Journal of Applied Psychology Copyright 1991 by the American Psychological Association, Inc.

1991, Vol. 76, No. 6, 845-855 0021-9010/91/$3.00

Relationship Between Organizational Justice and Organizational
Citizenship Behaviors: Do Fairness Perceptions Influence
Employee Citizenship?
Robert H. Moorman
Department of Management
West Virginia University

Although the study of organizational justice has increased markedly in the past few years, little
work has focused on the relationship between justice perceptions and extrarole behaviors. This
study examined the relationship between perceptions of fairness and organizational citizenship
behaviors in a sample drawn from two firms in the midwestern United States. A theoretical basis
for a relationship between fairness and citizenship was drawn from equity theory and other theories
of social exchange. Structural equation analysis with LISREL 7 found support for four hypotheses,
including support for a relationship between perceptions of procedural justice and four of five
citizenship dimensions. Conversely, perceptions of distributive justice failed to influence any di-
mension of citizenship. Implications for the relationship between procedural justice and citizen-
ship are discussed.

In an article assessing the past, present, and future states of Konovsky (1989), Fryxell and Gordon (1989), and Gordon and
research on organizational justice, Greenberg (1990b) sug- Fryxell (1989).
gested that organizational justice research may potentially ex- One research direction that has been recommended but not
plain many organizational behavior outcome variables. Organi- fully exploited is research on the relationship between justice
zational justice is the term used to describe the role of fairness perceptions and work behavior (Greenberg, 1990b; Lind &
as it directly relates to the workplace. Specifically, organiza- Tyler, 1988). Both early and more recent work on equity theory
tional justice is concerned with the ways in which employees (Adams, 1965; Greenberg, 1988a, 1989) has shown that em-
determine if they have been treated fairly in their jobs and the ployee job performance may increase or decrease in relation to
ways in which those determinations influence other work-re- perceptions of inequitable outcomes. However, because job per-
lated variables. Two sources of organizational justice are rou- formance is often heavily influenced by situational contingen-
tinely cited: distributive justice, which describes the fairness of cies, finding an effect of employee attitudes like perceptions of
the outcomes an employee receives; and procedural justice, fairness has been difficult (Organ, 1977).
which describes the fairness of the procedures used to deter- A more fruitful avenue through which relationships between
mine those outcomes (Folger & Greenberg, 1985). In essence, perceptions of fairness and employee behavior might be found
the belief of researchers who support the value of organiza- includes more nontraditional types of job behavior. These non-
tional justice is that if employees believe they are treated fairly, traditional behaviors are on-the-job behaviors that are not
they will be more likely to hold positive attitudes about their usually captured by traditional job descriptions and thus are
work, their work outcomes, and their supervisors. As evidence more likely to be under personal control (Organ, 1977). One
for the relationship between procedural and distributive justice such example of nontraditional job behavior is organizational
and a variety of organizational variables, Greenberg (1990b) citizenship behavior (OCB). OCBs are defined as work-related
cited studies by Alexander and Ruderman (1987), Folger and behaviors that are discretionary, not related to the formal orga-
nizational reward system, and, in the aggregate, promote the
effective functioning of the organization (Organ, 1988a). A five
An earlier version of this article was presented at the 51st Annual dimensional model of OCB includes altruism, courtesy, sports-
Meeting of the Academy of Management, Miami, Florida, in August manship, conscientiousness, and civic virtue. Organ has sug-
1991. gested that OCB should be considered an important compo-
This article was based on a doctoral dissertation submitted to the nent of job performance because citizenship behaviors are part
Department of Management, Indiana University. I gratefully acknowl- of the spontaneous and innovative behaviors noted by Katz and
edge the helpful comments of Dennis W Organ, Janet P. Near, Philip
Kahn (1966) as being instrumental for effective organizations.
M. Podsakoff, Scott B. MacKenzie, and two anonymous reviewers on
The purpose of this research was to test for relationships
earlier drafts of this article.
Correspondence concerning this article should be sent to Robert H. between organizational justice and organizational citizenship
Moorman, Department of Management, College of Business and Eco- behavior. Specifically, causal modeling was used to assess
nomics, West Virginia University, Morgantown, West Virginia 26506- causal paths from justice perceptions to the five dimensions of
6025. organizational citizenship. Evidence for such causal paths could

causal paths between the dimensions of fairness and job OCB in response to inequity would very likely be safer than satisfaction. Folger and Konovsky Therefore. mal in structure are the ones most likely to be exchanged for tional justice and specific job behaviors. tionship between the two. Given then that job satisfaction may be made up of a large Scholl et al. Studies by Dittrich and Carroll (1979) tween perceptions of fairness and OCB. most work individual to contribute in discretionary fashion without think. Organ (1988a) suggested that OCB could be consid. job satisfaction will not influence the A second reason why perceptions of fairness could be related dimensions of OCB. 1990) provide support for a relationship between per- reasons why fairness could predict citizenship. the other four The basis for Organ's view that perceptions of fairness are resources are proximal in some way to OCB. For exam- rewards are equally exchangeable and established a resource ple. Perceptions of organizational justice will positively influ- between economic and social exchange. fairness was found to be related to a two-factor model perceptions of fairness. would be re. Organ (1988b. causal relationship between procedural and distributive jus- lowing for discretionary. Organ (1988b) proposed that the cogni. when job satisfaction and perceptions of fairness sakoff. cally supported relationship between job satisfaction and OCB Some empirical support exists for the influence of percep- may be better described as one reflecting a relationship be. Because social exchange exists out. each other. why would fairness itself be related to (1990) and the conceptual rationale proposed by Organ (1988a. Therefore. Organ (1988b) believed ence the dimensions of OCB. Cooper. (1987). to OCB originates from Blau's (1964) definition of a difference 2. In addition. 36). leader supportiveness. the studies by Dittrich and Carroll (1979). 553). as one of social exchange. lated to OCB. One final point of interest in this study was the possible side strict contracts. exchange is the relative ease or difficulty of exchanging social However. Organ (1988a) noted that "the latter [to between fairness. 1990) suggested two 1988b. Foa and Foa (1974. On the basis of a review and Scholl. 1988b. Perceptions of organizational justice will positively influ- employee to define his or her relationship with the organization ence job satisfaction. can exchange the social rewards brought on by perceptions of zenship. For example. Adams ceptions of fairness and OCB. which he or she will attempt tice (in the form of distributive justice and procedural justice) to resolve. and OCB. OCB? In his recent work. with organizational justice has not explored any causal rela- ing that this will be acquiescence to exploitation" (p. distributive justice. Those resources that are proxi. but they looked at each fairness source sepa- Central to this idea that citizenship may be part of social rately. MOORMAN then be cited as support for a relationship between organiza. Leventhal (1980) suggested that procedural justice percep- configuration representing the relative likelihood that specific tions influenced subsequent perceptions of distributive justice. Pod- Furthermore. Organ (1988a. Therefore. OCB related to OCB can be found in his reinterpretation of the rela. Even though relatively high correlations have been re- (1988b) wrote "the inherent ambiguity of such a system frees the ported between distributive and procedural justice.846 ROBERT H. fairness component. satisfaction. that fairness perceptions may influence OCB by prompting an 3. The trying to change behavior in line with formal role requirements specific hypotheses were as follows: and. Though fairness was the degree it more cleanly taps cognition] will explain the more measured indirectly from reports of leader contingent reward variance in OCB" (p. Organ tice. 1980) noted that not all social tionship between procedural and distributive justice. Konovsky & Folger (1991) tive component of job satisfaction that appears to be related to presented preliminary evidence for a relationship between pro- OCB probably reflects the influence of perceptions of fairness. if not safer. and dimensions of citizenship behavior. The purpose of this study was to (1965) proposed in equity theory that conditions of unfairness test the relationship between perceptions of organizational jus- will create tension within a person. prosocial acts by the employee. a change in OCB. First. resources might be exchanged. and job satisfaction and OCB were also tested. at least would be directly under personal con. because of Organ's (1988a) suggestion that (1988b) went further by pointing out that changing OCB could the relationship between job satisfaction and OCB may reflect be the strategy of choice because OCB is discretionary and lies instead a relationship between perceptions of fairness and outside of formal role requirements. on OCB are controlled. Konovsky and Folger (1991) and Farh et al. of OCB. When the effects of perceptions of organizational justice trol. some theoretical evidence exists for a causal rela- rewards. 