Back to Journal Cover Page


CURRENT RESEARCH IN SOCIAL PSYCHOLOGY


Volume 8, Number 12
Submitted: January 7, 2002
First Revision: October 9, 2002
Second Revision: January 6, 2003
Third
Revision: January 7, 2003
Accepted: January 7, 2003
Publication Date: January 7, 2003

THE EFFECTS OF GROUP AFFILIATION ON THIRD PARTIES' JUSTICE PERCEPTIONS

Jeffrey A. Miles
University of the Pacific

Stefanie E. Naumann
University of the Pacific

ABSTRACT

We investigated the justice perceptions of third parties, individuals who are not directly involved in a justice situation, either as the decision maker or the recipient of the justice outcome.  Specifically, we examined the effects of two forms of in-group affiliation on third parties' justice perceptions: 1) affiliation of decision maker and guilty group of students, and 2) affiliation of third parties reading about the scenario with the decision maker and guilty group of students.  In a scenario study, university students evaluated the fairness of a disciplinary situation used to punish a group of students for plagiarizing.  Different results were found for fraternity vs. non-fraternity subjects, depending on whether the decision maker was affiliated with a fraternity or not and whether the decision maker was a group or individual.

[162]
--------------
[163]

INTRODUCTION

According to social identity theory (Tajfel & Turner, 1979), people derive part of their self-image from the groups to which they belong.  Further, the theory suggests that individuals will favor members of their group over non-group members in allocating resources or rewards (e.g., Campbell, 1965; Sherif & Sherif, 1969).  Research has found that people categorize others in terms of the groups to which they belong; further, this categorization is thought to occur whether those doing the categorizing are active group member participants or are watching from the outside (Wegner, 1982).  When third parties are made aware of group membership (i.e., "focal group awareness"), they develop stereotypes about members of the group (Wegner, 1982).

To our knowledge, no existing research has explicitly examined third parties' evaluations of justice situations when group affiliation is made salient.  The purpose of the present study is to examine the effects of group affiliation on third parties' fairness evaluations of a punishment scenario.  Whereas a considerable amount of justice research has examined the justice perceptions of those actively participating in various procedures and outcomes, little research has examined the justice perceptions of those not directly involved in these justice judgment situations.  In the legal discipline, there are terms for those directly involved in a justice situation (e.g., first and second parties; prosecution and defense).  There is literature examining the role of mediators and arbitrators in dispute resolution (e.g., Ross & Conlon, 2000).  However, little research has explicitly examined those not directly involved in the justice situation, either as a decision maker (e.g., arbitrator) or recipient of a justice outcome.

In the current study we define third parties as individuals who are not directly involved in a justice situation, either as the decision maker or the recipient of the justice procedure or outcome.  Although third parties are not directly involved in a justice situation, we expect them to have some interest in the process or outcome (e.g., James & Cropanzano, 1990; Lind, Kray & Thompson, 1998; Naumann & Bennett, 2000).  In the workplace third party perceptions are often just as important as those of the individuals directly involved in the justice situation.  For instance, James and Cropanzano (1990) argued that individuals observe others in their focal groups and arrive at judgments concerning how procedures experienced by others affect them.  Lind, Kray, and Thompson (1998) have presented a parallel argument:

Much of the information any individual person has about the fairness of an organization or a supervisor comes not from personal experiences but instead from the broader collective experience of other people.  Collective experience provides a far larger pool of instances of fair or unfair actions than does the experience of any one individual, and it would be logical for people seeking to form a complete assessment of justice to make use of these actions as they generate an impression (p. 2).

We examined two forms of in-group affiliation: 1) affiliation of decision maker and guilty group of students in a hypothetical scenario, and 2) affiliation of third party subjects reading about the scenario.  We also explored the effects of whether the decision maker was a group or individual.  Three components of fairness evaluations were considered: distributive justice, trust, and neutrality.  Distributive justice concerns the perceived fairness of outcomes received (e.g., Greenberg, 1987).  Trust (e.g., Tyler, 1989) may be defined as the degree to which individuals believe the decision maker intends to act in a fair manner and that they can count on the decision maker to make a fair decision.  Neutrality (e.g., Tyler, 1989) involves the extent to which the decision maker is perceived as free from bias (i.e., uses openness and honesty instead of hidden agendas or opinions to make decisions).

Existing organizational justice research (e.g., Leventhal, 1976; Lind & Tyler, 1988) suggests that in order for individuals to perceive a justice situation as fair, they must perceive the decision maker as being unbiased, impartial, and trustworthy.  Thus, we would expect that in a situation where a decision maker is a member of the same in-group as a guilty party, third parties would perceive the situation less fairly than one where the decision maker is a neutral third party.

