Organizational Behavior and Human Decision Processes 100 (2006) 202–215

www.elsevier.com/locate/obhdp

Relationships between organizational justice, identiWcation with

organization and work unit, and group-related outcomes
Maria-Elena Olkkonen ¤, Jukka Lipponen
Department of Social Psychology, University of Helsinki, Finland

Received 13 July 2004

Available online 20 October 2005

Abstract

This research examined the diVerential antecedents and consequences of organizational identiWcation and work-unit identiWca-
tion. SpeciWcally, we hypothesized that organization-focused procedural justice and distributive justice would be positively related to
organizational identiWcation, whereas supervisor-focused interactional justice would be positively related to work-unit identiWcation.
A further hypothesis was that organizational identiWcation would relate to organization-focused outcomes (turnover intentions and
extra-role behavior toward the organization), and work-unit identiWcation to work-unit-focused outcomes (extra-role behavior
toward the work unit). Our results from a sample of 160 employees of a research institution supported these hypotheses. In addition,
we found some evidence that organizational identiWcation and work-unit identiWcation diVerentially mediated the relationships
between organization-focused and supervisor-focused justice, and organization-focused and work-unit-focused outcomes. We dis-
cuss our Wndings in terms of their implications for social-identity research on organizational identiWcation, and for research on orga-
nizational justice.
© 2005 Elsevier Inc. All rights reserved.

Keywords: Multiple organizational identiWcations; Social identity; Organizational justice; Extra-role behavior; Turnover intentions

Introduction (Van Knippenberg & van Schie, 2000), for instance.

Since the late 1990s, psychologists have expressed
widespread interest in organizational identiWcation,
which is generally deWned as “the perception of oneness
with or belongingness to” the organization (Ashforth &
Mael, 1989, p. 34; see Riketta, 2005, for a recent review).
One reason for this interest is certainly the positive
eVects that organizational identiWcation has been shown
to have on various work outcomes. It has been linked
with lowered turnover intentions (Abrams, Ando, &
Hinkle, 1998; Tyler & Blader, 2000), increased extra-role
behavior (Dukerich, Golden, & Shortell, 2002; Tyler &
Blader, 2000, 2001) and increased job satisfaction
*
Corresponding author. Fax: +358 9 24 13 496.
E-mail address: maria-elena.olkkonen@ttl.W (M.-E. Olkkonen).
Recently, however, researchers have begun to note one
signiWcant limitation in the literature. While previous
research has predominantly taken the organization as a
whole as the focus of identiWcation, the workplace con-
text also suggests multiple other foci such as the work
group and the department (e.g., Ashforth & Johnson,
2001; Christ, van Dick, Wagner, & Stellmacher, 2003;
Van Dick & Wagner, 2002; Van Dick, Wagner, Stellm-
acher, & Christ, 2004; Van Knippenberg & van Schie,
2000). It thus appears that organizational identiWcation
is, in fact, a multi-foci construct.
This notion of multiple organizational identiWcations,
in turn, has led to the emergence of two new research
areas. The Wrst focuses on the consequences, and speciW-
cally on whether diVerent identiWcations are diVeren-
tially related to work outcomes. A few studies have
already addressed these issues with promising results

