How do people judge fairness in supervisor and peer relationships? Another assessment of the dimensions of justice

Abstract

The ultimate goal of organizational justice research is to help create fairer workplaces. This goal may have been slowed by an inattention to the criteria that workers themselves use to ascertain what they believe is fair. Referred to as 'justice rules', these were originally determined by theoretical considerations and organized in four dimensions (distributive, procedural, interpersonal and informational justice). There have been few attempts to investigate how far these classical norms represent fairness experiences and concerns in modern workplaces, especially in the context of working with peers. In
a person-centric study, we investigate which rules people use when judging the fairness
of interactions with supervisors and peers. This allows us to identify 14 new justice
rules that are not taken into account by traditional measures. When subjected to factor
analysis in follow-up studies, the enlarged set of rules suggests a more parsimonious
structure for organizational justice, with only three dimensions apiece for supervisor
and peer justice. We term these factors relationship, task, and distributive justice.
Furthermore, we find that the resulting model of justice rules is a good predictor of
attitudes in relation to supervisors and peers and can provide additional insights into
how to understand and manage justice.

Full text

human relations
https://doi.org/10.1177/0018726719875497
human relations
2020, Vol. 73(12) 1632 –1663
© The Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0018726719875497
journals.sagepub.com/home/hum
How do people judge fairness in
supervisor and peer relationships?
Another assessment of the
dimensions of justice
Marion Fortin
TSM-Research, University of Toulouse Capitole, CNRS, France
Russell Cropanzano
Leeds School of Business at the University of Colorado, USA
Natàlia Cugueró-Escofet
Department of Economics and Business Science, Universitat Oberta de Catalunya, Spain and
IESE Business School, Universidad de Navarra, Spain
Thierry Nadisic
emlyon business school, France
Hunter Van Wagoner
Leeds School of Business at the University of Colorado, USA
Abstract
The ultimate goal of organizational justice research is to help create fairer workplaces.
This goal may have been slowed by an inattention to the criteria that workers themselves
use to ascertain what they believe is fair. Referred to as ‘justice rules’, these were
originally determined by theoretical considerations and organized in four dimensions
(distributive, procedural, interpersonal and informational justice). There have been few
attempts to investigate how far these classical norms represent fairness experiences
Corresponding author:
Marion Fortin, TSM-Research (UMR CNRS 5303), University of Toulouse Capitole, 2 Rue du doyen Gabriel
Marty, Toulouse Cedex 9, Toulouse, 31042, France.
Email: marion.fortin@ut-capitole.fr
875497HUM0010.1177/0018726719875497Human RelationsFortin et al.
research-article2019

Fortin et al. 1633
and concerns in modern workplaces, especially in the context of working with peers. In
a person-centric study, we investigate which rules people use when judging the fairness
of interactions with supervisors and peers. This allows us to identify 14 new justice
rules that are not taken into account by traditional measures. When subjected to factor
analysis in follow-up studies, the enlarged set of rules suggests a more parsimonious
structure for organizational justice, with only three dimensions apiece for supervisor
and peer justice. We term these factors relationship, task, and distributive justice.
Furthermore, we find that the resulting model of justice rules is a good predictor of
attitudes in relation to supervisors and peers and can provide additional insights into
how to understand and manage justice.
Keywords
Justice dimensions, justice rules, organizational justice, peer justice, person-centric
Over the past few decades, empirical research has consistently shown that perceived fair-
ness provides numerous benefits to organizations and to their employees (e.g. Colquitt
et al., 2013; Rupp et al., 2014). Generally speaking, individuals make justice evaluations
by comparing an actual event, such as supervisory behavior, to a set of subjective criteria
(Hollensbe et al., 2008). When an event violates these criteria, it is likely to be judged as
unfair. These criteria are called ‘justice rules’, a term coined by Leventhal (1980: 30),
because they provide standards for judging whether behavior is just or unjust.
Given the centrality of employees’ subjective judgments, it is surprising that over the
past two decades, justice research has concentrated on a ‘traditional’ set of rules estab-
lished by early research and often through the introspection or theoretical work of
researchers, without empirically checking whether these are still largely shared and rea-
sonably complete. We argue that these rules do not always correspond to the experience
and criteria used by working people today (Rupp et al., 2017). The present article
addresses this concern by presenting a person-centric approach to identifying the ‘justice
rules’ that people use (Hollensbe et al., 2008: 1099), as then they can be built into actual
work systems, thereby promoting justice (Cugueró-Escofet and Fortin, 2014; Cugueró-
Escofet and Rosanas, 2013). We have three primary objectives. First, we use person-
centric methods to identify previously unmeasured justice rules. Second, we examine
how the dimensional structure of workplace justice changes once these new rules are
included. Third, we investigate a predictive model of justice, which takes these new rules
into consideration.
Organizing justice rules into justice dimensions
A ‘justice rule’ is a normative standard that acts as a criterion in order to determine if an
event is fair. Once a set of justice rules is identified, they can then be organized into
dimensions. Several justice researchers have approached this by factor-analyzing the
rules in order to group them. These factor solutions vary somewhat (cf. Blader and Tyler,

1634 Human Relations 73(12)
2003a, 2003b; Roch and Shanock, 2006). Perhaps the most common is the four-factor
model. This model views justice rules as referring to the fairness of outcomes (distribu-
tive justice), decision-making processes (procedural justice), dignity and respect (inter-
personal justice), and communication and information (informational justice) (Colquitt,
2001; Colquitt and Rodell, 2015). This allows researchers to organize many different
rules into a manageable number of dimensions. Justice scholars have embraced these
‘dimensions’ with so much enthusiasm that Rupp et al. (2017) recently cautioned
researchers to remember that the dimensions are not justice but only conceptualizations
and operationalizations of justice in organizations. Indeed, as new justice rules have been
discovered, the set of accepted dimensions has grown in phases. These phases have been
termed “waves” (Colquitt et al., 2005).
The first wave concerned the fairness of outcomes – distributive justice. When judging
the fairness of outcomes, I can select individuals according to at least three main rules
(Deutsch, 1985). According to Colquitt and Rodell (2015), these are (a) equity (alloca-
tions based on relative contributions or effort), (b) equality (the same allocations for all),
and (c) need (allocations based on necessity). The second wave of research concentrated
on criteria of fair decision-making procedures. Based on the seminal work of Leventhal
(1980), in order to be considered fair, a process must be: (a) consistently applied, (b) free
from personal bias, (c) correctable if in error, (d) accurate, (e) representative of the con-
cerns of all parties affected, and (f) consistent with ethical standards. Finally, a third wave
focused on rules people may use to judge the interpersonal treatment and communication
they receive at work (e.g. Bobocel and Holmvall, 2001). Colquitt (2001) lists four criteria
for interpersonal justice: (a) politeness, (b) dignity, (c) respect, and (d) refraining from
improper remarks. He adds five rules for informational justice: (a) candidness, (b) thor-
ough explanations about procedures, (c) reasonable justifications for decisions, (d) timely
communication, and (e) adaptation of communication to each employee’s needs. This is
an influential set of justice rules, organized into four dimensions, that has become almost
standardized (for exceptions, see Bies and Moag, 1986; Hollensbe et al., 2008).
Consequently, the present consensus may constrain our understanding of actual work-
place justice experiences. There may be important rules that we have missed, and when
these rules are considered, the dimensions of justice may need to be modified.
Apart from simply missing some rules, there is another reason to believe that our
present list may be incomplete. At the time these traditional rules were ‘discovered’,
justice researchers typically limited their work to hierarchical relationships (employee–
supervisor or employee–organization). Justice evaluations are, however, ubiquitous in
social life and not limited to interactions with supervisors, and this has become even
more true as the nature of organizations has changed (Ambrose and Schminke, 2001;
Rupp et al., 2017). Recently, organizational justice researchers have introduced the peer
or colleague as an alternative source of justice. Several studies by Li, Cropanzano and
colleagues have furthermore confirmed that team members do differentiate between fair-
ness from their supervisor and fairness amongst team members (Cropanzano et al., 2011;
Li et al., 2013, 2015). In modern organizations, where many employees work in teams,
some of which are self-managed teams, peer fairness may sometimes be more important
for daily work experience than supervisor fairness. While it is therefore important to
understand peer fairness, there is no guarantee that our present set of rules captures it. It

Fortin et al. 1635
is possible and indeed likely that individuals do not use exactly the same rules for every
type of target (Lavelle et al., 2009a).
Regardless of the specific cause, there are good grounds to believe that the present set
of rules typically found in the organizational justice literature is not comprehensive. For
example, a review of the literature by Törnblom and Kazemi (2015) suggests that there
are quite a few additional distributive rules that have been neglected. Moreover, Sheppard
and Lewicki (1987) interviewed 44 executives in order to identify how fairness judg-
ments are made at work regarding managers’ decisions. They identified nine ‘new’
norms, such as ‘giving fair recognition’, ‘not expecting from employees things that are
inconsistent with their role definition’, or the golden rule consisting in ‘doing things that
can benefit others and not damage them’. More recently, Hollensbe et al. (2008) asked
33 job entrants to describe how they decided that their employing organizations and
workplace supervisors were fair or unfair. Under half of the respondents used the tradi-
tional justice rules, and fewer than 30% relied on them exclusively. If we also take into
account employees’ relations with their peers, then it is even more likely that the estab-
lished canon of justice rules needs to be adapted and extended.
The research of Hollensbe et al. (2008) and Shepard and Lewicki (1987) provides
important support for further investigations of justice rules. Nevertheless, for all its mer-
its, this prior work is limited with respect to our present objectives. First, previous work
on justice rules does not consider justice amongst peers. Furthermore, while new rules
have been identified, these have not been integrated with the existing set of legacy rules
(cf. Colquitt, 2001). Thus, we do not know how the dimensional structure of justice
might change. Lacking revised dimensions, of course, we are unable to ascertain their
theoretical impactions. Consequently, this research provides a good first step, but the
implications of these new rules are in need of further development.
The present studies
In view of these issues, we follow the lead of Guo et al. (2011) and Weiss and Rupp
(2011), who recommend a person-centric approach to the study of organizational justice.
This approach emphasizes working individuals’ phenomenological experiences on the
job. It focuses on their personal and subjective understanding of their work life.
Undertaking our research from this perspective should allow us to meet our first objec-
tive, identifying additional justice rules, if these exist (Study 1). Following from this, we
will turn to our second objective. If the set of rules is changed, then the subsequent
dimensional structure may also change – this issue will be explored in subsequent stud-
ies. As for our third objective, we consider how these new rules and revised dimensional
structure impact various workplace criteria (Studies 3–4).
Study 1: Qualitative investigation of emergent rules
This first study examined whether there are emergent justice rules that have been over-
looked by previous research. Respondents were voluntary participants recruited via
university-sponsored executive education programs. Our sample consists of 62 full-
time employees (40 men and 22 women) who were working in Europe. On average,

