Marko Hakonen and Jukka Lipponen. 2007. Antecedents and consequences of
identification with virtual teams: Structural characteristics and justice concerns.
The Journal of e-Working, volume 1, number 2, pages 137-153.
© 2007 by authors
The Journal of E-working
Antecedents and consequences of identification with virtual
teams: Structural characteristics and justice concerns
Helsinki University of Technology
University of Helsinki
This research examined the antecedents and consequences of identification with virtual
teams. Specifically, we hypothesized that two structural characteristics (number of face-
to-face meetings and task interdependence) and perceived quality of interaction
(procedural and distributive justice) would be positively related to identification with the
virtual team. A further hypothesis was that team identification would have a positive
relationship to extra-role behaviors towards the virtual team. The results from our study,
based on a sample of 102 employees of Finnish-based companies, gave partial support for
these hypotheses. We found that task interdependence and procedural justice were
positively related to team identification. Moreover, team identification mediated the
relationship between task interdependence and extra-role behaviors and the relationship
between procedural justice and extra-role behaviors as we predicted. We discuss our
findings in terms of research on virtual teams, social identity, and organizational justice.
Keywords: Virtual teams, social identification, organizational justice, extra-role
Literature about virtual teams (VTs) has been rapidly accumulating during the past
decade. However, there is still amazingly little empirical research about VTs in real
working-life situations, as two recent reviews of the research area have highlighted
The Journal of E-working
Pages 137-153, Vol 1, December 2007
The Journal of E-working
(Hertel, Geister, & Konradt, 2005; Martins, Gilson, & Maynard, 2004). Furthermore,
previous studies on VTs have mainly used qualitative methodology and there is a clear
need for quantitative research conducted in field settings (Hertel et al., 2005). The aim of
this paper is to partially fill this gap by presenting survey results and their quantitative
analysis from real VTs. Moreover, our aim is to incorporate previous literature on social
identification (e.g., Tajfel & Turner, 1979) and organizational justice (e.g., Lind & Tyler,
1988) into the research on VTs.
Most VTs usually consist of sub-groups or individuals working in different locations, and
the members of distributed groups also have many other affiliations. Besides the VT
membership they are members of local work groups, networks, line and matrix
organizations and the whole company (e.g., Hinds & Kiesler, 2002). In practice, this
often means that creating cooperation between distributed sub-groups and individuals is
one major challenge for VTs. Building cooperation between the members of VTs may
also be much more challenging than in traditional teams, purely because of the nature of
VTs: out of sight is easily out of mind (e.g., Gibson & Cohen, 2003; Brown, 1988).
Many authors in the VT literature point out that the shared VT identity is crucial for VT
success because it provides a sense of belonging, despite the relative lack of face-to-face
interaction (Gibson & Cohen, 2003; Hinds & Kiesler, 2002; Lipnack & Stamps, 2000).
However, there are few empirical studies on identification with VTs (Fiol & O’Connor,
2005). The existing research (e.g., Bouas & Arrow, 1996; Wiesenfeld, Raghuram, &
Garud, 1999) is promising. For instance, Wiesenfeld et al. (1999) studied separate virtual
workers rather than virtual teams. Still, their results suggest that identification is
important since it enhanced, for example, cooperation.