0749-5978/$ - see front matter © 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.obhdp.2005.08.007
 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
Fig. 1. Summary of the hypothesized pattern of relationships.
(Christ et al., 2003; Van Dick et al., 2004). However,
given the potential implications in terms of fostering
positive work outcomes, further research is still needed.
The second research area concerns the antecedents of
multiple identiWcations, and speciWcally whether diVer-
ent identiWcations have diVerent sets of predictors.
Indeed, we cannot assume that the factors found to
inXuence organizational identiWcation would similarly
inXuence identiWcation with other foci. Further, estab-
lishing the diVerential antecedents of diVerent identiWca-
tions may shed more light on the strategies that
organizations should employ when fostering employee
identiWcation. Yet, these issues have been largely
neglected in previous research.
The aim of the present study was therefore to extend
the research on multiple organizational identiWcations
by examining the diVerential antecedents and conse-
quences of organizational identiWcation and work-unit
identiWcation. First, given the premise of the group
engagement model that perceived organizational justice
is positively related to organizational identiWcation
(Tyler & Blader, 2000, 2003), we consider employees’
perceptions of procedural justice, distributive justice,
and interactional justice (Greenberg & Colquitt, 2005) as
the key antecedent. SpeciWcally, we wanted to Wnd out
whether organization-focused perceptions of procedural
justice and distributive justice are related to organiza-
tional identiWcation, and whether, on the other hand,
supervisor-focused perceptions of interactional justice
are related to work-unit identiWcation.
Second, we investigated whether organizational iden-
tiWcation was further related to organization-focused
outcomes (turnover intentions and extra-role behavior
toward the organization), and whether work-unit identi-
Wcation was related to work-unit-focused outcomes
(extra-role behavior toward the work unit). Finally, we
looked at whether organizational identiWcation medi-
ated the relationship between organization-focused jus-
tice perceptions and our organization-focused outcomes,
and whether work-unit-identiWcation mediated the rela-
203
tionship between supervisor-focused justice perceptions
and our work-unit-focused outcomes.1
The speciWc pattern of relationships tested in our
study is summarized in Fig. 1 and the literature on which
they are based is reviewed below. First, however, we will
introduce our theoretical framework for examining mul-
tiple organizational identiWcations.
The social-identity approach and multiple identiWcations
According to the social-identity approach, which
comprises the theories of the social identity (Hogg &
Abrams, 1988; Tajfel & Turner, 1979) and self-categori-
zation (Turner, Hogg, Oakes, Reicher, & Wetherell,
1987), people deWne themselves, to a large extent, in
terms of their social-group memberships. This group-
based deWnition of the self forms an individual’s social
identity: “that part of an individual’s self-concept which
derives from his knowledge of his membership of a
social group (or groups) together with the value and
emotional signiWcance attached to that membership”
(Tajfel, 1978, p. 63). The importance of social identity, in
turn, is reXected in the degree of identiWcation with the
group in question. SpeciWcally, identiWcation refers to “a
relatively enduring state that reXects an individual’s
readiness to deWne him- or herself as a member of a par-
ticular social group” (Haslam, 2001, p. 383).
The social-identity approach also holds that social-
group memberships have signiWcant eVects on an indi-
vidual’s perceptions, emotions, and behavior. More
speciWcally, social identity leads to a tendency to act
1
Note that throughout this paper we treat perceptions of organiza-
tional justice as antecedents and turnover intentions and extra-role be-
havior as consequences of organizational identiWcation and work-unit
identiWcation. However, given that we analyze correlational and cross-
sectional data, we are not able to answer questions of causality on the
basis of our Wndings. Thus, by using the terms antecedents and conse-
quences for certain variables, we merely want to emphasize that these
variables may be antecedents and consequences of organizational iden-
tiWcation and work-unit identiWcation.
204 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
and feel in accordance with one’s group membership
and, as a consequence, is associated with more positive
feelings about the group members, cooperation, and
positive group-related attitudes and behaviors (Hogg &
Abrams, 1988; Hogg & Terry, 2001). In addition, and
what is most relevant to the present study, this frame-
work suggests that an individual’s social identity is
actually a collection of multiple more-or-less disparate
identities (Deaux, 1996; Hogg & Abrams, 1988; Pratt,
2001). In principle, an individual may have as many
social identities as he or she has social-group member-
ships (Hogg & Abrams, 1988). Finally, the social-iden-
tity approach also assumes that people often feel
multiple identiWcations simultaneously, and that iden-
tiWcation with one group does not have to exclude
identiWcation with another (Deaux, 1996; Hogg &
Abrams, 1988).
Recently, these propositions concerning the social-
identity approach have been successfully applied to
identiWcation processes in an organizational context
(see Haslam, 2001; Haslam, van Knippenberg, Platow,
& Ellemers, 2003; Hogg & Terry, 2001; for recent
reviews). Research has shown that organizations con-
stitute an important source of an individual’s identity
(Hogg & Terry, 2001). Similarly, as previously noted,
organizational identiWcation has been linked with vari-
ous positive group-related outcomes, such as extra-role
behavior (e.g., Dukerich et al., 2002; Tyler & Blader,
2000). Further, there is evidence that individuals distin-
guish between multiple foci of identiWcation in the
workplace, and that they simultaneously identify with
them (Christ et al., 2003; Lipponen, Helkama, & Juslin,
2003; Reade, 2001a; Van Dick et al., 2004; Van
Knippenberg & van Schie, 2000). It should be noted,
however, that much of this evidence concerns the dis-
tinctiveness of identiWcation foci that are structurally
distant from one another, such as the immediate work
group and the organization as a whole (Christ et al.,
2003; Van Dick et al., 2004; Van Knippenberg & van
Schie, 2000), or the local and the global organization
(Reade, 2001a). Thus, we do not yet know whether
individuals can also distinguish between entities that
are more closely nested within one another and conse-
quently particularly likely to blur (Ashforth &
Johnson, 2001). Therefore, in the present study we con-
sidered the identiWcation foci that are hierarchically
right next to each other, the organization as a whole
and the work unit.
Perceptions of organizational justice as antecedents of
organizational identiWcation and work-unit identiWcation
Although a number of studies have examined the
antecedents of organizational identiWcation (e.g.,
Lipponen, Olkkonen, & Moilanen, 2004; Mael &
Ashforth, 1992; Smidts, Pruyn, & van Riel, 2001), little
research has simultaneously considered the antecedents
of multiple organizational identiWcations. However,
Reade (2001b) considered in her seminal study whether
identiWcation with a global multinational corporation
and its local subsidiary was diVerentially explained by
prestige, distinctiveness, and group-formation factors
such as interpersonal similarity and liking, which were
measured as focusing on both the global and the local
organization. The results were promising. IdentiWca-
tion with the global organization was best explained by
global antecedents, whereas identiWcation with the
local subsidiary was best explained by local anteced-
ents (Reade, 2001b). This suggests that multiple organi-
zational identiWcations may have diVerential
antecedents, and speciWcally that there may be a corre-
spondence between the focus of identiWcation and its
antecedents. It nevertheless remains an empirical ques-
tion whether a similar pattern of results emerges with
respect to other foci of identiWcation and other sets of
antecedents. The present study addresses this question
by focusing on perceptions of organizational justice as
the antecedents of organizational identiWcation and
work-unit identiWcation.
According to recent research (e.g., Colquitt, 2001;
Greenberg & Colquitt, 2005), the concept of organiza-
tional justice can be deWned in terms of three distinct
dimensions: (1) distributive justice, referring to the per-
ceived fairness of the outcomes and the allocation of
resources in the workplace, (2) procedural justice,
referring to the perceived fairness of the formal deci-
sion-making procedures used in the organization, and
(3) interactional justice, referring to the perceived fair-
ness of the interpersonal treatment received from the
supervisor. These justice perceptions have been linked
meta-analytically to a variety of important work out-
comes, such as job satisfaction, organizational com-
mitment, and organizational-citizenship behaviors
(Cohen-Charash & Spector, 2001; Colquitt, Conlon,
Wesson, Porter, & Ng, 2001) There is also some evi-
dence of an association between justice perceptions
and organizational identiWcation (Lipponen et al.,
2004; Tyler & Blader, 2000, 2002; Tyler, Degoey, &
Smith, 1996).
Why, then, should perceptions of organizational jus-
tice be related to employee identiWcation with the orga-
nizational foci? Following the lines of the group
engagement model developed by Tyler and Blader (2000,
2003), we argue that justice perceptions should aVect
organizational identiWcations given the positive social-
identity-relevant information that justice communicates
to individuals. More speciWcally, justice communicates
to individuals that they are respected members within
their group, and that they can be proud of their group
membership. Furthermore, through its link to these
feelings of respect and pride, it should be further
related to increased identiWcation with the group
 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
(Tyler & Blader, 2000, 2003).2 It thus seems plausible
that perceptions of organizational justice should be posi-
tively associated with organizational identiWcation and
work-unit identiWcation.
However, we also suggest that perceptions of proce-
dural justice, distributive justice, and interactional jus-
tice should not be similarly related to organizational
identiWcation and work-unit identiWcation. This would
seem plausible given the recent multi-foci research in the
literature on organizational justice suggesting that, like
identiWcation, employee perceptions of justice have mul-
tiple foci such as the organization as a whole and the
supervisor (Cropanzano, Prehar, & Chen, 2002; Master-
son, Lewis, Goldman, & Taylor, 2000). Further, given
that organizations often establish general guidelines for
decision-making procedures, it is traditionally assumed
that perceptions of procedural justice focus on the orga-
nization as a whole. On the other hand, given that super-
visors usually administer interpersonal treatment quite
independently, perceptions of interactional justice are
suggested to focus on the immediate supervisor
(Cropanzano et al., 2002; Masterson et al., 2000). It thus
seems plausible to conceptualize procedural justice as
organization-focused and interactional justice as super-
visor-focused.
It should be noted, however, that recent studies by
Liao and Rupp (2005) and Rupp and Cropanzano
(2002) have demonstrated that perceptions of proce-
dural justice and interactional justice can, in fact, con-
cern both the organization as a whole and the
supervisor. Nevertheless, in terms of outcomes of justice
perceptions, these and the earlier multi-foci studies
(Cropanzano et al., 2002; Masterson et al., 2000) have
oVered rather unequivocal evidence. Accordingly, orga-
nization-focused justice perceptions have been shown to
predict organization-focused outcomes such as turnover
intentions and organization-directed citizenship behav-
ior, while supervisor-focused justice perceptions have
been shown to predict supervisor-focused outcomes such
as supervisor-directed citizenship behavior and satisfac-
tion (Cropanzano et al., 2002; Masterson et al., 2000;
Rupp & Cropanzano, 2002).
In the light of these Wndings, it is thus plausible to
expect organization-focused procedural justice to be
related to organizational identiWcation. However,
although this has not been discussed in previous work
on multi-foci justice, we suggest that distributive justice
could also be related to organizational identiWcation.
Indeed, just as organizations establish general decision-
making procedures, they often lay down general guide-
2
Although Tyler and Blader (2002) found evidence that relationships
between justice perceptions and organizational identiWcation are medi-
ated by feelings of pride and respect, given the complexity of our model,
we focus only on the direct relationships between justice perceptions and
organizational identiWcation and work-unit identiWcation.
205
lines for the allocation of rewards and resources, which
may also be beyond the control of a single supervisor.
Thus, it is plausible to assume that perceptions of dis-
tributive justice are organization- rather than supervi-
sor-based and, as a consequence, they should relate to
organizational identiWcation rather than work-unit iden-
tiWcation. Supervisor-focused interactional justice, in
contrast, is likely to be related to work-unit identiWca-
tion. This would seem plausible, since at least in organi-
zations that consist of several diVerentiated work units,
the supervisor is probably seen more as a representative
of the work unit than of the organization as a whole (see
also Sousa & Vala, 2002). We therefore hypothesize that:
Hypothesis 1. Organization-focused procedural justice
and distributive justice are positively related to organiza-
tional identiWcation.
Hypothesis 2. Supervisor-focused interactional justice is
positively related to work-unit identiWcation.
The consequences and mediating role of organizational
identiWcation and work-unit identiWcation
Although the consequences of organizational identiW-
cation have been widely studied (e.g., Riketta, 2005),
only recently have researchers begun to consider the
consequences of multiple identiWcations, and speciWcally
whether diVerent identiWcations are related to diVerent
outcomes. Nevertheless, there is already empirical evi-
dence that this could be the case (Christ et al., 2003; Van
Dick et al., 2004). For instance, Christ et al. (2003) found
that school teachers’ team identiWcation predicted orga-
nizational-citizenship behaviors (OCBs) towards their
team, organizational identiWcation predicted OCBs
towards their organization, whereas career identiWcation
predicted OCBs towards their own qualiWcations. Van
Dick et al. (2004), in turn, replicated these Wndings, and
also showed that team identiWcation predicted team cli-
mate, whereas organizational identiWcation predicted
turnover intentions and job satisfaction. It thus appears
that identiWcation with a certain organizational focus
may only predict outcomes relevant to it (see also Van
Knippenberg & Ellemers, 2003).
One should nevertheless be aware of the limited num-
ber of studies supporting this interesting pattern of rela-
tionships. Further, there is at least one study that
contradicts the correspondence between the focus of
identiWcation and its outcomes: Van Knippenberg and
van Schie (2000) found work-group identiWcation rather
than organizational identiWcation to be associated with
turnover intentions, which is generally considered to be
an organization-focused outcome (e.g., Masterson et al.,
2000; Van Dick et al., 2004). Therefore, to provide fur-
ther insight into the outcomes of multiple identiWcations,
we investigated whether organizational identiWcation
and work-unit identiWcation were diVerentially related to
206 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215

three organization-focused and work-unit-focused found to be mediated by supervisor-focused social-