Hypotheses 1a, b, and, c:  Subjects' distributive justice, trust, and neutrality perceptions will be lower when there is a common fraternity affiliation between decision maker and guilty party compared to when there is no affiliation.

[163]
--------------
[164]

A second question we addressed concerns third parties' group affiliation: What happens when the scenarios third parties read about involve the negative actions of their own group members: Will their fairness evaluations be the same as those of third parties not in their group?  The social psychology literature offers a basis from which to make predictions.  When unfavorable characteristics of one's group are made salient (e.g., the group has done something objectionable), it is thought that members of one's group perceive this as threatening to their social identity, even if they had not personally behaved in an objectionable way (Doosje, Branscombe, Spears, & Manstead, 1998).  As a result, they are motivated to maintain a positive group image (e.g., Kuhl, 1997).

Although the scenarios are hypothetical and, thus, the subjects were not personally acquainted with the individuals in the scenarios, there is some evidence that even when group members do not have an interpersonal relationship with other group members, they are motivated to maintain a positive image of the group (Marques, Abrams, Paez, & Martinez-Taboada, 1998).  Thus, we expected in-group subjects (i.e., fraternity members) to identify with the people in the scenario since they were presented as being in the same fraternity.

Fraternity members may be less likely to view the affiliation scenario as unfair because they identify strongly with the group members in the scenario.   They may have higher levels of trust than non-fraternity members that a fraternity member decision maker will make a fair decision.  Researchers have suggested that threats (e.g., punishments) aimed at members of one's group are viewed as shared threats and engender the view that one must help to protect the group's "common fate" (Lee & Ottati, 1995).  Further, shared threats facilitate the likelihood that group members will perceive themselves as similar to other group members (Lee & Ottati, 1995).

Social identity theory posits that, in order to increase their social identity, group members tend to give more favorable allocations to fellow group members than to out-group members.  However, it should be noted that the theory does not require the individual to actually make the allocation in order to yield the subsequent increase in social identity.  Group members observing another member making favorable allocations to a third group member should experience the same increase in social identity as if they made the allocations themselves (Platow, O'Connell, Shave, & Hanning, 1995). 

From the preceding discussion we hypothesize the following:

Hypotheses 2a, b, and c:  Subjects who are not members of the fraternity will have lower distributive justice, trust, and neutrality perceptions than will those subjects who are members of the fraternity.

Miles and Palmer (2001) found that students reading a scenario about a group of students being punished had higher distributive justice perceptions when the decision maker was a group compared to when the decision maker was an individual.  In that study, there was no affiliation among the guilty students or the decision maker(s) of any kind.  We expect that subjects' justice perceptions in the current study will be influenced by whether the decision maker is a group or an individual.  As noted earlier, social identity theory suggests that, in order to enhance their social identity, group members are likely to give more favorable allocations to fellow group members than to out-group members.

Much of our legal system is based on the belief that juries of one's peers are better suited to make fair judgments than one individual (Few, 1993; Forbes, 1995).  If the decision maker is a group, we expect fraternity subjects to view the scenario more fairly than non-fraternity subjects when the decision maker is a group affiliated with the subjects' fraternity as opposed to an individual not affiliated with the fraternity.  Non-fraternity subjects are less likely to identify with the decision-making group as their peers and, thus, are expected to view such a scenario as unfair; instead, they would likely view a non-affiliated individual as better able to make an unbiased decision.

Hypotheses 3a, b, and c: A significant three-way interaction is expected among affiliation of decision maker (in fraternity of not), type of decision maker (group or individual), and type of subject (in fraternity or not) in influencing the distributive justice, trust, and neutrality perceptions of subjects.   That is, the highest levels of distributive justice, trust, and neutrality perceptions should be found for fraternity subjects in the condition where the decision maker is a group affiliated with a fraternity as opposed to every other condition that lacks all or part of these conditions.

[164]
--------------
[165]

METHOD

Subjects

Participants were 579 undergraduate business students from six universities.  Approximately half of the students were members of a co-educational, professional, business fraternity (n = 281), and half of the students were not members of the fraternity (n = 298).  It should be noted that the level of identification of group members in a fraternity was expected to be higher than in most groups.  Implications of our choice of a fraternity to operationalize group affiliation will be addressed in the discussion section.  57.4% of the participants were female, and the average age of the sample was 20.5 years.  For race, participants reported themselves: 2.5% as "Black/African American," 11.8% as "Asian/Pacific Islander," 77.7% as "White/Caucasian," 3.6% as  "Hispanic/Latino," 4.3% as "Other," and 4.6% did not respond to this question.  There were no statistically significant differences in any of the demographic variables between the students who were fraternity members and those who were not, or across the universities.

