identification can take.
Fifth, although a body of research and theory indicates that health is influenced by social
identification with a single group at work, little research has examined health as a function of
social identification in relation to various other (work and leisure) groups (Ramarajan & Reid,
2013; van Steenbergen et al., 2008). This is important because emerging evidence suggests that
people’s well-being as well as their capacity to adjust to new life circumstances is determined by
the multiple identities that they see as important to self (Cruwys et al., 2013; Haslam et al., 2008;
Jetten et al., 2012; Ramarajan, 2014; van Dick, van Knippenberg, Kerschreiter, Hertel, &
Wieseke, 2008). In relation to individuals’ health in organizations, future research might examine
in more detail the health consequences that flow not just from workplace identification but also
from simultaneous identifications with other relevant groups (e.g., recreational, religious, family
groups) as well as the interrelationships between multiple identities.
Conclusion
The present meta-analysis is the first to provide a comprehensive quantitative examination

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 45
of the large body of empirical data on the relationship between social identifications and health in
organizations. In this, it helps clarify the question of the extent to which increased social
identification with a workgroup and with an organization has beneficial or detrimental
implications for individuals’ health. Considered as a whole, evidence indicates that both
workgroup and organizational identification are associated with individuals’ experience of
reduced strain and burnout as well as greater health and well-being. Yet, beyond these main
effects, the present analysis also uncovers a range of important factors that fortify or attenuate the
invigorating impact of such identification.
The present meta-analysis is set against the backdrop of a large body of research that has
paid considerable attention to the ways in which social identification is a determinant of
significant organizational outcomes such as leadership, motivation, and communication. To date
however, this work has devoted far less energy to questions of individuals’ health and well-being
in organizations. The present findings are therefore noteworthy because they suggest that social
identifications in organizations are as important for individuals’ health and well-being as they are
for their productivity and performance. Indeed, in many ways we see these two correlates of
social identifications not only to be equally important, but also to be intertwined. Moreover, it is
precisely in helping us to understand the nature of this relationship between health and social
endeavor that much of the value of a social identity approach resides.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 46
Footnote
1. Visual inspection of the forest plot indicates some outliers to the left and the right side of the
summary effect that may account for considerable heterogeneity in effect sizes. Analysis
excluding the three most extreme effect sizes to either side of the summary effect indicates an
effect size of, r = .21, 95%CIs: [.18, .25]; Z = 11.71, and a heterogeneity of effect sizes across
samples of, Q(51) = 269.61, p < . 001, T = .111, with a 95% prediction interval of –.01 to .42.
2. In additional analyses, we also coded for Hofstede’s culture dimensions: power distance,
individualism, masculinity, uncertainty avoidance, pragmatism, and indulgence (Hofstede,
2001; Hofstede, Hofstede, & Minkov, 2010). For van Dick and colleagues’ (2007) sample
which included participants from different countries, sample scores were weighted by the
number of participants from a particular country. There was no evidence that the
identification–health relationship was moderated by any of the culture dimensions: power
distance: b = .0004, 95%CIs [–.0022, .0093], Z = 1.22; individualism: b = .0044, 95%CIs [–
.0007, .0095], Z = 1.68; masculinity: b = –.0003, 95%CIs [–.0061, .0056], Z = .08;
uncertainty avoidance: b = –.0008, 95%CIs [–.0037, .0020], Z = .57; pragmatism: b = .0004,
95%CIs [–.0028, .0036], Z = .23; or indulgence: b = –.0007, 95%CIs [–.0065, .0052], Z = .22.
3. If the previous moderators are added to this model (and two studies are removed due to
dependent subgroups in health valence and index), the overall model combining multiple
outcomes was significant, Q(7) = 34.04, p < .001, R2 = .641. The beta-weights were largely
similar to those revealed in previous analyses with the exceptions that health index and
identification measure were not significant predictors the strength of the effect sizes in this
model: SD in identification (b = –.58, 95%CIs [–.92, –.23], Z = 2.43), mean in identification
(b = –.05, 95%CIs [–.20, .09], Z = .76), health valence (b = –.10, 95%CIs [–.22, .02], Z =
1.66), health index (b = .19, 95%CIs [.02, .35], Z = 2.22), identification scale (Q[2] = 5.57, p

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 47
= .062; note that none of the samples in this subset used the measure by van Dick et al.
[2004]), and gender (b = –.0026, 95%CIs [–.0046, –.0006], Z = 2.56).

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 48
References
Abrams, D., Ando, K., & Hinkle, S. (1998). Psychological attachment to the group: Cross-
cultural differences in organizational identification and subjective norms as predictors of
workers' turnover intentions. Personality and Social Psychology Bulletin, 24, 1027-39.
Adler, J. M., Lodi-Smith, J., Philippe, F. L., & Houle, I. (2016). The incremental validity of
narrative identity in predicting well-being: A review of the field and recommendations for the
future. Personality and Social Psychology Review, 20, 142-175.
Albert, S., & Whetten, D. A. (1985). Organizational identity. In L. L. Cummings, & B. M. Staw
(Eds.), Research in Organizational Behavior, Vol. 7: 263-295. Greenwich, CT: JAI Press.
Alnabulsi, H., & Drury, J. (2014). Social identification moderates the effect of crowd density on
safety at the Hajj. Proceedings of the National Academy of Sciences, 111, 9091-9096.
Amiot, C. E., Terry, D. J., & Callan, V. J. (2007). Status, equity and social identification during
an intergroup merger: A longitudinal study. British Journal of Social Psychology, 46, 557-577.
Anisman, H. (2014). An introduction to stress and health. London: Sage.
Anderson, N., & Thomas, H. D. C. (1996). Workgroup socialization. In M. A. West (Ed.),
Handbook of workgroup psychology (pp. 423–450). Chichester: Wiley.
APA Publications and Communications Board Working Group on Journal Article Reporting
Standards (2008). Reporting standards for research in psychology: Why do we need them?
What might they be? American Psychologist, 63, 839–851.
Ashforth, B. E., & Johnson, S. A. (2001). Which hat to wear. In M. A. Hogg & D. J. Terry
(Eds.), Social identity processes in organizational contexts (pp. 31-48). New York US:
Psychology Press.
Ashforth, B. E., & Kreiner, G. E. (1999). “How can you do it?”: Dirty work and the challenge of
constructing a positive identity. Academy of Management Review, 24, 413-434.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 49
Ashforth, B. E., Harrison, S. H., & Corley, K. G. (2008). Identification in organizations: An
examination of four fundamental questions. Journal of Management, 34, 325-374.
Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy of
Management Review, 14, 20-39.
Ashforth, B. E., Rogers, K. M., Pratt, M. G., & Pradies, C. (2014). Ambivalence in organizations:
A multilevel approach. Organization Science, 25, 1453-1578.
Ashmore, R. D., Deaux, K., & McLaughlin-Volpe, T. (2004). An organizing framework for
collective identity: Articulation and significance of multidimensionality. Psychological
Bulletin, 130, 80-114.
*Avanzi, L., Schuh, S. C., Fraccaroli, F., van Dick, R. (2015). Why does organizational
identification relate to reduced employee burnout? The mediating influence of social support
and collective efficacy. Work & Stress, 29, 1-10.
*Avanzi, L., Fraccaroli, F., Sarchielli, G., Ullrich, J., & van Dick, R. (2014). Staying or leaving:
A combined social identity and social exchange approach to predicting employee turnover
intentions. International Journal of Productivity and Performance Management, 63, 272-289.
*Avanzi, L., van Dick, R., Fraccaroli, F., & Sarchielli, G. (2012). The downside of organizational
identification: Relationships between identification, workaholism and well-being. Work &
Stress, 26, 289-307.
Bakker, A. B., Schaufeli, W. B., Leiter, M. P., & Taris, T. W. (2008). Work engagement: An
emerging concept in occupational health psychology. Work & Stress, 22, 187-200.
*Barbier, M., Dardenne, B., & Hansez, I. (2013). A longitudinal test of the Job Demands–
Resources model using perceived stigma and social identity. European Journal of Work and
Organizational Psychology, 22, 532-546.
Barker, J., & Tompkins, P. (1994). Identification in the self-managing organization:

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 50
Characteristics of target and tenure. Human Communication Research, 21, 223-240.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117, 497-529.
*Bedeian, A. G. (2007). Even if the tower is “ivory,” it isn't “white:” Understanding the
consequences of faculty cynicism. Academy of Management Learning & Education, 6, 9-32.
Bergami, M., & Bagozzi, R. P. (2000). Self‐categorization, affective commitment and group self‐
esteem as distinct aspects of social identity in the organization. British Journal of Social
Psychology, 39, 555-577.
Berkman, L. F. (2001). Social ties and mental health. Journal of Urban health, 78, 458-467.
*Bizumic, B., Reynolds, K. J., Turner, J. C., Bromhead, D., & Subasic, E. (2009). The role of the
group in individual functioning: School identification and the psychological well‐being of
staff and students. Applied Psychology, 58, 171-192.
*Bjerregaard, K., Haslam, S. A., Morton, T. A., Ryan, M. K. (2014). The motivation of care
workers: Clarifying the importance of social and relational identification. Unpublished
manuscript, Psychology: College of Life and Environmental Science, University of Exeter,
UK.
Blader, S. L., & Tyler, T. R. (2009). Testing and extending the group engagement model:
Linkages between social identity, procedural justice, economic outcomes, and extrarole
behavior. Journal of Applied Psychology, 94, 445-464.
Blanchard, C. M., Amiot, C. E., Perreault, S., Vallerand, R. J., & Provencher, P. (2009).
Cohesiveness, coach's interpersonal style and psychological needs: Their effects on self-
determination and athletes' subjective well-being. Psychology of Sport and Exercise, 10, 545-
551.
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to meta-

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 51
analysis. Hoboken, NJ: Wiley.
Branscombe, N. R., Schmitt, M. T., & Harvey, R. D. (1999). Perceiving pervasive discrimination
among African Americans: Implications for group identification and well-being. Journal of
Personality and Social Psychology, 77, 135-149.
Brähler, E., Hinz, A., Scheer, J. W. (2008). GBB-24: The Giessener Symptom Checklist [GBB-
24: Der Giessener Beschwerdebogen]. Bern: Huber.
Brown, R., Condor, S., Mathews, A., Wade, G., & Williams, J. (1986). Explaining intergroup
differentiation in an industrial organization. Journal of Occupational Psychology, 59, 273-286.
Cacioppo, J. T., Hawkley, L. C., & Berntson, G. G. (2003). The anatomy of loneliness. Current
Directions in Psychological Science, 12, 71-74.
Cacioppo, J. T., Berntson, G. G., Sheridan, J. F., & McClintock, M. K. (2000). Multilevel
integrative analyses of human behavior: Social neuroscience and the complementing nature of
social and biological approaches. Psychological Bulletin, 126, 829-843.
Cameron, J. E. (2004). A three-factor model of social identity. Self and identity, 3, 239-262.
Campbell, D. T. (1958). Common fate, similarity, and other indices of the status of aggregates of
persons as social entities. Behavioral Science, 3, 14-25.
Cantor, N., Kemmelmeier, M., Basten, J., & Prentice, D. A. (2002). Life task pursuit in social
groups: Balancing self-exploration and social integration. Self & Identity, 1, 177-184.
Carron, A. V., & Spink, K. S. (1995). The group size-cohesion relationship in minimal
groups. Small Group Research, 26, 86-105.
*Christ, O. (2004). An examination of the transactional stress theory in teachers in training [Die
Überprüfung der transaktionalen Stresstheorie im Lehramtsreferendariat]. Unpublished raw
data, Department of Psychology, Goethe University Frankfurt, Germany.
*Cicero, L., Pierro, A., & van Knippenberg, D. (2007). Leader group prototypicality and job