1636 Human Relations 73(12)
they had 4 years’ work tenure and their mean age was 39. There was a wide range of
different professions from different industries. Participants were asked to think of one
unjust incident that they had personally experienced at work.1
The interview instructions were organized into a 2x 2 matrix, with the two dimensions
being hierarchical relationship (supervisor vs peer) and relationship closeness (close vs
distant). We asked respondents: ‘Recall an incident where you were treated unfairly by a
distant (close) superior (peer/colleague) at work.’ We chose closeness in addition to hier-
archy to get a larger variety of workplace relationships and therefore of incidents. We
then asked follow-up questions, including why they thought the incident was unfair,
what it was that made it unfair, how it could have been made fair and whether there was
any specific rule that was not respected.
Upon completion of the interview, participants filled in a brief questionnaire with
demographic questions and overall fairness assessments. These measures confirmed that
the overall fairness assessments of participants did not significantly differ among the
four groups.2
Data analysis
The transcripts of the interviews were content analyzed by three of the authors. First, we
coded the different justice rules already established in the literature (Colquitt, 2001).
These rules were coded when the participants’ statements directly included reference to
them or when they described their meaning. For example, when a participant explained
that it was unfair for a supervisor to shout at him because this was a lack of respect (direct
mention of the rule), we coded this statement as representing ‘respect’. When a partici-
pant explained that it was unfair for her not to receive a raise although she had worked
harder (description of the rule), we coded this as representing ‘equity’. As anticipated,
participants described rules that were not covered by the set found in the justice litera-
ture. In these cases, we added new coding categories.
After having coded each interview at least once, we agreed on the definitions of
our emergent rules and then found that a few of our new categories could be com-
bined as they seemed strongly linked in the participants’ narratives. For example,
‘being professional’ was combined with ‘doing one’s job properly’ and with ‘han-
dling one’s mission properly’. Similarly, ‘respecting one’s word’ was combined with
‘respecting promises’. In addition to the classical justice rules, this process resulted
in an initial set of 14 emergent justice rules that were combined into a final 11 emer-
gent justice rules (see the list of the 11 combined rules in Supplementary Table 1 in
the online resources3).
At this point, two of the authors independently recoded all 62 interviews with the
final coding scheme. The inter-coder agreement was 95% for traditional distributive
justice rules, 87% for traditional procedural justice rules, 84% for traditional interper-
sonal justice rules, 87% for traditional informational justice rules, and 97% for the
emergent justice rules. In cases where the coding differed, agreement was reached
through discussion. The results reported here are based on the final agreed coding.
Supplementary Table 1 provides an overview of the number of occurrences of tradi-
tional and emergent justice rules in relation to supervisors and peers.

Fortin et al. 1637
Traditional justice rules
Distributive justice rules. The classical rule ‘equity’ was mentioned by seven out of the 62
participants. This rule was mentioned exclusively in the case of a relationship between a
manager and an employee. This is consistent with the role of managers as reward
allocators.
Procedural justice rules. The classical procedural norms identified in earlier justice
research were also found in the interviews (Leventhal, 1980). ‘Voice’, ‘representative-
ness’, ‘accuracy’, ‘consistency’, ‘no bias’, ‘possibility to appeal’, and ‘ethicality’ were
important procedural justice norms in this sample. Noticeably, although the justice litera-
ture has emphasized ‘voice’ (Hunton et al., 1998) and ‘accuracy’, in our sample ‘ethical-
ity’ was the most frequently used norm for evaluating justice at work (mentioned by 15
participants, see Supplementary Table 1). This is consistent with recent contributions,
suggesting that ethicality is at the heart of procedural justice judgments (Crawshaw et al.,
2013; Schminke et al., 1997). Furthermore, two of the procedural justice norms were
only mentioned for supervisors. These were ‘accuracy’ (four participants) and ‘consist-
ency’ (six participants). Similarly, the ‘no bias’ rule was mentioned five times for super-
visors and only once for peers. These norms tend to relate to formal decision-making
processes, which are more typical of hierarchical relationships.
Interpersonal justice rules. The established normative rules of ‘respect’, ‘politeness’, and
‘dignity’ were used to describe incidents in relation to both supervisors and peers (14
participants). We noticed that mentions of ‘improper remarks’ were made about incidents
only in peer relationships (three participants), where being careful about personal bound-
aries may be more important.
Informational justice rules. Pertaining to this group of rules, ‘openness and frankness’ (17
occurrences), ‘thorough and reasonable explanations’ (18 occurrences), ‘timely commu-
nication’ (one occurrence), and ‘tailoring communication to specific needs’ (two occur-
rences) emerged. These are also consistent with previous research (Colquitt, 2001).
Emergent justice rules
In addition to replicating the traditional rules, we identified a set of 14 emergent justice
rules that we reduced to 11 after having combined those that were close to each other and
had only a few occurrences. Illustrations of each of these 11 new rules can be found in
Table 1.
‘Recognition’. One of the new rules could be described as ‘giving recognition and praise
when due’. In relationships with managers this rule was mentioned twice. It is similar to
the equity norm but concerns immaterial benefits. Recognition can be seen as a symbolic
reward. This norm is also important in relationships among colleagues (for nine partici-
pants). For example, ‘being grateful’ was one manifestation of peer recognition.

1638 Human Relations 73(12)
Table 1. Sample of quotations for the emergent justice norms of Study 1.
New justice norms Sample quotes Interview code and
comments
New distributive justice rules
Recognition Giving recognition
and praise
I was in a position where I thought I’m getting the new job anyway, so I’m happy. Just give me the mark so I
have this feeling of confirmation. It’s more for myself, so that I can close this chapter and be very happy [. . .] I
really wanted that good mark as a kind of recognition basically [. . .] I think we should have gotten the reward
because we worked quite hard when we were only two instead of three.
Interview 10
This example
regards a
supervisor
New interpersonal justice rules
Non-harassment Not harassing
people
For some reason my manager at the time downgraded us. I don’t know why. It was two of us out of six got
downgraded and you immediately feel a bit resentful and, you know, blame. You think somebody, you know,
what they’re up to and maybe even want to get rid of you [. . .] They did some nasty things. [. . .] I think it was
purely the company using it to isolate you [. . .] You’re sitting with a team of people [. . .] They’re not really
interested in your problem [. . .] So you do lose self-respect. I think you probably lose confidence as well. And
I think stress, pressure comes on.
Interview 6
This example
regards a
supervisor
Non-exploitation Not exploiting
others
She was not taking her responsibilities, according to me. And so she was like, I think, make some profit of our,
of my work consciousness [. . .] The problem is that I felt I put a lot of energy. I was very tired sometimes and
. . . in order to compensate let’s say this lack of . . . or this injustice for me.
Interview 19
This example
regards a
supervisor
Separation of
personal/professional
Not mixing personal
and professional
matters
Lovely man. He was, you know, on the outside [laughs] very friendly, very supportive [. . .] It was only on the
day that I was interviewed here that I found out that he had applied for the same post. And it was done very
secretively [. . .] He didn’t actually do anything other than what, upon reflection, seemed to me extracting a lot
of information from me [. . .] Whereas with hindsight now it turned out that he was taking information from me
to support his application for the same post [. . .] You know, once bitten, twice shy. I think it took three or four
bites from him for me to say okay then, this is not cool. I need to change the way . . . [I] need to have some sort
of boundaries and limitations and not be as naïve and forthcoming with information as I have been in the past.
Interview 41
This example
regards a peer
Empathy Giving empathy and
support
I was actually in line for the next creative director that came up. I was associate director and had been told
that that position was mine. And when the spot came up it was actually given to somebody else, which in itself
seemed unfair on a personal basis. Actually, what I thought was awful about it was the fact that he came in. It
was a cube environment, came into my cube, and told me instead of asking me to come into his office which
had a door [. . .] I mean in terms of just human relationships [. . .] there are rules of engagement which may
be unspoken which I think have to be followed. [. . .] On a human level, that’s the higher level, I think [he] just
didn’t show any humanity [. . .] I think that [one has] just to understand other people’s feelings, to be decent.
Interview 8
This example
regards a
supervisor
Equality Treating people
equally
She had some favorites who she, you know, she let them get away with coming late, whereas everybody else
would get a scorning for it. So I don’t think it could. I don’t think I was there long enough to get that close to
her in order to start getting the same treatment as my colleagues. [. . .] She would brush people off and she’ll
give time to some people and not to other people [. . .] She didn’t really spread her time amongst everybody.
She only focused her time on certain people.
Interview 26
This example
regards a
supervisor
(Continued)

Fortin et al. 1639
New justice norms Sample quotes Interview code and
comments
Loyalty Being loyal And when our manager asked us why we were behind three weeks and a half, four weeks, she basically just
said okay . . . she just pointed the finger at me. [. . .] That’s it. She said no, no this is R’s fault [. . .] So I was
really disappointed. Extremely disappointed. I felt I was let down. I lost confidence in me and her. In a way I’m
kind of . . . I’m a really impulsive person. So I kind of always stay quiet, but honestly right in that moment, I
mean I thought I was going to punch her in the face [. . .] She did apologize, but I really did take it to the heart.
It was kind of bad. It was kind of bad. I felt really bad.
Interview 40
This example
regards a peer
Special needs Taking special needs
into account
The change is just enormous, completely came as a shock to me and I was very kind of disorientated. I had to
. . . I actually had very, very little support from my line manager and I felt that if I had been given the . . . if I
had been given more support, been allocated more time with my line manager because he is genuinely a very,
very busy person with all the roles and tasks he has to do. But he actually knew that I was from a different
background. [. . .] I just felt it was really unfair that I was being . . . I wasn’t given the correct amount of
support.
Interview 31
This example
regards a
supervisor
New integrity rules
Responsibility Taking full
responsibility for
one’s actions
Working on a piece of design, I wanted a peer review of the initial design. So I asked one of my colleagues
within the department who had a wealth of experience and had been with the organization for quite a while,
so was well respected within the department as being very knowledgeable. However, his reaction was quite
off-handish, very off-handish, which caused some personal offence [. . .] there’s an implicit agreement that our
work is reviewed by kinds of IPAs or someone knowledgeable in the system and I felt this was a legitimate part
of that review process.
Interview 60
This example
regards a peer
Professionalism, job
properly done and
mission properly
handled
Remaining
professional and
handling one’s
job and mission
properly
[That was] unfair because I did part of his job. He didn’t appreciate it and he didn’t care about my negative
feelings about, for example, future communications with that client and so on. Besides, I don’t know. It looked
like he didn’t even notice that I did his job [. . .] Well the ideal situation if I present results of my work like all
the proposal itself and he says ‘thank you’, checks it, and picks up [his] responsibility because there is not much
left to do. It’s just a matter of a couple of hours of work. In this case it didn’t happen and if we get the client
and his was zero efforts, then he gets the project. So that I feel is unfair.
Interview 44
This example
regards a peer
Respecting word and
promise keeping
Respecting one’s
word and promises
I was very emotional to be honest because I’d left a very solid situation behind to move across on the premise
that I would get, you know, I was on the fast track for promotion or I thought I was. And this guy had just
stopped all my ambitions in that training track, you know, without just cause in my opinion and I got quite
emotional and started shouting at him and I had to leave and go home for a couple of days to calm down
before I came back [. . .] I’d left a position that I was very experienced in. I was well within my comfort zone.
I’d moved outside my comfort zone to go across to this other job because, you know, in my interview for the
job it was being promised that because of my qualifications, experience, etc., that I would be on a fast track
for promotion and that’s what I wanted to do at that time. So it took all that away from me and I had to start
looking again. I couldn’t go back to my old job. So I had to start looking for something new and it just . . . It
just stopped what I was doing in this track, you know.
Interview 21
This example
regards a
supervisor
Table 1. (Continued)