outcomes. We concentrate on turnover intentions and exchange relationships (Cropanzano et al., 2002; Mast-
extra-role behavior toward the organization in terms of erson et al., 2000; Rupp & Cropanzano, 2002). It thus
organization-focused outcomes, and on extra-role also seems plausible that organizational identiWcation
behavior toward the work unit in terms of work-unit- mediates the relationships between organization-focused
focused outcomes. Moreover, in line with previous justice and outcomes, whereas work-unit identiWcation
research Wndings (e.g., Van Dick et al., 2004), we propose mediates the relationship between supervisor-focused
that organizational identiWcation is negatively related to justice and work-unit-focused outcomes. We therefore
turnover intentions and positively related to extra-role hypothesize that:
behavior toward the organization, while work-unit
Hypothesis 5. Organizational identiWcation mediates the
identiWcation is positively related to extra-role behavior
relationship between organization-focused procedural
toward the work unit.
justice and distributive justice and organization-focused
Hypothesis 3. Organizational identiWcation is negatively outcomes (turnover intentions and extra-role behavior
related to turnover intentions and positively related to toward the organization).
extra-role behavior toward the organization.
Hypothesis 6. Work-unit identiWcation mediates the rela-
Hypothesis 4. Work-unit identiWcation is positively tionship between supervisor-focused interactional justice
related to extra-role behavior toward the work unit. and work-unit-focused outcomes (extra-role behavior
towards the work unit).
As noted above, previous research has linked percep-
tions of organizational justice with organizational iden-
tiWcation, which in turn has been linked with turnover
Method
intentions and extra-role behaviors (e.g., Tyler & Blader,
2000). It thus seems plausible that organizational identi-
Organizational setting and procedure
Wcation and work-unit identiWcation at least partially
mediate the relationship between justice perceptions and
The context for this research was a Finnish research
our three outcomes. Indeed, this proposition concerning
institution with 270 employees working in one geo-
the mediating role of identiWcation is implicit in the
graphically placed facility. In terms of organizational
group engagement model, according to which organiza-
structure, the level immediately below the institution as a
tional identiWcation may be a second mediating factor,
whole was that of the work units, which were further
in addition to pride and respect, between organizational
divided into work groups. Three of the four work units
justice and group-related outcomes (Tyler & Blader,
based their activities on the diVerent research Welds cov-
2000, 2003).3 On the other hand, the mediating role of
ered by the institution, and their members were mainly
identiWcation is also consistent with the proposition
research professionals, while the fourth work unit ful-
inherent in the social-identity approach that people’s
Wlled supportive functions. The sizes of the units varied
experiences in the social world are related to their group
from 32 to 112 employees (M D 67). Given the identiWca-
identiWcation, which in turn guides their behavior in the
tion focus of this study, organizational identiWcation
social world (Tajfel & Turner, 1979).
thus referred to identiWcation with the research institu-
We further suggest that organizational identiWcation
tion as a whole, and work-unit identiWcation to identiW-
and work-unit identiWcation should diVerentially medi-
cation with the particular work unit.
ate the relationship between organization-focused and
The data for the study were collected by means of
supervisor-focused justice perceptions and our out-
self-administered questionnaires delivered in person to
comes. This would seem plausible given that the links
all employees by the Wrst author in spring 2001. All of
between organization-focused justice perceptions and
the participants were assured that their responses would
organization-focused outcomes have been found to be
be anonymous and would not be made available to com-
mediated by organization-focused social-exchange rela-
pany personnel. Of the 270 questionnaires delivered, 160
tionships (Masterson et al., 2000; Rupp & Cropanzano,
usable ones were returned and the eVective response rate
2002). On the other hand, the links between supervisor-
was thus 59%. The age of the respondents varied from 22
focused justice perceptions and outcomes have been
to 58 years (M D 34.6), and 74.4% of them were men.
Most of them (66.3%) had a university degree. Unfortu-
3
Tyler and Blader also empirically tested this proposition on a sam- nately, since the organization did not provide us with
ple of 404 Chicago residents, but unfortunately, when testing the medi- statistical information regarding all the demographic
ating role of identiWcation, they combined organizational identiWcation
variables, we were not able to assess statistically whether
with feelings of pride and respect to form the single construct of status
judgments (see Tyler & Blader, 2000, p. 196). Thus, the present study there was response bias. However, according to the com-
represents an attempt to provide more direct evidence of the mediating pany representatives, our sample was representative of
role of identiWcation in itself. the personnel.
 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
Measures
Distributive justice
Employee perceptions of organization-focused dis-
tributive justice were measured on a six-item scale. Five
of the items were taken directly from Moorman’s (1991)
Distributive Justice Scale and they assessed the degree to
which the rewards received by employees were perceived
to be fair in relation to their responsibilities, work expe-
rience, amount of eVort, and other work-related contri-
butions (e.g., “In my organization I am fairly rewarded
for the amount of eVort I have put in”). The scale also
included one item assessing the degree to which the
rewards were perceived to be consistent with the rewards
received by other employees engaged in a similar work
(“In my organization I am fairly rewarded with respect
to the rewards received by others doing a similar job”).
To highlight the organization as a whole as the focus of
distributive justice, all the items referred to the organiza-
tion as a whole as the context of reward allocation.
Procedural justice
Employee perceptions of organization-focused proce-
dural justice were measured on the Finnish version
(developed and tested by Elovainio, Kivimäki, & Helk-
ama, 2001) of Moorman’s (1991) seven-item Procedural
Justice Scale. These items were based on Leventhal’s
(1980) six procedural justice criteria (consistency, bias
suppression, accuracy, correctability, representativeness,
and ethicality), and focused on the structural aspects of
the decision-making process (e.g., “Everyone who is
aVected by the decision has a chance to voice his or her
opinion”). To highlight the organization as a whole as
the focus of procedural justice, the respondents were
instructed to think about the items in this respect.
Interactional justice
Employee perceptions of supervisor-focused interac-
tional justice were measured on seven items, six of which
were taken directly from the Finnish version of Moor-
man’s (1991) Interactional Justice Scale (developed and
tested by Elovainio et al., 2001). The items were based on
the work of Bies and Moag (1986), and assessed the
extent to which the supervisor provided explanations of
and justiWcation for his or her decisions, and treated his
or her subordinates with dignity and respect (e.g., “My
supervisor shows concern for my rights as an
employee”). The scale also included one item assessing
the extent to which the supervisor provided honest
explanations for his or her decisions. To conWrm the
empirical distinctiveness of distributive justice, proce-
dural justice, and interactional justice, a principal-com-
ponent analysis (PCA) with oblimin rotation was
computed. As expected, the PCA yielded three factors
(Eigenvalue of factor 1, 6.7; factor 2, 3.24; factor 3, 1.96;
variance explained by factor 1, 35.0%; factor 2, 17.0%;
207
factor 3, 10.3%), which accounted for 62.3% of the vari-
ance. All the distributive justice items loaded on the Wrst
factor, the procedural justice items on the second, and
the interactional justice items on the third, and there
were no cross-loadings of .30 or higher.
Organizational identiWcation and work-unit identiWcation
In keeping with the recent multidimensional
approach to the conceptualization of organizational
identiWcation (Christ et al., 2003; Van Dick et al., 2004),
organizational identiWcation and work-unit identiWca-
tion were conceptualized here as two-dimensional
constructs comprising both cognitive (indicating the self-
deWnitional role of group membership) and aVective
(indicating the emotional signiWcance of group member-
ship) aspects. Both of these constructs were measured on
six items, of which three assessed the aVective aspects
and the other three the cognitive aspects.
The items measuring aVective identiWcation were
derived from Allen and Meyer (1990) AVective Commit-
ment Scale (ACS) (e.g., “I feel emotionally attached to
(name of the organization or the work unit)”), while the
items measuring cognitive identiWcation were derived
from the Finnish version of Mael and Ashforth (1992)
Organizational IdentiWcation Questionnaire (OID) (devel-
oped and tested by Lipponen et al., 2003) (e.g., “When I
talk about (name of the organization), I usually say ‘we’
rather than ‘they”’). The items on the scales for organiza-
tional identiWcation and work-unit identiWcation were
identical, and only their referent targets were diVerent
(respectively, the name of the organization and the
respondents’ own work units). We carried out a principal-
component analysis (PCA) with oblimin rotation to
conWrm the empirical distinctiveness of organizational
identiWcation and work-unit identiWcation. As expected,
the PCA yielded two factors (Eigenvalue of factor 1, 5.5;
factor 2, 2.2; variance explained by factor 1, 45.9%; factor
2, 18.0%), which accounted for 63.9% of the variance. All
the work-unit identiWcation items loaded on the Wrst fac-
tor and the organizational identiWcation items on the sec-
ond, and there were no cross-loadings of .30 or higher.
Turnover intentions
Employees’ turnover intentions were measured on
four items derived from the measures developed by
Bozeman and Perrewe (2001) and Sager, GriVeth, and
Hom (1998) (e.g., “I have often thought about quitting
working in (name of the organization)”). Given this
explicit focus on the organization as a whole and the
views expressed in previous research (e.g., Van Dick
et al., 2004), we treated turnover intentions as an organi-
zation-focused outcome.
Extra-role behaviors
In line with Van Dyne, Cummings, and McLean
Parks (1995, p. 218), we deWned extra-role behavior as
208 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215

