JOURNAL OF ORGANIZATIONAL BEHAVIOR, VOL. 19, 131±141 (1998)
A multilevel analysis of procedural justice context KEVIN W. MOSSHOLDER 1 , NATHAN BENNETT 1 AND CHRISTOPHER L. MARTIN 2 1
Department of Management, Louisiana State University, U.S.A. Department of Management and Marketing, Louisiana State University in Shreveport, U.S.A.
2
Summary
Using a multilevel framework, we hypothesized that both employee perceptions of procedural justice and a work unit level measure of procedural justice context would be associated with employee reports of job satisfaction and organizational commitment. Three hundred and twenty-three employees from 53 branches of a ®nancial services organization were used to test this hypothesis. Hierarchical linear modeling analyses revealed that procedural justice context explained variance in employee job satisfaction beyond that accounted for by individual perceptions of procedural justice. With regard to organizational commitment, this hypothesis was not supported. The results are discussed in connection with contextual aspects of procedural justice. # 1998 John Wiley & Sons, Ltd. J. Organiz. Behav. 19: 131±141 (1998)
Introduction In the decades since Thibaut and Walker's (1975) treatise, researchers in several disciplines have found that procedural justice concerns extend to a variety of settings. Evidence shows that perceptions of justice are in¯uenced not only by the outcomes individuals receive, but also the procedures through which outcomes have been determined (Cropanzano and Greenberg, in press). Individuals tend to perceive greater procedural justice when they believe they have had the chance to participate in the decision-making process and can ascertain that organizational authorities have been neutral and unbiased (Tyler, 1990). The enactment of procedures also aects justice perceptions: treating people with respect, communicating in a trustful manner, and suciently justifying decisions reinforces their sense of justice (Bies, 1987; Tyler and Bies, 1990). Some researchers have also suggested that the nature of organizational treatment has implications for employees' identi®cation with their co-workers (Lind and Tyler, 1988).
Addressee for correspondence: Kevin W. Mossholder, Department of Management, Louisiana State University, Baton Rouge, LA 70803-6312. E-mail:
[email protected]. The authors would like to thank Robert Bies and Mark Gavin for comments on an earlier draft of this paper.
CCC 0894±3796/98/020131±11$17.50 # 1998 John Wiley & Sons, Ltd.
Received 16 July 1996 Accepted 17 February 1997
132
K. W. MOSSHOLDER, N. BENNETT AND C. L. MARTIN
Given its roots in social exchange theory (Leventhal, Karuza and Fry, 1980; Thibaut and Walker, 1975), it is not surprising that procedural justice has been conceptualized and examined at the individual level of analysis. More recently, however, some justice researchers have recognized the in¯uence of normative (e.g. Greenberg, 1990a) and contextual (Cropanzano and Greenberg, in press) elements on justice dynamics. In light of such recognition, it is puzzling that there has been no systematic eort to understand procedural justice as anything other than an individual level phenomenon. Therefore, the present study examined whether procedural justice is tenable as a contextual variable and whether procedural justice context is associated with individual work attitudes. After presenting a rationale for considering procedural justice as a contextual variable, we use a multilevel design to investigate whether procedural justice has implications as both an individual and a contextual level variable.
The Case for Procedural Justice Context By their very nature, organizations involve multiple levels of nested relationships. As a result, many organizational constructs may operate at more than one level (Klein, Dansereau and Hall, 1994). The issue of levels has only been partially broached with regard to procedural justice. To this point, the literature has generally acknowledged that antecedents of individuals' procedural justice perceptions may be found at many levels. For example, Greenberg (1990a) states that it is logical that aspects of the work environment may in¯uence perceptions of fairness. Sheppard, Lewicki and Minton (1992) emphasize that when organizations are seen as comprised of groups pursuing multiple interests, a central problem is identifying procedures for deciding which interests to include and how they should be included. Expectations established by culture and local norms may aect fairness perceptions and reactions to these perceptions; general determinants of procedural justice may come to life by being altered to ®t in their particular context (Cropanzano and Greenberg, in press). Procedures have an enduring quality that makes their appraisal important to the group as a whole. Tyler and Lind (1992) note that when authorities violate individual relational concerns, such as neutrality or trust, this may be interpreted as prejudice against all members of the unit. Moreover, when a sucient number of unit members perceive themselves to be subjected to similar treatment, positive or negative, procedural justice perceptions and reactions may emerge in the aggregate. Certainly, individual units must cope with their external environments while attempting to interface with the procedural and policy demands of the organization. Depending on the demands of their particular environment, units may feel that policies/procedures thwart their attempts to service internal or external customers, constrain innovation within the unit, or increase social and economic process costs to the unit. This will almost guarantee that some units will feel that the procedures applied to them are not in their best interest (Sheppard et al., 1992). As may occur with other key situational referents (cf. James and James, 1989), we propose that individuals include procedural justice considerations when cognitively appraising their work context. Where individuals are arranged in separate structural or functional units and interact regularly within them, we expect that divergent appraisals concerning the fairness context will emerge from the units. This expectation is in line with Cropanzano and Greenberg's (in press) assertion that structural determinantsÐthose dealing with the environmental context in which people interactÐare important in shaping their fairness evaluations. # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