1990) suggested that the empiri. fairness. He wrote . and McKenna (1987) found that percep- of the life satisfaction literature and a review of current job tions of job equity and pay equity were significantly correlated satisfaction measures. Because these two resources are opposite each other in Foa and Foa's (1974. ior.1980) configuration. Causal models con- ered an input for one's equity ratio and that raising or lowering taining paths between dimensions of fairness and OCB were one's level of OCB could be a response to inequity. cedural justice and altruism. with extrarole behavior. 1. Organ tested. appears to be a reasonable and likely way in which an employee tionship found between job satisfaction and organizational citi. and not job satisfaction. and participative leader behav- satisfaction and perceptions of fairness were both measured. tions of fairness on OCB. In summary. The value of OCB is that specific acts of citizenship can be described as examples of either information resources or ser- Relationship Between Justice and Citizenship vice resources. the exchange tends toward ambiguity. This conclusion suggests that. and Organ (1990) specifically studied the relationship are measured together. if job behavior. recent work by Farh. if employees consider themselves in conditions of (1989) tested for differential effects between procedural and social exchange. al. Finally. In addition. they may be more likely to exhibit OCB.

Company B. These studies suggest that separately by asking the supervisors to complete an OCB survey and perceptions of procedural justice can originate from an organization's send it directly to the researcher. procedures and from the way in which those procedures are carried out. Distributive justice was measured with the Dis- Taken together. 1990). employee citizenship behaviors were measured Bies & Moag. Second. 55% of the respondents from Company B were paid an annual salary then the final distribution is likely to be accepted as fair even over $30. It consists of two factors. and 34% were in upper management. All told. the were made. RELATIONSHIP BETWEEN JUSTICE AND CITIZENSHIP 847 An individual readily evaluates the fairness of procedural compo. Each item asks for the degree to which the respondent perceptions of distributive justice. formal proce- medium-sized companies in the midwestern United States. First. effort. 225 matched surveys were used in all the subsequent analyses. Also. sure the degree to which fair procedures are used in the organizations. Method and the scale has shown discriminant validity in relation to job satisfac- Participants and Settings tion and organizational commitment. (1986) reported reliability coefficients ranging from . tion's formal procedures. by sending the surveys through the company mail system and having The second dimension of procedural justice. Company B was and Fetter (1990). All reliabilities reported have been above . believes that he or she is fairly rewarded on the basis of some compari- son with education level. Some of the items included in this scale of employees and asking them to complete a questionnaire containing were based on the work of Folger and Konovsky (1989) and Konovsky the justice and satisfaction scales. and ethicality. the items included in the scale focused on procedures designed to in the study. Therefore. only the exempt by Leventhal (1980) and Leventhal. I developed items to tap the whereas Company B reported an employee response rate of 65% and a fairness perceptions of the interactions that accompanied an organiza- managerial response rate of 55%. . and whether the supervisor dealt with the All in all.99 and con- dents could be considered upper management. Data were collected in Company B and Folger (1989). Distributive justice. Bies. and . was more evenly represented in terms of organizational Organizational citizenship behavior. 36) upper-level managers in the sample from Company B. the final distribution of reward. definitions of the five dimensions of OCB described by Organ (1988a). other hand. These interactions that enacted those formal procedures. 1990). and age. and only 13% of the respon.999. and Fry (1980). Also. Managerial ratings were col. promote consistency. Because of the different data-collection procedures. that were consistent with recent multi- Company A. these findings suggest that a causal path from tributive Justice Index. just procedures prompted an increase in the Measures mean perceived fairness rating of an outcome. 29% were technical Organizational Citizenship Behavior Scale. . The measure of procedural justice was a mea- The sample for this project was drawn from the employees of two sure designed for this study. demographic differences Rothe's (1951) Job Satisfaction Scale. Price and Mueller and the department supervisor level (34%). Tyler & Bies. labeled interactional the respondents mail the completed surveys directly to the researcher. developed by Price and Mueller (1986). Greenberg (1987) found that when outcomes were consid- ered low (unfair). 1986. recently developed six-item scale measures the degree to which re- Therefore. the majority of Podsakoff and MacKenzie (1989). performance. dimensional models of procedural justice (Greenberg. over half (58%) of the respon- nents. (p. 1990b. This procedural justice to distributive justice may be appropriate. to tap the importance of causal Company B completed the first survey. I included items that measured the lected for 140 of the 169 Company A employees and 85 of the 101 degree to which the supervisor adequately explained the decisions that Company B employees. dures and interactional justice. procedural justice existed. This difference appears to reflect the inclusion of more though it may be disadvantageous.000. Company B. pare respondents with nonrespondents in Company B. Karuza. applies paints and other coatings to rolled steel. For exam- and nonexempt salaried employees from the companies were included ple. 169 employees from Company A and 101 employees from employee in a truthful manner. race. was suggested by the work of Bies and colleagues (Bies. two aspects of procedural justice were measured: final sample was 270 employee surveys and 225 surveys for which both the fairness of the formal procedures used. such evaluations affect the perceived fairness of dents from Company A were paid an annual salary less than $29. OCBs were measured with the level: 22% of the respondents were clerical workers.90. recently developed by workers. Moorman. with 41 % of the respondents reporting the com. The sec. Because of the These items originated from the rules of procedural justice developed nature of this study and pending labor negotiations. whether the supervisor consid- gender. accuracy. This scale is a global measure of between the two samples also existed. Items tapping formal procedures were designed to mea- some of which are used in the processes developed by Company A. the response Bies found that the actions taken by managers as they enacted proce- rates from the two companies differed. If the procedures are seen as fair. Although there was no way to com. 1987.78 to . Tyler & ond. When the two companies were combined. a fair procedure was perceived as fair regardless of the outcome level. Company A reported an em. Finally. only 15% of the sample had completed a 4-year degree. justice. on the cluded that this scale has adequate validity and reliability. The first. and the fairness of the employee responses and managerial ratings could be matched. These companies exist in a supplier-client relationship. archival data Items for this factor included questions that focused on the interper- collected by the human resources office