In this paper we study the interplay of identification and related variables from the
perspective of social identity approach (Tajfel & Turner, 1979). Among others, Fiol and
O’Connor (2005) have demonstrated the potential fruitfulness of the social identity
approach also in VT research. They state that prior research on identification with VTs is
too focused on VT characteristics and on the communication technology used. We
contribute to this line of research by examining how identification with a VT is related to
such structural factors as the number of face-to-face (FTF) meetings and task
interdependence. In addition, we test how organizational justice variables interplay with
identification. Some organizational justice researchers argue that identification mediates
the relationship between perceived justice and cooperative or extra-role behaviors (Tyler
& Blader, 2000; Blader & Tyler, 2005). Organizational justice has been understudied in
the VT literature (see Hakonen and Lipponen, in press and Kurland and Egan, 1999, for
exceptions) and our purpose is to examine whether similar effects can be found in VTs as
have been found in co-located settings. Finally, we bring the central element of
cooperation, called here extra-role behaviors, as our main dependent variable. In doing
this, we examine some solutions to the above-mentioned challenge of collaboration in
The Journal of E-working
2. Theoretical framework
In the VT literature, one of the elementary debates concerns the definition of a VT. The
recent reviews of VT literature (Hertel et al., 2005, Martins et al., 2004) concur in the
notion that virtuality is a matter of degree. Indeed, there are more and less virtual teams
and we also share the view that virtuality should be seen as a continuum rather than as an
absolute state. In addition, different authors name different aspects of VTs as definitional.
It seems rather clear that, for a VT to be a team, it should consist of more than one person
collaborating towards a common goal (e.g., Hertel at al., 2005; Lipnack & Stamps, 2000).
Very often virtual teams are characterized by the fact that members communicate with
each other mainly via information and communication technology (Axtell et. al., 2004;
Gibson & Gibbs, 2006). This can be put in another way: lack of face-to-face meetings is
typical feature of VTs. Usually it is also proposed (e.g., Bosch-Sijtsema, 2003; Duarte &
Snyder, 1999; Hertel et al., 2005) that for a team to be virtual at least one of its members
must work in a different location form the others. Furthermore, many authors also include
other features, such as crossing temporal, cultural, and organizational boundaries in their
definitions of VTs and virtuality (see Gibson & Gibbs, 2006, for review). We do not,
however, take a stance in the ongoing debate of what attributes of VTs constitute the core
of virtuality. We simply study the effects of one, obviously rather important (Axtell et al.,
2004), feature of VTs: the number of face-to-face meetings (see Kirkman, Rosen, Tesluk,
& Gibson, 2004, for similar use).
Social identity approach to virtual teams
The social identity approach provides a theoretical framework for the relationship
between the individual and the group. Specifically, it consists of two distinct theories: the
original social identity theory (e.g., Tajfel & Turner, 1979), and the more recent self-
categorization theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). Despite certain
differences, both theories share the same fundamental assumption that individuals define
themselves in terms of their social group memberships and that group-defined self-
perception produces distinctive effects on social behavior and inter-group relations (Hogg
& Abrams, 1988; Turner, 1999). This means that the more an individual conceives of him
or herself in terms of membership of a group, or, in other words, identifies with the
group, the more his or her attitudes and behavior are governed by this group membership
(Hogg & Abrams, 1988; Van Knippenberg & van Schie, 2000).
During the past ten years, social identity principles have been increasingly applied to the
study of organizational psychological processes (e.g., Haslam, 2001; Hogg & Terry,
2001). In this context, organizational or team membership is understood to reflect on the
self-concept in the same way as other social memberships do (Ashforth & Mael, 1989;
Hogg & Terry, 2001). Thus, organizational identification is often defined as “the
perception of oneness with or belonging to a group” (Ashforth & Mael, 1989, p. 34).
The Journal of E-working
Moreover, this group-based self-conception is proposed to lead to activities that are
congruent with this identity.
According to the self-categorization theory (Turner et al., 1987) different levels of self-
definition (e.g., self as individual or self as group member) should be related to a distinct
set of needs or motivators. When people categorize themselves at a personal level, they
should be motivated to do things which promote their personal identity as individuals
(e.g., personal advancement and growth). When group-level categorization and social
identity are salient, they should be associated with motivation to do things which promote
the individuals’ social identity as group members, for example, through cooperation and
enhancement of group goals. Accordingly, empirical studies have shown that group
identification is linked to various important outcomes, such as high levels of extra-role
behaviors (e.g., Tyler & Blader, 2000; 2001; see Riketta, 2005, for a review). We define
extra-role behavior as behavior which benefits the team and/or is intended to benefit the
team, which is discretionary, and which goes beyond the existing role expectations (see
van Dune, Cummings, & McLean Parks, 1995). Hence in VTs, the shared or
superordinate VT-level identity could be one tentative answer to the obvious challenge of
sub-group cooperation outlined above. Based on the argumentation presented above, our
first hypothesis is:
H1: Identification with a VT is positively related to the team members’ extra-role
behaviors within the VT.