Design

A 2 X 2 X 2 full factorial design was used.  The independent variables and their levels were as follows: affiliation of decision maker (in a fraternity or not), type of decision maker (individual or group), and fraternity membership of subject (in a fraternity or not).  Each subject was randomly assigned to complete one of eight versions of a survey containing hypothetical scenarios derived from an actual plagiarism situation at a university.  The most valid scenario studies involve situations the respondents have experienced and understand (Lind & Tyler, 1988).  In each of the scenarios a group of students was punished for plagiarizing another group's work.  This situation was seen as an ideal one for examining the perceptions of students reading about a justice incident.  Plagiarism is a key aspect of college life and a behavior familiar to many college students.  The subjects could personally identify with the situation and might have plagiarized themselves.  It has been estimated that between 36 and 55% of students have plagiarized (Hale, 1987; Roig, 1997).

Procedure

Non-fraternity participants completed the survey during their regular class sessions in a business school classroom; fraternity participants completed the survey during their regular chapter meetings held in a business school classroom.  Individual subjects received no rewards for their participation in the survey.  However, each of the fraternity chapters received a check for $50 for the chapter participating in the research project.

All survey versions contained the following introductory text:

A professor was grading students' 20-page group term papers for Zoology 101.  The professor noticed that one group's paper was particularly well-written, especially considering their prior poor work in the class.  The professor also thought the paper seemed familiar to a paper turned in last year.  The professor checked the files and found that an identical paper had been turned in last year by another group.

Manipulations of the Independent Variables

Affiliation of decision maker.  Next, students in the fraternity affiliation condition read the following text:

The professor turned over the case to the Decision maker, who handles all such cases at that university. The Decision maker carefully examined the evidence.  Next, the Decision maker called in the group of students to get their side of the story, too. It turned out that the Decision maker and the group of students under investigation were all Brothers in Delta Sigma Pi.  All wore their fraternity pins to the meeting. After weighing all the evidence, the Decision maker made a decision.  The decision of the Decision maker was to have the group redo the entire paper.

[165]
--------------
[166]

Students in the non-affiliation condition read the following text:

The professor turned over the case to the Decision maker, who handles all such cases at that university.  The Decision maker carefully examined the evidence.  Next, the Decision maker called in the group of students to get their side of the story, too.  After weighing all the evidence, the Decision maker made a decision.  The decision of The Decision maker was to have the group of students redo the entire paper.

Type of decision maker.  Half of the subjects read about an individual decision maker (Dean of Students) and the other half read about a group of decision makers in the scenario (Faculty Judiciary Committee).

Manipulation checks.  Two questions were included as manipulation checks.  Given the importance of establishing that role-playing subjects fully understood the situations to which they were asked to respond (Greenberg & Eskew, 1990; Greenberg & Folger, 1988), subjects were asked whether the students in the scenario were members of a fraternity or not and whether the decision maker was the Dean of Students or a Faculty Judiciary Committee.

Subjects' demographic information.  Subjects reported their age, gender, and race.

Measurement of the Dependent Variables

After reading one of the scenarios, subjects responded to items with 7-point scales ranging from  "None/Not at all" (1) to "Extremely/To a great extent" (7).

Organizational justice.  To assess how fair participants viewed the different scenario conditions, three justice scales were developed using items consistent with previous justice research (e.g., Bies & Shapiro, 1987; Greenberg, 1987; Tyler, 1989; 1994).  Two items (Cronbach's alpha = .70) assessed trust (e.g., "To what extent would the Dean of Students act the same way toward you if you were involved in a similar situation?") and four items (Cronbach's alpha = .71) assessed neutrality (e.g., "To what extent did the Dean of Students make the decision without being biased in any way?").  Five items (Cronbach's alpha = .83) assessed distributive justice (e.g., "How fair was the outcome that the student received").

RESULTS

Manipulation Checks

A separate group of 40 fraternity members and 40 non-fraternity members were asked to read the scenario in which the decision maker and the group of students were both affiliated with the fraternity.  The students in the two groups used in this part of the study did not differ significantly from those in the main study, or from each other on any of the demographic variables, F(610) = .571 or less in all cases, ns.

These 80 students were asked to read the scenario and to answer two questions about it.  The first question read, "To what extent do you believe that this situation could have occurred at this University?"  The mean response for this question on a 7-point scale (ranging from "1" none/not at all to "7" extremely/to a great extent) was 5.2.  The second question read, "To what extent do you believe this to be a plausible situation?"  The mean response for this question was 5.5.  Thus, based on these responses, the students found the scenario to be both a plausible and possible one for them at their institution.   This result provides support that the scenario used in the current study was a good one for creating the conditions that were desired to address the posited hypotheses.