did show that the composition sonal behavior of the supervisor. and so forth. Conversely. Work cited in Price and Mueller (1986) attests to the measure's psy- chometric properties. . manufactures paints and other types of coatings. ered the employee's rights. Specific items asked whether the of respondents was similar to the breakdown of total employees by supervisor was considerate and kind. Job satisfaction. MacKenzie. accounts to perceptions of fairness. the majority of the sample job satisfaction that assesses the degree to which respondents agree or from Company A was evenly divided between the clerical level (36%) disagree with a series of evaluative statements. dures and explained decisions were instrumental in determining if ployee response rate of 98% and a managerial response rate of 81%. This survey is a modified version of the sample from Company A had some college education (59%). correctability. Hypothesis 4 was as follows: wards received by employees are perceived to be related to perfor- 4. The items included in this scale were based on the more highly educated. but the measure used and validated by Podsakoff. Job satisfaction was measured with Brayfield and Beyond the difference in response rate. In both companies. Perceptions of procedural justice will positively influence mance inputs. pletion of at least a 4-year degree. repre- Data were collected in Company A by holding meetings with groups sentativeness. bias suppression. Procedural justice.

& Gerbing. formal procedures to job satisfaction. Second. and uct of the square root of the scale reliability and the scale standard civic virtue. This test compares the covariance matrices from each group and determines Scale scores for each OCB dimension. and the best fitting structural model were run with each sample. dural justice to distributive justice. justice and the OCB dimensions were found in both samples. Tucker-Lewis fit index of .94. Organ. ence between the variable-level constructs. which restricted the paths from procedural justice to ture of the citizenship dimensions. Williams (1988) specifically ad. However. This measure asks supervisors to rate the OCBs of subordinates. This approach separates the analysis of a measurement model representing the relationships between individual indicators and latent theoretical definitions of the dimensions of citizenship. using supervisors alone to measure citi. nested model was assessed to show which particular paths described zenship appears to be a reasonable alternative. OCB contains a wide variety of behaviors. Significant changes surement difference between supervisor ratings and self-report rat. nested models were evaluated: Model 2. First. confir. 1989). and Model 5. researchers are often im. was used to test Hypothesis 1. Because the specific factor structures of the variables structural equation modeling was followed (Anderson & Gerbing. The only differences factor analysis with the individual indicators of the three justice di. How. the significance of the individual paths in the best-fitting method variance. First. when relationships between job attitudes and OCB are cance of the restricted paths and their corresponding hypotheses. were yet to be determined. was vide relatively accurate and complete pictures of an employee's OCB. MOORMAN namely. and Burton (1990). altruism. The full structural model determine the fit of a single dimension measuring job satisfaction. Organ & Konovsky. Next. deviation. were in the significance levels of the paths from distributive justice to mensions. the samples model of citizenship. when compared with co-workers. Therefore. The relationships between interactional set the path from the latent variable to the indicator equal to the prod. 3. used to test Hypothesis 2. The first con- best to have citizenship rated by a number of different sources.95. 1983. sisted of comparing the change in chi-square associated with the re- ever.70 for civic virtue to . justice dimension. Brayfield and Rothe's Job and measured variables existed.848 ROBERT H. organizational sources. This technique has been presented by Podsakoff et al. Given this result. it was con- include more than 30 indicators are exceedingly difficult to fit even cluded that all subsequent analyses should be based on a com- with strong theoretical support. and formal pro- To include an adjustment for measurement error in the scale scores. and Finally. However. the relationships found in the nested model. . Confir. Williams and Hazer (1986). Figure 1 shows the first model evaluated (Model 1). supervisors have been the source of choice in the literature justice to OCB. not between single Second.' pelled to specify models that include a mixture of single-item indica- tors and scale scores. through a chi-square test whether one or more sets of relationships and job satisfaction were calculated to generate the covariance exist between the variables. Model 4. bined sample of 225. Such an approach was taken here with the more 1 established measures of job satisfaction and citizenship. and proce- (Bateman & Organ. a two-step approach for confirmatory factor analysis and items. that. variables. and only some Two significance tests were used to assess the structural relation- may be within the purview of the supervisor. sportsmanship. can be tested. The significance of the individual paths showed which specific paths accounted for the signifi- cant change in chi-square and also showed whether the change was Data Analyses positive or negative. with a tion for the error variance by Netemeyer. job satisfaction. was used to test Hypothesis 4. the scale scores were based on the 1988). which restricted the paths from organizational justice to OCB. and Joreskog each factor ranging from . Smith. because two Results companies were used to generate this sample. justice. the firms participating allowed measures of citizenship to be striction of certain paths to zero in a series of nested models (Anderson taken only from supervisors. They reported reliabilities for explained by Kenny (1979). Therefore. Though it is always better to obtain ratings of OCB from a variety of which contains paths from job satisfaction to OCB. organizational justice to job satisfaction. By Structural Equations Modeling definition. distributive justice. A nonsignificant chi-square indicates that matrices for each company used in the two-groups analysis. self-reports may contaminate the relationships with common Second. a two-groups analysis with LISREL 7 was used to determine if the groups were similar Two-Groups Analysis of Differences Between Companies enough to be combined for the subsequent data analyses. in the chi-square from Model 1 indicate general support for the signifi- ings. The result of the two-groups analysis was a chi-square statis- Of issue in a study involving a large number of variables is the limits tic that was not large enough to reject the null hypothesis that imposed by the LISREL algorithm in the number of indicators that one general group accounted for both covariance matrices. four 1990. The data were analyzed in two distinct steps. and these items were then included in the scale were combined for the study. Podsakoff et al. and SOrbom (1989) and has been shown to be a reasonable approxima- matory factor analysis showed evidence for a five-factor model. differences in demographics scores for the citizenship behaviors. p > . it is probably ships between the justice dimensions and citizenship. 1989. which restricted the paths from The supervisor ratings were more likely to distinguish between inrole organizational justice to job satisfaction. was used to test Hypothesis and extrarole behavior and exhibited less variance in the factor struc. conscientiousness. courtesy. all these scale scores were included in an overall confirmatory little real difference was found between samples. supervisors were able to pro. Johnston. he found from job satisfaction to OCB. N= 225) = 52. a single model accounts for the covariance structures within each Scale scores were used because the interest here was the differ- group (Joreskog & S6rbom. 1983). Because the two-groups analysis does not reject the null hypothesis matory factor analysis was used to determine the fit of a five-factor that one population is accounting for the two samples. 1988). Therefore. tested.. which restricted the paths dressed the question of the source of OCB measures. variables from the analysis of the structural paths between the latent and job satisfaction. & Near. Williams found little mea. Joreskog and Sorbom (1986) noted that models that X2(45.85 for altruism. (1990). First.05. The error variance was set equal to the variance of the scale The psychometric properties of the earlier version of this scale were score multiplied by 1 minus the reliability. so it was decided to see if similar Satisfaction Scale was analyzed with confirmatory factor analysis to findings would be obtained in both samples. I cedures to distributive justice. From this saturated model. Model 3.