Structural characteristics – FTF meetings and task interdependence
From the social identity viewpoint, the number of FTF meetings is related to social
category salience (Turner, 1987; Hogg & Terry, 2001). We tend to form social categories
more easily to the groups which we meet often. Therefore, it seems plausible that the
more there are FTF meetings, the higher the category salience. Salient group categories,
in turn, facilitate group members’ identification with the group (Fiol & O’Connor, 2005).
Hence we hypothesize as follows:
H2: The number of face-to-face meeting in a VT is positively related to identification
with the VT.
Task interdependence has been named even as a management practice of VTs (Hertel,
Konradt, & Orlikowski, 2004). The reasoning behind the claim is appealing. The more
the tasks of VT members are coupled with each other, the stronger are the demands for
team members to coordinate, communicate, and cooperate. Thus, by structuring the task
in an interdependent manner, the management can foster collaboration within the team
(see also Bell & Kozlowski, 2002). In a field study of VTs, Hertel et al. (2004) found also
empirical support for their hypothesis, according to which task interdependence was
positively related to VT effectiveness. They found that during the start-up period of one
year the hypothesized relationship was rather strong (r = .49, p < .03) but after that, the
VTs having established their collaboration patterns, the relationship was no more
The Journal of E-working
significant. The authors explain this effect by noting that, at the beginning of a group’s
life, keeping the tasks strongly interdependent helps the VT to build collaboration
patterns, but it can cause process losses later on due to high transaction costs.
A complementary view to the above reasoning can be found from the social category
salience effects that task interdependence may produce (Hogg & Terry, 2001). It is
plausible to claim that the coordination needs created by strong task interdependence help
VT members to perceive their VT as a salient social category with which they can
identify. Taking a closer look at the hypotheses and findings of Hertel et al. (2004) we
find that they expected and found motivational factors to (partially) mediate the
relationship between structural factors (e.g., task interdependence) and outcomes (i.e.,
effectiveness). Similarly, the number of FTF meetings can be seen as a structural factor
of VTs which is not necessarily directly related to behavioral outcomes but the effect is
mediated by a motivational factor. As previously noted, based on social identity approach
(Turner et al., 1987), identification with a VT could be regarded as a powerful
motivational force, which may also serve as a mediator between the structural factors and
extra-role behaviors. Therefore, we propose:
H3: Task interdependence is positively related to identification with the VT.
H4: Identification with a VT mediates both (a) the relationship between the number of
FTF meetings and extra-role behaviors and (b) the relationship between task
interdependence and extra-role behaviors.
After many years of research, it is now well-acknowledged that employees’ perceptions
of organizational justice are critical factors influencing various important work outcomes,
such as organizational commitment, job satisfaction, organizational citizenship behavior
and turnover intentions (see Cohen-Charash & Spector, 2001 and Colquitt, Conlon,
Wesson, Porter, & Ng, 2001, for recent reviews). These attitudes and behaviors have
been found to be related to two aspects of organizational justice: (1) distributive justice -
the perceived fairness of outcome distributions and (2) procedural justice – the perceived
fairness of the decision-making procedures (Cropanzano & Greenberg, 1997). Given
these important consequences of perceived justice, researchers have been trying to
explain why people care about justice.
In order to explain the justice effects on extra-role behaviors and the mediating role of
VT identification in that relationship we will use the group engagement model (GEM).