For the main study, to ensure the success of the affiliation/non-affiliation condition, participants were asked whether the students in the scenario were members of a fraternity or not.  None of the participants in the main study failed to answer this question correctly.  Subjects were also asked whether the decision maker was the Dean or a Faculty Committee.  Three subjects failed to answer this question correctly.  Results were run both with and without these three subjects, but did not significantly change the results, so their scores were left in the analyses. 

[166]
--------------
[167]

Hypothesis Tests

Descriptive statistics for the three dependent variables (distributive justice, trust, and neutrality) appear in Table 1. 

Table 1: Descriptive Statistics

Mean

SD

1

2

3

1.  Distributive justice

4.21

1.46

(.83)

2.  Trust

3.96

1.45

.50***

(.70)

3.  Neutrality

4.25

1.23

.48***

.52***

(.71)

n = 571; ***p < .001; Coefficient alphas appear on the diagonal.

Means, standard deviations, and planned comparisons for all experimental conditions are reported in Tables 2 through 4. 

Table 2: Means, Standard Deviations, and Planned Comparisons for Distributive Justice

Scenario Affiliation

Decision Maker

Subject in Fraternity

Row simple main effects
F(1, 135-154)

Yes

No

Decision maker(s)
affiliated with fraternity

Individual

4.14
(1.85)

3.61
(1.43)

3.76*

Group

4.83
(1.64)

3.74
(1.37)

17.84***

Decision maker(s) not affiliated with fraternity

Individual

4.34
(1.21)

4.15
(1.26)

0.83

Group

4.41
(1.43)

4.47
(1.06)

0.12

Column simple main effects F(3, 269-302)

2.33

7.10***

Note: Standard deviations are in parentheses. In each cell, n = 64-84.
  * p < .05;  ** p < .01; *** p < .001

[167]
--------------
[168]

Table 3: Means, Standard Deviations, and Planned Comparisons for Trust

Scenario Affiliation

Decision Maker

Subject in Fraternity

Row simple main effects

F(1, 135-154)

Yes

No

Decision maker(s)
a Affiliated with fraternity

Individual

4.44

(1.01)

3.41
(1.54)

17.16***

Group

4.41
(1.27)

3.57
(1.45)

11.95**

Decision maker(s) not affiliated with fraternity

Individual

3.62
(1.46)

3.88
(1.49)

1.04

Group

4.16
(1.39)

4.15
(1.24)

0.00

Column simple main effects F(3, 269-302)

4.56*

3.74*

Note: Standard deviations are in parentheses. In each cell, n = 64-84.
* p < .05;  ** p < .01; *** p < .001

[168]
--------------
[169]

Table 4: Means, Standard Deviations, and Planned Comparisons for Neutrality

Scenario Affiliation

Decision Maker

Subject in Fraternity

Row simple main effects

F(1, 135-154)

Yes

No

Decision maker(s)
affiliated with fraternity

Individual

4.24
(1.01)

3.95
(1.38)

1.91

Group

4.11
(1.27)

3.55
(1.27)

6.65*

Decision maker(s) not affiliated with fraternity

Individual

4.29
(1.22)

4.46
(1.23)

0.65

Group

4.55
(0.92)

4.71
(1.16)

0.80

Column simple main effects F(3, 269-302)

2.03

19.00***

Note: Standard deviations are in parentheses. In each cell, n = 64-84.
  * p < .05;  ** p < .01; *** p < .001

The results of a multivariate analysis of variance (MANOVA) for decision maker affiliation (in fraternity or not), type of decision maker (individual or group), and subject (in fraternity or not) with distributive justice, trust, and neutrality as dependent variables are reported in Table 5.

Table 5: Analysis of Variance of Dependent Variables

Analysis of variance
F(1, 534)

Source

MANOVA Pillais
F(7, 1070)

02

Distributive
Justice

Neutrality

Trust

Affiliation (A)

14.38***

.08

4.64*

29.47***

5.11*

Decision Maker (D)

3.04*

.02

14.28**

0.06

3.20

Fraternity (F)

4.59**

.02

12.10**

5.12*

10.09**

A X D

4.10**

.03

0.64

6.22*

1.81

A X F

5.71**

.05

12.84**

6.75*

6.61*

D X F

0.32

.00

0.37

0.86

0.34

A X D X F

2.76*

.02

2.94

0.42

0.87

MSE

1.64

1.23

1.65

* p < .05, ** p < .01, *** p < .001
Affiliation = decision maker in the scenario was a member of the fraternity or not
Decision maker = decision maker was an individual or a group
Fraternity = subject reading the scenario was a member of the fraternity or not

[169]
--------------
[170]

Hypotheses 1a, 1b, and 1c predicted that subjects' perceptions of distributive justice, trust, and neutrality would be lower when there was a common fraternity affiliation between decision maker and guilty party compared to when there was no affiliation.  The MANOVA supported all three of these hypotheses (Please see Table 5).  Subsequent univariate ANOVAs yielded the following results.