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 52
satisfaction: The moderating role of job stress and team identification. Group Dynamics:
Theory, Research, and Practice, 11, 165-175.
Cohen, S. (2004). Social relationships and health. American Psychologist, 59, 676.
Cohen, S., Doyle, W. J., Skoner, D. P., Rabin, B. S., & Gwaltney, J. M. (1997). Social ties and
susceptibility to the common cold. JAMA, 277, 1940-1944.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived
stress. Journal of Health and Social Behavior, 24, 385-396.
Cooper, H. M. (1998). Synthesizing research: A guide for literature reviews (3rd ed.). Thousand
Oaks, CA: Sage.
Cooper, H., Robinson, J. C., & Patall, E. A. (2006). Does homework improve academic
achievement? A synthesis of research, 1987–2003. Review of Educational Research, 76, 1-62.
Crabtree, J. W., Haslam, S. A., Postmes, T., & Haslam, C. (2010). Mental health support groups,
stigma, and self‐esteem: Positive and negative implications of group identification. Journal of
Social Issues, 66, 553-569.
Cruwys, T., Dingle, G. A., Haslam, C., Haslam, S. A., Jetten, J., & Morton, T. A. (2013). Social
group memberships protect against future depression, alleviate depression symptoms and
prevent depression relapse. Social Science & Medicine, 98, 179-186.
Cruwys, T., Haslam, S. A., Dingle, G. A., Haslam, C., & Jetten, J. (2014). Depression and social
identity: An integrative review. Personality and Social Psychology Review, 18, 215-238.
Cruwys, T., South, E. I., Greenaway, K. H., & Haslam, S. A. (2015). Social identity reduces
depression by fostering positive attributions. Social Psychological and Personality Science, 6,
65-74.
Cumming, G. (2014). The new statistics: Why and how. Psychological Science, 25, 7–29.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 53
*Das, D., Dharwadkar, R., & Brandes, P. (2008). The importance of being ‘Indian': Identity
centrality and work outcomes in an off-shored call center in India. Human Relations, 61,
1499-1530.
Diener, E. (2000). Subjective well-being: The science of happiness and a proposal for a national
index. American Psychologist, 55, 34-43.
Diener, E., & Biswas-Diener, R. (2008). Happiness: Unlocking the mysteries of psychological
wealth. Malden US: Blackwell.
Doosje, B., Ellemers, N., & Spears, R. (1995). Perceived intragroup variability as a function of
group status and identification. Journal of Experimental Social Psychology, 31, 410-436.
Drury, J. (2012). Collective resilience in mass emergencies and disasters. In J. Jetten, C. Haslam,
& S. A. Haslam (Eds.), The Social Cure: Identity, Health and Well-being (pp. 195–215). New
York: Psychology Press.
Drury, J., Cocking, C., & Reicher, S. (2009). Everyone for themselves? A comparative study of
crowd solidarity among emergency survivors. British Journal of Social Psychology, 48, 487-
506.
Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organizational images and member
identification. Administrative Science Quarterly, 39, 239-263.
Duval, S., & Tweedie, R. (2000). Trim and fill: A simple funnel‐plot–based method of testing
and adjusting for publication bias in meta‐analysis. Biometrics, 56, 455-463.
Egger, M., Juni, P., Bartlett, C., Holenstein, F., & Sterne, J. (2003). How important are
comprehensive literature searches and the assessment of trial quality in systematic reviews?
Empirical study. Health Technology Assessment, 7, 1-76.
Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a
simple, graphical test. British Medical Journal, 315, 629-634.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 54
Ellemers, N. (1993). The influence of socio-structural variables on identity management
strategies. European Review of Social Psychology, 4, 27-57.
Ellemers, N. (2012). The group self. Science, 336, 848-852.
Ellemers, N., De Gilder, D., & Haslam, S. A. (2004). Motivating individuals and groups at work:
A social identity perspective on leadership and group performance. Academy of Management
Review, 29, 459-478.
Ellemers, N., Kortekaas, P., & Ouwerkerk, J. W. (1999). Self-categorisation, commitment to the
group and group self-esteem as related but distinct aspects of social identity. European
Journal of Social Psychology, 29, 371-389.
Escartín, J., Ullrich, J., Zapf, D., Schlüter, E., & van Dick, R. (2013). Individual‐and group‐level
effects of social identification on workplace bullying. European Journal of Work and
Organizational Psychology, 22, 182-193.
Eschleman, K. J., Bowling, N. A., & Alarcon, G. M. (2010). A meta-analytic examination of
hardiness. International Journal of Stress Management, 17, 277-307.
*Fampri, R. (2003). Does questionnaire framing affect levels of self-reported organizational
identification and burnout? Unpublished raw data, Psychology: College of Life and
Environmental Science, University of Exeter, UK.
Ferguson, C. J., & Brannick, M. T. (2012). Publication bias in psychological science: Prevalence,
methods for identifying and controlling, and implications for the use of meta-
analyses. Psychological Methods, 17, 120.
Fishbein, M., & Ajzen, I. (1974). Attitudes towards objects as predictors of single and multiple
behavioral criteria. Psychological Review, 81, 29–74.
*Frisch, J. U., Häusser, J. A., van Dick, R., & Mojzisch, A. (2014). Making support work: The

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 55
interplay between social support and social identity. Journal of Experimental Social
Psychology, 55, 154-161.
Gallo, L. C., & Matthews, K. A. (2003). Understanding the association between socioeconomic
status and physical health: Do negative emotions play a role? Psychological Bulletin, 129, 10-
51.
*Galang, M., & Jones, S. E. (2014). Workplace bullying and job satisfaction: The mediating role
of social identity. Unpublished Manuscript, Oxford Brookes University, UK.
Gallagher, S., Meaney, S., & Muldoon, O. T. (2014). Social identity influences stress appraisals
and cardiovascular reactions to acute stress exposure. British Journal of Health Psychology.
19, 566-579.
Gleibs, I. H., Haslam, C., Jones, J. M., Haslam, S. A., McNeill, J., & Connolly, H. (2011). No
country for old men? The role of a ‘Gentlemen's Club’ in promoting social engagement and
psychological well-being in residential care. Aging and Mental Health, 15, 456-466.
Golden, L., & Wiens-Tuers, B. (2006). To your happiness? Extra hours of labor supply and
worker well-being. The Journal of Socio-Economics, 35, 382-397
Goldberg, D. P. (1972). The detection of psychiatric illness by questionnaire. London: Oxford
University Press.
Graeser, S. (2011). Salutogenic factors for mental health promotion in work settings and
organizations. International Review of Psychiatry, 23, 508-515.
Greenaway, K., Haslam, S. A., Branscombe, N. R., Cruwys, T., Ysseldyk, R., & Heldreth, C.
(2015). From "we" to "me": Group identification enhances perceived personal control with
consequences for health and well-being. Journal of Personality and Social Psychology, 1, 53-
74.
Griffith, J. (2002). Multilevel analysis of cohesion's relation to stress, well-being, identification,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 56
disintegration, and perceived combat readiness. Military Psychology, 14, 217-239.
Grossman, P., Niemann, L., Schmidt, S., & Walach, H. (2004). Mindfulness-based stress
reduction and health benefits: A meta-analysis. Journal of Psychosomatic Research, 57, 35-
43.
*Grubba, C., Ahlswede, O. (2002). Relationships between job satisfaction, team climate,
identification, and engagement in the banking sector [Zusammenhänge zwischen
Arbeitszufriedenheit, Teamklima, Identifikation und Engangement im Bankenbereich].
Unpublished raw data, Department of Psychology, Goethe University Frankfurt, Germany.
Hamilton, D. L., & Sherman, S. J. (1996). Perceiving persons and groups. Psychological
Review, 103, 336-355.
*Harris, G. E., & Cameron, J. E. (2005). Multiple dimensions of organizational identification and
commitment as predictors of turnover intentions and psychological well-being. Canadian
Journal of Behavioural Science, 37, 159-169.
Harrison, D. A., & Klein, K. J. (2007). What's the difference? Diversity constructs as separation,
variety, or disparity in organizations. Academy of Management Review, 32, 1199-1228.
Haslam, C., Holme, A., Haslam, S. A., Iyer, A., Jetten, J., & Williams, W. H. (2008).
Maintaining group memberships: Social identity continuity predicts well-being after
stroke. Neuropsychological Rehabilitation, 18, 671-691.
Haslam, S. A. (1997). Stereotyping and social influence: Foundations of stereotype consensus.
In: R. Spears, P. J. Oakes, N. Ellemers & S. A. Haslam (Eds.), The social psychology of
stereotyping and group life (pp. 119-143). Oxford, UK: Blackwell.
Haslam, S. A. (2001). Psychology in organizations: The social identity approach. London &
Thousands Oaks, CA US: Sage.
Haslam, S. A. (2014). Making good theory practical: Five lessons for an Applied Social Identity