“behavior which beneWts the organization and/or is
intended to beneWt the organization, which is discretion-
ary and which goes beyond the existing role expecta-
tions.” Further, following the lines of previous research
(e.g., Christ et al., 2003; Masterson et al., 2000), we used
separate scales for extra-role behaviors toward the orga-
nization as a whole and toward the work unit, both of
which included four items: for example, one item on the
former was, “I have made suggestions to improve the
(name of the organization)” and, on the latter, “I have
volunteered to help others in my work unit when they
have a heavy workload.”
The items in both scales were derived from existing
scales of extra-role behavior developed by O’Reilly and
Chatman (1986) and Tyler and Blader (2000), and scales
of organizational altruism developed by Smith, Organ,
and Near (1983). We conducted a principal-component
analysis (PCA) with oblimin rotation to conWrm the
empirical distinctiveness of extra-role behavior toward
the organization and toward the work unit. As expected,
the PCA yielded two factors (Eigenvalue of factor 1, 3.8;
factor 2, 1.1, variance explained by factor 1, 47.8%; fac-
tor 2, 14.2%), which accounted for 62.0% of the variance.
However, one item on the scale for extra-role behavior
toward the work unit had a high cross-loading on the
other scale (.71). We therefore decided to conduct an
additional PCA by deleting this item. This analysis
yielded two factors (Eigenvalue of factor 1, 3.3; factor 2,
1.1, variance explained by factor 1, 46.7%; factor 2,
15.9%), which accounted for 62.6% of the variance. All
the items measuring extra-role behavior toward the
organization loaded on the Wrst factor and those mea-
suring extra-role behavior toward the work unit on the
second, and there were no cross-loadings of .30 or
higher.
We used a Wve-point Likert scale throughout for mea-
suring the respondents’ level of agreement with the state-
ment (1, strongly disagree; 5, strongly agree). The
descriptives, Cronbach’s s and correlations for the Wnal
scales are presented in Table 1.
Results
We tested our hypotheses using multiple regression
analysis. In testing Hypotheses 1 and 2, we Wrst regressed
organizational identiWcation and work-unit identiWcation
on perceptions of organizational justice. The results shown
in Table 2 reveal that procedural justice ( D .34, p < .001)
and distributive justice ( D .23, p < .01) were positively
related to organizational identiWcation, whereas the eVect
for interactional justice was non-signiWcant ( D ¡.05, ns).
Hypothesis 1 was therefore supported. Interactional justice
( D .54, p < .001), on the other hand, was positively related
to work-unit identiWcation, whereas the eVects for proce-
dural justice (D.12, ns) and distributive justice (D.11, ns)
were non-signiWcant. Hypothesis 2 was thus supported.
We then regressed turnover intentions, extra-role
behavior toward the organization and extra-role behavior

Table 1
Descriptives and correlations among studied variables (N D 160)
M SD 1 2 3 4 5 6 7 8
1. Distributive justice 3.35 0.79 (.87)
2. Procedural justice 3.34 0.71 .37¤¤¤ (.82)
3. Interactional justice 4.00 0.70 .25 ¤¤ .40¤¤¤ (.93)
4. Organizational identiWcation 3.62 0.71 .35¤¤¤ .41¤¤¤ .15 (.83)
5. Work-unit identiWcation 3.76 0.83 .29¤¤¤ .37¤¤¤ .61¤¤¤ .42¤¤¤ (.92)
6. Turnover intentions 2.24 0.92 ¡.29¤¤¤ ¡.41¤¤¤ ¡.09 ¡.55¤¤¤ ¡.18¤ (.81)
7. Extra-role behavior toward the organization 2.42 0.79 .02 ¡.02 ¡.02 .26¤¤¤ .08 ¡.01 (.75)
8. Extra-role behavior toward the work unit 2.99 0.79 .01 .09 .13 .22¤¤ .40¤¤¤ ¡.03 .48¤¤¤ (.75)
Note. Cronbach’s s are shown in parentheses on the diagonal.
¤
p < .05.
¤¤
p < .01.
¤¤¤
p < .001.

Table 2
Regressions predicting organizational identiWcation and work-unit identiWcation
Dependent variable Independent variables  B SE t R2 F df
¤¤ ¤¤ ¤¤¤
Organizational identiWcation Distributive justice .23 .20 .07 2.98 .22 14.23 3
Procedural justice .34¤¤¤ .33¤¤¤ .08 4.23
Interactional justice ¡.05 ¡.00 .08 ¡.57
Work-unit identiWcation Distributive justice .11 .12 .07 1.67 .40 34.87¤¤¤ 3
Procedural justice .12 .13 .08 1.62
Interactional justice .54¤¤¤ .64¤¤¤ .08 7.87
¤¤
p < .01.
¤¤¤
p < .001.
 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215 209

Table 3
Regressions predicting turnover intentions, extra-role behavior toward the organization and extra-role behavior toward the work unit
Dependent variable Independent variables  B SE t R2 F df
¤¤¤ ¤¤¤
Turnover intentions Organizational identiWcation ¡.57 ¡.75 .10 ¡7.82 .31 34.43¤¤¤ 2
Work-unit identiWcation .06 .00 .08 .80

Extra-role behavior toward the organization Organizational identiWcation .28¤¤¤ .31¤¤¤ .10 3.26 .07 5.89¤¤ 2
Work-unit identiWcation ¡.03 ¡.00 .08 ¡.40

Extra-role behavior toward the work unit Organizational identiWcation .07 .00 .09 .85 .16 15.38¤¤¤ 2
Work-unit identiWcation .37¤¤¤ .35¤¤¤ .08 4.61
¤¤
p < .01.
¤¤¤
p < .001.

toward the work unit on organizational identiWcation
and work-unit identiWcation to test Hypotheses 3 and 4.
As shown in Table 3, organizational identiWcation
( D ¡.57, p < .001) was negatively related to turnover
intentions, whereas the eVect for work-unit identiWcation
was non-signiWcant ( D .06, ns). Similarly, organiza-
tional identiWcation ( D .28, p < .001) was positively
related to extra-role behavior toward the organization,
whereas the eVect for work-unit identiWcation was non-
signiWcant ( D ¡.03, ns). Hypothesis 3 was thus sup-
ported. Work-unit identiWcation ( D .37, p < .001), in
turn, was positively related to extra-role behavior
toward the work unit, whereas the eVect for organiza-
tional identiWcation was non-signiWcant ( D .07, ns).
Hypothesis 4 was thus supported.
Finally, to test Hypotheses 5 and 6, we applied Baron
and Kenny’s (1986) four-step strategy for testing mediat-
ing eVects with multiple regression, according to which
support for mediation requires four conditions to be
met. First, the independent variable must be related to
the mediator. Second, the mediator must be related to
the dependent variable. Third, the independent variable
must be related to the dependent variable. Finally, the
eVect of the independent variable on the dependent vari-
able must be substantially reduced (partial mediation) or
eliminated (full mediation) after the eVect of the media-
tor has been taken into account. It should be noted,
however, that several researchers (e.g., Kenny, Kashy, &
Bolger, 1998; MacKinnon, Lockwood, HoVman, West,
& Sheets, 2002; Shrout & Bolger, 2002) have recently
questioned the necessity of testing and establishing an
association between the independent and dependent
variables, suggesting that this association is implied if
the Wrst and second conditions are met. The essential
conditions for establishing mediation thus appear to
be the Wrst and second conditions (Kenny et al., 1998,
p. 260).
With regard to the Wrst and second conditions, we
have shown that perceptions of organizational justice
are related to organizational identiWcation and work-
unit identiWcation and these, in turn, are related to our
three outcomes (Tables 2 and 3). In the Wnal step of the
analysis, we needed to establish whether or not percep-
tions of organizational justice were related to each of the
three outcomes. More speciWcally, we considered
whether organization-focused procedural justice and
distributive justice were related to organization-focused
outcomes (turnover intentions and extra-role behavior
toward the organization), whether supervisor-focused
interactional justice related to work-unit-focused out-
come (extra-role behavior towards the work unit), and
whether these eVects were reduced or eliminated after
the eVects of organizational identiWcation and work-unit
identiWcation had been taken into account. The results of
these analyses are shown in Table 4.
Table 4 shows that procedural justice ( D ¡.39,
p < .001) and distributive justice ( D ¡.17, p < .05) were
negatively related to turnover intentions, thus providing
support for the third condition (Step 1). It also shows
that distributive justice ( D ¡.07, ns) was not signiW-
cantly related to turnover intentions after the eVect of
organizational identiWcation was taken into account
(Step 2), thus its relationship to turnover intentions
seems to be fully mediated by organizational identiWca-
tion. Procedural justice ( D ¡.24, p < .01), on the other
hand, was still signiWcantly related to turnover inten-
tions over and above the eVect of organizational identiW-
cation, even though this relationship was reduced (Step
2). This suggests that its relationship to turnover inten-
tions was partially mediated by organizational identiW-
cation. Thus, Hypothesis 5 was supported with regard to
turnover intentions as an outcome.
Further, procedural justice ( D ¡.03, ns), distributive
justice ( D .03, ns), and interactional justice ( D ¡.01,
ns) were found not to be signiWcantly related to extra-
role behavior toward the organization (Step 1), and sim-
ilarly, procedural justice ( D .06, ns), distributive justice
( D ¡.04, ns) and interactional justice ( D .12, ns) were
not signiWcantly related to extra-role behavior toward
the work unit (Step 1). These Wndings thus contradict the
traditionally acknowledged third condition of media-
tion. Nevertheless, as noted above, the Wrst and second
conditions of mediation, considered the most essential in
recent research, were met.
Finally, to substantiate further our mediation Wnd-
ings, we also tested our hypotheses by calculating the
210 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215