Specifically, for Hypothesis 1a, subjects' distributive justice perceptions were lower when there was a decision maker affiliation (M = 4.07) compared to when there was not an affiliation (M = 4.33), F(1,578) = 4.64, p < .05.  For Hypothesis 1b, subjects' trust perceptions were lower when there was a decision maker affiliation (M = 3.93) compared to when there was not an affiliation (M = 4.19), F(1,561) = 5.11, p < .05.  For Hypothesis 1c, subjects' neutrality perceptions were lower when there was a decision maker affiliation (M = 3.96) compared to when there was not an affiliation (M = 4.51), F(1,557) = 29.47, p < .001.

Hypotheses 2a, 2b, and 2c predicted that subjects who were not members of the fraternity would have lower distributive justice, trust, and neutrality perceptions than would those subjects who were members of the fraternity.  Again, the MANOVA supported all three of these hypotheses (Please see Table 5).  Subsequent univariate ANOVAs yielded the following results.

More specifically, for hypothesis 2a, subjects who were not members of the fraternity had lower distributive justice perceptions (M = 4.01) compared to those who were members (M = 4.43), F(1,578) = 12.10, p < .01.  For hypothesis 2b, subjects who were not members of the Fraternity had lower trust perceptions (M = 3.89) compared those who were members (M = 4.26), F(1,561) = 10.10, p < .01.  For hypothesis 2c, subjects who were not members of the Fraternity had lower neutrality perceptions (M = 4.19) compared those who were members (M = 4.40), F(1,557) = 5.12, p < .05.

Hypotheses 3a, 3b, and 3c predicted a significant three-way interaction among affiliation of decision maker (in fraternity of not), type of decision maker (group or individual), and type of subject (in fraternity or not) in influencing the distributive justice, trust, and neutrality perceptions of subjects.   As shown in Table 5, the three-way interaction was significant.  Subsequent univariate ANOVAs yielded the following results.

Specifically, for the fraternity subjects, we found significant interactions for type of decision maker (individual vs. group) and affiliation of decision maker (in fraternity or not) in the distributive justice, F(1, 79) = 11.07**, p < .01, and neutrality perceptions, F(1,19) = 10.56**, p < .01, of these subjects.  However for the trust dependent variable, only a main effect for the affiliation independent variable was detected F(1,79) = 27.84**, p < .01.  On the other hand, for the non-fraternity subjects, we found no significant interactions.  Instead, we found main effects of type of decision maker and affiliation on the distributive justice and neutrality perceptions of non-fraternity subjects.  We also found a main effect of type of decision maker on the trust perceptions of non-fraternity subjects.

DISCUSSION

The purpose of the present study was to examine two forms of in-group affiliation: 1) affiliation of decision maker and guilty group of students in a hypothetical scenario, and 2) affiliation of third party subjects reading about the scenario.  We also explored the effects of whether the decision maker was a group or individual.  Three components of fairness evaluations were considered: distributive justice, trust, and neutrality.  Our findings are consistent with social identity theory.

Overall, non-fraternity member subjects had lower fairness perceptions than fraternity member subjects when the students in the scenario were fraternity members compared with when the students in the scenario were not fraternity members.  Non-fraternity member subjects may have believed that the fraternity member scenario violated justice rules that the procedure be unbiased.  Further, they would be less likely to identify with the fraternity members in the scenario.  Thus, their fairness perceptions were lower.

[170]
--------------
[171]

The significant interactions we found for the fraternity subjects for the type of decision maker (individual vs. group) and the affiliation of decision maker (in fraternity or not) are also consistent with social identity theory.  Some social identity research suggests that when group members are exposed to a stereotypic threat (e.g., the perception that fraternity members will soften the rules to help out one another), they are motivated to maintain a positive sense of social identity (e.g., Lee & Ottati, 1995).  Thus, the fraternity members may have been motivated to view the decision maker as fair and impartial (neutral).

Fraternity member subjects also had higher trust perceptions when the people in the scenario were fraternity members.  Social identity theory suggests that when a negative outcome is directed at one's group, this leads to a "negative social identity" that spurs attempts at enhancing the group's status.  Further, the theory suggests that one's own social standing depends on the group's status (e.g., Tyler, Kramer, & John, 1999).  Thus, subjects may have been more trusting of a fraternity decision maker to ensure that the group's status as a whole would not be downgraded as a result of the actions of the students in the scenario.