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 57
Approach to challenges of organizational, health, and clinical psychology. British Journal of
Social Psychology, 53, 1-20.
Haslam, S. A., Eggins, R. A., & Reynolds, K. J. (2003). The ASPIRe model: Actualizing Social
and Personal Identity Resources to enhance organizational outcomes. Journal of Occupational
and Organizational Psychology, 76, 83-113.
Haslam, S. A., Jetten, J., Postmes, T., & Haslam, C. (2009). Social identity, health and well-
being: An emerging agenda for applied psychology. Applied Psychology: An International
Review, 58, 1-23.
*Haslam, S. A., Jetten, J., & Waghorn, C. (2009). Social identification, stress and citizenship in
teams: A five‐phase longitudinal study. Stress and Health, 25, 21-30.
*Haslam, S. A., O'Brien, A., Jetten, J., Vormedal, K., & Penna, S. (2005). Taking the strain:
Social identity, social support, and the experience of stress. British Journal of Social
Psychology, 44, 355-370.
*Haslam, S. A., & Reicher, S. (2006). Stressing the group: Social identity and the unfolding
dynamics of responses to stress. Journal of Applied Psychology, 91, 1037-1052.
Haslam, S. A., Reicher, S. D., & Platow, M. J. (2011). The new psychology of leadership:
Identity, influence and power. London & New York: Psychology Press.
*Häusser, J. A., Kattenstroth, M., van Dick, R., & Mojzisch, A. (2012). “We” are not stressed:
Social identity in groups buffers neuroendocrine stress reactions. Journal of Experimental
Social Psychology, 48, 973-977.
Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis. Orlando, FL: Academic
Press.
Helliwell, J. F., & Putnam, R. D. (2004). The social context of well-being. Philosophical
Transactions of the Royal Society London, Biological Sciences, 359, 1435-1446.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 58
Helliwell, J. F., Layard, R., & Sachs, J. (eds.) (2013). World Happiness Report 2013. New York:
UN Sustainable Development Solutions Network.
*Herrbach, O. (2006). A matter of feeling? The affective tone of organizational commitment and
identification. Journal of Organizational Behavior, 27, 629-643.
Higgins, J., Thompson, S. G., & Spiegelhalter, D. J. (2009). A re‐evaluation of random‐effects
meta‐analysis. Journal of the Royal Statistical Society: Series A (Statistics in Society), 172,
137-159.
Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions, and
organizations across nations (2nd Edition). Thousand Oaks CA: Sage Publications.
Hofstede, G., Hofstede, G. J., & Minkov, M. (2010). Culture and organizations, Software of the
mind: Intercultural cooperation and its importance for survival (3rd Ed). New York: McGraw
Hill.
Hogg, M. A., Abrams, D. (1988). Social Identifications: A social psychology of intergroup
relations and group processes. London & New York: Routledge.
Hogg, M. A., & Terry, D. J. (Eds.) (2001). Social identity processes in organizational contexts.
Philadelphia, Hove & PA US: Psychology Press.
Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A
meta-analytic review. PLoS Medicine, 7, e1000316.
Hopkins, N., & Reicher, S. (2016). The psychology of health and well-being in mass gatherings:
A review and a research agenda. Journal of Epidemiology and Global Health, 6, 49-57.
Hornsey, M. J., & Hogg, M. A. (2000). Assimilation and diversity: An integrative model of
subgroup relations. Personality and Social Psychology Review, 4, 143-156.
*Horton, K. E., McClelland, C. R., & Griffin, M. A. (2014). Defined by our hierarchy? How
hierarchical positions shape our identifications and well-being at work. Human Relations, 67,

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 59
1167-1188.
Howell, J. L., Koudenburg, N., Loschelder, D. D., Weston, D., Fransen, K., De Dominicis, S., ...
& Haslam, S. A. (2014). Happy but unhealthy: The relationship between social ties and health
in an emerging network. European Journal of Social Psychology, 44, 612-621.
Hunter, J. E., & Schmidt, F. L. (2000). Fixed effects vs. random effects meta‐analysis models:
Implications for cumulative research knowledge. International Journal of Selection and
Assessment, 8, 275-292.
Inglehart, R., & Baker, W. E. (2000). Modernization, cultural change, and the persistence of
traditional values. American Sociological Review, 65, 19-51.
*Ishii, K. (2012). Dual organizational identification among Japanese expatriates: The role of
communication in cultivating subsidiary identification and outcomes. The International
Journal of Human Resource Management, 23, 1113-1128.
Iyer, A., Jetten, J., Tsivrikos, D., Postmes, T., & Haslam, S. A. (2009). The more (and the more
compatible) the merrier: Multiple group memberships and identity compatibility as predictors
of adjustment after life transitions. British Journal of Social Psychology, 48, 707-733.
Jackson, J. W. (2002). Intergroup attitudes as a function of different dimensions of group
identification and perceived intergroup conflict. Self and Identity, 1, 11-33.
Jans, L., Leach, C. W., Garcia, R. L., & Postmes, T. (2015). The development of group influence
on in-group identification: A multilevel approach. Group Processes & Intergroup Relations,
18, 190-209.
Jans, L., Postmes, T., & Van der Zee, K. I. (2011). The induction of shared identity: The positive
role of individual distinctiveness for groups. Personality and Social Psychology Bulletin, 37,
1130-1141.
Jetten, J., Branscombe, N. R., Schmitt, M. T., & Spears, R. (2001). Rebels with a cause: Group