The reliabilities of the other measures used though the TLI was below the . In each model. Second.16.51 correlations offer partial support for the hypotheses in this (df= 123. In terms of goodness of fit. Below the diagonal are the correla- fit was appropriate. (The path labeled 2 represents the paths restricted to zero in Model 2. indicating Balla. ated. which have been shown in Monte Carlo studies to Table 2 are the reliabilities (Cronbach's alpha) of the three orga- be more resistant to sample size effects (Bentler. and internal reliabilities for Rentier.) Confirmatory Factor Analysis for the Scale Scores fit the data as measured. = distribution. Distrib. then support is indicated for reinstating those paths into the model. The confirmatory mated $ matrix from the LISREL printout. the test statistics used here offered conflicting evidence. 1988). tions calculated from scales scores for all the latent variables. a CFI of .55 (df= 216. if restriction of the paths to zero results in a significant difference in the chi-square. p < . significant cross loadings existed.001). Of special interest in Lewis. study because significant correlations exist between the justice These factor analyses are available on request.90. the path labeled 3 represents the paths restricted to zero in Model 3. were also over the . standard deviations. 1973). x2(233.001. Theoretical relationships between organizational justice. and organizational citizenship behavior and the nested models used to assess the significance of those relationships. the comparative fit index (CFI. Overall Confirmatory Factor Analysis Analyses of Nested Models The results of the overall confirmatory factor analysis are reported in Table 1. This confirmatory model offered evidence Both aggregate relationships in nested models and individual for the convergent and discriminant validity of the nine latent relationships between the variables were tested to determine variables in this study and also assessed whether this full model the significance of paths between organizational justice.96. p < . RELATIONSHIP BETWEEN JUSTICE AND CITIZENSHIP 849 Organizational Organizational Justice Citizenship Behavior Altruism Courtesy Sportsmanship Conscientiousness Civic Virtue Figure 1. job satisfaction. the interest in maintaining a content-valid (1978). Results suggesting convergent and discriminant validity showed that all the indicators had signifi- Confirmatory factor analysis was first used to determine cant loadings on their hypothesized latent variables and that no which items should contribute to the scale scores in the mea. Satis. The chi-square for this satisfaction. job . and the path labeled 5 represents the paths restricted to zero in Model 5. Marsh. = interactive. strong reliability. respectively. N = 225) = 554. Tucker & the scale scores are reported in Table 2. the justice measures and job satisfaction. measurement model was 320. were . All three were over . However.93. N= 225. a unidimensional model for Brayfield and Rothe's Above the diagonal are the correlations as reported in the esti- (1951) Job Satisfaction Scale was evaluated. p < . Al. and a TLI of .001). the path labeled 4 represents the paths restricted to zero in Model 4.90 and . Goodness of fit was indi- surement model representing the citizenship behaviors and job cated by a CFI of .70 minimum established by Nunnally old for a good fit.97 and a TLI of . 1990. Descriptive Statistics surement model indicated a poor fit.90 level considered the thresh. and Interact. measures and the citizenship dimensions. The means. a five-factor model for citizenship was evalu. = satisfaction.91. 1990) and the Tucker-Lewis index (TLI. First. nizational justice scales. & MacDonald. and procedural justice and distributive justice. The chi-square score for the mea. These two sets of factor analysis of all 18 items resulted in a chi-square of 222. Table 2 also reports two different correlations between measure of citizenship suggested that accepting this marginal the constructs in this study.88. N= 225.

The first model cance of the individual paths suggests that this relationship in Table 3 is the measurement model. .87 model. Model 5 was evaluated to test Hypothesis 4: Percep- Your supervisor took steps to deal with you in a truthful tions of procedural justice will influence perceptions of distrib- manner. . have all sides affected by the decision represented.01 level. N = 225) = eter estimates for Model 2 are presented because this model was 320. hear the concerns of all those affected by the decision. provide opportunities to appeal or challenge the decision. justice perceptions and four of the five OCB dimensions.93 ses were also tested by evaluating the individual paths in the Conscientiousness . . . . This model resulted in a change in chi-square and degrees of freedom were the same for this model chi-square of 19.91 Civic virtue . the signifi- nested models analysis are reported in Table 3. Thus. which was not decisions. With the best fitting of the nested models. The results in Table 4 aid interpretation of the results of the nested-models comparison. Interactional justice Model 4 was evaluated to test Hypothesis 3: Perceptions of Your supervisor considered your viewpoint.05 level. . and both di- cedural justice because of the relatively high correlations be. and organizational citizenship. . .67 significant at the p = .69.42 for a change of 3 degrees Your supervisor treated you with kindness and of freedom. The change in chi-square resulting from the re- Distributive justice striction of the paths from procedural justice to distributive Fairly rewarded considering the responsibilities. which .90 paths from justice to citizenship resulted in a change of chi- . . allow for requests for clarification or additional was significant at the p = . .01 level. the param- Note.01 level. x2 (216. . . offering support for Fairly rewarded for the amount of effort you put forth. . generate standards so that decisions could be made organizational justice will positively influence organizational with consistency. Fairly rewarded for the work you have done well. general support was found . and Citizenship that the paths from job satisfaction to the citizenship dimen- sions were restricted to zero. . Conversely. The Your supervisor provided you with timely feedback about change in chi-square resulting from the restriction of the jus- the decision and its implications. Job Satisfaction. and distributive justice and formal procedures were not justice and job satisfaction. the restriction to zero of the . . . supporting Hypothesis 3. .48. Model 2 differed from Model 1 in Justice. For example.89 square (for a change of 15 degrees of freedom) of 41.91 Hypothesis 4. . provide useful feedback regarding the decision and its implementation. Correlations were maintained between the five no need to reinterpret Hypotheses 3 and 4. . The second model is Model 1.80 for Hypothesis 1. .77 organizational justice will influence job satisfaction. the individual path analysis shows tive justice. N= 225) = 4575. . This have. . and procedural justice and distribu.90 Brayfield and Rothe's (1951) Job Satisfaction Scale . the change in chi-square for .90 Courtesy . actional justice predicted all the OCB dimensions but civic tion and OCB. . x2 (276. MOORMAN Table 1 will not be related to citizenship behavior when perceptions of Overall Confirmatory Factor Analysis Model for Procedural fairness are also measured.73 tice to job satisfaction paths was 57.86 . Your supervisor showed concern for your rights as an employee. which sub.81 utive justice. lending support Model 2 was evaluated to test Hypothesis 1: Job satisfaction for the lack of significance of the excluded paths. The results for the ceptions of organizational justice and citizenship.84 Model 3 was evaluated to test Hypothesis 2: Perceptions of . the change in chi- Item A square between Model 1 and Model 2 reflects the effect of re- Formal procedures moving those paths and thus is a test of their significance to the Procedures designed t o . which allows all the latent may be best explained as a relationship between interactional variables to correlate. perceptions of virtue. . . although support was found in the nested-models test for a relationship between per- satisfaction. Tucker-Lewis index = . . .86 found for Hypothesis 2. . As can be seen in Table 3.05 level.850 ROBERT H. the null model. organizational justice influenced job satisfaction.88 for a change of 2 degrees of freedom.93 Organizational citizenship behaviors Significance of Individual Paths Altruism .92 Fairly rewarded for the stresses and strains of your job. Thus. containing only the significant paths in Model 2 was evaluated tional paths were measured between all latent variables. In Table 4. Therefore. . the and compared with Model 1. . model.82 change was significant at the p = . the hypothe- Sportsmanship . . . These tests allow for the determination of the direction of the effects as well as their significance.87 citizenship. . collect accurate information necessary for making the change in 5 degrees of freedom was 9. general support was information about the decision. .84 Your supervisor was able to suppress personal biases.55. comparative fit index = .97. Because either causal or correla. Distributive Justice. All dimensions of OCB dimensions and also between the two dimensions of pro. This and the measurement model.87 Finally. This change in chi-square was significant at the p = consideration. mensions of procedural justice influenced distributive justice.07 for a change of 16 degrees of freedom.96. tween them and because of the lack of theoretical rational for As a final check on the significance of the findings. directly related. As reported in Table 3. change was not significant at the p = . . Inter- stituted structural paths for correlations between job satisfac.94 Because of the results of the model comparisons.85 Fairly rewarded in view of the amount of experience you justice was 97. . a model the presence of causal paths. perceptions of justice and OCB. Results are completely standardized.55.

Conscientiousness 5.11 . In this case.57** (.16* .50** .94) . Even though a satisfaction measures include job fairness.42 .17* .15* .88) . The follow.93) . formal to citizenship. Job satisfaction 3.14* . the employee may be less likely to believe that citizenship be- 1990) stated that.17* .55. procedures.11 4. This finding equation modeling. 1988b.97 219 57.29** .62 1. Estimated $ matrix is above the diagonal.95 Null model 4. Formal procedures 3.42* .45 .42** . been treated fairly by the organization.34** .18* .97 Model 3: Justice -* OCB 362.37** . support was found for the four hypotheses provided support for Organ's (1988b.93* Note. Difference scores were taken from x2 (216. RELATIONSHIP BETWEEN JUSTICE AND CITIZENSHIP 851 Table 2 Means.55** .43** .94) . job fairness measures may subject to exploitation. In essence. CFI = comparative fit index.32** .10 231 41.48 1.48 276 4.96 Model 4: Justice -* job satisfaction 377.87) . OCB = organizational citizenship behavior. a closer examination of the fairness to OCB link between job satisfaction and OCB reported in the literature suggests that the type of fairness perception may have been may be spurious and merely reflects the degree to which job important in predicting the occurrence of OCB.41** 9.82 .42** .96 . If treated fairly.43 218 97.95 .14* . because job satisfaction measures appear to havior outside his or her prescribed role is inappropriate and include a large fairness component.69 .05 3. Correlations.57 0.08 5.16* . *p<.22** .96 .55 216 . Courtesy 5. Though most work with citizenship one of economic exchange to social exchange. interactional Table 3 Nested Model Comparisons Model x2 df X2 change TLI CFI Measurement model 320.62** (.07 .66** . has included job satisfaction as a cause.20* . whereas employees who believe they are fairly treated will see Job Satisfaction and Organizational Citizenship continued citizenship as a reasonable contribution to the sys- tem.45 1. 1990) view that the deci- pertaining to how perceptions of organizational justice influ.62** .15 .22** .09 .18* .23** .17* .14* .96 .55* . sion to behave as an organizational citizen may be a function of ence an employee's reported citizenship behavior. and Reliabilities for the Combined Sample Variable M SD 1 1. TLI = Tucker-Lewis index. Distributive justice 3.575. Through structural perceptions of organizational justice and OCB. Support for this view general relation was found.32** . and interactional justice.08 . Altruism 5.55 216 . Organ (1988a.64** (. **p<.32** .24** . Discussion Perceptions of Fairness and Organizational Citizenship The purpose of this research was to examine the relationship Support was also found for a causal relationship between between perceptions of fairness and OCBs.01.254.37 . Sportsmanship 4.38** . be more strongly related to OCB.75 1.24 221 9.68 0.96 .97 Model 2: Job satisfaction -*• OCB 330.38** . .83) .23** .70 0.10 .50** .16* (.59** . Reliabilities are reported along the diagonal.41** (.88* .17* . the degree to which an employee believes that he or she has ing is a discussion of the primary findings of this study.50** (.29** . the relationship However.05. N= 225) = 320. Interactive justice 3.96 .14* . Civic virtue 5. This finding is also consistent with Organ's view that fair- Hypotheses 1 and 3 were included in this study to test ness may influence citizenship by allowing for a redefinition of whether job satisfaction is actually a primary antecedent of the exchange between the organization and the employee from organizational citizenship.08 .49** .97 Model 1 : Saturated model with all theoretical paths 320.18* 2.22** 8.45** (.57** .22** .96 Model 5: Procedural justice -»• distributive justice 418.50** 6.76) Note.82 .08 .58** .43** .42** 7.81) .43** .37** (.60** . Standard Deviations.58** .94 .42 0. When perceptions of fairness were measured between the three dimensions of fairness and OCB resulted in separately from job satisfaction.49 (.05 .18* . analyses of the individual relations was found here. Scale correlations are below the diagonal.15* .22** . This result is consistent with equity theory in that employees who perceive unfairness may reduce the frequency or magnitude of their citizenship. job satisfaction was not related differential effects attributable to distributive justice.86) .88 1.58** .