We also examined whether subjects' justice perceptions would be influenced by whether the decision maker in the scenario was an individual or a group.  The highest levels of distributive justice, trust, and neutrality perceptions were found for fraternity subjects in the condition where the decision maker was a group affiliated with the fraternity.  Non-fraternity subjects appeared to be less likely to identify with the decision-making group as their peers and, thus, considered such a scenario to be less fair.

Areas for Future Research

The present study introduces numerous avenues for future research.  The focal group examined in the present study was that of a fraternity.  It may be argued that the level of identification of group members in a fraternity is higher than in other groups.  Given that recent research has distinguished between high and low identification group members (e.g., Perreault & Bourhis, 1999; Simon & Pettigrew, 1990), future research should examine whether the present results hold for other types of groups with varying levels of group identity.  For example, some research has found that when participants voluntarily choose their group membership, they identify more with their group than those who are randomly assigned to groups (Perreault & Bourhis, 1999).   Other research has found that group members identify more strongly in minority groups than in majority groups (e.g., Simon & Pettigrew, 1990).  Further, another study found that group members with low levels of identification are more likely to acknowledge negative aspects of their group than group members with high levels of identification (Doosje, Branscombe, Spears, & Manstead, 1998).  Thus, examining varying levels of identity with the variables tested in the current study would be an interesting avenue for subsequent research.  In addition, future research might look at different types of group affiliation, such as union membership or demographics (e.g., Lau & Murnighan, 1998).

Other areas for future research include the addition of alternate scenarios.  The present study investigated scenarios in which either both the decision maker and guilty defendants were affiliated with a group or neither was affiliated with a group.  In reality, the decision maker might be affiliated or not, and the guilty/innocent defendant might be affiliated or not.  Also, the current study examined a group of defendants.  Future research could determine if the same results would hold if the defendant were a sole individual.  Further, future research might examine different types of outcomes with various types and sizes of decision-making and defendant groups.  It would be interesting for future investigations to compare the justice perceptions of those going through a justice process themselves versus third parties merely observing the process.

The present study examined only a negative procedure, namely a disciplinary one.  Other organizational procedures and processes should also be examined.  Many organizational procedures are positive in nature, such as giving rewards, bonuses, or promotions.  It is not known whether the findings of the current study will generalize to positive procedures.  As such, future research should also address the influences of affiliation on positive organizational procedures.

[171]
--------------
[172]

The current study used fraternity membership for the affiliation variable.  Many other types of in-group affiliation exist in organizations (e.g., age, sex, race, military experience).  It is not known whether the findings of the present study will generalize to these other types of affiliation.  As the current study demonstrated the potential influences of affiliation on fairness perceptions of third parties, future research should make note of and examine other types of affiliation influences on fairness perceptions.

Implications for Theory

The present study has implications for theory.  Most organizational justice research has focused on the justice evaluations of individuals directly involved in justice situations.  Yet, researchers have begun to acknowledge the role of individuals as observers or third parties in others' justice situations (e.g., Lind, Kray, & Thompson, 1998).  Further, most existing organizational justice research has emphasized reactions to decisions about grievances and other processes, often ignoring the people who deliver or administer the justice.   The present study builds on this research by examining the role that the group affiliation of the decision maker plays in the justice perceptions of third parties.  Future justice research designs should describe and possibly control for who the decision maker, defendants, and third parties are.  In addition, the antecedents reported in the current study should be included in future justice models.

Implications for Practice

The present study also has implications for practice.  The workplace of the new millennium is one of teamwork with ever-growing diversity among team members.  Considerable organizational resources are being devoted to the development of effective teams and identification with one's team is generally viewed as being critical to effective teamwork.  Yet, given the need for integration and cooperation between and within teams, more attention should be given to group dynamics and social identity theory.  The perceptions of third parties would be especially relevant in organizations with work groups whose members have the opportunity to interact, perform similar tasks, and share a supervisor:

To the extent that supervisors can be readily seen implementing organizational procedures, enforcing organizational policies, and acting as deliverers of justice in organizations, individuals sharing a supervisor should hold uniform perceptions on how the group as a whole is treated (Naumann & Bennett, 2000, p. 883).