The model is closely related to the social identity approach since it focuses on the identity
implications of perceived fairness (Blader & Tyler, 2005). The propositions of how
justice perceptions enhance cooperation were originally developed in the group-value
model (Lind & Tyler, 1988) and in the relational model of authority (Tyler & Lind,
The Journal of E-working
According to the GEM, the team members’ identification is shaped by procedural justice
perceptions (Tyler & Blader, 2000; 2001). The model suggests that procedural justice
matters because it communicates information to group members about the quality of their
relationship with authority and with other group members. In particular, fair procedures
and treatment indicate a positive, respectful position within the group and promote pride
in group membership. These feelings of respect and pride, in turn, are suggested to be
related to group identification and other group-related attitudes and behavior. In other
words, the proponents of the GEM claim that procedural justice perceptions in an
organization create positive identification with that organization and that identification
mediates the relationship between procedural justice judgments and cooperative
behaviors towards the organization (e.g., Blader & Tyler, 2005). To our knowledge,
GEM has not been previously tested in virtual settings.
GEM holds that it is primarily procedural fairness, in contrast to distributive fairness, that
is related to identification. There is, however, reason to believe that distributive justice is
also related to identification. Equity theory (e.g., Adams, 1963) assumes that an
individual judges the fairness of his or her own or others’ rewards based on an equity
principle that dictates that persons with greater contributions should receive higher
outcomes (Leventhal, 1980). In most previous studies (e.g., Cropanzano & Greenberg,
1997; Moorman, 1991), distributive justice has been operationalized based on the equity
principle, because it is generally considered appropriate, especially in achievement-
related contexts such as work-organizations. If outcomes are distributed in a way which is
not in accordance with one’s effort, it is very likely that this leads to feelings that one is
not valued and respected by the group (see also De Cremer, 2002). This idea is also in
line with Deutsch’s (1985) argument that outcomes can have either economic or socio-
emotional consequences. The argument is also supported by a study of De Cremer
(2002), which showed that equity perceptions are strongly related to self-esteem and
acceptance, concepts closely linked to respect, pride, and identification. In addition,
recent meta-analyses show that distributive justice has significant correlations with
affective commitment, a concept closely related to identification (Cohen-Charash &
Spector, 2001; Colquitt at al., 2001).
Taken the above theories and empirical findings together we hypothesize as follows:
H5: Perceptions of procedural justice are positively related to identification with a VT.
H6: Perceptions of distributive justice are positively related to identification with a VT.
H7: Identification with a VT mediates (a) the relationship between procedural justice and
extra-role behaviors and (b) the relationship between distributive justice and extra-role
The Journal of E-working
Figure 1. Summary of the hypotheses of this study
Procedure and respondents
The set-up of our study is based on a cross-sectional survey methodology. The data was
gathered with a web-based questionnaire during the year 2005 from six companies
participating in research projects carried out by the first author’s university department.
The questionnaires were sent to the members of the 14 VTs in the participating Finnish-
based companies. These companies represented different lines of business, but the
majority of the data was gathered from the members of VTs in multinational IT
companies (9 teams). All of the VTs consisted of specialists conducting non-routine
tasks. The respondent teams were selected in collaboration with the contact person in
each company and following the agreement of the team leaders. The minimal conditions
for selection were the following features of VTs: the teams had more than one member
collaborating towards a common goal, (at least one of) the team members were located in
different towns, and they communicated mainly via ICT (not FTF).
Altogether 172 respondents received individual e-mails with an introduction to the study
and a web address, through which they could confidentially complete the questionnaire.
Number of FTF
The Journal of E-working
In the e-mail and in the questionnaire the respondents were prompted to answer all of the
questions relating to their named VT. It was stressed that, even though in the items the
term ‘team’ was consistently used, the respondents should think about their VT named in
the e-mail and in the questionnaire cover page while answering. 102 acceptable
questionnaires were returned, a response rate of 59.3%. Respondents were predominantly
male (71.2%), with an average age of 38.5 years (SD = 8.1). Their mean team tenure was
24.6 (SD = 20.5) months and they represented altogether 14 nationalities. Team size
ranged from 6 to 29 members (M = 14.5; it did not correlate significantly with any of the
variables of our study).