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 60
identification as a response to perceived discrimination from the mainstream. Personality and
Social Psychology Bulletin, 27, 1204-1213.
Jetten, J., Haslam, C., & Haslam, S. A. (Eds.) (2012). The social cure: Identity, health and well-
being. New York, NY US: Psychology Press.
Jetten, J., Haslam, C., Haslam, S. A., Dingle, G., & Jones, J. M. (2014). How groups affect our
health and well‐being: The path from theory to policy. Social Issues and Policy Review, 8,
103-130.
*Jimmieson, N. L., McKimmie, B. M., Hannam, R. L., & Gallagher, J. (2010). An investigation
of the stress-buffering effects of social support in the occupational stress process as a function
of team identification. Group Dynamics: Theory, Research, and Practice, 14, 350-367.
Karasawa, M. (1991). Toward an assessment of social identity: The structure of group
identification and its effects on in‐group evaluations. British Journal of Social Psychology, 30,
293-307.
Kepes, S., Banks, G. C., McDaniel, M., & Whetzel, D. L. (2012). Publication bias in the
organizational sciences. Organizational Research Methods, 15, 624-662.
Kendler, K. S., Myers, J., & Prescott, C. A. (2005). Sex differences in the relationship between
social support and risk for major depression: a longitudinal study of opposite-sex twin
pairs. American Journal of Psychiatry, 162, 250-256.
*Knight, C., & Haslam, S. A. (2010a). Your place or mine? Organizational identification and
comfort as mediators of relationships between the managerial control of workspace and
employees' satisfaction and well‐being. British Journal of Management, 21, 717-735.
*Knight, C., & Haslam, S. A. (2010b). The relative merits of lean, enriched, and empowered
offices: An experimental examination of the impact of workspace management strategies on
well-being and productivity. Journal of Experimental Psychology: Applied, 16, 158.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 61
*Knowles, M., & Smith, L. G. E. (2013). Occupational stress and perceived job performance:
The moderating role of team identification. Unpublished Manuscript, University of Bath, UK.
*Kreiner, G. E., & Ashforth, B. E. (2004). Evidence toward an expanded model of organizational
identification. Journal of Organizational Behavior, 25, 1-27.
Lambert, P. S., & Prandy, K. (2014) CAMSIS project webpages: Cambridge Social Interaction
and Stratification Scales, Retrieved 15 January 2015 from http://www.camsis.stir.ac.uk/ .
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
Leach, C. W., van Zomeren, M., Zebel, S., Vliek, M. L., Pennekamp, S. F., Doosje, B., ... &
Spears, R. (2008). Group-level self-definition and self-investment: a hierarchical
(multicomponent) model of in-group identification. Journal of Personality and Social
Psychology, 95, 144-165.
Leary, M. R. (1990). Responses to social exclusion: Social anxiety, jealousy, loneliness,
depression, and low self-esteem. Journal of Social and Clinical Psychology, 9, 221-229.
Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic examination of the correlates of the three
dimensions of job burnout. Journal of Applied Psychology, 81, 123-133.
Levine, M., Cassidy, C., Brazier, G., & Reicher, S. D. (2002). Self‐categorization and bystander
non‐intervention: Two experimental studies. Journal of Applied Social Psychology, 32, 1452-
1463.
Levine, M., Prosser, A., Evans, D., & Reicher, S. D. (2005). Identity and emergency intervention:
How social group membership and inclusiveness of group boundaries shape helping
behavior. Personality and Social Psychology Bulletin, 31, 443-453.
Levine, R. M., & Reicher, S. D. (1996). Making sense of symptoms: Self‐categorization and the
meaning of illness and injury. British Journal of Social Psychology, 35, 245-256.
Lickel, B., Hamilton, D. L., Wieczorkowska, G., Lewis, A., Sherman, S. J., & Uhles, A. N.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 62
(2000). Varieties of groups and the perception of group entitativity. Journal of Personality and
Social Psychology, 78, 223-246.
Liao, H., Liu, D., & Loi, R. (2010). Looking at both sides of the social exchange coin: A social
cognitive perspective on the joint effects of relationship quality and differentiation on
creativity. Academy of Management Journal, 53, 1090-1109.
Lindström, B., & Eriksson, M. (2006). Contextualizing salutogenesis and Antonovsky in public
health development. Health Promotion International, 21, 238-244.
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. London, England: Sage.
Lundberg, U., & Cooper, C. L. (2011). The science of occupational health: stress, psychobiology,
and the new world of work. Oxford, UK: John Wiley & Sons.
Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: Self-evaluation of one's social
identity. Personality and Social Psychology Bulletin, 18, 302-318.
Lyubomirsky, S., King, L., & Diener, E. (2005). The benefits of frequent positive affect: Does
happiness lead to success? Psychological Bulletin, 131, 803-855.
MacDonald, G., & Leary, M. R. (2005). Why does social exclusion hurt? The relationship
between social and physical pain. Psychological Bulletin, 131, 202-223.
Mael, F., & Ashforth, B. E. (1992). Alumni and their alma mater: A partial test of the
reformulated model of organizational identification. Journal of Organizational Behavior, 13,
103-123.
Mael, F. A., & Tetrick, L. E. (1992). Identifying organizational identification. Educational and
Psychological Measurement, 52, 813-824.
Markus, H. R., & Kitayama, S. (1994). A collective fear of the collective: Implications for selves
and theories of selves. Personality and Social Psychology Bulletin, 20, 568-579.
Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 63
Organizational Behavior, 2, 99-113.
Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of
Psychology, 52, 397-422.
*Matheson, K., & Cole, B. M. (2004). Coping with a threatened group identity: Psychosocial and
neuroendocrine responses. Journal of Experimental Social Psychology, 40, 777-786.
McAuley, L., Tugwell, P., & Moher, D. (2000). Does the inclusion of grey literature influence
estimates of intervention effectiveness reported in meta-analyses? The Lancet, 356, 1228-
1231.
*Menzel, B. (2007). Work motivation and identification in nurses [Arbetismotivation und
Identifikation bei Krankenpflegepersonal]. Unpublished raw data, Department of Psychology,
Goethe University Frankfurt, Germany.
*Merecz, D., & Andysz, A. (2012). Relationship between Person-Organization fit and objective
and subjective health status (Person-Organization fit and health). International Journal of
Occupational Medicine and Environmental Health, 25, 166-177.
Mesmer-Magnus, J., Glew, D. J., & Viswesvaran, C. (2012). A meta-analysis of positive humor
in the workplace. Journal of Managerial Psychology, 27, 155-190.
Meyer, J. P., Becker, T. E., & van Dick, R. (2006). Social identities and commitments at work:
Toward an integrative model. Journal of Organizational Behavior, 27, 665-683.
*Mishra, S. K. (2013). Perceived external prestige and employee outcomes: Mediation effect of
organizational identification. Corporate Reputation Review, 16, 220-233.
*Mishra, S. K., & Bhatnagar, D. (2010). Linking emotional dissonance and organizational
identification to turnover intention and emotional well‐being: A study of medical
representatives in India. Human Resource Management, 49, 401-419.
Moreland, R. L., & Levine, J. M. (2001). Socialization in organizations and workgroups. In M. E.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 64
Turner (Ed.), Groups at work: theory and research (pp. 69–112). Hillsdale, NJ: Erlbaum.
Muldoon, O. T., & Downes, C. (2007). Social identification and post-traumatic stress symptoms
in post-conflict Northern Ireland. The British Journal of Psychiatry, 191, 146-149.
Mühlhaus, J. & Bouwmeester, O. (2016). The paradoxical effect of self-categorization on work
stress in a high-status occupation: Insights from management consulting. Human Relations.
Advance online publication. Doi:10.1177/0018726715626255.
Ng, T. W. (2015). The incremental validity of organizational commitment, organizational trust,
and organizational identification. Journal of Vocational Behavior, 88, 154-163.
Ng, T. W., & Feldman, D. C. (2008). Long work hours: a social identity perspective on meta‐
analysis data. Journal of Organizational Behavior, 29, 853-880.
*Nieuwenhuis, M., Knight, C., Postmes, T., & Haslam, S. A. (2014). The relative benefits of
green versus lean office space: Three field experiments. Journal of Experimental Psychology:
Applied, 20, 199-214.
*O’Brien, A. T., & Haslam, S. A. (2004). Shaping the future: The 2003 Stress Survey of Staff in
the West Dorset General Hospitals Trust. Psychology: College of Life and Environmental
Science, University of Exeter, UK.
Orwin, R. G. (1983). A fail-safe N for effect size in meta-analysis. Journal of Educational
Statistics, 8, 157-159.
Oyserman, D., Fryberg, S. A., & Yoder, N. (2007). Identity-based motivation and health. Journal
of Personality and Social Psychology, 93, 1011-1027.
Pandey, K., Stevenson, C., Shankar, S., Hopkins, N. P., & Reicher, S. D. (2014). Cold comfort at
the Magh Mela: Social identity processes and physical hardship. British Journal of Social
Psychology, 53, 675-690.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 65
Parker, S. K. (2014). Beyond motivation: Job and work design for development, health,
ambidexterity, and more. Annual Review of Psychology, 65, 661-691.
Parker, D. F., & DeCotiis, T. A. (1983). Organizational determinants of job stress.
Organizational Behavior and Human Performance, 32, 160-177.
Pascoe, E. A., & Smart Richman, L. (2009). Perceived discrimination and health: A meta-
analytic review. Psychological Bulletin, 135, 531-554.
Patchen, M. (1970). Participation, achievement and involvement on the job. Englewood Cliffs
NJ: Prentice-Hall.
Perkins, H. W., & Wechsler, H. (1996). Variation in perceived college drinking norms and its
impact on alcohol abuse: A nationwide study. Journal of Drug Issues, 26, 961-974.
Peters, K., Ryan, M. K., Haslam, S. A., & Fernandes, H. (2012). To belong or not to belong:
Evidence that women’s occupational disidentification is promoted by lack of fit with
masculine occupational prototypes. Journal of Personnel Psychology, 11, 148-158.
Phinney, J. S. (1992). The multigroup ethnic identity measure a new scale for use with diverse
groups. Journal of Adolescent Research, 7, 156-176.
*Pisarski, A., Lawrence, S. A., Bohle, P., & Brook, C. (2008). Organizational influences on the
work life conflict and health of shiftworkers. Applied Ergonomics, 39, 580-588.
Platow, M. J., Voudouris, N. J., Gilford, N., Jamieson, R., Najdovski, L., Papaleo, N., ... & Terry,
L. (2007). In‐group reassurance in a pain setting produces lower levels of physiological
arousal: Direct support for a self‐categorization analysis of social influence. European Journal
of Social Psychology, 37, 649-660.
Podsakoff, P. M., MacKenzie, S. B., Lee, J., & Podsakoff, N. P. (2003). Common method biases
in behavioral research: A critical review of the literature and recommended remedies. Journal
of Applied Psychology, 88, 879-903.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 66
*Portaluri, S. (2013). Work and well-being: A two-wave study exploring the relationship
between leadership effectiveness, social identity, psychological well-being and job
involvement. Unpublished Manuscript, Australian National University, Australia.
Postmes, T., & Branscombe, N. R. (2002). Influence of long-term racial environmental
composition on subjective well-being in African Americans. Journal of Personality and Social
Psychology, 83, 735-751.
Postmes, T., & Branscombe, N. R. (2010). Rediscovering social identity. New York: Psychology
Press.
Postmes, T., Haslam, S. A., & Jans, L. (2013). A single‐item measure of social identification:
Reliability, validity, and utility. British Journal of Social Psychology, 52, 597-617.
Prandy, K. (1990). The revised Cambridge scale of occupations. Sociology, 24, 629-655.
Pratt, M. G. (2000). The good, the bad, and the ambivalent: Managing identification among
Amway distributors. Administrative Science Quarterly, 45, 456-493.
Ramarajan, L. (2014). Past, present and future research on multiple identities: Toward an
intrapersonal network approach. The Academy of Management Annals, 8, 589-659.
Ramarajan, L., & Reid, E. (2013). Shattering the myth of separate worlds: Negotiating nonwork
identities at work. Academy of Management Review, 38, 621-644.
Reicher, S. D., Cassidy, C., Wolpert. I., Hopkins, N. & Levine, M. (2006). Saving Bulgaria’s
Jews: An analysis of social identity and the mobilisation of social solidarity. European
Journal of Social Psychology, 36, 49-72.
Reicher, S. D., & Haslam, S. A., (2006). On the agency of individuals and groups: Lessons from
the BBC Prison Study. In: T. Postmes & J. Jetten (Eds.), Individuality and the group:
Advances in social identity (pp.237-257). London: Sage.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 67
Reicher, S. D., Spears, R., & Haslam, S. A. (2010). The social identity approach in social
psychology. In M. S. Wetherell & C. T. Mohanty (Eds.), Sage Identities Handbook (pp. 45-
62). London: Sage.
Richardson, K. M., & Rothstein, H. R. (2008). Effects of occupational stress management
intervention programs: A meta-analysis. Journal of Occupational Health Psychology, 13, 69-
93.
Riketta, M. (2005). Organizational identification: A meta-analysis. Journal of Vocational
Behavior, 66, 358-384.
Riketta, M., & van Dick, R. (2005). Foci of attachment in organizations: A meta-analytic
comparison of the strength and correlates of workgroup versus organizational identification
and commitment. Journal of Vocational Behavior, 67, 490-510.
Roccas, S., Sagiv, L., Schwartz, S., Halevy, N., & Eidelson, R. (2008). Toward a unifying model
of identification with groups: Integrating theoretical perspectives. Personality and Social
Psychology Review, 12, 280-306.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic
motivation, social development, and well-being. American Psychologist, 55, 68-78.
Ryff, C. D., & Singer, B. (1998). The contours of positive human health. Psychological
Inquiry, 9, 1-28.
Sacker, A., Firth, D., Fitzpatrick, R., Lynch, K., & Bartley, M. (2000). Comparing health
inequality in men and women: Prospective study of mortality 1986-96. British Medical
Journal, 320, 1303-1307.
Sani, F. (Ed.). (2008). Self continuity: Individual and collective perspectives. New York, NY US:
Psychology Press.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 68
Sani, F., Bowe, M., & Herrera, M. (2008). Perceived collective continuity and social well‐being:
Exploring the connections. European Journal of Social Psychology, 38, 365-374.
*Sani, F., Herrera, M., Wakefield, J. R., Boroch, O., & Gulyas, C. (2012). Comparing social
contact and group identification as predictors of mental health. British Journal of Social
Psychology, 51, 781-790.
*Sani, F., Magrin, M. E., Scrignaro, M., & McCollum, R. (2010). In‐group identification
mediates the effects of subjective in‐group status on mental health. British Journal of Social
Psychology, 49, 883-893.
Schaufeli, W. B., Salanova, M., González-Romá, V., & Bakker, A. B. (2002). The measurement
of engagement and burnout: A two sample confirmatory factor analytic approach. Journal of
Happiness Studies, 3, 71-92.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological and physical well-
being: Theoretical overview and empirical update. Cognitive Therapy and Research, 16, 201-
228.
Schmitt, M. T., Branscombe, N. R., Postmes, T., & Garcia, A. (2014). The consequences of
perceived discrimination for psychological well-being: A meta-analytic review. Psychological
Bulletin, 140, 921-948.
Seligman, M. E., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction.
American Psychologist, 55, 5-14.
Sellers, R. M., Smith, M. A., Shelton, J. N., Rowley, S. A., & Chavous, T. M. (1998).
Multidimensional model of racial identity: A reconceptualization of African American racial
identity. Personality and Social Psychology Review, 2, 18-39.
Shamir, B., & Kark, R. (2004). A single‐item graphic scale for the measurement of organizational