1640 Human Relations 73(12)
‘Responsibility’, ‘professionalism’, and ‘respecting word and promises’. Three new rules related
to meeting agreed obligations and behaving professionally. The first one, ‘taking full
responsibility for one’s actions’ (11 participants), included an instance where an inter-
viewee was asked to act unethically by his managers, who did not adhere to their respon-
sibilities regarding the security and health of their consumers. The second rule concerned
‘remaining professional and handling one’s job and mission properly’ and was mentioned
by 11 participants. This shows that participants did not only care about negative conse-
quences that any particular injustice could bring for themselves, but also held perpetrators
accountable for general professional standards of conduct that were seen as a duty. The
third rule, ‘respecting one’s word and promise keeping’, was mentioned by eight partici-
pants and is aligned with the previously identified rule of ‘following through with prom-
ises’ that is included in Blader and Tyler’s measure of the fairness of quality of treatment
(Blader and Tyler, 2003b). We viewed this set of rules as adhering to and enforcing psy-
chological contracts (Robinson, 1996; Rousseau, 1995). Previous research has suggested
that maintaining psychological contracts promotes fairness (e.g. Morrison and Robinson,
1997). Consequently, these respondents were telling us about a rule that has been previ-
ously identified, but is under-discussed, in studies of organizational justice.
‘Empathy’, ‘being loyal’, and ‘equality and consistency of treatment’. These new rules
appeared to concern guidelines of good interpersonal treatment and maintaining good
relationships. ‘Giving empathy and support’ was mentioned by 16 participants. It is
close to the norm of ‘emotional support’ that was proposed by Greenberg (2006) and
emerged in the work of Hollensbe et al. (2008). Empathy also appeared as an antecedent
of fairness in the work of Patient and Skarlicki (2010). The rule ‘being loyal’ was also
evoked by five participants and concerned, for example, not blaming a colleague in
front of the supervisor or defending the team against others in the organization. The
loyalty rule has already been identified by Bies (2001), but to our knowledge has not
been used in subsequent research. Finally, four participants also mentioned ‘equality
and consistency of treatment’ as a rule for fairly managing relationships. For example,
a participant stated that a manager should treat everyone in the same way and ‘not treat
friends better than others’. With regards to relationships between peers, one participant
said that an employee should treat all same-level colleagues in the same way.
‘No-harassment’, ‘non-exploitation’, and ‘not mixing personal and professional matters’. These
three rules concerned the avoidance of particularly negative treatment or behavior.
Although ‘no harassment’ (three participants) is not included in traditional measures of
fairness, it has been identified in research on interpersonal justice (Bies, 2001). ‘Not
exploiting others’ (three participants) is not used in current scales of justice, although it
was originally identified by Lind (2001). ‘Not mixing personal and professional matters’
(seven participants) refers, for example, to not letting personal grudges or likes and dis-
likes interfere with work decisions. This set of rules was somewhat distinct, as it focused
on what not to do rather than what should be done.
‘Special needs’. One final emergent rule was about ‘taking employees’ special needs into
account’. It has a wider scope than the traditional informational justice norm regarding

Fortin et al. 1641
‘tailoring communication to employees’ special needs’. Indeed, it not only consists of
adapting communication but also adapting behavior and decisions. For example, it was
perceived to be unfair when a manager did not take into account that an employee was
new to the role.
Study 2: Exploring the factors of peer and supervisor justice
Our second study had four goals. First, we wanted to see if the use by employees of these
new rules could be replicated using a different methodology. Second, we wanted to
assess whether these new rules loaded on the traditional four-factor model or whether
they changed the factor structure of justice. Third, we wanted to examine whether the
new rules changed our interpretation of these factors. Fourth, we sought to test the gen-
eralizability of the new factor structure across two sets of targets: supervisors and col-
leagues/peers. With these goals in mind, we conducted an exploratory factor analysis
using Colquitt’s (2001) items as well as the new items found in Study 1. We conducted
separate analyses for justice from supervisors and justice from peers to develop a distinct
factor model for each of these two sources.
Method
Two hundred and twenty-eight participants in executive education programs at three
business schools in Europe participated in this study. One hundred and seventy-five par-
ticipants responded to all justice measures. Of those who elected to provide their gender,
64 were male and 95 were female. Our participants were on average 32 years old and had
5 years’ work experience.
Measuring justice rules. We used all the 14 items which emerged from our qualitative study
(and not only the 11 combined) in Study 2 because we wanted to assess whether an inde-
pendent sample of people had a similar interpretation. We also wanted to assess how these
emergent rules were organized. As a measure of traditional justice rules, we included
items from Colquitt’s (2001) measure of justice. This measure includes 20 items that have
been found to usually cluster into four factors. We omitted one of the distributive justice
items because it was not applicable to both managers and colleagues as a source of justice.
(The deleted item was: ‘Is your outcome justified, given your performance?’)
All of the items, both traditional and emergent, were qualitatively pretested with several
employees. We asked pretest respondents, who were equally executive education participants
who worked full time, to verbally comment on the items as they went through the question-
naire. This resulted in slight adjustment to the three distributive justice items used when refer-
ring to colleagues. For all other items, we simply replaced the word ‘manager’ or ‘organization’
of the original scale with ‘colleagues’ in order to refer to peers as the source of justice.
Study 2: Results
Confirmatory checks. To ensure that the measures were behaving in accordance with ear-
lier validation evidence, we conducted a confirmatory factor analysis (CFA) of Colquitt’s

1642 Human Relations 73(12)
(2001) measure. The fit indices, which are in Supplementary Table 2, show that the four-
factor model provides a good representation of these 19 items. For supervisors, the TFI
= .942, CFI = .950, and RMSEA = .063. For peers, the TFI = .901, CFI = .915, and
RMSEA = .080. While the fit when peers are evaluated is a bit lower than when supervi-
sors are evaluated, both sets of items are generally consistent with the original predic-
tions. Further evidence can be found in Supplementary Table 3, which includes estimates
for all 19 justice rules. As in Colquitt’s (2001) original validation study, covariances
between the four factors ranged between .54 and .95. Overall, these analyses replicate
Colquitt’s earlier findings for these 19 traditional rules. What was left was to examine
the impact of the emergent rules for supervisors and for peers.
Results for supervisory justice rules
We then added the 14 emergent rules for supervisors. Because we lacked an a priori
theoretical structure (Bryant and Yarnold, 1995) and were interested in letting our
respondents define the relevant dimensions, we conducted an exploratory factor analy-
sis. The factor matrix for this complete set of rules (traditional + emergent) is included
in Supplementary Table 4. Factor loadings below .40 were suppressed. Given the afore-
mentioned correlations among the dimensions, we subjected the data to an oblique rota-
tion (Bandalos and Boehm-Kaufman, 2009). Similar to some earlier work, items loaded
onto three factors. Only these factors had eigenvalues greater than 1.00. Collectively,
they accounted for an impressive 86.76% of the variance. Covariances ranged between
.56 and .69. Table 2 provides an overview of the items that were retained for the three
factors.
First factor: Relationship justice. The first factor we extracted includes Colquitt’s (2001)
interpersonal justice dimension. This factor had an eigenvalue of 14.99 and accounted
for 72.29% of the variance. All four of the traditional interpersonal justice items used in
Colquitt’s (2001) scale load onto this factor (‘respect’, ‘politeness’, ‘dignity’, and ‘non-
impropriety’). Additionally, this factor also includes several emergent rules identified in
Study 1. These involve the avoidance of negative behaviors that people should not
engage in at work: ‘non-exploitation’, ‘non-harassment’, and ‘not mixing personal/pro-
fessional matters’. Four other emergent rules, which were related to proper actions, also
loaded on this factor: ‘respecting word’, ‘doing job properly’, and ‘professionalism’. In
addition, this first factor includes four emergent justice rules that are related to support-
ing people. These are ‘empathy’, ‘equal and consistent treatment’, ‘recognition’, and
‘loyalty’. Among this set of rules, however, ‘recognition’ loads more strongly on Factor
3, while ‘does job properly’ loaded more strongly on Factor 2, as is discussed below. It
also bears mention that even if most traditional procedural justice items are represented
on the second factor, two procedural justice rules, ‘able to express your views and feel-
ings’ (voice) and ‘upheld ethical and moral standards’ loaded on this first factor. The
voice item loaded somewhat even more strongly on distributive justice, and so it was not
retained for this first factor.
If we look at it closely, this first factor contains items that regard the fair management
of the relationship. Communication should be fair: the manager should respect the four

Fortin et al. 1643
Table 2. Confirmatory factor analysis structure.
Item
Relationship justice
1 My colleagues treat me with equality in comparison with other colleagues.
2 They are empathetic and supportive.
3 My colleagues don’t harass me.
4 My colleagues respect their word.
5 My colleagues don’t inappropriately mix personal and professional matters.
6 My colleagues are loyal to me.
7 My colleagues don’t exploit me.
8 My colleagues remain professional at all times.
9 My colleagues treat me in a polite manner.
10 My colleagues treat me with dignity.
11 My colleagues treat me with respect.
12 My colleagues do not make improper remarks or comments.
13 I am able to express my views to my colleagues.
14 My colleagues’ decision-making procedures comply with ethical and moral standards.
Task justice
1 In general, my colleagues apply procedures consistently.
2 Decisions that they make are free of bias.
3 My colleagues base decisions on accurate information.
4 I can discuss and appeal the decisions that my colleagues make.
5 My colleagues take my special needs into account.
6 My colleagues keep their promises.
7 My colleagues take full responsibility for their actions.
8 My colleagues handle their mission properly.
9 My colleagues do their job properly.
10 My colleagues are open and frank in their communications with me.
11 My colleagues explain the decisions they make thoroughly.
12 My colleagues give me reasonable explanations regarding the decisions they make.
13 My colleagues communicate details in a timely manner.
14 My colleagues tailor their communications to my specific needs.
Distributive justice
1 I feel I have influence over decisions taken by my colleagues.
2 Does your (outcome) reflect the effort you have put into your work?
3 Is your (outcome) appropriate for the work you have completed?
4 Does your (outcome) reflect what you have contributed to the organization?
5 They give me recognition and praise where due.
traditional interpersonal justice rules (respect, politeness, dignity, and non-impropriety).
However, there is more to the story than that. First, the manager should be empathetic.
Second, the emergent rules included in this factor place an emphasis on meeting obliga-
tions by behaving professionally, remaining loyal, and treating everyone consistently. It
also seems important to avoid illegitimate demands. Managers should not mix personal
and professional matters and should neither harass nor exploit their subordinates.