Since the companies did not allow us access to the respondent VTs’ full demographical
data we could not test statistically whether there was response bias. Due to the small
sample size (102 respondents from 14 teams) we were, unfortunately, not able to perform
multilevel analyses. Consequently, the data is analyzed at individual level.
Number of FTF meetings. Since the numbers of both formal and informal team meetings
were asked, the responses were summed and averaged to create the measure of number of
FTF meetings. The response scale in each question was: never (1), less than monthly (2),
monthly (3), weekly (4) and daily (5).
Task interdependence. The three-item measure developed by Campion, Medsker, and
Higgs (1993) was used to assess the task interdependence (e.g., “My team cannot
accomplish its tasks without information or materials from all the members in the team”).
The response scale ranged from strongly disagree (1) to strongly agree (5). The measure
reached moderate reliability (alpha = .64).
Distributive justice. Perceptions of distributive justice were measured with three items
based upon Moorman’s (1991) measure. The evaluations focused on the team level by
sharing a common opening: “When resources and rewards are distributed in our team”.
The claims after the lead (e.g., “... I get a fair share considering the stresses and strains of
my job”) tapped the perceived equity of distributions from different angles.
Procedural justice. Employees’ perceptions of procedural justice were measured with
eight items derived from previous scales by Moorman (1991) and Tyler and Blader’s
(2000) procedural justice scale. The items reflect the three aspects of fair procedures and
treatment suggested by the group value model: neutrality, trustworthiness, and status
recognition, and they focus on both the quality of the team’s decision-making procedures
and the quality of treatment at the team level. The response focus was highlighted to be at
the team level by opening all the questions with “When decisions are made in our team”.
The claims after the focus-creating lead tapped the procedural aspects outlined above
(e.g., “… they are based on accurate information”, or “… I am dealt with in a truthful
manner”). In the distributive and procedural justice measures, the items were answered
on a five-point scale ranging from strongly disagree (1) to strongly agree (5).
The Journal of E-working
To test the empirical distinctiveness of the two justice dimensions, a principal-
components analysis (PCA) with oblimin rotation was conducted. PCA yielded two
components (Eigenvalue of component 1 = 5.39; component 2 = 2.0; variance explained
by component 1 = 49.0%; component 2 = 18.3%) which accounted for 67.3% of the total
variance. All of the procedural justice items loaded on the first component and all the
distributive justice items on the second component. There were no cross-loadings above
.35. The Cronbach’s alpha for the distributive justice measure was .94 and for procedural
Team identification. Identification with VT was measured using a modified version of the
organizational identification scale developed by Mael and Ashforth (1992). One original
item regarding public opinions was not used, since teams seldom attract the same
publicity as organizations. In addition, the questions were modified to assess the team
level identification (e.g., “When I talk about this team, I usually say ‘we’ rather than
‘they’”). The response scale was similar to the justice items. The Cronbach’s alpha for
this five-item scale was .81.
Extra-role behaviors. To measure the extra-role behaviors, a scale developed and tested
by Olkkonen and Lipponen (2006) was used. They used items from three existing scales
(Smith, Organ & Near, 1983; O’Reilly & Chatman, 1986; Tyler & Blader, 2000) to
develop their six-item measure (e.g., “I have volunteered to do tasks beyond my job
description in order to help my team to succeed”). One item regarding information and
innovation sharing (“I have shared information and innovations in my team even when I
was not required to do so”) was added to the scale, since the work in all the VTs of the
sample was expert work and information-intensive. The measure covered helping and
innovation dimensions of OCB (Moon, Van Dyne, & Wrobel, 2005). The scope of
respondent’s thinking was limited to the last month by a common question (“How often
have you engaged in the following behaviors during last month?”) before the items. The
final seven items were answered in a five-point scale: never (1), seldom (2), sometimes
(3), often (4), and very often (5). The reliability of the measure was good (alpha = .84).