When justice situations arise where employees perceive salient affiliations (such as age, sex, race, etc.) supervisors should be extra vigilant in their efforts on being fair (e.g., in terms of trust and neutrality) to both in-group and out-group members.  They should pay particular attention to acting as fairness lenses to third parties in instances when it may be perceived that a decision maker and employee defendant are affiliated in some way.  In other words, they should focus on both being fair and looking fair (Greenberg, 1988; Greenberg, Bies, & Eskew, 1991).  The findings of the present study suggest this appears to be especially important in cases where the decision maker is a group of individuals affiliated with the defendant.  In these cases, managers should make a concentrated effort to both be fair and look fair to those in situations as well as to those third parties merely observing situations.

Finally, perhaps one of the more important reasons why third party perceptions are important in today's workplace involves the consequences of unfair fair treatment to both those directly involved and to third parties.  For instance, it has been suggested that after a supervisor treats one group member unfairly, both the poorly treated employee and unaffected others who witnessed or heard about the mistreatment would be expected to exhibit lower levels of cooperation (Naumann & Bennett, 2000).

In sum, the current study suggests that how supervisors treat employees affects not only those directly involved, but also those indirectly involved.  Decision makers should make the effort to consider how their treatment of one subordinate might affect the perceptions of the rest of the subordinates, especially when the subordinate is affiliated with the decision maker.  For example, someone who plays on the company softball team with the boss might be perceived as being given a break.  As such, decision makers should demonstrate how all are treated fairly regardless of affiliations.

[172]
--------------
[173]

Limitations

The present study is not without limitations.  A scenario research design was used, sacrificing external validity for internal validity.  It has been argued that in certain instances, scenario studies are most appropriate precisely because of their hypothetical nature (Lind & Tyler, 1988).  For example, in studies of topics such as cheating where social desirability bias may contaminate results, scenario designs are useful.   Scenario studies are thought to be most fitting for topics related to subjective reactions to procedures (e.g., procedural preferences and attitudes; Lind & Tyler, 1988).  Clearly, the findings of scenario studies are most meaningful when they are confirmed with subsequent studies employing other research methods. 

In fact, some research has found that when the same phenomenon has been subjected to multiple designs (e.g., correlational; scenario), stronger effects occur most frequently in correlational studies (Lind & Tyler, 1988).  Thus, the results of the present study may be viewed as conservative estimates of the true relationships examined.  Given that effects were found for third parties' affiliation, it would be interesting to determine in future research if group affiliation effects are even stronger for individuals going through the procedure themselves rather than reading about it. 

In conclusion, despite these limitations, this study was successful in demonstrating the influences of in-group affiliation on third-party justice perceptions.  Paying careful attention to both being fair and looking fair in those situations in which affiliations are salient may help improve organizational fairness perceptions.

REFERENCES

Bies, R. J., and Shapiro, D. L. (1987).  "Interactional fairness judgements:  The influence of causal accounts."  Social Justice Research, 1: 199-218.

Campbell, D.T. (1965).  "Ethnocentric and other altruistic motives."  In D. Levine (Ed.), Nebraska symposium on motivation.  Lincoln: University of Nebraska Press.

Doosje, B., Branscombe, N. R., Spears, R., and Manstead, A. S. (1998).  "Guilty by association: When one's group has a negative history."  Journal of Personality and Social Psychology, 75: 872-886.

Few, J. K. (1993).  Trial by jury.  New York: American Jury Trial Foundation.

Forbes, M. (1995).  "Juries and jurors: Juries on trial: constitutional right versus judicial burden; an analysis of jury effectiveness and alternative methods for deciding cases."  Oklahoma Law Review, 48/3: 563-592.

Greenberg, J. (1987).  "A taxonomy of organizational justice theories."  Academy of Management Review, 12: 9-22.

Greenberg, J. (1988).  "Cultivating an image of justice: Looking fair on the job."  Academy of Management Executive, 2: 155-157.

Greenberg, J., Bies, R. J., and Eskew, D. E. (1991).  "Establishing fairness in the eye of the beholder: Managing impressions of organizational justice. " In R. A. Giacalone and P. Rosenfeld (Eds.), Applied impression management: How image-making affects managerial decisions (Vol. 135, pp. 111-132). 

Greenberg, J., and Eskew, D. E. (1990).  "The role of role-playing in organizational research."  Paper presented at the meeting of the Academy of Management, San Francisco.

[173]
--------------
[174]

Greenberg, J., and Folger, R. (1988).  Controversial issues in social science research.  New York:  Springer-Verlag.

Hale, J. L. (1987).  "Plagiarism in the classroom."  Communication Research Reports, 4: 66-70.

James, K., and Cropanzano, R. (1990).  "Focus of attention and locus of control as moderators of fraternal justice effects."  Social Justice Research, 4: 169-185.

Kuhl, V. (1997).  "Disparities in judgments of the O.J. Simpson case: A social identity perspective."  Journal of Social Issues, 53: 531-545.