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 69
identification. Journal of Occupational and Organizational Psychology, 77, 115-123.
Sheldon, K. M., & Bettencourt, B. (2002). Psychological need‐satisfaction and subjective well‐
being within social groups. British Journal of Social Psychology, 41, 25-38.
Sherman, D. K., Kinias, Z., Major, B., Kim, H. S., & Prenovost, M. (2007). The group as a
resource: Reducing biased attributions for group success and failure via group
affirmation. Personality and Social Psychology Bulletin, 33, 1100-1112.
Simonsohn, U., Nelson, L. D., & Simmons, J. P. (2014). P-curve: A key to the file-
drawer. Journal of Experimental Psychology: General, 143, 534-547.
Simonsohn, U., Simmons, J. P., & Nelson, L. D. (2015). Better P-curves: Making P-curve
analysis more robust to errors, fraud, and ambitious P-hacking, a Reply to Ulrich and Miller
(2015). Journal of Experimental Psychology: General, 144, 1146-1152.
Singh, J., Goolsby, J. R., & Rhoads, G. K. (1994). Behavioral and psychological consequences of
boundary spanning burnout for customer service representatives. Journal of Marketing
Research, 31, 558-569.
Steffens, N. K., Haslam, S. A., Reicher, S. D., Platow, M. J., Fransen, K., Yang, J., Jetten, J.,
Ryan, M. K., Peters, K., & Boen, F. (2014). Leadership as social identity management:
Introducing the Identity Leadership Inventory (ILI) to assess and validate a four-dimensional
model. The Leadership Quarterly, 25, 1001-1024.
Steffens, N. K., Shemla, M., Wegge, J., & Diestel, S. (2014). Organizational tenure and
employee performance: A multilevel analysis. Group & Organization Management, 39, 664-
690.
*Steffens, N. K., Yang, J., Jetten, J., Haslam, S. A., & Lipponen, J. (2014). Leader identity
entrepreneurship and employee health: Lagged effects on well-being and turnover intentions.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 70
Unpublished Manuscript, The University of Queensland, Australia.
Stott, C., & Drury, J. (2000). Crowds, context and identity: Dynamic categorization processes in
the 'poll tax riot'. Human Relations, 53, 247-273.
*Suh, T., Houston, M. B., Barney, S. M., & Kwon, I. W. G. (2011). The impact of mission
fulfillment on the internal audience: Psychological job outcomes in a services setting. Journal
of Service Research, 14, 76-92.
Tajfel, H. (1981). Social stereotypes and social groups. In J. C. Turner & H. Giles (Eds.),
Intergroup behaviour (pp.144-167). Oxford: Blackwell.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin &
S. Worchel (Eds). The social psychology of intergroup relations. Brooks / Cole.
Tanis, M., & Postmes, T. (2005). A social identity approach to trust: Interpersonal perception,
group membership and trusting behaviour. European Journal of Social Psychology, 35, 413-
424.
Tarrant, M., Haggar, M. S., & Farrow, C. V. (2012). Promoting positive orientation towards
health through social identity. In J. Jetten, C. Haslam, & S. A. Haslam (Eds.), The Social
Cure: Identity, Health and Well-being (pp. 39–54). New York: Psychology Press.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological perspective
on mental health. Psychological Bulletin, 103, 193-210.
Thoits, P. A. (1986). Social support as coping assistance. Journal of Consulting and Clinical
Psychology, 54, 416-423.
Thoits, P. A. (1994). Stressors and problem-solving: The individual as psychological
activist. Journal of Health and Social Behavior, 35, 143-160.
Thoits, P. A. (1995). Stress, coping, and social support processes: Where are we? What next?
Journal of Health and Social Behavior, 35, 53-79.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 71
*Topa, G., & Moriano, J. A. (2013). Stress and nurses' horizontal mobbing: Moderating effects of
group identity and group support. Nursing Outlook, 61, e25-e31.
Turner, J. C. (1982). Towards a redefinition of the social group. In H. Tajfel (Ed.), Social Identity
and Intergroup Relations (pp. 15-40). Cambridge: Cambridge University Press.
Turner, J. C. (1991). Social influence. Milton Keynes: Open University Press.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987).
Rediscovering the social group: A self-categorization theory. Cambridge, MA, US: Basil
Blackwell.
Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. (1994). Self and collective: Cognition
and social context. Personality and Social Psychology Bulletin, 20, 454-463.
Turner, J. C., & Reynolds, K. J. (2001). The social identity perspective in intergroup relations:
Theories, themes and controversies. In R. Brown & S. Gaertner (Eds.), Handbook of social
psychology: Intergroup processes (Vol. 4). Oxford, UK and Cambridge, USA: Blackwell
Tyler, T. R., & Blader, S. L. (2003). The group engagement model: Procedural justice, social
identity, and cooperative behavior. Personality and Social Psychology Review, 7, 349-361.
Uchino, B. N. (2009). Understanding the links between social support and physical health: A life-
span perspective with emphasis on the separability of perceived and received
support. Perspectives on Psychological Science, 4, 236-255.
Uchino, B. N., Cacioppo, J. T., & Kiecolt-Glaser, J. K. (1996). The relationship between social
support and physiological processes: A review with emphasis on underlying mechanisms and
implications for health. Psychological Bulletin, 119, 488-531.
van Dick, R. (2004). My job is my castle: Identification in organizational contexts. International
Review of Industrial and Organizational Psychology, 19, 171-204.
van Dick, R., & Haslam, S. A. (2012). Stress and well-being in the workplace: Support for key

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 72
propositions from the social identity approach. In J. Jetten, C. Haslam, & S. A. Haslam (Eds.),
The social cure: Identity, health and well-being (pp. 175-194). New York, NY US:
Psychology Press.
*van Dick, R., Stierle, C., Govaris, H., Tissington, P. A., Kodakos, A. (2007). School teachers'
well-being and job satisfaction: An analysis across countries and levels of explanation. In J.
Löwstedt, P. Larsson, S. Karsten, & R. van Dick (Eds). From Intensified work to professional
development. Brussels (BE): PIE Peter Lang.
van Dick, R., van Knippenberg, D., Kerschreiter, R., Hertel, G., & Wieseke, J. (2008). Interactive
effects of work group and organizational identification on job satisfaction and extra-role
behavior. Journal of Vocational Behavior, 72, 388-399.
*van Dick, R., & Wagner, U. (2002). Social identification among school teachers: Dimensions,
foci, and correlates. European Journal of Work and Organizational Psychology, 11, 129-149.
*van Dick, R., Wagner, U., & Lemmer, G. (2004). Research note: The winds of change—
Multiple identifications in the case of organizational mergers. European Journal of Work and
Organizational Psychology, 13, 121-138.
*van Dick, R., Wagner, U., Stellmacher, J., & Christ, O. (2004). The utility of a broader
conceptualization of organizational identification: Which aspects really matter? Journal of
Occupational and Organizational Psychology, 77, 171-191.
van Knippenberg, D., & van Schie, E. (2000). Foci and correlates of organizational
identification. Journal of Occupational and Organizational Psychology, 73, 137-147.
van Knippenberg, D., & Sleebos, E. (2006). Organizational identification versus organizational
commitment: Self‐definition, social exchange, and job attitudes. Journal of Organizational
Behavior, 27, 571-584.
van Steenbergen, E. F., Ellemers, N., Haslam, S. A., & Urlings, F. (2008). There is nothing either

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 73
good or bad but thinking makes it so: Informational support and cognitive appraisal of the
work–family interface. Journal of Occupational and Organizational Psychology, 81, 349-367.
van Zomeren, M., Leach, C. W., & Spears, R. (2012). Protesters as “passionate economists” a
dynamic dual pathway model of approach coping with collective disadvantage. Personality
and Social Psychology Review, 16, 180-199.
Viswesvaran, C., Sanchez, J. I., & Fisher, J. (1999). The role of social support in the process of
work stress: A meta-analysis. Journal of Vocational Behavior, 54, 314-334.
Warr, P. B. (1990). Decision latitude, job demands, and employee well-being. Work & Stress, 4,
285-294.
*Waszkowska, M., Andysz, A., & Merecz, D. (2014). Person-organization fit as a mediator of
the relationship between work environment and stress among social workers. Medycyna
Pracy, 65, 219-228.
*Wegge, J., Schuh, S. C., & van Dick, R. (2012). ‘I feel bad’, ‘We feel good’? Emotions as a
driver for personal and organizational identity and organizational identification as a resource
for serving unfriendly customers. Stress & Health, 28, 123-136.
*Wegge, J., van Dick, R., Fisher, G. K., Wecking, C., & Moltzen, K. (2006). Work motivation,
organisational identification, and well-being in call centre work. Work & Stress, 20, 60-83.
*West, A. (2005). Organizational communication, bullying, and stress. Unpublished raw data,
Psychology: College of Life and Environmental Science, University of Exeter, UK.
World Values Survey (2014). Retrieved 03 May 2016 from
http://www.worldvaluessurvey.org/WVSDocumentationWV6.jsp
Zhang, S., Chen, G., Chen, X. P., Liu, D., & Johnson, M. D. (2014). Relational versus collective
identification within workgroups: Conceptualization, measurement development, and
nomological network building. Journal of Management, 40, 1700-1731.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 74
*Zhang, R. Y., Liu, X. M., Wang, H. Z., & Shen, L. (2011). Service climate and employee
service performance: Exploring the moderating role of job stress and organizational
identification. The Service Industries Journal, 31, 2355-2372.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 75
Table 1 Literatures (and Illustrative References) Related to Competing Hypotheses about the Nature of the Identification–Health Relationship
Summary effect
Hypothesis
Literatures (with illustrative references)
Identification–health relationship
Invigoration
(Positive relationship)
• Belonging (Baumeister & Leary, 1995; Ryan & Deci, 2000)
• Meaning and purpose (van Dick & Wagner, 2002; Wegge, van Dick, Fisher, Wecking, &
Moltzen, 2006)
• Control and agency (Greenaway et al., 2015; Reicher & Haslam, 2006)
• Social support (Levine, Cassidy, Brazier, & Reicher, 2002; Haslam, O’Brien, Jetten,
Vormedal, & Penna, 2005)
• Collective self-efficacy (Avanzi, Schuh, Fraccaroli, & van Dick, 2015; van Zomeren,
Spears, & Leach, 2012)
• Self-affirmation and positive attributions (Cruwys, South, Greenaway, & Haslam, 2015;
Sherman, Kinias, Major, Kim, & Prenovost, 2007)
Exhaustion
(Negative relationship)
• Excessive involvement and pressure to perform (Herrbach, 2006; Mühlhaus &
Bouwmeester, 2016)
• Working long hours and presenteeism (Escartín, Ullrich, Zapf, Schlüter, & van Dick,
2013; Ng & Feldman, 2008)
• Dirty work and stigma (Ashforth & Kreiner, 1999; Branscombe, Schmitt, & Harvey,
1999)
• Health-damaging norms and motivation (Oyserman, Fryberg, & Yoder, 2007; Perkins &
Wechsler, 1996)

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 76
Table 2 Literatures (and Illustrative References) Related to Competing Hypotheses about the Impact of A-Priori Moderators on the
Identification–Health Relationship
Moderator
Hypothesis
(stronger positive effects for)
Literatures (with illustrative references)
Identification focus
Workgroup
• Proximity and salience of group (Ashforth & Johnson, 2001; van Knippenberg & van
Schie, 2000)
• Familiarity with members and cohesion (Carron & Spink, 1995; Moreland & Levine,
2001)
Organization
• Stability and continuity (Albert & Whetten, 1985; Sani, Bowe, & Herrera, 2008)
• Public regard (Dutton, Dukerich, & Harquail, 1994; Luhtanen & Crocker, 1992)
Health valence
Presence of well-being
• Positive human health (Ryff & Singer, 1998; Thoits, 1994)
Absence of stress
• Stress buffering (Branscombe, Schmitt, & Harvey, 1999; Thoits, 1986)
Health index
Psychological health
• Attitude-behavior compatibility in levels of analysis (Fishbein & Aijzen, 1974; Riketta &
van Dick, 2005)
• Connection with others jeopardizing physical health (Hopkins & Reicher, 2016; Howell et
al., 2014)
Physical health
—
Sharedness in identification
Shared
• Solidarity (Drury, Cocking, & Reicher, 2009; Reicher, Cassidy, Wolpert, Hopkins, &
Levine, 2006)
• Cohesion (Blanchard, Amiot, Perreault, Vallerand, & Provencher, 2009; Griffith, 2002)
Non-shared
• Individual distinctiveness (Hornsey & Jetten, 2004; Sheldon & Bettencourt, 2002)