Table 4
Hierarchical regressions predicting turnover intentions, extra-role behavior toward the organization and extra-role behavior toward the work unit
Dependent variable Independent variables  B SE t R2 R2 F df
Turnover intentions
Step 1 Distributive justice ¡.17¤ ¡.19¤ .09 ¡2.15 .20 12.93¤¤¤ 3
Procedural justice ¡.39¤¤¤ ¡.49¤¤¤ .10 ¡4.75
Interactional justice .11 .14 .10 1.33
Step 2 Distributive justice ¡.07 ¡.00 .08 ¡.97 .36 .16¤¤¤ 17.20¤¤¤ 5
Procedural justice ¡.24¤¤ ¡.30¤¤ .10 ¡3.06
Interactional justice .02 .00 .11 .27
Organizational identiWcation ¡.48¤¤¤ ¡.62¤¤¤ .10 ¡6.08
Work-unit identiWcation .11 .12 .10 1.25
Extra-role behavior toward the organization
Step 1 Distributive justice .03 .00 .09 .37 .00 .08 3
Procedural justice ¡.03 ¡.00 .10 ¡.32
Interactional justice ¡.01 ¡.00 .10 ¡.15
Step 2 Distributive justice ¡.05 ¡.00 .09 ¡.53 .09 .09¤¤¤ 3.08¤ 5
Procedural justice ¡.15 ¡.16 .10 ¡1.56
Interactional justice ¡.00 ¡.00 .12 ¡.04
Organizational identiWcation .33¤¤¤ .37¤¤¤ .10 3.57
Work-unit identiWcation .01 .00 .10 .11
Extra-role behavior toward the work unit
Step 1 Distributive justice ¡.04 ¡.00 .09 ¡.47 .02 1.13 3
Procedural justice .06 .00 .10 .62
Interactional justice .12 .14 .10 1.40
Step 2 Distributive justice ¡.12 ¡.12 .08 ¡1.45 .19 .17¤¤¤ 7.43¤¤¤ 5
Procedural justice ¡.03 ¡.00 .09 ¡.36
Interactional justice ¡.14 ¡.15 .11 ¡1.41
Organizational identiWcation .09 .10 .10 1.05
Work-unit identiWcation .49¤¤¤ .46¤¤¤ .10 4.86
¤
p < .05.
¤¤
p < .01.
¤¤¤
p < .001.