The descriptive statistics and correlations between the studied variables are presented in
Table 1. As can be seen, the respondents were relatively highly identified with the VT (M
= 3.90). In addition, identification with the VT was strongly (r = .41, p < .001) related to
extra-role behaviors within the VT, as we predicted (H1). Interestingly, none of the other
variables had a significant correlation with extra-role behaviors.
The Journal of E-working
Table 1. Descriptive statistics and Pearson correlations among the variables (N = 102)
Variables M SD 1
1 No. of FTF meetings 2.94 1.01
2 Task interdependence 3.50 0.86 .08
3 Distributive justice 3.51 1.05 – .04
4 Procedural justice 3.80 0.72 .04
5 Identification 3.90 0.75 – .05
6 Extra-role behaviors 3.67 0.77 .04
* p < .05; ** p < .001; two-tailed
To test the main effect hypotheses, H2, H3, H5, and H6, we regressed identification with
VT on both the structural variables and the justice variables. The results are shown in
Table 2. In order to investigate the relative importance of structural and justice variables,
we entered them into analysis in two steps. Our structural variables accounted 7% of the
variance in identification, and adding the justice variables significantly increased the
amount of variance explained (R2 = .22, p < .001; R2change = .15, p < .001). Task
interdependence (H3) and procedural justice (H5) were, in turn, positively related to
identification (β = .22, p < .05 and β = .40, p < 001), just as we expected. Contrary to our
hypothesis H2 and H6, the number of FTF meetings and distributive justice were not
related to identification (β = – .08; n.s. and β = – .06, n.s.).
Table 2. Hierarchical regressions predicting identification with VT (N = 102)
Number of FTF meetings – .08
Task interdependence .26*
* p < .05; ** p < .001; two-tailed
We used hierarchical regression to test the mediation hypotheses H4 and H7 (Table 3). In
the first step we entered the structural variables (the number of FTF meetings and task
interdependence) and the two justice variables (distributive justice and procedural justice)
2 3 4 5
The Journal of E-working
into the equation. In the second step, to test the mediation, identification was entered into
the regression model. As can be seen from the Table 3, neither the structural nor justice
variables had a main effect on extra-role behaviors. Rather, as we predicted,
identification had a strong relationship with extra-role behaviors (β = .46, p < .001).
Since the independent variables had no main effect on extra-role behaviors, their beta
coefficients did not change significantly after the entry of the expected mediator. This
pattern violates the classic third rule of the four-step mediation testing strategy, that is,
the independent variables should be related to the dependent variable (Baron & Kenny,
1986). However, several subsequent authors (e.g., Kenny, Kashy, & Bolger, 1998; Shrout
& Bolger, 2002) have suggested that the relationship between independent variables and
the dependent variable is not a necessary condition for the mediation. The essential
criteria for establishing mediation were met here (Kenny at al., 1998), since task
interdependence and procedural justice (independent variables) were related to
identification (mediator; see Table 2) and identification was related to extra-role
behaviors (dependent variable; see Table 3). The mediation hypotheses regarding the
number of FTF meetings (H4a) and distributive justice (H7b) were naturally rejected,
because these variables were not related to identification.
Table 3. Hierarchical regression predicting extra-role behaviors (N = 102)
Number of FTF meetings .03
Task interdependence .10
Distributive justice .01
Procedural justice .05
* p < .05; ** p < .001; two tailed
To substantiate the mediation findings we computed the Sobel’s statistics for the two
mediation paths. For the task interdependence – identification – extra-role behaviors path
the Sobel’s test value z was 2.01 (p < .05) and for the procedural justice – identification –
extra-role behaviors path z was 3.02 (p < .01). Thus, the mediation hypotheses regarding
these two paths (H4b and H7a) gained further support.