The focus of this scale was on the degree to ship.481 Interactive justice -*• job satisfaction .005 0.49 . MOORMAN Table 4 Parameter Estimates for the Paths in Model 2 Path description Unstandardized /3 Standardized /3 Distributive justice -*• job satisfaction .86 . fairness of the procedures used to determine outcomes may rise fessional jobs in which interactions with superiors were fre.50 . it seems reasonable to suggest that. these results add to the growing realization teractions with their supervisors communicated more informa. In their study.071 -0. an employee could easily be. which the behavior of the supervisor enacted the formal proce.477** 2.30 -.84 .493** 5. perceptions of fairness should include assessments of the con- Similar value could be communicated through formal proce. employee theft could be considered a negative form of citizen- tive justice.290 Interactive justice -*• conscientiousness .03 .69 .01. First. a study by Greenberg Differential Effects of Interactional and Procedural Justice (1990a) on the incidence of employee theft found that employee theft increased with pay cuts.006 Formal procedures -*• civic virtue .034 -0.93 -.422 Interactive justice -»• sportsmanship .103 Formal procedures -*• sportsmanship -. effective and compelling communicator of an employee's value (e.530** 3. Differential Effects of Procedural and Distributive Justice Three recent studies can be cited as support for the impor. Perceptions of the absence of fair procedures. employees' impressions of the fairness of their in. Thus.118 -1.092 Distributive justice -* conscientiousness .96 .05.g.008 0. First. interactional justice was relate to citizenship behaviors and appeared to mediate a rela- defined as the fairness of the manner in which the procedures tionship between transformational leader behaviors and citizen- were carried out. Taken together. support for the importance of interactions in an em- organization.060 -0.080 Distributive justice -*• courtesy .139** 2.194 Interactive justice -»• courtesy .089 Formal procedures -*• conscientiousness . a study by Greenberg (1988b) on the impor- to organizational citizenship. employees who believed tance of managing impressions of fairness found that supervi- that their supervisor personally treated them fairly appeared to sors were more likely to be seen as fair if they actively communi- be more likely to exhibit citizenship behaviors.103 0.148 1.406** 2.79 . ence an appraisal of supervisor trust because it focuses on the Relatively speaking. Of the three sources of fairness tested in the present study.87 . In terms of the differential relationship between procedural tance of supervisor interactions over the presence of formal justice and distributive justice. that the interpersonal context of procedural fairness is a potent tion to them regarding trust and equity than did the presence or source of influence (Tyler & Bies. The focus of the items in this scale was on the ployee's decision to be a good citizen was reported by Podsakoff organization as a whole and the degree to which fair procedures et al.35 . as well as fair procedures and fair dures. future studies of the influence of lieve that the organization considered him or her important.107** 2.079 Distributive justice -* civic virtue -.. this sample.082 Interactive justice -» distributive justice . **p<. dures are enacted.246 Interactive justice -*• altruism . similar support for the larger . 1990).173 Formal procedures -»• job satisfaction . tribution of fair interactions.69 .09 -. The jobs surveyed were mostly pro.134 -0.60 -.195 Distributive justice -*• altruism . actions speak louder than words).69 .004 Distributive justice -»• sportsmanship .852 ROBERT H.036 */><.151 Formal procedures -*• altruism -. (1990). interactional justice appears to be the one most likely to influ- dures in a fair manner.07 . decisions were made by supervisors. Through such interactions. In comparison. in actions of the supervisor specifically. Second.306 Interactive justice -»• civic virtue -.162* 2.072 0. The results of the present tween interactional justice and formal procedures and in the study are consistent with Greenberg's (1990a) study because differences between procedural justice in general and distribu.20 .078* 1. Therefore. formal procedures were denned here as the ship behavior. cated that fairness through interactions rather than merely rely- ing on actual fair behavior.061 Formal procedures -» courtesy -.66 -. but the actions of the supervisor are probably the most outcomes. justice was the only dimension of fairness to significantly relate procedures. trust in leadership was found to were at least present.099 0.78 . or fall depending only on the manner in which those proce- quent.065 Formal procedures -*• distributive justice . degree to which fair procedures were present and used in the Finally. but that the theft rates were re- Reasons for why interactional justice was the only source of duced when thorough and sensitive explanations of the pay-cut justice found to relate to OCB may lie in the differences be.245* 1.

LISREL is perhaps the most sophisticated lated to interactional justice because the choice to be a citizen is method for making causal inferences. The strongest implication of this study is that supervisors tem. the dynamics through which fairness perceptions render OCB tional study. a second way to interpret this could be to say that signs are never more than inferences. In bution could produce a halo effect so that organization proce- studies comparing the two perceptions of fairness. It was suggested here that citizenship may be re- lytic technique. the First. Citizen- interactional justice). the behaviors. manager's interest to be fair. future research should be concerned with studying limit confidence in the findings. to evaluations of the specific outcomes in question. (1990). The study was a cross-sec. fairly by their supervisors and therefore performed citizenship tice perceptions predisposed employees to perform citizenship behaviors that would benefit the supervisors. The organization policy and not a manager's intentions. 1989). the fairness of the interactions between managers and em- in a city in the midwestern United States. Similarly. whereas distributive justice appears to be related cedure was explained by Podsakoffet al. 1988a). Perhaps the relationships found here would be dif. vertical dyad linkage model of leadership (Dansereau. These vari- (Folger & Konovsky. the OCBs measured were prob- opposite direction. yet causal relationships were inferred. Therefore. strained by forces outside the manager's control. Such findings prompted Lind and Tyler (1988) to conclude be used to overcome this by modeling a method factor on which that procedural justice appears to be related to more general all indicators presented to the employees would load. By compari- two companies were both involved with the chemical industry son. though coming from two companies. mal procedures. 