Sobel statistics for each of the mediated paths (Sobel,
1982). In support of Hypothesis 5, the results of these
analyses revealed that the mediating inXuence of organi-
zational identiWcation on the relationship between dis-
tributive justice and turnover intentions (z D ¡3.93,
p < .001), as well as on the relationship between proce-
dural justice and this outcome (z D ¡4.78, p < .001), was
signiWcant. Further, the mediating inXuence of organiza-
tional identiWcation on the relationship between distrib-
utive justice and extra-role behavior toward the
organization (z D 2.83, p < .01), as well as on the relation-
ship between procedural justice and this outcome
(z D 3.22, p < .001), was also signiWcant. Finally, the medi-
ating inXuence of work-unit identiWcation on the rela-
tionship between interactional justice and extra-role
behavior toward the work unit was signiWcant (z D 4.21,
p < .001), which supports Hypothesis 6.
Discussion
The present research investigated whether organiza-
tion-focused and supervisor-focused justice perceptions
were diVerentially associated with organizational identi-
Wcation and work-unit identiWcation. We also examined
whether organizational identiWcation and work-unit
identiWcation were further diVerentially related to orga-
nization-focused and work-unit-focused outcomes, and
whether they diVerentially mediated the relationships
between organization-focused and supervisor-focused
justice perceptions and these outcomes. In general, our
Wndings supported our hypotheses, although less
strongly with respect to the diVerential mediating role of
identiWcation. In the following, we summarize our Wnd-
ings and consider their theoretical and practical implica-
tions. We then discuss the limitations of our study and
oVer suggestions for future research.
Implications for social-identity research on organizational
identiWcation
Our Wndings have several implications for social-iden-
tity research on multiple organizational identiWcations.
First, we found organization-focused and supervisor-
focused justice perceptions to be diVerentially associated
with organizational identiWcation and work-unit identi-
Wcation. SpeciWcally, whereas organization-focused per-
ceptions of procedural justice and distributive justice
were associated with organizational identiWcation,
supervisor-focused perceptions of interactional justice
 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
were associated with work-unit identiWcation. This Wnd-
ing is signiWcant because it supports the seminal Wndings
of Reade (2001b) that identiWcation with a certain orga-
nizational focus may only be inXuenced by antecedents
relevant to it. However, our study also extends this pat-
tern of relationships with respect to other sets of ante-
cedents and other foci of identiWcation. Researchers of
employee identiWcation may thus greatly beneWt from
matching the correspondence between the focus of iden-
tiWcation and its antecedents.
On the other hand, our results are relevant to social-
identity research, given our focus on perceptions of
organizational justice as antecedents. Previous research
following the social-identity approach has tended to
explain organizational identiWcation in terms of struc-
tural features of an intergroup context such as interorga-
nizational competition or the organization’s relative
status and size (e.g., Mael & Ashforth, 1992). One prob-
lem with this approach may be the fact that these kinds
of intergroup factors are relatively abstract and distant,
and may thus turn out to be rather poor predictors of
identiWcation (Lipponen, Helkama, Olkkonen, & Juslin,
2005). This has led researchers recently to consider ante-
cedents pertaining to the intraorganizational context
and employees’ everyday work experiences (e.g., the con-
tent of communication and the communication climate
in the workplace; Smidts et al., 2001). By showing that
perceptions of justice with regard to decision-making
procedures, outcome distribution and treatment in the
workplace are rather strongly related to organizational
identiWcation and work-unit identiWcation, our study
provides further evidence of the importance of this
newly emergent line of research. An interesting task
for the future would be to compare the relative predic-
tive power of diVerential intragroup and intergroup
antecedents.
Our study further revealed that organizational identi-
Wcation and work-unit identiWcation were diVerentially
related to our organization-focused and work-unit-
focused outcomes. In particular, our results suggest that,
whereas organizational identiWcation is associated with
organizationally relevant outcomes, work-unit identiW-
cation is associated with work-unit-relevant outcomes.
This proposition of the correspondence between the
focus of identiWcation and its consequences has also
been supported in studies conducted by Christ et al.
(2003) and Van Dick et al. (2004) with respect to extra-
role behavior as an outcome. Our study has one speciWc
strength worth mentioning, however, which is related to
the operationalization of identiWcation. Christ et al.
(2003) and Van Dick et al. (2004) measured identiWca-
tion using a multidimensional scale including items indi-
cating the readiness to behave in a group-supportive
way. One potential problem with this kind of measure is
its potential conceptual overlap with extra-role behavior,
which in turn may contaminate results concerning the
211
relationship between identiWcation and extra-role behav-
ior. Our measure of identiWcation did not include any
items referring to behavioral intentions on behalf of the
group, however, and our results may thus oVer more reli-
able evidence of the correspondence between the focus
of identiWcation and its outcomes than that obtained in
the two studies mentioned above.
Finally, although the main focus of this study was on
the diVerential antecedents and consequences of organi-
zational identiWcation and work-unit identiWcation, our
Wndings concerning the distinctiveness of these two
forms of identiWcation are also of interest. At least to our
knowledge, it is the Wrst time that it has been shown that
employees can diVerentiate between such closely nested
foci as the organization as a whole and the work unit
(cf., Reade, 2001a; Van Knippenberg & van Schie, 2000).
Thus, our study also yields important new evidence of
the distinctiveness of foci of identiWcation in the
workplace.
Implications for research on organizational justice
Our Wndings also oVer some interesting insights into
the study of organizational justice, and speciWcally into
the group engagement model. Namely, given the positive
links between justice perceptions and organizational
identiWcation and work-unit identiWcation, our study
provides further support for the notion that justice per-
ceptions communicate important social-identity-rele-
vant information in terms of increased identiWcation
with the group (Tyler & Blader, 2000, 2003). It should
nevertheless be borne in mind that previous research fol-
lowing the group engagement model has not considered
the possibility that organizational identiWcation is a
multi-foci construct. For example, Tyler and Blader
(2002) only measured identiWcation with the work orga-
nization as a whole. The pattern of relationships
revealed here clearly suggests, however, that employees
simultaneously identify with both their organization as a
whole and their work units, and that justice perceptions
inXuence both of these identiWcations. Therefore, it
would seem valuable if researchers following the group
engagement model would extend their approach to con-
sidering multiple foci of identiWcation simultaneously.
It is also worth noting that we found organization-
focused procedural justice and supervisor-focused
interactional justice to be the strongest predictors of
organizational identiWcation and work-unit identiWca-
tion. Our study thus corroborates the proposition within
the group engagement model, and replicates previous
Wndings that fair procedures and interpersonal treat-
ment are the key sources of social-identity-relevant
information (Tyler & Blader, 2000, 2003). However, in
our data organization-focused distributive justice was
also rather strongly associated with organizational iden-
tiWcation. Our results thus suggest that fair outcomes
212 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
may have important social-identity-relevant implica-
tions at least when they are deWned in terms of their per-
ceived consistency with a person’s work-related
contributions. It is clear that future research should cor-
roborate our pattern of relationships and extend it with
respect to other conceptualizations of distributive justice
as well as other social-identity-relevant outcomes to
establish the importance of fair outcomes. Nonetheless,
our results could be considered a promising starting
point and potential source of inspiration for more care-
ful investigation of the role of distributive justice in the
group engagement model.
In this regard, it is also worth noting that in our data
organization-focused justice perceptions were associated
with organizational identiWcation, whereas supervisor-
focused justice perceptions were associated with work-
unit identiWcation. This Wnding is in accordance with the
multi-foci proposition and previous Wndings that organi-
zation-focused and supervisor-focused justice percep-
tions have diVerential consequences and, as a
consequence, the organization as a whole and the super-
visor should be treated as distinct sources of justice per-
ceptions (Masterson et al., 2000; Rupp & Cropanzano,
2002). This distinctiveness of the organization and the
supervisor is also implicit in the group engagement
model (Blader & Tyler, 2003; Tyler & Blader, 2000,
2003), according to which people distinguish between
justice information originating from the authorities
(i.e., informal source) and the organization as a whole
(i.e., formal source). Nevertheless, researchers following
the group engagement model have not considered the
possibly diVerential eVects of organization-focused and
supervisor-focused justice perceptions. Our Wndings
would seem to suggest, however, that devoting attention
to this possibility may substantially advance our under-
standing of the links between justice and identity-rele-
vant outcomes. SpeciWcally, it would appear that
organization-focused justice perceptions are relevant for
organization-focused identity constructs, while supervi-
sor-focused justice perceptions are relevant for work-
unit-focused identity constructs.
Finally, our results provide evidence that organiza-
tional identiWcation and work-unit identiWcation medi-
ate the relationships between perceptions of
organizational justice and work outcomes. Our study
thus supports the proposition within the group engage-
ment model that the eVects of justice perceptions on
group-related outcomes can be explained in terms of the
social-identity-relevant information they communicate
(Tyler & Blader, 2000, 2003). Moreover, we were also
able to extend previous research Wndings in this Weld.
Namely, at least to our knowledge, it has not previously
been demonstrated that the relationships between
organization-focused justice perceptions and organiza-
tion-focused outcomes are mediated speciWcally by
organizational identiWcation, whereas the relationships
between supervisor-focused justice perceptions and
work-unit-focused outcomes are mediated speciWcally by