The Journal of E-working
We also conducted some additional analyses in order to create more confidence on our
findings. For example, as our data was gathered from 14 different teams we also
controlled for teams in our analyses. Teams were added as dummy variables into
regression models, and the results relating to the variables of our study were virtually
identical to those presented in Tables 2 and 3.
Our finding, according to which the number of face-to-face meetings of the VT, was not
related to identification is, at first glance, rather surprising. It seems to contradict our
assumption that rare meetings blur VT level social category salience and hence impedes
identification formation with VT. In the light of current literature, we must note that our
data sets us limitations in capturing the essential features of VTs. Even though the
number of FTF meetings is obviously relevant in VTs, it certainly is not the only relevant
attribute of such teams. Hertel et al. (2005) suggest, for example, the relative amount of
face-to-face communication and mediated communication, and the average distance
between team members, as key attributes of VTs. Furthermore, Gibson & Gibbs (2006)
propose geographical dispersion, electronic dependence, structural dynamism, and
national diversity to be associated with VTs or virtuality. Our measure might simply not
have been sufficiently wide to capture the complex nature of VTs. Thus, it remains the
task of further research to use more covering sets of variables that characterize VTs and
study their effects on identification and extra-role behaviors.
Based on our results, it seems that task interdependence is an important factor in VTs. A
practical implication of this is that it can be used as a management practice, forcing the
remote sub-groups and individuals to coordinate their activities and to encourage them to
cooperate effectively as suggested by Hertel et al. (2004). Our outcome variable was a
self report of extra-role behaviors, whereas Hertel et al. (2004) used manager-rated
effectiveness as their outcome measure. Still, the pattern between these two studies is
interestingly similar. In both studies it was assumed, and found, that task interdependence
has an mediated effect on behavioral outcomes and that the mediator is a motivational
construct. Our approach highlights the importance of VT-level identity in shaping the
cooperative behaviors, as suggested by earlier research on organizational identity (see
e.g., Riketta, 2005, for a review). Hence our theorizing based on the social identity
approach provides a complementary explanatory mechanism to the one suggested by
Hertel et al. (2004). Unfortunately, we were not able to control the objective team age
and its effects on task interdependence as Hertel et al. (2004) did.
This study gave a strong indication that procedural justice matters also in VTs.
Procedural justice was positively related to identification with the VT, which, in turn, was
strongly related to cooperative or extra-role behaviors. This result is in-line with
assumptions of the group engagement model (GEM) of organizational justice (Tyler &
Blader, 2000, 2001) and suggests that the perception of fair decision-making and the
The Journal of E-working
perceived quality of interaction are important in VTs. Concretely, these results imply that
leaders and members striving to make their VTs successful, should enhance fair and
transparent decision-making procedures as well as respectful interpersonal treatment.
Contrary to our expectations, distributive justice perceptions were neither related to
identification, nor to extra-role behaviors. One possible explanation for this may be that
we measured distributive justice at the team level and in this particular context the team
members had limited power to allocate rewards and resources, especially when compared
to the authorities that represent the whole organization (e.g., immediate supervisors in the
traditional line organization). Thus, the relative importance of distributive justice
perceptions in carrying identity-relevant information may therefore be much weaker.
As already noted above, we found that identification is an important factor in predicting
behavioral outcomes in VTs, at least extra-role behaviors. In the light of the GEM, and
the social identity approach more generally, our results suggest that similar identity
dynamics appear in co-located and virtual settings. As Martins et al. (2004) state, VTs are
primarily teams, and the researchers should try to find to what extent the results found in
co-located teams apply also to VTs. Simultaneously, we should be cautious in expecting
similar dynamics in all kinds of teams and try to map the new phenomena in VTs.