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 77
Table 3 Overview of Studies and Effect Sizes Concerning the Organizational Identification–Health Relationship (58 independent samples, 102
effect sizes, N = 19,799)
Study
CO
Sample
N
r
LL/UL
IF
IS
Val
Ind
SM
%♀
Age
SS
SD Id
Ashforth & Saks (1996)
CA
Business graduates (workers in
various industries)
294
.08
–.03/.19
O
M
S
Phys
E/L
59.90
23.50
68.40
–
Avanzi, Fraccaroli, Sarchielli,
Ullrich, & van Dick (2014)
IT
Workers in food industry
195
.16
.02/.29
O
M
S
Psy
C-S
66.30
21.44
34.18
1.41
Avanzi, Schuh, Fraccaroli, & van
Dick (2014)
IT
Teachers
192
.24
.10/.37
O
M
S
Psy
C-S
71.40
47.01
77.42
–
Effect size b
IT
Teachers
192
.15
.01/.29
O
M
S
Psy
C-S
71.40
47.01
77.42
–
Effect size c
IT
Teachers
192
.14
.00/.28
O
M
S
Psy
C-S
71.40
47.01
77.42
–
Avanzi, van Dick, Fraccaroli, &
Sarchielli (2012) Study 1
IT
Court employees
195
.08
–.06/.22
O
M
WB
Psy
C-S
76.92
48.10
52.69
–
Avanzi, van Dick, Fraccaroli, &
Sarchielli (2012) Study 2
IT
Teachers
140
.08
–.09/.24
O
M
WB
Psy
C-S
82.86
41.20
69.95
–
Barbier, Dardenne, & Hansez
(2013)
BE
Public administration workers
473
.33
.25/.41
O
RO
S
Psy
E/L
56.03
48.71
52.30
1.32
Effect size b
BE
Public administration workers
473
.13
.04/.22
O
RO
S
Phys
E/L
56.03
48.71
52.30
1.32
Effect size c
BE
Public administration workers
473
.30
.22/.38
O
RO
WB
Psy
E/L
56.03
48.71
52.30
1.32
Bedeian (2007)
US
Academics
379
.32
.23/.41
O
M
S
Psy
C-S
36.88
45.00
80.62
–
Bizumic, Reynolds, Turner,
Bromhead, & Subasic (2009)
AU
Teachers
113
.16
–.03/.33
O
RO
S
Psy
C-S
79.09
34.41
75.30
–
Bjerregaard, Haslam, Morton, &
Ryan (2014)
GB
Care staff
1274
.43
.38/.47
O
D
S
Psy
C-S
89.68
42.57
43.09
.99
Effect size b
GB
Care staff
1274
.41
.36/.45
W
D
S
Psy
C-S
89.68
42.57
43.09
–
Christ (2004)
DE
Teachers in training internship
93
.08
–.13/.28
O
vD
S
Phys
C-S
73.12
27.56
65.87
–
Cicero, Pierro, & van Knippenberg
(2007) Study 1
IT
Call center workers, nurses,
military officers
329
.29
.19/.39
W
M
S
Psy
C-S
50.15
35.26
43.72
–
Cicero, Pierro, & van Knippenberg
IT
Workers in consumer electronics
89
.11
–.10/.31
W
M
S
Phys
C-S
62.92
38.65
42.80
–

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 78
(2007) Study 2
company
Das, Dharwadkar, & Brandes (2008)
IN
Call center workers
132
.19
.02/.35
O
M
S
Psy
C-S
30.00
23.00
35.05
1.20
Effect size b
IN
Call center workers
132
.20
.03/.36
O
M
S
Psy
C-S
30.00
23.00
35.05
1.20
Fampri (2003)
GR
Workers electronics companies
68
.20
–.04/.42
O
D
S
Phys
C-S
54.41
30.00
–
–
Effect size b
GR
Workers electronics companies
68
.38
.16/.57
O
D
S
Psy
C-S
54.41
30.00
–
–
Effect size c
GR
Workers electronics companies
68
.01
–.23/.25
O
D
S
Psy
C-S
54.41
30.00
–
–
Frisch, Häusser, van Dick, &
Mojzisch (2014)
DE
Ad-hoc workgroup members
90
.09
–.12/.29
W
D
S
Phys
E/L
54.44
22.00
–
–
Effect size b
DE
Ad-hoc workgroup members
90
.01
–.20/.21
W
D
S
Psy
E/L
54.44
22.00
–
–
Galang & Jones (2014)
GB
Workers in various industries
65
–.41
–.59/–.19
W
RO
S
Psy
C-S
61.54
21.50
–
–
Grubba & Ahlswede (2002)
DE
Bank assistants
338
.17
.06/.27
O
vD
S
Phys
C-S
51.51
40.00
52.25
–
Effect size b
DE
Bank assistants
338
.16
.05/.26
W
vD
S
Phys
C-S
51.51
40.00
52.25
–
Harris & Cameron (2005)
CA
Workers in food producing
company
60
.14
–.12/.38
W
D
WB
Psy
C-S
40.00
39.37
28.38
1.47
Haslam, Jetten, & Waghorn (2009)
GB
Theatre employees
30
.47
.17/.69
W
D
S
Psy
C-S
40.00
20.80
58.16
–
Haslam, O’Brien, Jetten, Vormedal,
& Penna (2005) Study 2
GB
Bomb disposal workers and bar
staff
40
.63
.40/.79
W
D
S
Psy
C-S
25.00
25.58
47.55
1.25
Haslam & Reicher (2006)
GB
Simulated guards and prisoners
15
.43
–.02/.74
W
D
S
Phys
E/L
0.00
33.00
25.89
–
Effect size b
GB
Simulated guards and prisoners
15
.50
.08/.77
W
D
S
Psy
E/L
0.00
33.00
25.89
–
Häusser, Kattenstroth, van Dick, &
Mojzisch (2012)
DE
Ad-hoc workgroup members
96
.18
–.01/.37
W
RO
S
Phys
E/L
50.00
22.80
–
–
Effect size b
DE
Ad-hoc workgroup members
96
.03
–.17/.23
W
RO
S
Psy
E/L
50.00
22.80
–
–
Herrbach (2006)
FR
Engineers
365
.29
.19/.38
O
M
WB
Psy
C-S
36.00
32.70
66.98
–
Effect size b
FR
Engineers
365
–.22
–.32/–.12
O
M
S
Psy
C-S
36.00
32.70
66.98
–
Horton, McClelland, & Griffin
(2014) Study 1
GB
Navy strategic officers
65
.44
.22/.62
O
M
WB
Psy
C-S
13.00
31.62
72.70
1.02
Effect size b
GB
Navy strategic officers
65
.27
.03/.48
W
M
WB
Psy
C-S
13.00
31.62
72.70
–
Horton, McClelland, & Griffin
(2014) Study 2
GB
Navy mid-level officers
381
.20
.10/.29
O
M
WB
Psy
C-S
8.00
33.71
72.70
1.35
Effect size b
GB
Navy mid-level officers
381
.20
.10/.29
W
M
WB
Psy
C-S
8.00
33.71
72.70
–
Horton, McClelland, & Griffin
(2014) Study 3
GB
Navy operational officers
343
.24
.14/.34
O
M
WB
Psy
C-S
15.00
23.61
62.96
1.35

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 79
Effect size b
GB
Navy operational officers
343
.38
.29/.47
W
M
WB
Psy
C-S
15.00
23.61
62.96
–
Ishii (2012)
US
Japanese expatriates in US
159
.41
.27/.53
O
RO
S
Psy
C-S
1.26
40.00
–
–
Jimmieson, McKimmie, Hannam, &
Gallagher (2010)
AU
Students in employment
155
.36
.21/.49
W
RO
WB
Psy
C-S
77.26
21.60
–
–
Effect size b
AU
Students in employment
155
–.12
–.27/.04
W
RO
S
Psy
C-S
77.26
21.60
–
–
Knight & Haslam (2010a) Study 1
GB
Office workers
288
.26
.15/.36
O
RO
WB
Phys
C-S
40.28
32.75
55.60
–
Effect size b
GB
Office workers
288
.24
.13/.35
O
RO
WB
Psy
C-S
40.28
32.75
55.60
–
Knight & Haslam (2010a) Study 2
GB
Office workers
1643
.20
.15/.25
O
RO
WB
Phys
C-S
66.22
28.84
55.60
–
Effect size b
GB
Office workers
1643
.43
.39/.47
O
RO
WB
Psy
C-S
66.22
28.84
55.60
–
Knight & Haslam (2010b) Study 1
GB
Workers in different industries
112
.15
–.04/.33
O
D
WB
Phys
C-S
64.29
37.55
–
–
Effect size b
GB
Workers in different industries
112
.14
–.05/.32
O
D
WB
Psy
C-S
64.29
37.55
–
–
Knight & Haslam (2010b) Study 2
GB
Office workers
47
.67
.47/.80
O
D
WB
Phys
C-S
40.43
36.23
55.60
–
Effect size b
GB
Office workers
47
.64
.43/.78
O
D
WB
Psy
C-S
40.43
36.23
55.60
–
Knowles & Smith (2013)
GB
Office workers
139
.11
–.06/.27
W
RO
S
Psy
C-S
48.20
30.50
55.60
–
Kreiner & Ashforth (2004)
US
University graduates
330
.43
.34/.51
O
M
WB
Psy
C-S
56.06
43.60
–
–
Effect size b
US
University graduates
330
.25
.15/.35
O
M
S
Psy
C-S
56.06
43.60
–
–
Effect size c
US
University graduates
330
.15
.04/.25
O
M
S
Psy
C-S
56.06
43.60
–
–
Matheson & Cole (2004)
CA
Ad-hoc university group
72
.20
–.04/.41
O
RO
S
Phys
C-S
66.67
20.00
–
–
Effect size b
CA
Ad-hoc university group
72
–.04
–.27/.19
O
RO
S
Psy
C-S
66.67
20.00
–
–
Effect size c
CA
Ad-hoc university group
72
.33
.11/.52
O
RO
WB
Psy
C-S
66.67
20.00
–
–
Menzel (2007)
DE
Nurses
220
.11
–.02/.24
O
RO
S
Phys
C-S
76.82
30.89
80.11
1.27
Effect size b
DE
Nurses
220
.08
–.05/.21
W
RO
S
Phys
C-S
76.82
30.89
80.11
–
Merecz & Andysz (2012)
PL
Workers in different industries
576
.18
.10/.26
O
RO
WB
Psy
C-S
52.2
39.48
–
–
Mishra & Bhatnagar (2010)
IN
Medical sales employees
468
.39
.31/.46
O
M
S
Psy
C-S
4.91
26.99
58.95
–
Effect size b
IN
Medical sales employees
468
.23
.14/.31
O
M
WB
Psy
C-S
4.91
26.99
58.95
–
Effect size c (reported in Mishra,
2013)
IN
Medical sales employees
468
.43
.35/.50
O
M
S
Psy
C-S
5.00
26.99
58.95
–
Nieuwenhuis, Knight, Postmes, &
Haslam (2014) Study 1
GB
Consultants
167
.38
.24/.50
O
D
WB
Psy
C-S
40.74
29.88
70.38
1.24
Effect size b
GB
Consultants
167
.44
.31/.55
W
D
WB
Psy
C-S
40.74
29.88
70.38
–
Effect size c
GB
Consultants
167
.27
.12/.41
O
D
S
Psy
C-S
40.74
29.88
70.38
1.24
Effect size d
GB
Consultants
167
.27
.13/.41
W
D
S
Psy
C-S
40.74
29.88
70.38
–