1644 Human Relations 73(12)
Second factor: Task justice. The second factor was smaller than the first, having an eigen-
value of 1.61 and accounting for 7.77% of the variance. The second factor includes all
of the traditional informational justice items used in Colquitt’s (2001) justice scale
(‘frankness’, ‘timely communication’, ‘thorough explanation’, ‘reasonable explana-
tion’, and ‘tailored communication’). It also includes three procedural justice items
from Colquitt’s (2001) justice scale. These are linked to the way in which work deci-
sions are made: ‘no bias’, ‘consistency’, and ‘accuracy.’ This is consistent with the work
of Blader and Tyler (2003a, 2003b), who found that fairness in decision-making includes
both implementation of decision procedures and appropriate informal sharing of infor-
mation pertaining to them. However, as with our first factor, inclusion of the emergent
justice rules changes the interpretation somewhat.
Various emergent rules also loaded on this second factor. ‘Handling mission prop-
erly’, ‘doing job properly’, ‘taking responsibility’, and ‘professionalism’ are four rules
that represent the ability of the managers to apply the vision and the strategy of the
company in a responsible and professional way. ‘Respects word’, ‘loyalty,’ and ‘keeps
promises’ pertain to the capacity of the managers to respect the psychological contract
they have negotiated with their employees regarding the way the work should be done
and what they can expect from the company in return (Robinson, 1996). Finally, ‘tak-
ing into account special needs’ allows to renegotiate this psychological contract when
some specific situation occurs (Rousseau, 1995). Among these emergent rules, notice
that three of them loaded more strongly on the first factor (relationship justice) and
were also discussed above (‘respects word’, ‘loyalty’, and ‘professionalism’). For our
participants, respecting one’s word, being loyal and acting professionally are seen as
pertaining both to fair treatment and fair decision making.
Overall, the scope of this dimension includes traditional informational justice items
but goes beyond that point. This factor includes predictability, consistency between
words and actions over time, and professional integrity. Therefore, it also captures the
way in which managers build a cognitive and affective frame for the future by showing
the vision of the company and helping employees to understand how they can best
coordinate their work behaviors within this frame.
Third factor: Distributive justice. The third factor had an eigenvalue of 1.34 and accounted
for 6.46% of the variance. This final factor consists of the three items of distributive
justice that we included from Colquitt’s (2001) scale, which all refer to the equity rule,
plus a single emergent rule that was identified in Study 1 – ‘recognition and praise’. This
factor is about the fairness of the allocation of outcomes. Recognition and praise are also
considered by our participants as outcomes, albeit immaterial ones, which over time
(when coming from one’s superiors) tend to be linked to material outcomes. Thus, fol-
lowing the terminology established in the literature, we call this third factor ‘distributive
justice,’ as it pertains to the outcomes that one receives. Notice that these findings are
generally consistent with earlier notions of distributive justice, though with additional
emphasis placed on non-material rewards.
Nevertheless, inclusion of the emergent rules did slightly qualify interpretation of this
factor. In this regard, it bears mention that the first two procedural justice items crossed
over and loaded on this third factor. These were ‘able to express views’ (voice) and

Fortin et al. 1645
‘influence over outcomes’ (outcome control). The participants seem to have considered
that the possibility to express their views at work is a right that they deserve, thus closer
to an outcome than to a process (see Supplementary Table 4; see also Cropanzano and
Ambrose, 2001), or they may have considered that the most important attribute of voice
is to influence outcomes.
Results for peer justice rules
We next conducted an exploratory factor analysis of the traditional and emergent rules for
ratings of peer justice. Once again, the data were subjected to an oblique rotation (Bandalos
and Boehm-Kaufman, 2009). The factor matrix is presented in Supplementary Table 5. As
in our analysis for the supervisor items, we suppressed factor loadings for emergent justice
rules that were smaller than .40. The total variance accounted for was 85.3% for supervi-
sors and 84.2% for peers. Our analysis of the scree plot extracted three factors, all of which
have eigenvalues greater than 1.0. These were recovered in a different order, but corre-
spond to task justice, relationship justice, and distributive justice. For peers, the strongest
(first) factor was task justice, followed by relationship, and then distributive. For managers,
as we have previously seen, the strongest (first) factor was relationship justice, which was
followed by task and distributive, respectively. Covariances ranged between .56 and .57.
First factor: Task justice. This factor had an eigenvalue of 15.62 and accounted for 73.36%
of the variance. As with supervisor justice, the factor also includes three procedural jus-
tice items, specifically those that are linked with the way decisions are made: ‘no bias’,
‘consistency’, and ‘accuracy’. A fourth, ‘upholds moral standards’, also loaded strongly.
Likewise, all of the traditional informational rules items loaded on this first factor,
though two loaded a bit more strongly on Factor 3. We also found four emergent integ-
rity norms relating to the respect of psychological contracts: ‘respects word’, ‘keeps
promises’, ‘takes responsibility’, and ‘does not mix personal and professional’. Finally,
the items relating to professionalism also load onto this factor: ‘handles missions prop-
erly’, ‘does job properly’, and ‘remains professional’. While there was some switching
of rules between manager and peer ratings, this factor is similar regarding its meaning,
reflecting elements of the fairness of task behavior and also of psychological contracting
(Robinson, 1996; Rousseau, 1994), further linking our findings with previous research
in related areas of study (Morrison and Robinson, 1997).
Second factor: Relationship justice. The second factor had an eigenvalue of 1.77 and
accounted for 8.31% of the variance. This factor for peers is similar to the relationship
factor that we found for supervisors, which included the traditional norms of ‘interper-
sonal justice’ but went beyond them. This factor again includes all of Colquitt’s (2001)
interpersonal justice items, though the fourth (‘refrained from improper comments’)
loaded weakly. Paralleling the first factor for supervisors, it also includes two emergent
rules identified in Study 1. These concern negative behaviors that people should not
engage in at work – ‘non-exploitation’ and ‘non-harassment’. In addition, and as is the
case for supervisors, this factor includes the newly identified rules that are related to
care. These are ‘empathy’ (weakly), ‘equal treatment’, and ‘loyalty’. We find that our

1646 Human Relations 73(12)
respondents use similar criteria to judge the fairness of the relationship with their peers
and the fairness of the relationship with their managers.
Third factor: Distributive justice. The third factor had an eigenvalue of 1.08 and accounted
for 5.1% of the variance. All three of Colquitt’s (2001) distributive justice items loaded
on this third factor, though the first (‘effort put into work’) also loaded with interpersonal
justice. ‘Recognition’ also loaded on this factor as it is the case for managers. Interest-
ingly, two other emergent rules loaded primarily on this third factor: ‘taking into account
special needs’ and ‘empathy’. This may reflect that individuals think they deserve the
right to be treated interpersonally fairly by their peers and see this right as an outcome
that is due to them at work, and taking into account needs becomes an important rule for
distributions amongst peers. Besides, in the same way as for supervisors, this factor also
included Colquitt’s (2001) procedural justice items of ‘voice’ and ‘outcome control’.
Finally, ‘appeals’ loaded on this factor, suggesting that employees also view this criterion
as a right that is also deserved in their relationship with peers or a way to have a better
chance to get due outcomes.
Study 2: Conclusion
The results of Study 2 suggest that employees draw on three meaningful groups of justice
rules when judging the fairness of their supervisors and peers – task justice, relationship
justice, and distributive justice. The inclusion of 14 new justice rules somewhat changed
how justice rules organize into dimensions, as well as changing the meaning of the indi-
vidual dimensions. The task justice factor includes informational as well as procedural jus-
tice. At the same time, it is much broader than these traditional dimensions, as it includes
rules pertaining to psychological contracting. In a similar way, relationship justice goes
beyond interpersonal justice, by extending the spectrum of justice to include both the avoid-
ance of negative behavior, for example ‘not harassing people’, and to include taking care of
others, for example ‘showing empathy’. Overall, adding 14 new justice rules had the effect
of making the factor solution more parsimonious: we recovered three justice dimensions,
rather than the usual four, while categorizing substantially more rules (33 vs 20).
Study 3: Replicating Study 2 and predicting reactions
Our third study had three goals. First, we sought to replicate our findings from Study 2.
Second, we wished to extend these results by going beyond an investigation of the factor
structure and investigate their predictive validity. Third, we wanted to consider the impli-
cations of our emergent rules for research on overall justice. Fourth, we wanted to exam-
ine whether the rule sets for supervisors versus peers differentially relate to attitudes and
behaviors towards these two sources.
Literature and hypothesis development
In Study 3, we replicate our earlier findings for peer and supervisor justice. This distinc-
tion has implications for predicting work behaviors. According to the target similarity

Fortin et al. 1647
model (Lavelle et al., 2007, 2009a, 2009b, 2015), justice judgments are made in relation
to a specific party and will in turn predominantly influence attitudes and behaviors
towards that same party (Rupp et al., 2014). Thus, the justice rules related to supervisors
should be most consistently related to supervisory-targeted reactions, whereas the justice
rules related to peers should be most consistently related to peer-targeted reactions (see
Holtz and Harold, 2009, for a supportive empirical test).
Ambrose and her colleagues (Ambrose and Schminke, 2009; Ambrose et al., 2015)
further maintain that these target similarity effects may be both indirect and direct. The
indirect effects are partially mediated by overall justice (Holtz and Harold, 2009; Jones
and Martens, 2009), which conveys the impact of the justice rules to various criterion
variables. Ambrose and Schminke (2009) further predicted and found that justice rules
showed a direct (i.e. an unmediated) relationship to work behavior. Integrating the target
similarity model with overall justice yields the following two predictions:
Hypothesis 1: Task justice, relationship justice, and distributive justice judgments
related to the supervisor are positively related to trust towards the supervisor and
perceived supervisor support. These relationships are partially mediated by supervi-
sor overall justice.
Hypothesis 2: Task justice, relationship justice and distributive justice related to peers
are positively related to trust in peers and perceived workgroup support. These rela-
tionships are partially mediated by peer overall justice.
These predictions are summarized in Figure 1. In addition to testing the hypotheses,
we also used confirmatory factor analysis in an attempt to replicate the structure obtained
Figure 1. Hypothesized relationships for peers and supervisors.
Note: Direct paths from emergent justice factors to criterion variables are in gray. Paths that constitute the
mediated effect through overall justice are in black.