Nevertheless, our study supports the applicability of the social identity approach also to
There are also some limitations in this study which are worth mentioning. The problems
of using a single measure to characterize VTs are discussed above. Even though we have
used the terms like antecedent, consequence, and effect in the title and some points of our
article for the sake of simplicity, we naturally can not infer causality from the present
cross-sectional study. Longitudinal studies of VTs could overcome this problem. In
addition, the use of solely self-report measures naturally risks the reliability and validity
of the findings (common method variance).
Since the quantitative field research of VTs is rather rare and has been called for (Hertel
et al., 2005), our study helps in filling this gap. A look at the regression statistics shows
that we were able to explain 22 per cent of the variance of identification and 19 per cent
of the variance of extra-role behaviors. The figures are not necessarily impressive, but we
strove for theoretical coherence and novelty, not for maximizing the explained variance.
Our small sample forced us to keep our analyses at individual level and prevented us
from performing group-level analysis (see e.g., Liao & Rupp, 2005). In addition, the
small sample may have hidden some relationships, such as the one between task
interdependence and extra-role behaviors, simply due to the lack of statistical power.
Taken together, the results of the present study suggest that VT researchers might want to
explore identification with and justice within VTs in more detail. The wider application
of the social identity approach is certainly worthwhile for further VT research. This work
has begun (e.g., Mortensen & Hinds, 2001), but the field studies are still rare (Hertel et al,
2005; Martins at al., 2004). Moreover, in the previous literature on VTs the concept of
trust is often mentioned and the relationship between trust and team success has been
The Journal of E-working
rather frequently studied (e.g., Jarvenpaa & Leidner, 1999; Zolin, Hinds, Fruchter, &
Levitt, 2004), but the nearby construct of justice has thus far been largely neglected.
Therefore, one potential direction for future research would be to study the complex
relationship between trust and justice in virtual settings.
Adams, J. S. (1963) Toward an understanding of inequity. Journal of Abnormal and
Personality Psychology, 67, 422-436.
Ashforth, B. E., & Mael, F. A. (1989). Social identity theory and the organization.
Academy of Management Review, 14, 20-39.
Axtell, C. M., Fleck S. J., & Turner, N. (2004). Virtual teams: collaborating across
distance. In Cooper C. and Robertson I. (Eds.), International review of industrial
and organizational psychology (pp. 205-248). John Wiley & Sons, Chichester.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1173-1182.
Bell, B. S., & Kozlowski, S. W. J. (2002). A typology of virtual teams: Implications for
effective leadership. Group and Organization Management, 27, 14-49.
Blader, S. L., & Tyler, T. R. (2005). How can theories of organizational justice explain
the effects of fairness? In J. Greenberg & J. Collquitt (Eds.), Handbook of
organizational justice (pp. 329-354). New Jersey, NJ: Lawrence Erlbaum.
Bosch-Sijtsema, P. (2003). Virtualness: A new organizational dimension. Ridderkerk:
Bouas, K. S., & Arrow, H. (1996). The development of group identity in computer and
face-to-face groups with membership change. Computer Supported Cooperative
Work, 4, 153-178.
Brown, R. (1988). Group Processes: Dynamics within and between groups. Oxford:
Campion, M. A., Medsker, G. J., & Higgs, A. C. (1993). Relations between work group
characteristics and effectiveness: Implications for designing effective work groups.
Personnel Psychology, 46, 823-850.
Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A
meta-analysis. Organizational Behavior and Human Decision Processes, 86,
Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng, K. Y. (2001).
Justice at the Millennium: A meta-analytic review of 25 years of organizational
justice research. Journal of Applied Psychology, 86, 424-445.
Cropanzano, R., & Greenberg, J. (1997). Progress in organizational justice: Tunneling
through the maze. In: Cooper, C. L., & Robertson, I. T. (Eds.), International Review
of Industrial and Organizational Psychology (pp. 317-372). New York, NY: Wiley.