1975). In practice. interactional fairness was important because the employee was An example of this limitation is that it would be possible to deciding to exhibit behaviors that would benefit the supervisor. response bias on the part of the respondent (because supervi- ported that procedural justice better predicted organizational sors completed the OCB measure. LISREL can tice. 1983). one must al. organiza. the analytical techniques. a measure of the fairness of supervision. among their employees. whereas procedural justice was re. 1975. both ployees is often a matter of a manager's being sensitive to the companies did not appear to have a strong problem with a lack interests of the employees and convincing them that it is in the of fairness. ways note that causal inferences made from cross-sectional de. justice and distributive justice were found to predict different Another limitation is that the relationships between the per- attitudes. As an ana. the methods used to gather and analyze the data and OCB. fined by the target of the citizenship behaviors. several limitations deserve mention. not a concern for those relationships). institutions. the nature of the supervisor's behavior while enacting those sory commitment (Gregersen. In essence. Poor distribution could prompt employees to Citizenship directed toward co-workers should then be related . However. but because subordinates who were rated highly on decision to perform OCB could be the function of the evalua- citizenship were members of the in-group characterized by the tion of the fairness received from the target of that OCB. interpret the direction of causality between constructs in the Because supervisors rated OCB. 1990b). For example. and authorities evoked by procedural justice can directly influence employees' citizenship behaviors. Citizenship group might be rated by managers more positively (high OCB) directed toward the supervisor should be related to interac- and. one would need to believe that the for Future Research decision to behave as an organizational citizen was more a re- sult of a general positive evaluation of the organizational sys. is presence or absence of fair procedures may be a function of unique enough to prompt concerns over generalizability. 1989). might rate their managers more positively (high tional justice. Because of the relationship between perceptions of fairness sive. Distribution of outcomes may be con- measures in this study. 1986). Folger and Konovsky also re. common method bias was commitment and trust in supervision than did distributive jus. Graen & Cashman. A similar case could be made for the di. they should work to increase the fair- ness of their interactions with employees. appropriate. If managers want to increase citizenship behavior be supported. so any relationship that existed could be attributed to a authorities (Lind & Tyler. Graen. ables were measured from one source (the employee) at one lated to evaluations of organizational systems. this conclusion appears to procedures. RELATIONSHIP BETWEEN JUSTICE AND CITIZENSHIP 853 relationship found between procedural justice and OCB has question an organization's procedures. dural justice. whereas a positive distri- been reported by a number of studies (cf. institution. Members of the in. based on a more general evaluation of the working atmosphere. However. interactional justice may be ably directed toward those supervisors. the causes of citizenship behavior could be de- & Haga. Employees were treated related to the OCB dimensions not because interactional jus. such as pay satisfaction job satisfaction included common method variance. In addition. This pro- evaluations. dures that were designed to promote fairness and was based on tional commitment (O'Reilly & Chatman. procedural dures were seen as fair (Organ. and the fairness to manage. 1988). ship directed toward the organization should be related to for- rectionality of the relationship between distributive and proce. and time. in turn. To say that the results of this study are consistent with the Managerial Implications and Directions results discussed above. this sample. ferent in a company in which a lack of fairness was more perva. Distributive justice predicted attitudes that related ceptions of fairness and between the perceptions of fairness and directly to the outcome in question. Limitations A corollary of this implication is that perceptions of fairness based on interactional justice may be the easiest perceptions of Because of the subjects. a measure of the fairness of the organization. Second. Greenberg. The than an evaluation of the fairness of specific outcomes. and supervi. Given perception of fairness that originated from interactional justice that OCB and extrarole behavior have been found to be related was based on whether the supervisor correctly used the proce- to general job satisfaction (Bateman & Organ.

Comparative fit indexes in structural models. S. J. Unpub- linkage approach to leadership within formal organizations: A lon. L. Cummings & B. H. M. Kent. R. Multiple commitments at work and their rela- Bateman. 2.. CT: JAI gram and applications (2nd ed. (1979). U. & Gordon. (1964). F. E. structural relationships by maximum likelihood. Tulane Univer- gitudinal investigation of the role making process. G. R. F. & M. (1990). did not support Folger. (1974). A. Greenwich. (1991). A role making model of leadership should be aware of the benefits of behaving toward subordi. Academy of Management Executive. H. G. B. D. perceptions of that treatment could affect the occurrence of Greenberg. Structural equation modeling 75. the target of behavior. B. Larson (Eds. man (Eds. Willis (Eds. Social exchange: over rates. Managers should be con. (1989). Bies. An index of job satisfaction. 7. S. 267-299). Kenny. (1987). Resource theory: Interpersonal behav- represent citizenship behavior directed toward individuals. R. R. criteria of fairness.). & Cashman. J. Cognitive reevaluation of outcomes in response Diego. Karuza. (1989). B. Press. 2. M. instrumental vari- Psychological Bulletin. 289-319). R. S. M. L. Journal of Applied Psychology. F. Social exchange: Advances in theory and research (pp. & Fryxell. however. IL: Thomas. A.. R. 238-246. 32. E. & Sorbom. particularly Gordon.). ables and least squares methods (4th ed. 993-1009. Leventhal. (1988a). S. San Greenberg. Research on negotiations in organizations (Vol. In K. In L. J. M. inequity: The hidden cost of pay cuts. New \brk: Plenum 29-40. A. In G. S. 307-311. J. The effects of procedure and Dansereau. 32. E. J. 46-78. Gergen. D. 73. (1980). 167-218).. D. M. J. New \brk: Wiley. Mooresville. J. Greenwich. in practice: A review and recommended two-step approach. Greenberg. Psycho. Brayfield. (1988b). & Organ. Accounting for organi. Ferris sion to behave as a good citizen depends on the target of that (Eds.). & Ruderman. J. managers Graen. 399-432. social interaction (pp. 24. (1975). distributions: Do the means justify the ends? Journal of Applied Psy- chology. & SSrbom. CT: JAI Press. J. Press.. B. OCBO to represent citizenship behavior directed toward the New \brk: Plenum Press. M. Folger. 1. Reactions to procedural injustice in payment citizenship behaviors.. In R. J. J. Fryxell. & Greenberg. Greenberg. G. Unpublished manuscript A. & Folger.. Correlation and causality. K. Greenberg. (1989). 107. 72. M. Freeman School of Business.. 