work-unit identiWcation. This pattern of relationships is
also interesting from the perspective of research on
multi-foci justice (e.g., Masterson et al., 2000; Rupp &
Cropanzano, 2002). Namely, it suggests that, in addition
to multi-foci social-exchange relationships, the eVects of
multi-foci justice perceptions on multi-foci outcomes
may be mediated by multi-foci identiWcation in the
workplace. An interesting task for the future would be to
compare the relative mediating power of these diVeren-
tial sets of mediating factors.
It should be acknowledged, however, that there were
some problems in our mediation analyses with regard to
extra-role behaviors as an outcome. Namely, there were
no associations between justice perceptions and extra-
role behavior toward the organization and the work
unit. As a consequence, the third of the generally
acknowledged conditions for mediation (i.e., the signiW-
cant association of the independent variable with the
dependent variable) was not met (e.g., Baron & Kenny,
1986; Holmbeck, 1997). However, given that the other
and perhaps more essential conditions of mediation were
met in these data (e.g., Kenny et al., 1998), and that the
results of our Sobel tests were signiWcant, we consider
our Wndings to provide at least preliminary support for
the proposed diVerential mediating role of organiza-
tional identiWcation and work-unit identiWcation. Never-
theless, it is clear that future research should replicate
them before Wrm conclusions can be drawn.
Practical implications
Complementing the above-mentioned theoretical
implications, the present study also oVers insights into
the practical side of organizational development. First,
our Wndings corroborate the picture emerging from pre-
vious studies (e.g., Riketta, 2005) that fostering employee
identiWcation with organizational entities is an eYcient
way of increasing positive work-related attitudes and
behaviors. Our results also carry more speciWc implica-
tions in this regard, however, suggesting that when the
aim is to enhance positive attitudes and behaviors
directed at the organization as a whole, the best strategy
may be to foster identiWcation with the organization as a
whole. On the other hand, if the aim is to enhance posi-
tive attitudes and behaviors directed at the work unit, it
may be more beneWcial to foster identiWcation with the
work unit. Nevertheless, it is worth pointing out that a
high degree of work-unit identiWcation may also turn
out to be detrimental if it is combined with a low degree
of organizational identiWcation. Indeed, high identiWca-
tion with an organizational subgroup has been shown to
be related to negative attitudes toward the members
of the other subgroups (Lipponen et al., 2003). Thus,
despite its beneWts, work-unit identiWcation should
 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
never be fostered at the expense of organizational
identiWcation.
Our results also suggest speciWc means through which
organizational identiWcation and work-unit identiWca-
tion could be fostered. One eYcient way of fostering
organizational identiWcation would be to develop fair
decision-making procedures at the organizational level.
In practice, this would mean providing employees with a
chance to voice their opinion on important organiza-
tional matters, as well as providing them with timely
information, ensuring the consistency and impartiality
of the decision-making process, and basing the decisions
on accurate information. In this regard, it might also be
useful to develop organizational guidelines for the allo-
cation of rewards and resources that maximize the corre-
spondence between the allocated outcomes and each
individual’s work-related contributions. On the other
hand, work-unit identiWcation could be fostered by
adopting training programs that teach supervisors to
treat their subordinates with dignity and respect as well
as to be eYcient communicators.
Limitations and suggestions for future research
As with any research, our study has several limita-
tions that should be acknowledged. First, we employed a
cross-sectional design and so conclusions regarding cau-
sality must be made cautiously. This particularly applies
to our Wndings concerning organizational justice and
identiWcation, since it is likely that the relationships
between these concepts are much more complex than
they appear here. On the one hand, identiWcation may
well be an antecedent of justice perceptions instead of
being their outcome. This would seem plausible given
the proposition within the social-identity approach that
identiWcation with a certain group leads to ingroup
favoritism and positive attitudes towards the group (Taj-
fel & Turner, 1979). Therefore, it would seem likely that
highly identiWed organizational members would also
evaluate their organization and their supervisor as fair,
given that fairness is generally considered a positive
group characteristic. Further, a possibly reversed causal
relationship between justice and identiWcation is also
suggested in recent literature on organizational justice.
For example, the Accessible Identity Model of Justice
Reasoning developed by Skitka (2003) suggests that the
salience of identity (material, personal or social) may
inXuence perceptions of justice. On the other hand, iden-
tiWcation may well be a moderating factor in the eVects
of justice perceptions (e.g., Smith, Tyler, Huo, Ortiz, &
Lind, 1998). It is thus clear that before any Wrm conclu-
sions can be reached, our results need to be replicated in
longitudinal and experimental research.
Another limitation of our study relates to the concep-
tualization and measurement of justice perceptions. In
line with Masterson et al.’s (2000) and Cropanzano
213
et al.’s (2002) approaches, we considered procedural jus-
tice an organization-focused type of justice, and interac-
tional justice a supervisor-focused type of justice.
However, as noted, recent work by Rupp and Cropanz-
ano (2002) and Liao and Rupp (2005) has shown that
perceptions of procedural justice and interactional jus-
tice can in fact focus both on the organization as a whole
and on the supervisor. One could thus argue that our
research has excluded some potentially signiWcant
aspects of procedural justice and interactional justice,
that could also have been important antecedents of
organizational identiWcation and work-unit identiWca-
tion. Furthermore, it is possible that perceptions of dis-
tributive justice may similarly focus not only on the
organization as a whole, but also on the supervisor.
Future research should therefore extend our model by
simultaneously measuring perceptions of procedural jus-
tice, distributive justice, and interactional justice as
focusing on both. Researchers might also wish to con-
sider the role of co-worker-focused justice perceptions as
the antecedents of organizational identiWcation and
work-unit identiWcation. Indeed, co-worker-focused jus-
tice could turn out to be an important antecedent of
work-unit identiWcation in particular, because it is even
more directly work-unit-related antecedent than supervi-
sor-focused justice.
Finally, although the group engagement model sug-
gests that relationships between justice perceptions and
identiWcation are mediated by feelings of pride and
respect, we did not investigate this possibility (Tyler &
Blader, 2000, 2003). Our study therefore does not pro-
vide a complete test for the group engagement model,
and also cannot speak about why justice aVects identiW-
cation. In addition, given this omission of pride and
respect, our study may also have missed some potentially
relevant antecedents of organizational identiWcation and
work-unit identiWcation. We therefore hope that future
research would extend our model by considering
pride and respect as the mediators between justice per-
ceptions and organizational identiWcation and work-unit
identiWcation. It would also be interesting to examine
whether pride and respect diVerentially mediate the
eVects of organization-focused, supervisor-focused, and
co-worker-focused justice perceptions on organizational
identiWcation and work-unit identiWcation. We suspect
this might well be the case, since in previous research
feelings of pride have been treated as focusing on the
organization as a whole, whereas feelings of respect have
been treated as focusing on one’s supervisor and one’s
co-workers, which in practice refer to one’s work unit
(e.g., Tyler & Blader, 2000). We would therefore hypoth-
esize that pride would mediate the eVects of organiza-
tion-focused justice perceptions on organizational
identiWcation, whereas respect would mediate the eVects
of supervisor- and co-worker-focused justice perceptions
on work-unit identiWcation.
214 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
However, given the recent evidence of the multiple
foci of justice perceptions and identiWcation (e.g., Rupp
& Cropanzano, 2002; Van Dick et al., 2004), it would
also be plausible that pride and respect may simulta-
neously have both the organization as a whole and the
work unit as their foci. Therefore, our Wnal proposition
would be that organization-focused justice perceptions
aVect organizational identiWcation through organiza-
tion-focused pride and respect, and that organizational
identiWcation, in turn, mediates their eVects on organiza-
tion-focused outcomes. On the other hand, we would
hypothesize that supervisor- and co-worker-focused jus-
tice perceptions aVect work-unit identiWcation through
work-unit-focused pride and respect, and that work-unit
identiWcation, in turn, mediates their eVects on work-
unit-focused outcomes. Given that support for these
hypotheses would advance our understanding on both
the eVects of multi-foci justice perceptions and the ante-
cedents of multiple identiWcations, we hope that future
research will test their accuracy.
Conclusion
To conclude, this research aimed at extending recent
social-identity research on multiple foci of organiza-
tional identiWcation. Despite the above-mentioned limi-
tations, we believe our Wndings make valuable
contributions in this regard. First, in suggesting that
organizational identiWcation and work-unit identiWca-
tion can only be predicted by justice perceptions relevant
to their foci, our research increases understanding of the
antecedents of multiple foci of identiWcation. Second, by
showing that organizational identiWcation and work-
unit identiWcation are further associated with outcomes
relevant to their foci, it augments the rather limited evi-
dence of the consequences of multiple foci of organiza-
tional identiWcation. Finally, we have produced some
evidence of the diVerential mediating role of organiza-
tional identiWcation and work-unit identiWcation in the
relationship between organization-focused and supervi-
sor-focused justice and organization-focused and work-
unit-focused outcomes. Taken together, our study thus
clearly highlights the importance of adopting the multi-
foci framework in research on both organizational iden-
tiWcation and organizational justice.
References
Abrams, D., Ando, K., & Hinkle, S. (1998). Psychological attachment
to the group: Cross-cultural diVerences in organizational identiWca-
tion and subjective norms as predictors of workers’ turnover inten-
tions. Personality and Social Psychology Bulletin, 24, 1027–1039.
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents
of aVective, continuance and normative commitment to the organi-