1648 Human Relations 73(12)
in Study 2 (for details, see Bryant and Yarnold, 1995). Once the structure fit was found
we combined the measurement model with the structural model containing the hypoth-
esized relationships.
Study 3: Method
Currently employed alumni of a university in the United Kingdom voluntarily partici-
pated in this study. Of the 2000 individuals contacted, 458 participated in the survey, and
376 completed all justice measures. Of those who elected to provide their gender, 141
were male and 233 were female. Participants had an average age of 34 years and an aver-
age tenure of 6 and a half years.
Measures
We included the same justice rules as in Study 2 and added measures of trust and citizen-
ship behaviors for both supervisors and peers. Overall justice was measured using five
items from Ambrose and Schminke (2009). To assess trust in supervisors and peers we
included six items from Robinson (1996) and five from Spreitzer and Mishra’s (1999)
integrity trust scales. Perceived workgroup/supervisor support was measured with eight
items adapted from Eisenberger et al. (2002). Some of these items were also used by
Lavelle et al. (2009b). All of these scales were presented twice, referenced once for
supervisors and once for peers.
Study 3: Results
Confirmatory analyses. As with Study 2, we began by assessing whether we could repli-
cate the traditional four-factor model of justice without including the emergent new jus-
tice rules. Results of confirmatory factor analyses are shown in Supplementary Tables
8–9. Findings were supportive. For supervisors, the TFI = .926, CFI = .937, and
RMSEA = .074. For peers, the TFI = .916, CFI = .928, and RMSEA = .075. Likewise,
Supplementary Tables 6 and 7 show that all of the traditional items loaded significantly
on the predictor factors and that the four-factor model provides a good representation of
these 19 items. Covariances between the four factors ranged between .56 and .90. We
then added the emergent rules and conducted a confirmatory factor analysis based on the
groupings of rules shown in Table 2.
Test of the measurement model for supervisors and peers. Following Ambrose and
Schminke (2009), we compared this model to a three-factor model which only included
our three justice dimensions, a two-factor model which collapsed all specific justice
items into a single factor and had a second overall justice factor, and finally to a single-
factor model with all justice items. This process was conducted twice – once for supervi-
sors and once for peers. Finally, we fit an eight-factor model with our three justice
dimensions for both peers and supervisors as well as overall justice for peers and
supervisors.

Fortin et al. 1649
We next assessed the fit of a six-factor model reflecting our three justice dimensions,
overall justice, and our two criterion measures. We compared this to a five-factor model
where trust and perceived support were collapsed into a single factor. This process was
conducted twice – once for supervisors and once for peers. Results are shown in Table 3.
Both factor structures have a moderate fit, though they are generally supportive. For
supervisors, the TFI = .87, CFI = .88, and RMSEA = .076. For peers, the TFI = .86,
CFI = .87, and RMSEA = .078. In addition to overall justice, we obtained the three
predictor factors – task justice, relationship justice, and distributive justice. Supplementary
Table 8 shows that each of the rules had a significant loading on the predictor factor.
Covariances between the three factors ranged between .34 and .60 for supervisors and
between .24 and .39 for peers.
Test of the mediated model for supervisors and peers. A correlation matrix is presented in
Table 4. The reliabilities are all above .70. They are a bit higher for our new scales than
for the original four-factor model. This makes sense, given that we added additional
items and revised the factor structure. We tested our full theoretical model, presented in
Figure 1. In so doing, we examined the predictive power of our six types of justice (three
for supervisors, three for peers).
As expected, the partial mediation model provided a much better fit than the fully
mediated model. Fit indices for the fully mediated model were TLI = .746, CFI = .684,
and RMSEA = .24. In contrast, fit indices for the partially mediated model were
TLI = .91, CFI = .95, and RMSEA = .125. The delta chi-squared between our
Table 3. Measurement model fit indices.
Item χ2d.f. RMSEA TLI CFI
Justice only supervisors
One-factor 2968.6 665 .094 .8 .8
Two-factor 2765.3 664 .09 .8 .8
Three-factor – emergent only 1809.1 492 .83 .9 .9
Four-factor – emergent and overall 2135 659 .076 .9 .9
Justice and attitudes supervisors
Five-factor 5330.1 1880 .069 .8 .8
Seven-factor 4698 1869 .062 .9 .9
Justice only peers
One-factor 2604.4 629 .091 .8 .8
Two-factor 2322.2 628 .084 .8 .9
Three-factor – emergent only 1660.1 461 .083 .9 .9
Four-factor – emergent and overall 2044.5 623 .078 .9 .9
Justice and attitudes peers
Five-factor 4617.8 1759 .066 .8 .8
Seven-factor 4267.9 1748 .062 .9 .9
Justice only peers and supervisors
Eight-factor 6850.02 2746 .063 .83 .83

1650 Human Relations 73(12)
Table 4. Correlation matrix.
Measure 1 2 3 4 5 6 7 8 9 10 11
1. Procedural justice .86, .84 .74 .72 .77 .73 .87 .82 .88 .78 .79 .39
2. Distributive justice .62 .92, .80 .59 .72 .64 .75 .95 .77 .73 .77 .33
3. Interpersonal justice .76 .48 .87, .85 .65 .8 .9 .62 .71 .69 .72 .32
4. Informational justice .76 .58 .69 .88, .88 .68 .76 .74 .93 .75 .72 .33
5. Overall justice .83 .59 .82 .74 .92, .89 .84 .64 .75 .75 .77 .32
6. Relationship justice .88 .59 .91 .76 .89 .94, .93 .77 .86 .84 .84 .37
7. Distributive justice .74 .95 .58 .67 .67 .68 .88, .84 .78 .74 .81 .38
8. Task justice .91 .65 .76 .92 .84 .88 .74 .94, .94 .83 .78 .33
9. Trust .82 .6 .73 .74 .83 .84 .67 .85 .94, .94 .81 .32
10. Support .83 .62 .73 .74 .83 .83 .72 .84 .83 .89, .87 .43
11. OCB .49 .31 .41 .4 .44 .45 .41 .44 .45 .55 .86, .85
Note: Supervisor correlations are below the diagonal and peer correlations are above the diagonal. Alpha reliabilities are along the diagonal. Supervisor reliabilities
are shown first and peer reliabilities are shown second.

Fortin et al. 1651
theoretical model (partial mediation) and the full mediation model was significant; χ2
difference (d.f. = 12) = 732.06, p < .01. Given that the test of partial mediation pro-
vided significantly better fit, we concluded that our findings were consistent with our
prediction. Additional evidence for the target similarity model comes from inspection of
the path coefficients displayed in Supplementary Figure 1, which displays the standard-
ized path coefficients for our hypothesis tests.
Results for supervisors. Two of the justice dimensions, relationship justice (.79, 95% CI =
.66, .91) and task justice (.26, 95% CI = .13, .39), were related to overall justice in super-
visors. Contrary to Hypothesis 1, distributive justice was unrelated to overall justice in
supervisors (.06, 95% CI = –.01, .14). Supervisor overall justice, in turn, was related to
trust toward supervisors (.29, 95% CI = .17, .41) and perceived supervisor support (.26,
95% CI = .15, .36).
Bootstrapped confidence intervals of the direct, indirect and total effects of the super-
visor justice dimensions with 5000 repetitions are reported in Supplementary Table 9.
Both the relationship and task justice dimensions demonstrated complementary partial
mediation on trust in supervisors and perceived supervisor support. In contrast, the dis-
tributive justice dimension only demonstrated a single significant direct effect on per-
ceived supervisor support.
As expected, the cross-paths from supervisor justice to peer overall justice were non-
significant for relationship and task justice. There was a weak relationship between
supervisor distributive justice and peer overall justice (–.07, p < .05). However, given
that it was in the wrong direction and of very low magnitude, it is probably not interpret-
able. Taken together, the weak effects for the cross-paths are consistent with the target
similarity model.
Results for peers. Bootstrapped confidence intervals of the direct, indirect and total
effects of the peer justice dimensions with 5000 repetitions are reported in Supplemen-
tary Table 10. One of the justice dimensions, relationship justice (.86, 95% CI = .70,
1.01), was related to overall justice from supervisors. Only relationship justice demon-
strated significant indirect effects on both trust in peers (.09, 95% CI = .003, .19) and
perceived workgroup support (.16, 95% CI = .08, .23). Distributive justice and task
justice did not show the predicted indirect effects. The findings for task justice and dis-
tributive justice, therefore, are inconsistent with Ambrose and Schminke’s (2009) earlier
findings. Peer overall justice, in turn, was related to trust in peers (.11, 95% CI = .004,
.22) and perceived workgroup support (.19, 95% CI = .1, .27). Each of the three types of
justice was directly related to trust in peers, which is not inconsistent with the target simi-
larity model.
Study 4: Further investigations of the emergent rules
In Study 4, we decided to consider two additional issues to strengthen the results of
Study 3. First, Colquitt and Rodell (2015) have argued that measures of justice should
take into account injustice. To address this, Colquitt et al. (2015) have proposed a full
range assessment, which includes the same items as Colquitt’s (2001) previous measure,

1652 Human Relations 73(12)
but this time negatively worded (as injustice). Second, we wish to again replicate and
expand our causal model.
Method
Participants were given a link to an online survey, which asked questions about all of our
measures, once for peers and once for supervisors. The order of survey questions was
randomized. Half of the participants responded first to supervisor referenced items and
then to peer referenced items, and vice versa for the second half of participants. The
sample consisted of 277 adults recruited through Mturk, a marketplace for online work-
ers. Participants were based in the USA and all had work experience. The age of partici-
pants ranged from 21 to 74 (M = 36, SD = 11.2), and 59% of participants were male.
Among the participants, 87.23% reported that they currently hold a full-time position.
Forty-two participants were removed from the dataset for missing data or failing at least
one of several attention checks. We included all the same measures as were included in
Study 2.
Validity check: Justice and injustice
As discussed above, our initial qualitative study asked respondents to speak about an inci-
dent in which they had experienced injustice. This could have biased our emergent rules, if
they pertain more strongly to injustice than they do to justice. We examined this possibility
in two steps. First, we attempted to replicate the four-factor full range model proposed by
Colquitt et al. (2015). Second, we examined the causal model proposed in Figure 1.
Full range model of justice/injustice. We initially considered the full range model of justice,
developed by Colquitt et al. (2015). The full range model contains both justice and injus-
tice items for each family of traditional rules (distributive, procedural, interpersonal, and
informational). Each of these new dimensions ranges from an unjust low to a just high.
In an attempt to replicate this model, we submitted all Colquitt et al.’s (2015) items to a
confirmatory factor analysis. Our emergent rules were, of course, excluded from this
initial attempt at replication. In line with the target similarity model (Lavelle et al., 2007),
we conducted two separate CFAs, one for supervisory justice and one for peer justice.
For supervisors, the model showed poor fit: TLI = .51, CFI = .54, and RMSEA = .14.
This was also the case for peers, TLI = .51, CFI = .54, and RMSEA = .13. Given these
findings, we could not utilize the full range model as it was originally proposed. Thus,
we needed to find an alternative.
Because model fit was lower than anticipated, we next conducted two exploratory
factor analyses, one for supervisors and another for peers. Again, we used only the
Colquitt et al. (2015) items, both for justice and injustice. Contrary to these earlier results,
we found that for both supervisors and peers two very strong factors emerged, one for
justice and one for injustice. These two factors accounted for 84% of the variance for
supervisors and 83% of the variance for peers. For supervisors, a third weak factor which
included four positive interpersonal justice items also had an eigenvalue greater than 1.0.
Interestingly, it appears that inclusion of the injustice items impacted the factor structure.