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 80
Nieuwenhuis, Knight, Postmes, &
Haslam (2014) Study 2
NL
Insurance service center workers
128
.11
–.06/.28
O
D
WB
Psy
C-S
78.13
34.65
45.48
1.01
Effect size b
NL
Insurance service center workers
128
.07
–.11/.24
W
D
WB
Psy
C-S
78.13
34.65
45.48
–
Effect size c
NL
Insurance service center workers
128
.27
.10/.43
O
D
WB
Psy
C-S
78.13
34.65
45.48
1.02
Effect size d
NL
Insurance service center workers
128
.37
.21/.51
W
D
WB
Psy
C-S
78.13
34.65
45.48
–
O’Brien & Haslam (2004)
GB
Hospital staff
1090
.17
.11/.23
W
D
WB
Psy
C-S
81.45
50.12
56.40
–
Effect size b
GB
Hospital staff
1090
.25
.19/.30
W
D
S
Psy
C-S
81.45
50.12
56.40
–
Pisarski, Lawrence, Bohle, & Brook
(2008)
AU
Nurses
530
.00
–.09/.09
W
M
WB
Psy
C-S
92.64
35.00
36.60
–
Effect size b
AU
Nurses
530
–.13
–.21/–.05
W
M
S
Phys
C-S
92.64
35.00
36.60
–
Portaluri (2013)
AU
Teachers
84
–.14
–.34/.08
O
RO
S
Psy
E/L
72.29
31.44
75.30
–
Sani, Herrera, Wakefield, Boroch, &
Gulyas (2012)
EE
Army members
150
.47
.33/.59
O
RO
WB
Psy
C-S
32.00
39.11
52.00
.74
Sani, Magrin, Scrignari, &
McCollum (2010)
IT
Prison guards
93
.55
.39/.68
W
D
S
Psy
C-S
16.30
33.00
43.97
1.15
Effect size b
IT
Prison guards
93
.47
.29/61
W
D
S
Psy
C-S
16.30
33.00
43.97
1.15
Steffens, Yang, Jetten, Haslam, &
Lipponen (2014)
CN
Engineering workers
140
.18
.01/.34
W
RO
WB
Psy
E/L
24.81
34.81
63.27
–
Effect size b
CN
Engineering workers
140
.24
.08/.39
W
RO
WB
Psy
E/L
24.81
34.81
63.27
–
Effect size c
CN
Engineering workers
140
.28
.12/.43
W
RO
S
Psy
E/L
24.81
34.81
63.27
–
Suh, Houston, Barney, & Kwon
(2011)
US
Health care service workers
3999
.27
.24/.30
O
M
S
Psy
C-S
81.17
43.50
42.64
1.08
Topa & Moriano (2013)
ES
Nurses
388
.27
.18/.36
W
M
S
Psy
C-S
77.60
36.30
69.30
–
van Dick, Stierle, Govaris,
Tissington, & Kodakos
(2007)
GR,
DE,
GB
Teachers
367
.08
–.02/.18
O
RO
S
Phys
C-S
44.78
42.73
73.77
–
Effect size b
DE
Teachers
367
.05
–.05/.15
W
RO
S
Phys
C-S
44.78
42.73
73.77
–
van Dick & Wagner (2002) Study 1a
DE
Teachers
201
.27
.14/.39
O
RO
S
Phys
C-S
54.73
45.00
77.65
–
van Dick & Wagner (2002) Study 2
DE
Teachers
283
.28
.17/.38
W
RO
S
Phys
C-S
54.77
46.00
77.64
–
van Dick, Wagner, & Lemmer
(2004)
DE
Hospital staff
459
.05
–.04/.14
O
RO
S
Psy
C-S
38.00
41.74
84.51
1.82
van Dick, Wagner, Stellmacher, &
DE
Teachers
433
.06
–.03/.15
O
vD
S
Phys
C-S
61.43
46.08
77.94
–

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 81
Christ (2004) a
Effect size b
DE
Teachers
433
.10
.01/.19
W
vD
S
Phys
C-S
61.43
46.08
77.94
–
Waszkowska, Andysz, & Merecz
(2014)
PL
Social workers
500
.30
.21/.37
O
RO
S
Psy
C-S
91.20
39.60
40.04
–
Wegge, Schuh, & van Dick (2012)
DE
Call center workers
96
.07
–.13/.27
O
RO
S
Phys
C-S
56.25
29.80
31.08
–
Wegge, van Dick, Fisher, Wecking,
& Moltzen (2006)
DE
Call center workers
161
.27
.12/.41
O
vD
S
Phys
C-S
62.11
32.60
30.58
–
Effect size b
DE
Call center workers
161
.40
.26/.52
O
vD
S
Psy
C-S
62.11
32.60
30.58
–
Effect size c
DE
Call center workers
161
.28
.13/.42
O
vD
S
Psy
C-S
62.11
32.60
30.58
–
Effect size d
DE
Call center workers
161
.48
.35/.59
O
vD
S
Psy
C-S
62.11
32.60
30.58
–
West (2005)
GB
Workers in different industries
99
.36
.18/.52
O
RO
S
Psy
C-S
58.33
34.62
–
–
Zhang, Liu, Wang, & Shen (2011)
CN
Bank assistants
368
–.18
–.28/–.08
O
M
S
Psy
C-S
61.00
39.30
56.89
–
Note. Total N = 19,799 from 58 independent samples; Dashes indicate that particular indicator is not applicable in particular study; Study indicates an independent effect size;
Effect size indicates a non-independent sample effect size; CO = country; CA = Canada; IT = Italy; BE = Belgium; AU = Australia; US = United States; GB = United
Kingdom; DE = Germany; IN = India; GR = Greece; FR = France; PL = Poland; NL = Netherlands; EE = an Eastern European country (that authors of original research
cannot disclose); CN = China; ES = Spain; N indicates number of participants included in calculation of particular effect size; r indicates sample size weighted corrected
correlation between IV-DV; LL/UL indicate lower and upper 95% limit of confidence interval; IF = identification focus; O = organization; W = workgroup; IS =
identification scale; M = Mael & Ashforth’s (1992) scale; D = Doosje et al.’s (1995) scale; vD = van Dick et al.’s (2004) scale; RO = residual other scale; Val = health
valence; S = absence of stress; WB = presence of well-being; Ind = health index; Psy = psychological health; Phys = physical health; SM = study methodology; CS =
cross-sectional design; E/L = experimental / longitudinal design; %♀ = proportion of female participants in sample; Age = sample mean age; SS = social status of
profession based on social stratification index CAMSIS; SD Id = standardized sample standard deviation in identification in samples in which all participants shared
membership in the same entity. a physical symptoms measure is not published in indicated article.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 82
Table 4 Interpretation of Results in Terms of Robustness of Evidence after Estimating and Correcting for Measurement and Publication Bias
Key questions relating to robustness of evidence
Analysis
Answer provided by statistical
significance (or other content)
Q1: Is the overall summary effect statistically significant when correcting for
measurement bias due to sampling error?
Random-effects model
Yes (r = .21, 95%CIs [.17, .25])
Q2: How many independent studies with a correlation of zero would be needed
in file drawers to reduce summary effect to a small correlation of r = .05?
Fail-safe N for effect size analysis
217 studies
Q3: Is there evidence of publication bias as indicated by a statistically
significant linear relationship between effect size and standard error?
Funnel plot asymmetry test
No
Q4a: After computing effect sizes that are potentially missing due to
publication bias, is the unbiased summary effect statistically significant?
Trim and fill method
Yes
Q4b: Does unbiased summary effect estimate have the same substantive
implications than random-effects model estimate?
Yes (r = .18, 95%CIs [.14, .22])
Q5a: Is the overall summary effect in unpublished studies statistically
significant?
Subgroup analysis by publication status
Yes
Q5b: Does the summary effect based on published studies have larger
magnitude than the summary effect based on unpublished studies?
No
Q6a: Does the set of studies have evidential value?
P-curve analysis
Yes
Q6b: Is there evidence of publication bias due to p-hacking (or ambitious p-
hacking)?
No
Q6c: Is there evidence that the existing evidential value is inadequate (due to
underpowered studies)?
No
Q7a: Is there evidence of publication bias as indicated by a statistically
significant relationship between sample size and magnitude of effect size?
Cumulative meta-analysis
No
Q7b: Is there evidence of bias due to small-study effects (such that studies with
smaller samples have larger effect sizes)?
No
Q7c: Does summary effect estimate based on large samples have the same
substantive implications as the random-effects model estimate?
Yes (r = .20, 95%CIs [.15, .25])

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 83
Table 5 Overview of Effect Sizes and Tests of Categorical Moderation Concerning the
Organizational Identification–Health Relationship (58 independent samples, 102 effect sizes, N =
19,799)
Moderator
r
95% CIs
Z
k
Q
Identification focus
Workgroup identification
.21
.14, .28
5.90
16 (26)
202.90**
Organizational identification
.21
.17, .26
9.20
32 (42)
298.27**
Between-group Q
.14
Identification scale
Doosje et al. (1995)
.34
.25, .43
6.98
12
67.78**
Mael & Ashforth (1992)
.18
.11, .24
5.05
19
165.87**
van Dick et al. (2004)
.17
.02, .32
2.22
4
10.80**
Residual others
.18
.12, .25
5.54
23
119.16**
Between-group Q
9.80**
Health Valence
Absence of Stress
.18
.12, .23
6.43
35 (45)
443.96**
Presence of Well-Being
.27
.22, .33
9.37
13 (23)
129.05**
Between-group Q
7.71**
Health Index
Psychological Health
.23
.18, .28
9.44
36 (48)
407.17**
Physical Health
.16
.10, .22
5.39
10 (22)
90.69**
Between-group Q
4.87**
Study methodology
Cross-sectional
.22
.18, .26
10.04
50
377.65**
Longitudinal/experimental
.13
.01, .24
2.11
8
19.03**
Between-group Q
2.37
Note. * p < .10, ** p < .05; r indicates sample size weighted corrected correlation between organizational
identification and health; 95% CIs indicate lower and upper limits of 95% confidence interval; The k column
indicates total number of samples including dependent and independent subgroups; Number in parenthesis
indicates number of dependent samples; The Q column indicates heterogeneity within each level of a variable
(Qw) and the test for heterogeneity between levels of a variable (Qb).