zation. Journal of Occupational Psychology, 63, 1–18.
Ashforth, B., & Johnson, S. A. (2001). Which hat to wear? The relative
salience of multiple identities in organizational contexts. In M. A.
Hogg & D. J. Terry (Eds.), Social identity processes in organizational
contexts (pp. 31–48). Philadelphia, USA: Psychology Press.
Ashforth, B. E., & Mael, F. (1989). Social identity theory and the orga-
nization. Academy of Management Review, 14, 20–39.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator distinc-
tion in social psychological research: Conceptual, strategic, and sta-
tistical considerations. Journal of Personality and Social
Psychology, 51(6), 1173–1182.
Bies, R. J., & Moag, J. S. (1986). Interactional justice: Communication
criteria of fairness. In R. J. Lewicki, B. H. Sheppard, & M. H.
Bazerman (Eds.), Research in negotiations in organizations (Vol. 1,
pp. 43–55). Greenwich, CT: JAI Press.
Blader, S. L., & Tyler, T. R. (2003). A four-component model of proce-
dural justice: DeWning the meaning of a “fair” process. Personality
and Social Psychology Bulletin, 29, 747–758.
Bozeman, D. P., & Perrewe, P. L. (2001). The eVect of item content
overlap on organizational commitment questionnaire—turnover
cognition relationships. Journal of Applied Psychology, 86, 161–173.
Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in orga-
nizations: A meta-analysis. Organizational Behavior and Human
Decision Processes, 86, 278–324.
Colquitt, J. A. (2001). On the dimensionality of organizational justice:
A construct validation of a measure. Journal of Applied Psychology,
86, 386–400.
Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng,
K. Y. (2001). Justice at the millenium: A meta-analytic review of 25
years of organizational justice research. Journal of Applied Psychol-
ogy, 86, 425–445.
Christ, O., van Dick, R., Wagner, U., & Stellmacher, J. (2003). When
teachers go the extra-mile: Foci of organizational identiWcation as
determinants of diVerent forms of organizational citizenship behav-
ior among schoolteachers. British Journal of Educational Psychol-
ogy, 73, 329–341.
Cropanzano, R., Prehar, C. A., & Chen, P. Y. (2002). Using social
exchange theory to distinguish procedural from interactional jus-
tice. Group and Organization Management, 27, 324–351.
Deaux, K. (1996). Social identiWcation. In E. T. Higgins & A. W. Kru-
glanski (Eds.), Social psychology: Handbook of basic principles (pp.
777–798). New York: Guilford Press.
Dukerich, J. M., Golden, B. R., & Shortell, S. M. (2002). Beauty is in the
eye of the beholder: The impact of organizational identiWcation,
identity, and image on cooperative behaviours of physicians.
Administrative Science Quarterly, 47, 507–534.
Elovainio, M., Kivimäki, M., & Helkama, K. (2001). Organizational
justice evaluations, job control, and occupational strain. Journal of
Applied Psychology, 86, 418–424.
Greenberg, J., & Colquitt, J. A. (2005). Handbook of organizational
justice. Mahwah, NJ: Lawrence Erlbaum Associates.
Haslam, S. A. (2001). Psychology in organizations: The social identity
approach. London Newbury Park, CA: Sage.
Haslam, S. A., van Knippenberg, D., Platow, M. J., & Ellemers, N.
(Eds.). (2003). Social identity at work: Developing theory for organi-
zational practice. New York: Psychology Press.
Hogg, M. A., & Abrams, D. (1988). Social identiWcations: A social psychol-
ogy of intergroup relations and group processes. London: Routledge.
Hogg, M. A., & Terry, D. J. (2001). Social identity theory and organiza-
tional processes. In M. A. Hogg & D. J. Terry (Eds.), Social identity
processes in organizational contexts (pp. 1–12). Philadelphia, US:
Psychology Press.
Holmbeck, G. N. (1997). Toward terminological, conceptual, and sta-
tistical clarity in the study of mediators and moderators: Examples
from the child-clinical and pediatric psychology literatures. Journal
of Consulting and Clinical Psychology, 65, 599–610.
Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social
psychology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The
 M.-E. Olkkonen, J. Lipponen / Organizational Behavior and Human Decision Processes 100 (2006) 202–215
handbook of social psychology (Vol. 1, pp. 233–268). Oxford: Oxford
University Press.
Leventhal, G. S. (1980). What should be done with equity theory? New
approaches to the study of fairness in social relationships. In K. J.
Gergen, M. S. Greenberg, & R. H. Willis (Eds.), Social exchange:
Advances in theory and research (pp. 27–55). New York: Plenum.
Liao, H., & Rupp, D. E. (2005). The impact of justice climate and jus-
tice orientation on work outcomes: A cross-level multifoci frame-
work. Journal of Applied Psychology, 90, 242–256.
Lipponen, J., Helkama, K., & Juslin, M. (2003). Subgroup identiW-
cation, superordinate identiWcation and intergroup bias
between the subgroups. Group Processes and Intergroup Rela-
tions, 6, 239–250.
Lipponen, J., Olkkonen, M.-E., & Moilanen, M. (2004). Perceived pro-
cedural justice and employee responses to an organizational
merger. European Journal of Work and Organizational Psychology,
13, 391–413.
Lipponen, J., Helkama, K., Olkkonen, M.-E., & Juslin, M. (2005). Pre-
dicting the diVerent proWles of organizational identiWcation: A case
of shipyard subcontractors. Journal of Occupational and Organiza-
tional Psychology, 78, 97–112.
MacKinnon, D. P., Lockwood, C. M., HoVman, J. M., West, S. G., &
Sheets, V. (2002). A comparison of methods to test mediation and
other intervening variable eVects. Psychological Methods, 7, 83–104.
Mael, F. A., & Ashforth, B. E. (1992). Alumni and their alma mater: A
partial test of the reformulated model of organizational identiWca-
tion. Journal of Organizational Behavior, 13, 103–123.
Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. (2000).
Integrating justice and social exchange: The diVering eVects of fair
procedures and treatment on work relationships. Academy of Man-
agement Journal, 43, 738–748.
Moorman, R. H. (1991). Relationship between organizational justice
and organizational citizenship behaviors: Do fairness perceptions
inXuence employee citizenship? Journal of Applied Psychology, 76,
845–855.
O’Reilly, C., & Chatman, J. (1986). Organizational commitment and
psychological attachment: The eVects of compliance, identiWcation,
and internalization on prosocial behavior. Journal of Applied Psy-
chology, 71, 492–499.
Pratt, M. G. (2001). Social identity dynamics in modern organizations:
An organizational psychology/organizational behavior perspective.
In M. A. Hogg & D. J. Terry (Eds.), Social identity processes in orga-
nizational contexts (pp. 13–30). Philadelphia, PA: Psychology Press.
Reade, C. (2001a). Dual identiWcation in multinational corporations:
Local managers and their psychological attachment to the subsid-
iary versus the global organization. International Journal of Human
Resource Management, 12, 405–424.
Reade, C. (2001b). Antecedents of organizational identiWcation in mul-
tinational corporations: Fostering psychological attachment to the
local subsidiary and the global organization. International Journal
of Human Resource Management, 12, 1269–1291.
Riketta, M. (2005). Organizational identiWcation: A meta-analysis.
Journal of Vocational Behavior, 66, 358–384.
Rupp, D. E., & Cropanzano, R. (2002). The mediating eVects of social
exchange relationships in predicting workplace outcomes from
multifoci organizational justice. Organizational Behavior and
Human Decision Processes, 89, 925–936.
Sager, J. K., GriVeth, R. W., & Hom, P. W. (1998). A comparison of
structural models representing turnover cognitions. Journal of
Vocational Behavior, 53, 254–273.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and non-
experimental studies: New procedures and recommendations. Psy-
chological Methods, 7, 422–445.
215
Skitka, L. J. (2003). Of diVerent minds: An accessible identity model of jus-
tice reasoning. Personality and Social Psychology Review, 7, 286–297.
Smidts, A., Pruyn, A. H., & van Riel, C. B. M. (2001). The impact of
employee communication and perceived external prestige on orga-
nizational identiWcation. Academy of Management Journal, 44,
1051–1063.
Smith, C. A., Organ, D. W., & Near, J. P. (1983). Organizational citizen-
ship behavior: Its nature and antecedents. Journal of Applied Psy-
chology, 68, 653–663.
Smith, H. J., Tyler, T. R., Huo, Y. J., Ortiz, D. J., & Lind, E. A. (1998).
The self-relevant implications of the group-value model: Group
memberships, self-worth and procedural justice. Journal of Experi-
mental Social Psychology, 34, 480–493.
Sobel, M. E. (1982). Asymptotic conWdence intervals for indirect eVects
in structural equation models. In S. Leinhardt (Ed.), Sociological
methodology (pp. 290–312). Washington, DC: American Sociologi-
cal Association.
Sousa, F. H., & Vala, J. (2002). Relational justice in organizations: The
group-value model and support for change. Social Justice Research,
15, 99–121.
Tajfel, H. (1978). DiVerentiation between social groups: Studies in the
social psychology of intergroup relations. London: Academic Press.
Tajfel, H., & Turner, J. (1979). An integrative theory of intergroup con-
Xict. In W. G. Austin & S. Worschel (Eds.), The social psychology of
intergroup relations (pp. 33–47). Monterey, CA: Brooks/Cole.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M.
S. (1987). Rediscovering the social group: A self-categorization the-
ory. Oxford, UK: Blackwell.
Tyler, T. R., & Blader, S. L. (2000). Cooperation in groups: Procedural
justice, social identity and behavioral engagement. Philadelphia, US:
Psychology Press.
Tyler, T. R., & Blader, S. L. (2001). Identity and cooperative behavior
in groups. Group Processes and Intergroup Relations, 4, 207–226.
Tyler, T. R., & Blader, S. L. (2002). Autonomous vs. comparative status:
Must we be better than others to feel good about ourselves? Organi-
zational Behavior and Human Decision processes, 89, 813–838.
Tyler, T. R., & Blader, S. L. (2003). The group engagement model: Pro-
cedural justice, social identity, and cooperative behavior. Personal-
ity and Social Psychology Review, 7, 349–361.
Tyler, T. R., Degoey, P., & Smith, H. (1996). Understanding why the
justice of group procedures matters: A test of the psychological
dynamics of the group-value model. Journal of Personality and
Social Psychology, 70, 913–930.
Van Dick, R., & Wagner, U. (2002). Social identiWcation among school
teachers: Dimensions, foci, and correlates. European Journal of
Work and Organizational Psychology, 11, 129–149.
Van Dick, R., Wagner, U., Stellmacher, J., & Christ, O. (2004). The util-
ity of a broader conceptualization of organizational identiWcation:
Which aspects really matter? Journal of Occupational and Organiza-
tional Psychology, 77, 171–191.
Van Dyne, L., Cummings, L. L., & McLean Parks, J. (1995). Extra-role
behaviors: In pursuit of construct and deWnitional clarity (A bridge
over muddied waters). In L. L. Cummings & B. M. Staw (Eds.),
Research in organizational behavior (17, pp. 215–285). Greenwich,
CT: JAI Press.
Van Knippenberg, D., & van Schie, E. C. M. (2000). Foci and correlates
of organizational identiWcation. Journal of Occupational and Orga-
nizational Psychology, 73, 137–147.
Van Knippenberg, D., & Ellemers, N. (2003). Social identity and group
performance: IdentiWcation as the key to group-oriented behavior.
In S. Haslam, D. van Knippenberg, M. J. Platow, & N. Ellemers
(Eds.), Social identity at work: Developing theory for organizational
practice (pp. 29–42). New York: Psychology Press.