Fortin et al. 1653
Given these results, we followed our exploratory factor analysis for the negatively
worded (injustice) items, forming a separate scale. However, we were reluctant to do the
same for the positive worded (justice) items. As the four-factor structure has been widely
replicated for the positively worded items (e.g. Colquitt and Rodell, 2015), we retained
their original structure.
Comparison of correlations. Given that the full range factor structure was not replicated, we
averaged the injustice items into a single scale. This allowed us to compare them with our
emergent rules, on the one hand, and to the four traditional justice dimensions, on the
other. For simplicity, we also averaged our emergent rules into a single dimension. These
results appear in Supplementary Table 11. Findings for peers are above the diagonal, and
those for supervisors are below. As can be seen, our emergent rules were moderately cor-
related with the injustice scale at –.40 for supervisors and –.35 for peers. This is slightly
higher than the correlations between injustice and the four traditional scales. However, the
association of our emergent rules to injustice is consistently lower than are the associa-
tions among our emergent rules and justice. Thus, there is no evidence of a negativity bias.
Test of theoretical model
Given the aforementioned findings, we next turned to our theoretical model, which was
based on earlier work by Ambrose and Schminke (2009). In order to examine this frame-
work, we followed the two-step approach recommended by Anderson and Gerbing
(1988, 1992). Specifically, we first conducted CFA to test our measurement model. The
constructs investigated in Study 4, while correlated, should be empirically distinguisha-
ble. Given a supportive measurement model, we next conducted the causal analysis. Our
goal was to again replicate the factor structure obtained from Studies 2–3.
Examination of measurement model. In Study 4 we explored the following six constructs:
the three new justice dimensions (relationship, distributive, and task), overall justice,
support, and trust. As was the case in Studies 2–3, measures of the justice dimensions
included both the traditional and the emergent rules. All six of these variables were
assessed for both supervisors and peers, bringing the total number to 12. Consequently,
we expected that the best fitting model would have 12 factors. We compared this com-
prehensive model to two nested, alternative models. For the first alternative, we took a
closer look at the target similarity model, and pooled the indictors of each construct
across source. That is, we combined ratings of supervisors and peers. For the second
alternative, we explored the discriminant validity by examining whether our relationship
justice dimension was distinct from trust and support. Chi-squared difference tests show
that the best fitting model is the one which treats each variable in our causal model as a
distinct latent variable. Results of these analyses can be seen in Supplementary Table 12.
Test of our causal model
As in Study 3, we examined the predictive power of our three types of justice for peers
and supervisors, while their effects were partially mediated by overall justice. To

1654 Human Relations 73(12)
replicate this possibility, we compared a partially mediated model to a fully mediated
model using trust and support. We also omitted the injustice items from this analysis,
given our emergent items behaved similarly to the traditional justice rules. Fit indices for
the fully mediated model were TLI = .67, CFI = .73, and RMSEA = .21. In contrast, fit
indices for the partially mediated model were stronger: TLI = .86, CFI = .92, and
RMSEA = .14. The delta chi-squared between the partial mediation model and the full
mediation model was significant: χ2 difference (d.f. =18) = 526.2, p < .01. Given the
test of partial mediation provided significantly better fit, we concluded that our findings
were consistent with our predictions. However, the fit indices were lower than we would
have preferred – suggesting that, as in Study 3, some of the predicted paths were not sup-
portive of our expectations. We examine this below.
Results for supervisors. Bootstrapped confidence intervals of the direct, indirect, and total
effects of the supervisor justice dimensions with 5000 repetitions are reported in Sup-
plementary Table 13. Both the relationship and task justice dimensions demonstrated
complementary partial mediation on trust in supervisors and perceived supervisor sup-
port. In contrast, the distributive justice dimension did not demonstrate any significant
direct effects. Two of the justice dimensions, relationship justice (.72, 95% CI = .55, .88)
and task justice (.29, 95% CI = .09, .49), were related to overall justice in supervisors.
Contrary to Hypothesis 1, however, distributive justice was unrelated to overall justice in
supervisors (–.07, 95% CI = –.21, .08). Supervisor overall justice, in turn, was directly
related to trust toward supervisors (.39, 95% CI = .28, .50) and perceived supervisor
support (.47, 95% CI = .28, .50).
As in Study 3, a test of the target similarity framework was conducted by examining
a model with cross paths from supervisor justice to peer overall justice. While neither
model showed a good fit, the model with cross paths showed poorer fit than a model with
no cross paths (TLI = .70, CFI = .94, RMSEA = .21 versus TLI = .86, CFI = .92,
RMSEA = .14). However, a closer look at the model did find two significant cross paths.
One of these was supervisor relationship justice, which increased overall justice in peers
(.28, p < .01). The second was supervisor distributive justice, which decreased overall
justice in peers (–.31, p < .01).
This was surprising in two ways, first that these paths were significant and second that
distributive justice was negative. If replicated, the path from supervisor relationship jus-
tice to peer overall justice could be an example of a trickle-down model (Masterson,
2001). In a trickle-down model, the behavior of supervisors impacts the behavior of
subordinates. These effects have been found for interactional justice (Ambrose et al.,
2013), which is at least somewhat related to our relationship justice dimension. That
said, we are reluctant to interpret an unpredicted effect without replicates.
The second unexpected relationship was between supervisory distributive justice and
peer overall justice. Though unexpected, a similar effect was found in Study 3. This
raises the possibility that distributive justice judgments depend on the behavior of both
managers and peers. Intuitively this makes sense. For example, pay is often set by man-
agers, but its fairness is determined by comparison with peers. The odd finding in our
Study 4, however, is the negative coefficient. Perhaps a generous supervisor evokes envy
from peers or perhaps this is an example of over-reward inequity, whereby people feel

Fortin et al. 1655
inequity when they are paid more than their peers. Again, this effect is in need of future
research.
Results for peers. Bootstrapped confidence intervals of the direct, indirect, and total
effects of the peer justice dimensions with 5000 repetitions are reported in Supplemen-
tary Table 14. One of the justice dimensions, relationship justice (.88, 95% CI = .69,
1.08), was related to overall justice from peers. Similar to Study 3, only relationship
justice demonstrated significant indirect effects on both trust in peers (.33, 95% CI =
.22, .43) and perceived workgroup support (.27, 95% CI = .17, .38). Distributive justice
and task justice did not show the predicted indirect effects. The findings for task justice
and distributive justice, therefore, are inconsistent with Ambrose and Schminke’s (2009)
earlier findings. Peer overall justice, in turn, was related to trust in peers (.37, 95% CI =
.27, .46) and perceived workgroup support (.31, 95% CI = .20, .42). Only relationship
justice was directly related to trust in peers (.37, 95% CI = .18, .55).
One cross-path, that from peer task justice to supervisor overall justice, was signifi-
cant. As with supervisors, we are hesitant to interpret unpredicted effects, at least pend-
ing replication. Moreover, the overall model fit decreased when cross-paths were
included. Speculatively, it could be that task justice sometimes spans two organizational
levels. For example, if peers are not ‘doing their fair share’, and the supervisor fails to
intervene, then both entities have behaved poorly. That said, this possibility must await
future research.
General discussion and contribution
We have argued that the subjective and personal justice experiences of employees have
not been fully appreciated by researchers. As a result, the legacy justice rules identified
by prior contributions are incomplete. A fuller understanding can be gained by taking a
person-centric approach, which investigates the individual experience of working peo-
ple. Through this approach, we identified 14 emergent justice rules. Adding the 14 new
rules re-defines the dimensional structure of justice, so that it focuses less on procedures,
interpersonal interactions, and communication of information per se, and more on work
tasks and organizational relationships.
Theoretical implications: Contributions and future needs
Though we do not think that the list of justice rules is infinite, after these results we can
state that it is larger than previously imagined. There are a number of reasons why this is
important for researchers to recognize. Efforts to build fairer organizations could floun-
der on disagreements as to what constitutes ‘justice’. The range of issues that employees
consider when making their justice judgments may indeed be larger than anticipated,
taking into account both the avoidance of transgressions and the inclusion of care behav-
iors. Likewise, accepting to include new rules of justice may help to better take into
account different contexts, including cultural differences. Our findings also suggest ave-
nues for integration. Our new justice rules suggest that doing a proper job, promise keep-
ing (psychological contracting), empathy, non-harassment, and other behaviors are

1656 Human Relations 73(12)
important for fairness. In turn, these constructs have research literatures in their own
right. Our new justice rules could provide opportunities for integrative theory building.
While these prospects are promising, our findings also raise challenges. Perhaps the
most immediate involves measurement. In an intriguing article, Colquitt and Rodell
(2015) argue that some measures of justice are ‘indirect,’ assessing justice rules that
should produce fairness perceptions, whereas others are ‘direct,’ explicitly assessing
feelings of fairness. If our current list of justice rules is incomplete, then indirect meas-
ures of justice, by emphasizing legacy rules, are also incomplete. An alternative, of
course, would be to expand these instruments by adding additional rules. This approach
is limited, however, as the list of rules might change across cultures and over time. A
more straightforward solution would be to switch to a direct measure, such as overall
justice (Ambrose et al., 2015). Use of a direct measure would bring an additional concep-
tual advantage. As Ambrose and Schminke (2009) point out, justice rules have direct
effects on various work outcomes, even apart from their benefits to fairness. For exam-
ple, use of accurate procedures is certainly fair, but it also allows for better decision-
making (Cropanzano et al., 2015). Consequently, when an indirect measure is correlated
with a workplace criterion, we are not necessarily certain if this relationship is owing to
increased fairness, more valid business practices, or both. From a justice point of view,
direct measures are more precise. Despite these advantages of direct measures, the inclu-
sion of indirect measures – which represent a set of antecedents of overall justice – may
be particularly helpful when diagnosing where a justice problem stems from in a particu-
lar organizational context and where action is required. Furthermore, when systemati-
cally including the indirect measure in conjunction with direct measures, we can learn
more about the relative importance of different antecedents.
Another topic for future investigation concerns the relationship between justice and
job performance. Among the new rules, we found an interesting extension relating to the
importance of professional duties for justice, such as ‘doing your job properly’ and
‘being professional’. This could suggest that high performance causes justice, because
meeting one’s obligations is viewed as fair. This stands in contrast to the common inter-
pretation, that people perform better when they are treated justly. However, this contra-
diction may be more apparent than real. An organizational actor can demonstrate justice
by performing well, thereby meeting his or her obligations. This strong performance, in
turn, could inspire reciprocal efforts from a co-worker. This is fully consistent with the
traditional view, which maintains that just behavior motivates better performance from
others. Future research in non-work contexts (e.g. families) may investigate how far
performing and behaving according to one’s role expectations is related to being judged
as fair.
As was true for the emergent justice rules, the revised justice dimensions also connect
with research and theory in a range of neighboring fields: the strongest factors across our
four studies are relationship justice and task justice. These two dimensions show some
commonalities with the two fundamental dynamics of an organization (Katz and Kahn,
1978): the ‘techno-economic’ system oriented towards the economic efficiency in the
accomplishment of tasks (Taylor, 1911; Weber, 1921) and the ‘social maintenance’ sys-
tem (Roethlisberger and Dickson, 1939) oriented towards the building of organizational
solidarity. Various forms of this distinction have been conceptualized by traditional