77-94). (1989). Academy of Manage- Conclusions ment Journal. S. Greenberg. lished manuscript. G. 19. W R. H. Justice and satisfaction. tions. Journal of Applied Social Psychology. Equity and workplace status: A field experi- ment. 561-568. M. H. in formal organizations: A developmental approach. and logical Bulletin. & Fry.. (1986). Mikula (Ed. & Foa. J. and social accounts in mitigating the negative Blau. G. (1989). Cultivating an image of justice: Looking fair on Adams. K. R.). tomorrow. M. Berkowitz (Ed. E. Anderson. In L. Konovsky. H. Research in Joreskog.. P. J. L. J. In J. H. L.. Rowland & G. & R. 103. The effectiveness ofattributional 43-55). Leadership frontiers (pp. CT: JAI Press. Gregersen. Inequity in social exchange. S. A. pp. Greenwich. Interactional justice: Communication New York: Wiley.. 16. Behavior and Human Performance.). pp. B. approaches to the study of fairness in social relationships. cues. (1966). & Moag. W (1988). & Foa. Procedural justice: An interpretive this distinction.. He suggested two dimensions: OCBI to Foa.854 ROBERT H. & Konovsky. S. & Folger. In K. predicting the occurrence of citizenship. New York: Springer-Verlag. & Rothe. (1986). S. A. Social Justice Re. References Greenberg. Academy of Management Journal. 9. 851-866. Exchange and power in social life. (1975).). Joreskog. Williams (1988) Foa.. R. 32. are instrumental in tice. Job satisfaction and the good tionships with extra-role behavior. the job. Spring- found that OCB items could be factor analyzed on the basis of field. Jr. Organizational justice: Yesterday. Beyond fairness: A zational citizenship behavior: Leader fairness and task scope versus theory of allocation preferences.155-157. G. D. Konovsky. (1985). H. J. pp. T. (1989). C. & Haga. J. 27-55). Freeman School of Business. G. Lewicki. Journal of Applied Psychology. W (1990). Advances in experimental psychology (Vol. Journal of Management. (1987). 115-130. pp. 411-423. Research in personnel and human resource management (Vol. U. (1990b). P. 35. Sheppard. Farh.." sylvania State University. A vertical dyad distributive justice on organizational citizenship behavior. distributive justice in organizational behavior. W (1983). (1979). & Organ. procedural justice. Tulane University. 141-183). IN: Scientific Bies. & R. Journal of Management. analysis of personnel systems. employee job satisfaction. CA: Academic Press. (1989). M. Advances in theory and research (pp. and ior as exchange.177-198.. OH: cerned with how they treat their employees because employees' Kent State University Press. Leventhal. The predicament of injustice: The management of Software. Unpublished manuscript. & Carroll. M. Hunt & nates in a manner perceived as fair. 16. P. ment. Employee theft as a reaction to underpayment search. 13. Graen. consequences of involuntary job loss. Academy of Management Journal 26. Bazer.). G. and departmental absence and turn.). B. W J. A. (1980). Greenberg. 606-613. Societal structures of the mind. Chicago: SPSSX Inc. What should be done with equity theory? New Dittrich. & Kahn. The social psychology of organizations. today. D. E. Staw (Eds. organization. . Organizational Behavior and Human Performance. 143-166). E. LISREL IV: Analysis of linear Bentler.). His substantive tests. 587-595..). LISREL 7: A guide to the pro- organizational behavior (Vol. New "Vfork: Wiley. J. Willis (Eds. G.. MOORMAN to the fairness of treatment by co-workers. Alexander. The role of procedural and 174-184. Voluntariness of association as a moderator of the importance of procedural and distributive jus- those derived from interactional justice. to underpayment inequity. 705-722. (1965). Future research should assess whether the deci. zens. S. D.. 3. E. D. (1980). Katz. G. In K. Therefore. Penn- soldier: The relationship between affect and employee "citizenship. Effects of procedural and distrib- utive justice of reactions to pay raise decisions. Podsakoff. G. Journal of Applied Psychology. (1951). The results indicate that fairness perceptions. Workplace justice and job This research attempted to show that perceptions of fairness satisfaction as predictors of satisfaction with union and manage- influence employees' decisions to behave as organizational citi. M. (1990a). R. Gergen. & Gerbing. M. Academy of Management Journal. Organizational equity percep. 55-61. behavior. Organizational sity. moral outrage. (1987).

M. & Tyler.. Journal of Applied ogy. 547-557. Carroll (Ed. L. S. Unpublished manuscript. C. Marsh. D. Transformational leader behaviors and their effects on trust. Beyond formal procedures: The inter- view. 14. C.. satisfaction and organizational commitment as determinants of orga- Organ. Psychometric theory. A.1991 • . H. Moorman.1991 Indiana University. Tucker L.157-164. J. J.. B. & Chatman. R. chology. New York: McGraw-Hill. M. R. and internalization on prosocial behavior. D. mum likelihood factor analysis. & Lewis. The reliability coefficient for maxi- Organ.. T. J. 38. L. Lexington. J. W (1986). New York: Plenum Press. nizational citizenship and in-role behaviors. In B. 75.1-10. organizational behavior (Vol. P. pp. M. & MacKenzie. MA: Pittman. (1987). & Bies. R. MA: Lexington Books. 74. Goodness-of-fit satisfaction. R.. & Konovsky. Hillsdale. Antecedents and consequences of Press. 2. Journal of Management.. W (1988b). 43-72). Organizational commitment and Smith. tion causes performance hypothesis.. Psychometrika. J.. & Burton. Journal of Ap. M. (1978). 653-663. (1989). E. 12. C. Affective and nonaffective components of job hypothesis. in determining equity perceptions: Differential effects on behav- Nunnally. C. Bloomington. Bloomington. ship behavior: Its nature and antecedents. (1988).113-124. F. A. (1973). ioral and attitudinal outcomes.. (1986). Analysis of role surement. O'Reilly. 77-98). 46-53. D. personal context of procedural justice. Handbook of organizational mea- Netemeyer. Staw & L. & McKenna. identifica. Received December 10. ship Quarterly. 103. J. P. S. D. (1983). baum. R. Organizational citizenship behavior: The good social psychology in business settings (pp. Johnston. Psychological Bulletin. 71. Unpublished doctoral ship behavior. R. (1989). S. Academy of Management Re. Price. W (1977). W. C. The motivational basis of organizational citizen. Organ. Scholl. J. 71. Research in dissertation. Greenwich. T.1990 sure of organizational citizenship behavior.).. E. CT: JAI Williams. P. Indiana University. A reappraisal and reinterpretation of the satisfac. (1990).. justice. A restatement of the satisfaction-performance Williams. Tyler. & Hazer. R. and organizational citizenship behaviors. 40. W (1988a). The Leader- indexes in confirmatory factor analysis: The effect of sample size. NJ: Erl- soldier syndrome. 68. plied Psychology. H. Balla. Cooper. 107-142. Podsakoff. P. Revision received July 1. R. A. & MacDonald. Cognitive versus affective latent variable structural equation methods. Psychology. (1990). B. J. W (1990).. W. (1988). MacKenzie. & Mueller. S. D. 219-231. Applied Organ.. Journal of Applied Psychol- tion.148-157. W. Organizational citizen- psychological attachment: The effects of compliance. Journal of Applied Psy- determinants of organizational citizenship behavior. In J. (1990). M.. D. satisfaction and commitment in turnover models: A reanalysis using Organ. Marshfield. Organ. 492-499. L. L. conflict and role ambiguity in a structural equations framework. (1986). T. & Near. A. J. RELATIONSHIP BETWEEN JUSTICE AND CITIZENSHIP 855 Lind. Personnel Psychology. 1. A second generation mea. Referent selection Journal of Applied Psychology. Cummings (Eds. Accepted July 2. J. & Fetter. W. (1988). R.). 391-410. R. The social psychology of procedural Podsakoff..