Lau, D., and Murnighan, J. K. (1998).  "Demographic diversity and faultlines: The compositional dynamics of organizational groups."  Academy of Management Review, 23: 325-340.

Lee, Y., and Ottati, V. (1995).  "Perceived in-group homogeneity as a function of group membership salience and stereotype threat."  Personality and Social Psychology Bulletin, 21: 610-619.

Leventhal, G. S. (1976).  "Fairness in social relationships."  In J. W. Thibaut, J. T. Spence, and R. C. Carson (Eds.), Contemporary topics in social psychology (pp. 211-239).  Morristown, NJ:  General Learning Press.

Lind, E. A., Kray, L., and Thompson, L. (1998).  "The social construction of injustice: Fairness judgments in response to own and others' unfair treatment by authorities."  Organizational Behavior and Human Decision Processes, 75: 1-22.

Lind, E. A., and Tyler, T. R. (1988).  The social psychology of procedural justice.  New York:  Plenum.

Marques, J., Abrams, D., Paez, D., and Martinez-Taboada, C. (1998).  "The role of categorization and in-group norms in judgments of groups and their members."  Journal of Personality and Social Psychology, 75: 976-988.

Miles, J. A., and Palmer, T. S. (2001).  "Peer versus authority as decision maker: Are the demographics of the perceiver related to judgments of fairness?"  Psychological Reports, 88: 1107-1118.

Naumann, S. E., and Bennett, N. (2000).  "A case for procedural justice climate: Development and test of a multilevel model."  Academy of Management Journal, 43: 81-889.

Perreault, S., and Bourhis, R. Y. (1999).  "Ethnocentrism, social identification, and discrimination."  Personality and Social Psychology Bulletin, 25: 92-103.

Platow, M. J., O'Connell, A., Shave, R., and Hanning, P. (1995).  "Social evaluations of fair and unfair allocators in interpersonal and intergroup situations."  British Journal of Social Psychology, 34: 363-381.

Roig, M. (1997). "Can undergraduate students determine whether text has been plagiarized?"  Psychological Reports, 47: 113-122.

Sherif, M., and Sherif, C. W. (1969).  Social psychology.  New York: Harper & Row.

[162]
--------------
[163]

Simon, B., and Pettigrew, T. (1990).  "Social identity and perceived group homogeneity: Evidence for the in-group homogeneity effect."  European Journal of Social Psychology, 20: 269-286.

Tajfel, H., and Turner, J. C. (1979).  "An integrative theory of intergroup conflict."  In W. G. Austin and S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33-47).  Monterey, CA:  Brooks/Cole.

Tyler, T. R. (1989).  "The psychology of procedural justice:  A test of the group-value model."  Journal of Personality and Social Psychology, 57: 830-838.

Tyler, T. R. (1994).  "Psychological models of the justice motive:  Antecedents of distributive and procedural justice." Journal of Personality and Social Psychology, 67: 850-863.

Tyler, T. R., Kramer, R. M., and John, O. P. (1999).  The psychology of the social self.  New York: Lawrence Erlbaum Associates, Inc.

Wegner, D. M. (1982).  "Justice and the awareness of social entities."  In J. Greenberg and R. L. Cohen (Eds.), Equity and justice in social behavior. New York: Academic Press.

APPENDIX: NEUTRALITY, TRUST, AND DISTRIBUTIVE JUSTICE ITEMS

Neutrality

1.  To what extent did the Dean of Students make the decision without being biased in any way?

2.  To what extent do you believe the Dean of Students would make the same decision about the case regardless of who the student was?

3.  To what extent did the Dean of Students get the information needed to make a fair decision?

4.  To what extent did the Dean of Students do anything that seemed improper or dishonest?

Trust

1.  To what extent would the Dean of Students act the same way toward you if you were involved in a similar situation?

2.  In your view, how typical was the Dean of Students compared to the Dean of Students at other universities?

Distributive Justice

1.  How fair was the outcome that the student received?

2.  To what extent do you believe that the punishment was UNFAIR to the student?

3.  To what extent did the student receive what he/she deserved?

4.  How much do you feel that the punishment fit the crime?

5.  To what extent do you agree that the Dean of Students made the correct decision regarding the punishment for the student?

[174]
--------------
[175]

AUTHORS' BIOGRAPHIES

Jeffrey A. Miles, Ph.D., received his doctorate from Ohio State University.  He is an associate professor of management and organizational behavior at University of the Pacific.  E-mail: jmiles@uop.edu

Stefanie E. Naumann, Ph.D., received her doctorate from Louisiana State University.  She is an assistant professor of management and organizational behavior at University of the Pacific.  E-mail: snaumann@uop.edu

[175]
--------------
[176]