Fortin et al. 1657
leadership theories (e.g. Blake and Mouton, 1961; Fiedler, 1967; Likert, 1961; Stodgill
et al., 1962), and a similar dichotomy has been used in the conflict literature (Sandy
et al., 2006). More recently, Cuddy et al. (2011) summarized in an integrative review
what is known about how people judge others – mainly by drawing on two broad catego-
ries, described as ‘warmth’ and ‘competence’ (Cuddy et al., 2008). These theories and
our results seem to have in common that the focus on the task and the focus on the rela-
tionship are two important ways for people to judge interactions at work. This similarity
between our dimensions of task justice and relationship justice and previous research
suggests the need for additional investigations and the potential for integrative work.
Practical implications
A prevailing concern amongst organizational justice researchers is that our findings, though
suggesting that justice is very important to workers, have yielded few practical applications
(Greenberg, 2009). While admittedly speculative, the present studies speak to this issue. As
we have seen, justice theorists organize a limited set of rules into a larger set of dimensions
– distributive, procedural, informational, and interpersonal (Colquitt, 2001). Workers, it
seems, use a larger set of rules that are organized into a smaller set of dimensions – distribu-
tive, task, and relationships – which we could call “justice concerns”. Consequently, too
few rules are being applied too broadly. Instead of asking, ‘are our procedures fair?’ (tradi-
tional dimension), firms may ask whether tasks are justly assigned and conducted, etc. For
example, employees are not (only) seeking voice (a rule) in the service of procedural jus-
tice (a dimension). They also want their managers and coworkers to behave professionally
and to keep their word. Likewise, rather than asking, ‘do we communicate in a fair way?’,
firms may ask whether relationships are just (justice concern). For example, employees’
justice concerns extend beyond sensitive communication (a rule) in the service of interac-
tional justice (a dimension). They want empathy and loyalty and fear harassment. The
additional rules identified here may provide managers with a larger ‘tool kit’ in order to
achieve justice amongst the central justice concerns.
This is a positive way to see our findings, but additional rules also create challenges.
In their fashion, earlier theories were somewhat mechanistic. At least in principle, organ-
izations could build in a relatively small number of specific rules (e.g. consistency, accu-
racy, and the like). With more rules, and presumably more disagreements as to what rules
are applicable, then application will be more organic and fluid. To illustrate, let us con-
sider three challenges – the challenge of culture, the challenge of empowerment, and the
challenge of teams.
Everyone is unlikely to prioritize justice rules in the same fashion, and the more rules
we identify, the more grounds there are for disagreement. This is especially relevant
when one considers cross-cultural differences. Individuals from East Asia (Li and
Cropanzano, 2009), Africa (Beugré, 2002), or the Middle East (Kahn et al., 2015) may
each have a somewhat different point of view. Justice rules identified by North American
and western European scholars are unlikely to capture fully their concerns. Consequently,
our findings suggest that organizations cannot simply “lift” an intervention from one
nation and move it to another. Rather, researchers and managers may need to make extra
efforts to understand justice concerns and rules inside a specific culture, consulting with
employees from that culture.

1658 Human Relations 73(12)
Even within a single nation, additional rules create problems, as everyone is unlikely
to agree. Besides, more rules create more implementation challenges. This is especially
true when no comprehensive list exists. In this climate of diversity and uncertainty,
organizations may wish to think about justice from the perspective of empowerment.
Rather than, say, designing justice into procedures in a top-down fashion, firms could
empower workers to set priorities and build their own systems of justice. In this way,
employees would have voice in creating just tasks and relationships. Voice, for example,
would not only be a fair procedure, it would also be a tool for creating these procedures
in the first place. This broader and more empowered way to seek justice should be a topic
for future research.
It is important to recognize that our justice rules, both traditional and emergent, appear
to apply in both hierarchical and non-hierarchical relationships. To be specific, peers
appear to be a strong source of justice within work organizations. While managerial
behavior is certainly important, it is not the entire story. Organizations will wish to create
a climate of justice, whereby co-workers are encouraged to treat one another appropri-
ately. By promoting participation and collaborative teaming, firms could provide
employees with the skills for treating one another fairly. This learning-by-doing might be
especially relevant, given that all workers will likely not value the same justice rules to
the same degree. Practical experience working with people of different backgrounds
would prove useful and also stresses the need for education (schools, universities) to
foster the learning of peer fairness.
Finally, we wish to conclude our discussion of practical implications on a positive
note. We were struck by the importance our participants place on promise keeping and
honesty in their work relationships. This suggests an untapped opportunity for justice
researchers. Employees, it seems, understand that things cannot always go their way.
However, if managers can keep their word, then much ill feeling and conflict could be
averted. In a business world of diversity and change, honesty may go a long way.
Conclusions
The present studies took a fresh look at how people experience fairness at work. In our
first study, we employed a new research paradigm, the person-centric model, in order to
investigate which rules people use when judging workplace fairness in interactions with
supervisors and peers. This approach allowed us to identify several justice rules that are
not included in traditional measures of justice, even though some of these rules had pre-
viously been identified by justice researchers. Taken together, the addition of 14 new
rules broadens the range of justice. However, they reduced the number of justice dimen-
sions from four to three, providing for a more parsimonious structural model. We hope
that these findings will encourage researchers to take a new look at the dimensions of
justice and how these can be applied within work settings.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/
or publication of this article: Studies 1 and 2 were supported by grant no. RES-061-25-0147
(Economic and Social Research Council, UK).

Fortin et al. 1659
ORCID iD
Marion Fortin https://orcid.org/0000-0002-5568-875X
Supplemental material
Supplemental material for this article is available online.
Notes
1 We decided to ask our respondents about incidents of injustice (instead of justice), because
the critical incident approach requires that people can identify salient events. Justice is often
perceived as a stable ‘state’, whereby features of ‘justice’ may not even become salient until a
transgression (i.e. an injustice) occurs. Not surprisingly, then, this has also been the traditional
way to study justice, beginning from Bies and Moag’s (1986) seminal studies that discovered
the concept of interactional justice through analyzing injustice events. People usually tend to
explore unjust events in more depth to better understand what is fair. This can lead to identify
which justice rules were broken and thus discover new justice rules.
2 For managers, the mean fairness was slightly lower with 1.97 and the mean injustice slightly
higher with 4.29 than for colleagues, with 2.29 and 4.16, respectively. However, ‘morally
wrong’ was higher for colleagues, with 4.35 versus 4.06 for managers. Comparing the close
and distant groups, we see almost no differences (close: fair 2.17, unjust 4.23, morally wrong
4.20; distant: fair 2.09, unjust 4.22, morally wrong 4.22). None of the differences was statisti-
cally significant at p = .05.
3 Note: All supplementary tables are published online only.
References
Ambrose ML and Schminke M (2001) Are flexible organizations the death knell for the future
of procedural justice? In: Cropanzano R (ed.) Justice in the Workplace: From Theory to
Practice. Mahwah, NJ: Erlbaum, 229–244.
Ambrose ML and Schminke M (2009) The role of overall justice judgments in organizational jus-
tice research: A test of mediation. Journal of Applied Psychology 94(2): 491–500.
Ambrose ML, Schminke M and Mayer DM (2013) Trickle-down effects of supervisor percep-
tions of interactional justice: A moderated mediated approach. Journal of Applied Psychology
98(4): 678–689.
Ambrose ML, Wo DXH and Griffith MD (2015) Overall justice: Past, present, and future. In:
Cropanzano RS and Ambrose ML (eds) The Oxford Handbook of Justice in the Workplace.
New York: Oxford University Press, 109–135.
Anderson JC and Gerbing DW (1988) Structural equation modeling in practice: A review and
recommended two-step approach. Psychological Bulletin 103(3): 411–423.
Anderson JC and Gerbing DW (1992) Assumptions and comparative strengths of the two-step
approach: Comment on Fornell and Yi. Sociological Methods and Research 20(3): 321–333.
Bandalos DL and Boehm-Kaufman MR (2009) Four common misunderstandings in exploratory
factor analysis. In: Lance CE and Vandenberg RJ (eds) Statistical Myths and Urban Legends.
New York: Routledge, 61–87.
Beugré CD (2002) Understanding organizational justice and its impact on managing employees: An
African perspective. International Journal of Human Resource Management 13(7): 1091–1104.
Bies RJ (2001) Interactional (in)justice: The sacred and the profane. In: Greenberg J and
Cropanzano R (eds) Advances in Organizational Justice. Stanford, CA: Stanford University
Press, 89–118.

1660 Human Relations 73(12)
Bies RJ and Moag JS (1986) Interactional justice: Communication criteria of fairness. In: Lewicki
RJ, Sheppard BH and Bazerman MH (eds) Research on Negotiation in Organizations.
Greenwich, CT: JAI Press, 43–55.
Blader S and Tyler TR (2003a) What constitutes fairness in work settings? A four-component
model of procedural justice. Human Resource Management Review 12(1): 107–126.
Blader S and Tyler TR (2003b) A four component model of procedural justice: Defining the mean-
ing of a ‘fair’ process. Personality and Social Psychology Bulletin 29(6): 747–758.
Blake RR and Mouton JS (1961) Group Dynamics: Key to Decision Making. Houston: Gulf
Publishing Co.
Bobocel DR and Holmvall CM (2001) Are interactional justice and procedural justice different?
Framing the debate. In: Gilliland S, Steiner D and Skarlicki D (eds) Research in Social Issues
in Management: Theoretical and Cultural Perspectives on Organizational Justice, Vol. 1.
Greenwich, CT: Information Age Publishing 85–108.
Bryant FB and Yarnold PR (1995) Principal-components analysis and exploratory and con-
firmatory factor analysis. In: Grimm LG and Yarnold PR (eds) Reading and Understanding
Multivariate Statistics. New York: American Psychological Association, 99–136.
Colquitt JA (2001) On the dimensionality of organizational justice: A construct validation of a
measure. Journal of Applied Psychology 86(3): 386–400.
Colquitt JA and Rodell JB (2015) Measuring justice and fairness. In: Cropanzano R and Ambrose
MA (eds) The Oxford Handbook of Justice in the Workplace. Oxford: Oxford University
Press, 187–202.
Colquitt JA, Greenberg J and Scott BA (2005) Organizational justice: Where do we stand? In:
Greenberg J and Colquitt JA (eds) Handbook of Organizational Justice. Mahwah, NJ:
Lawrence Erlbaum, 589–619.
Colquitt JA, Scott BA, Rodell JB, et al. (2013) Justice at the millennium, a decade later: A meta-
analytic test of social exchange and affect-based perspectives. Journal of Applied Psychology
98(2): 199–236.
Colquitt JA, Long DM, Rodell JB and Halvorsen-Ganepola MDK (2015) Adding the ‘in’ to jus-
tice: A qualitative and quantitative investigation of the differential effects of justice rule
adherence and violation. Journal of Applied Psychology 100(2): 278–297.
Crawshaw JR, Cropanzano R, Bell C and Nadisic T (2013) Organisational justice: New insights
from behavioural ethics. Human Relations 66(7): 885–904.
Cropanzano R and Ambrose ML (2001) Procedural and distributive justice are more similar than
you think: A monistic perspective and a research agenda. In: Greenberg J and Cropanzano R
(eds) Advances in Organizational Justice. Stanford, CA: Stanford University Press, 119–151.
Cropanzano R, Li A and Benson L III (2011) Peer justice and teamwork processes. Group and
Organization Management 36(5): 567–576.
Cropanzano R, Fortin M and Kirk J (2015) How do we know when we are treated fairly? Justice
rules and fairness judgments. Research in Personnel and Human Resources Management 33:
279–350.
Cuddy AJC, Fiske ST and Glick P (2008) Warmth and competence as universal dimensions of
social perception: The stereotype content model and the BIAS map. In: Zanna MP (ed.)
Advances in Experimental Social Psychology, Vol. 40. New York: Academic Press, 61–149.
Cuddy AJC, Glick P and Beninger A (2011) The dynamics of warmth and competence judgments,
and their outcomes in organizations. Research in Organizational Behavior 31: 73–98.
Cugueró-Escofet N and Fortin M (2014) One justice or two? A model of reconciliation of norma-
tive justice theories and empirical research on organizational justice. Journal of Business
Ethics 124(3): 435–451.