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 84
Figure 1 Forest Plot Displaying Effect Sizes, Confidence Intervals, and Weights in Estimation of
Summary Effect for Independent Samples and Distribution of Effect Sizes around Summary Effect
of Identification–Health Relationship
Outcome Study name Statistics for each study Correlation and 95% CI
Lower Upper
Correlation limit limit
SPhySymptomsOI Ashforth & Saks (1996) 0.08 -0.03 0.19
SPsyExhaustionOI Avanzi, Fraccaroli et al. (2014) 0.16 0.02 0.29
Combined Avanzi, Schuh et al. (2014) 0.18 0.04 0.31
HPsyGHQOI Avanzi, van Dick et al. (2012) Study 1 0.08 -0.06 0.22
HPsyGHQOI Avanzi, van Dick et al. (2012) Study 2 0.08 -0.09 0.24
Combined Barbier, Dardenne et al. (2013) 0.26 0.17 0.34
SPsyDepersonalisationOIBedeian (2007) 0.32 0.23 0.41
SPsyStressOI Bizumic, Reynolds et al. (2009) 0.16 -0.03 0.33
Combined Bjerregaard, Haslam et al. (2014) 0.42 0.37 0.46
SPhySymptomsOI Christ (2004) 0.08 -0.13 0.28
SPsyStressWI Cicero, Pierro et al. (2007) Study 1 0.29 0.19 0.39
SPhySymptomsWI Cicero, Pierro et al. (2007) Study 2 0.11 -0.10 0.31
Combined Das, Dharwadkar et al. (2008) 0.20 0.02 0.35
Combined Fampri (2003) 0.20 -0.04 0.42
Combined Frisch et al. (2014) 0.05 -0.16 0.25
SPsyStressWI Galang & Jones (2014) -0.41 -0.59 -0.18
Combined Grubba & Ahlswede (2002) 0.17 0.06 0.27
HPsySatisfactionLifeOI Harris & Cameron (2005) 0.14 -0.12 0.38
Combined Haslam & Reicher (2006) 0.47 0.03 0.75
SPsyStressWI Haslam, Jetten et al. (2009) 0.47 0.17 0.69
SPsyStressWI Haslam, O'Brien et al. (2005) 0.63 0.40 0.79
Combined Hauesser, Kattenstroth et al. (2012) 0.11 -0.09 0.30
Combined Herrbach (2006) 0.04 -0.07 0.14
Combined Horton, McClelland et al. (2014) Study 1 0.36 0.12 0.55
Combined Horton, McClelland et al. (2014) Study 2 0.20 0.10 0.29
Combined Horton, McClelland et al. (2014) Study 3 0.31 0.21 0.40
SPsychStressOI Ishii (2012) 0.41 0.27 0.53
Combined Jimmieson, McKimmie et al. (2010) 0.13 -0.03 0.28
Combined Knight & Haslam (2010a) Study 1 0.25 0.14 0.36
Combined Knight & Haslam (2010a) Study 2 0.32 0.28 0.36
Combined Knight & Haslam (2010b) Study 1 0.15 -0.04 0.32
Combined Knight & Haslam (2010b) Study 2 0.66 0.45 0.79
SPsyStressWI Knowles & Smith (2013) 0.11 -0.06 0.27
Combined Kreiner & Ashforth (2004) 0.28 0.18 0.38
Combined Matheson & Cole (2004) 0.17 -0.07 0.38
Combined Menzel (2007) 0.10 -0.04 0.22
HPsyGHQOI Merecz & Andysz (2012) 0.18 0.10 0.26
Combined Mishra & Bhatnagar (2010) 0.35 0.27 0.43
Combined Nieuwenhuis, Knight et al. (2014) Study 1 0.34 0.20 0.47
Combined Nieuwenhuis, Knight et al. (2014) Study 2 0.21 0.04 0.37
Combined O'Brien & Haslam (2003) 0.21 0.15 0.27
Combined Pisarski, Lawrence et al. (2008) -0.07 -0.15 0.02
SPsychStressOI Portaluri (2013) -0.14 -0.34 0.08
HPsySatisfactionLifeOI Sani, Herrera et al. (2012) Study 2 0.47 0.33 0.59
Combined Sani, Magrin et al. (2010) 0.51 0.34 0.65
Combined Steffens, Yang et al. (2014) 0.23 0.07 0.38
SPsyExhaustionOI Suh, Houston et al. (2010) 0.27 0.24 0.30
SPsyStressWI Topa & Moriano (2013) 0.27 0.18 0.36
SPhySymptomsOI van Dick & Wagner (2002) Study 1 0.27 0.14 0.39
SPhySymptomsWI van Dick & Wagner (2002) Study 2 0.28 0.17 0.38
Combined van Dick, Stierle et al. (2007) 0.07 -0.04 0.17
SPsyNegAffectOI van Dick, Wagner, Lemmer (2004) 0.05 -0.04 0.14
Combined van Dick, Wagner, S., & C. (2004) 0.08 -0.01 0.17
SPsyStressOI Waszkowska, Andysz et al. (2014) 0.30 0.21 0.37
SPhysicalSalivaOI Wegge, Schuh et al. (2012) 0.07 -0.13 0.27
Combined Wegge, van Dick et al. (2006) Study 2 0.36 0.22 0.49
SPsyStressOI West (2005) 0.36 0.18 0.52
SPsyStressOI Zhang, Liu et al. (2011) -0.18 -0.28 -0.08
0.21 0.17 0.25 -1.00 -0.50 0.00 0.50 1.00
Exhaustion Invigoration
PlotofCorrelationsand95%CIsStudy Correlationsand95%CIs
Invigoration
Exhaustion

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 85
Figure 2 P-Curve Displaying Observed P-Curve and Expected P-Curve if Effect Size was Zero.

SOCIAL IDENTIFICATION AND HEALTH IN ORGANIZATIONS 86
Figure 3 Cumulative Forest Plot Displaying Effect Sizes and Confidence Intervals with Studies
Sorted by Sample Size (Smaller Studies are Cumulatively Added towards Bottom of Forest Plot)
Outcome Study name Cumulative correlation (95% CI)
Lower Upper
Point limit limit Total
0.27 0.24 0.30 SPsyExhaustionOI Suh, Houston et al. (2010) 3999
0.17 -0.05 0.37 SPsyNegAffectOI van Dick, Wagner, Lemmer (2004) 4458
0.14 -0.03 0.30 Combined van Dick, Wagner, S., & C. (2004) 4891
0.17 0.05 0.29 SPsyStressWI Topa & Moriano (2013) 5279
0.18 0.08 0.27 Combined Horton, McClelland et al. (2014) Study 2 5660
0.20 0.12 0.28 SPsyDepersonalisationOIBedeian (2007) 6039
0.15 0.03 0.27 SPsyStressOI Zhang, Liu et al. (2011) 6407
0.14 0.03 0.25 Combined van Dick, Stierle et al. (2007) 6774
0.13 0.02 0.23 Combined Herrbach (2006) 7139
0.15 0.05 0.24 Combined Horton, McClelland et al. (2014) Study 3 7482
0.15 0.06 0.23 Combined Grubba & Ahlswede (2002) 7820
0.16 0.08 0.24 Combined Knight & Haslam (2010a) Study 2 9463
0.17 0.10 0.25 Combined Kreiner & Ashforth (2004) 9793
0.18 0.11 0.25 SPsyStressWI Cicero, Pierro et al. (2007) Study 1 10122
0.18 0.11 0.24 SPhySymptomsOI Ashforth & Saks (1996) 10416
0.18 0.11 0.24 Combined Knight & Haslam (2010a) Study 1 10704
0.19 0.12 0.25 SPhySymptomsWI van Dick & Wagner (2002) Study 2 10987
0.18 0.12 0.24 Combined Menzel (2007) 11207
0.19 0.13 0.24 SPhySymptomsOI van Dick & Wagner (2002) Study 1 11408
0.18 0.13 0.24 SPsyExhaustionOI Avanzi, Fraccaroli et al. (2014) 11603
0.18 0.12 0.23 HPsyGHQOI Avanzi, van Dick et al. (2012) Study 1 11798
0.18 0.13 0.23 Combined Avanzi, Schuh et al. (2014) 11990
0.19 0.14 0.25 Combined Bjerregaard, Haslam et al. (2014) 13264
0.20 0.14 0.25 Combined Nieuwenhuis, Knight et al. (2014) Study 1 13431
0.20 0.15 0.26 Combined Wegge, van Dick et al. (2006) Study 2 13592
0.21 0.16 0.26 SPsychStressOI Ishii (2012) 13751
0.21 0.16 0.26 Combined Jimmieson, McKimmie et al. (2010) 13906
0.22 0.17 0.27 HPsySatisfactionLifeOI Sani, Herrera et al. (2012) Study 2 14056
0.21 0.16 0.26 HPsyGHQOI Avanzi, van Dick et al. (2012) Study 2 14196
0.21 0.16 0.26 Combined Steffens, Yang et al. (2014) 14336
0.21 0.16 0.26 SPsyStressWI Knowles & Smith (2013) 14475
0.21 0.16 0.26 Combined Das, Dharwadkar et al. (2008) 14607
0.21 0.16 0.26 Combined Nieuwenhuis, Knight et al. (2014) Study 2 14735
0.21 0.17 0.25 Combined O'Brien & Haslam (2003) 15825
0.21 0.16 0.25 SPsyStressOI Bizumic, Reynolds et al. (2009) 15938
0.21 0.16 0.25 Combined Knight & Haslam (2010b) Study 1 16050
0.21 0.17 0.25 SPsyStressOI West (2005) 16149
0.21 0.17 0.25 Combined Hauesser, Kattenstroth et al. (2012) 16245
0.21 0.16 0.25 SPhysicalSalivaOI Wegge, Schuh et al. (2012) 16341
0.20 0.16 0.25 SPhySymptomsOI Christ (2004) 16434
0.21 0.17 0.25 Combined Sani, Magrin et al. (2010) 16527
0.21 0.17 0.25 Combined Frisch et al. (2014) 16617
0.21 0.16 0.25 SPhySymptomsWI Cicero, Pierro et al. (2007) Study 2 16706
0.20 0.16 0.24 SPsychStressOI Portaluri (2013) 16790
0.20 0.16 0.24 HPsyGHQOI Merecz & Andysz (2012) 17366
0.20 0.16 0.24 Combined Matheson & Cole (2004) 17438
0.20 0.16 0.24 Combined Fampri (2003) 17506
0.19 0.15 0.23 SPsyStressWI Galang & Jones (2014) 17571
0.19 0.15 0.23 Combined Horton, McClelland et al. (2014) Study 1 17636
0.19 0.15 0.23 HPsySatisfactionLifeOI Harris & Cameron (2005) 17696
0.20 0.16 0.24 Combined Knight & Haslam (2010b) Study 2 17743
0.20 0.16 0.24 SPsyStressWI Haslam, O'Brien et al. (2005) 17783
0.21 0.17 0.25 SPsyStressWI Haslam, Jetten et al. (2009) 17813
0.21 0.17 0.25 Combined Haslam & Reicher (2006) 17828
0.20 0.16 0.24 Combined Pisarski, Lawrence et al. (2008) 18358
0.20 0.16 0.25 SPsyStressOI Waszkowska, Andysz et al. (2014) 18858
0.21 0.17 0.25 Combined Barbier, Dardenne et al. (2013) 19331
0.21 0.17 0.25 Combined Mishra & Bhatnagar (2010) 19799
0.21 0.17 0.25 -1.00 -0.50 0.00 0.50 1.00
Exhaustion Inv igoration
PlotofCumulativeCorrelationsand95%CIs
Study
Cumulative
SampleSize
Exhaustion Invigoration