Fortin et al. 1661
Cugueró-Escofet N and Rosanas JM (2013) The just design and use of management control systems
as requirements for goal congruence. Management Accounting Research 24(2013): 23–40.
Deutsch M (1985) Distributive Justice. New Haven, CT: Yale University Press.
Eisenberger R, Stinglhamber F, Vandenberghe C, et al. (2002) Perceived supervisor support:
Contributions to perceived organizational support and employee retention. Journal of Applied
Psychology 87(3): 565–573.
Fiedler FE (1967) A Theory of Leadership Effectiveness. New York: McGraw-Hill.
Greenberg J (2006) Losing sleep over organizational injustice: Attenuating insomniac reactions to
underpayment inequity with supervisory training in interactional justice. Journal of Applied
Psychology 91(1): 58–69.
Greenberg J (2009) Everybody talks about organizational justice, but nobody does anything about
it. Industrial and Organizational Psychology: Perspectives on Science and Practice 2(2):
181–195.
Guo J, Rupp DR, Weiss H and Trougakos J (2011) Justice in organizations: A personcen-
tric approach. In: Gilliland S, Steiner D and Skarlicki D (eds) Emerging Perspectives on
Organizational Justice and Ethics, Vol. 7. Charlotte, NC: Information Age Publishing, 3–32.
Hollensbe EC, Khazanchi S and Masterson SS (2008) How do I assess if my supervisor and
organization are fair? Identifying the rules underlying the entity-based justice perceptions.
Academy of Management Journal 51(6): 1099–1116.
Holtz BC and Harold CM (2009) Fair today, fair tomorrow? A longitudinal investigation of overall
justice perceptions. Journal of Applied Psychology 94(5): 1185–1199.
Hunton JE, Hall TW and Price KH (1998) The value of voice in participative decision making.
Journal of Applied Psychology 83(5): 788–797.
Jones DA and Martens ML (2009) The mediating role of overall fairness and the moderating role
of trust certainty in justice–criteria relationships: The formation and use of fairness heuristics
in the workplace. Journal of Organizational Behavior 30(8): 1025–1051.
Kahn K, Abbas M, Gul A and Raja U (2015) Organizational justice and job outcomes: Moderating
role of Islamic work ethics. Journal of Business Ethics 126(2): 235–246.
Katz D and Kahn RL (1978) The Social Psychology of Organizations, Vol. 2. New York: Wiley,
528.
Lavelle JJ, Rupp DE and Brockner J (2007) Taking a multifoci approach to the study of jus-
tice, social exchange, and citizenship behavior: The target similarity model. Journal of
Management 33(6): 841–866.
Lavelle JJ, Brockner J, Konovsky MA, et al. (2009a) Commitment, procedural fairness, and organ-
izational citizenship behavior: A multifoci analysis. Journal of Organizational Behavior
30(3): 337–357.
Lavelle JJ, McMahan GC and Harris CM (2009b) Fairness in human resource management, social
exchange relationships, and citizenship behavior: Testing linkages of the target similarity
model among nurses in the United States. The International Journal of Human Resource
Management 20(12): 2419–2434.
Lavelle JJ, Rupp DE, Manegold J and Thornton MA (2015) Multifoci justice and target similarity:
Emerging research and extensions. In: Cropanzano R and Ambrose MA (eds) The Oxford
Handbook of Justice in the Workplace. Oxford: Oxford University Press, 165–186.
Leventhal GS (1980) What should be done with equity theory? In: Gergen KJ, Greenberg MS and
Willis RH (eds) Social Exchange: Advances in Theory and Research. New York: Plenum,
27–55.
Li A and Cropanzano R (2009) Do East Asians respond more/less strongly to organizational justice
than North Americans? A meta-analysis. Journal of Management Studies 46(5): 787–805.

1662 Human Relations 73(12)
Li A, Cropanzano R and Bagger J (2013) Justice climate and peer justice climate: A closer look.
Small Groups Research 44(5): 563–592.
Li A, Cropanzano R and Molina A (2015) Fairness at the unity level: Justice climate, justice
climate strength, and peer justice. In: Cropanzano R and Ambrose MA (eds) The Oxford
Handbook of Justice in Work Organizations. Oxford: Oxford University Press, 137–164.
Likert RL (1961) The Human Organization. New York: McGraw-Hill.
Lind EA (2001) Fairness heuristic theory: Justice judgments as pivotal cognitions in organiza-
tional relations. In: Greenberg J and Cropanzano R (eds) Advances in Organizational Justice.
Stanford, CA: Stanford University Press, 56–88.
Masterson SS (2001) A trickle-down model of organizational justice: Relating employees’ and
customers’ perceptions of and reactions to fairness. Journal of Applied Psychology 86(4):
594–604.
Morrison EW and Robinson SL (1997) When employees feel betrayed: A model of how psycho-
logical contract violation develops. Academy of Management Review 22(1): 226–256.
Patient DL and Skarlicki DP (2010) Increasing interpersonal and informational justice when com-
municating negative news: The role of the manager’s empathic concern and moral develop-
ment. Journal of Management 36(2): 555–578.
Robinson SL (1996) Trust and breach of the psychological contract. Administrative Science
Quarterly 41(4): 574–599.
Roch SG and Shanock LR (2006) Organizational justice in an exchange framework: Clarifying
organizational justice distinctions. Journal of Management 32(2): 299–322.
Roethlisberger FJ and Dickson WJ (1939) Management and the Worker: An Account of a Research
Program Conducted by the Western Electric Company, Hawthorne Works. Chicago: Harvard
University Press.
Rousseau DM (1995) Psychological Contracts in Organizations: Understanding Written and
Unwritten Agreements. Thousand Oaks, CA: SAGE.
Rupp DE, Shao R, Jones KS and Liao H (2014) The utility of a multifoci approach to the study of
organizational justice: A meta-analytic investigation into the consideration of normative rules,
moral accountability, bandwidth-fidelity, and social exchange. Organizational Behavior and
Human Decision Processes 123(2): 159–185.
Rupp D, Shapiro D, Folger R, et al. (2017) A critical analysis of the conceptualization and meas-
urement of ‘Organizational justice: Is it time for reassessment?’ Academy of Management
Annals 12(2): 919–959.
Sandy SV, Boardman SK and Deutsch M (2006) Personality and conflict. In: Deutsch M and
Coleman PT (eds) The Handbook of Conflict Resolution: Theory and Practice: Jossey-Bass,
331–355.
Schminke M, Ambrose ML and Noel TW (1997) The effect of ethical frameworks on perceptions
of organizational justice. Academy of Management Journal 40(5): 1190–1207.
Sheppard BH and Lewicki RJ (1987) Toward general principles of managerial fairness. Social
Justice Research 2(2): 161–176.
Spreitzer GM and Mishra AK (1999) Giving up without losing control: Trust and its substitutes’
effects on managers’ involving employees in decision-making. Group and Organization
Management 24(2): 155–187.
Stodgill RM, Goode OS and Day DR (1962) New leader behavior description subscales. Journal
of Psychology 54(2): 259–269.
Taylor FW (1911) The Principles of Scientific Management. New York: Norton.
Törnblom K and Kazemi A (2015) Distributive justice: Revising past statements and reflecting on
future prospects. In: Cropanzano R and Ambrose MA (eds) The Oxford Handbook of Justice
in the Workplace. Oxford: Oxford University Press, 15–50.

Fortin et al. 1663
Weber M (1921–1984) Wirtschaft und Gesellschaft: Grundriss der verstehenden Soziologie.
Tübingen: Mohr JCB (Siebeck, Paul).
Weiss H and Rupp DE (2011) Experiencing work: An essay on a person-centric work psychol-
ogy. Industrial and Organizational Psychology: Perspectives on Science and Practice 4(1):
83–97.
Marion Fortin is a Professor of Human Resource Management and Organizational Behavior at
Toulouse School of Management (TSM), University of Toulouse 1, France. Prior to joining TSM,
she worked at the University of Durham, UK. She received her PhD from Trinity College Dublin,
Ireland. Fortin’s research focuses on fairness dynamics at work, for example in the context of
women’s careers, and on the links between morality and fairness. Her research has appeared in a
number of international journals, including Journal of Organizational Behavior, Journal of
Business Ethics, Journal of Applied Psychology, and Organizational Behavior and Human
Decision Processes. [Email: marionfortin@gmail.com ]
Russell Cropanzano is a Professor of Management and Entrepreneurship at the University of
Colorado. Previously, he has published over 140 scholarly articles and chapters. Cropanzano has also
authored two books and edited four others. Cropanzano was also a winner of the 2000 Outstanding
Paper Award from the Consulting Psychology Journal, the 2007 Best Paper Award from Academy
of Management Perspectives, and the 2010 Best Paper Award from the Journal of Management.
He is a past editor of the Journal of Management and a fellow in the Academy of Management,
the Society for Industrial/Organizational Psychology, the Southern Management Association,
and the Association for Psychological Science. [Email: russell.cropanzano@colorado.edu ]
Natàlia Cugueró-Escofet is a Professor of Management at the Open University of Catalonia,
Barcelona, where she is also the course director of the Master in Corporate Social Responsibility.
She collaborates with the Chair funded by Crèdit Andorrà of Markets, Humanism and Organizations
at IESE Business School in Barcelona, Spain. Her research interests include organizational justice,
ethics, corporate social responsibility, management control systems, and sharing economy. Prior
to joining the academic world, Natàlia worked for 15 years as a professional manager in several
companies, and still collaborates with several start-ups and SMEs. She has published in interna-
tional scholarly journals and her work has appeared in Journal of Business Ethics, Management
Accounting Research, Intangible Capital, and Social Justice Research. She has written several
books and book chapters about justice and management control systems. She is a regular contribu-
tor to the press in her areas of interest. [Email: ncuguero@iese.edu ]
Thierry Nadisic is an Associate Professor of Organizational Behavior at emlyon business school,
Paris campus, France. He completed his PhD in management at HEC Paris and won in 2008 the
HEC Foundation prize for dissertation of the year. He also received the 2016 ‘Best Research
paper’ prize awarded by the AGRH (French Speaking Academy of Research in Management and
Human Resources). Thierry’s research, which has been published widely in French- and English-
speaking outlets, addresses issues of fair management and well-being at work. [Email: nadisic@
em-lyon.com ]
Hunter Phoenix Van Wagoner is currently a PhD candidate studying organizational behavior at Leeds
School of Business at the University of Colorado. His primary research interests include emotions and
well-being, diversity and inclusion, and justice. [Email: hunter.vanwagoner@colorado.edu ]