A MULTILEVEL ANALYSIS
133
Procedural justice context: support from recent studies Indirect support for the concept of contextual procedural justice can be found in individual level procedural justice studies addressing context in terms of individuals' perceptions of and reactions to procedural justice. In such studies, context has been considered largely in terms of social referents. Social context implies relational dynamics involving multiple individuals that cannot be understood in terms of any one of them separately. For example, Folger, Rosen®eld, Grove and Corkran (1979) examined how a co-worker's opinion of allocation decisions aected procedural justice perceptions, ®nding that information regarding a co-worker's beliefs about the fairness of allocations decisions in¯uenced individuals' own perceptions of procedural fairness. In a laboratory study, Steil (1983) discovered that when study participants were informed that others doubted the fairness of the evaluation process by which participant performance was assessed, they too doubted the fairness of the evaluation process. Ambrose, Harland and Kulik (1991), noting that information about procedures as applied to others is an important aspect of the social context, found that social comparisons are involved in individuals' procedural justice determinations. Just as individuals in the same group compare their opinions or pay, they may also make comparisons concerning procedures aecting them. While the above studies used individuals as referents for justice appraisals, two other studies have considered how individuals' procedural justice perceptions and subsequent reactions are in¯uenced by group-based referents. Miller, Jackson, Mueller and Schersching (1987) found support for the importance of group considerations on fairness judgments, showing that decision fairness was more strongly associated with the degree to which the decision represented all group members' interests than those of members individually. James and Cropanzano (1990) maintained that individuals may focus more on the implications for themselves of procedures they see others experience. Testing this, they found support for the notion that individuals use information from observing others in the group to arrive at evaluations of procedural justice. Although the role of contextual factors can be extrapolated from the above studies, their direct generalizability to the current study is limited because all were framed in terms of the individual level of analysis.
Research hypothesis Two pre-conditions need to be suciently satis®ed before the hypothesis concerning procedural justice context can be addressed. The ®rst pre-condition concerned whether some degree of consensus in procedural justice perceptions occurred in organizational work units. Homogeneity or agreement implies a shared assignment of psychological meaning among group members, suggesting that they may be characterized as a whole (James, 1982; Klein et al., 1994). Although the issue of within-unit homogeneity of procedural justice perceptions has not been explored, based on the theoretical arguments and indirect empirical support discussed above, we anticipated that work units would exhibit homogeneity in their members' procedural justice assessments. The second pre-condition concerned whether there was systematic variance across units in their justice perceptions. Given the widespread salience of fairness concerns to individuals (Tyler, 1990; Tyler and Lind, 1992), it would be surprising if group members did not communicate their perceptions regarding fairness issues. Because information exchanges occur more frequently # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
134
K. W. MOSSHOLDER, N. BENNETT AND C. L. MARTIN
within groups rather than between them (Erickson, 1988), members will likely formulate and exchange justice-relevant information that has been ®ltered and processed in ways germane to the work unit. Once group identity is created, members become increasingly responsive to groupcentered motives regarding procedural justice (Tyler and Dawes, 1993). What is considered procedurally fair may depend partly upon whether procedures or policies coincide with dominant group motives. Thus, regarding beliefs about the fairness with which the organization treats them, ingroup dynamics may pull persons in a unit together and outgroup dynamics may create dierences across units. To address the question of whether variation in procedural justice perceptions across units was associated with other important individual level workplace variables, we focused on two frequently studied work-related attitudes that have been linked with procedural justiceÐjob satisfaction and organizational commitment. Procedural justice creates expectations of fair treatment in the long run, signaling to organization members that they are valued, which should encourage greater commitment in return. Procedural justice should also have repercussions for job satisfaction to the degree that it allows individuals to gain necessary input and feedback on task-related activities. Studies have found that individuals' procedural justice perceptions are associated with job satisfaction and organizational commitment (e.g. Alexander and Ruderman, 1987; Folger and Konovsky, 1989; Konovsky and Cropanzano, 1991; McFarlin and Sweeney, 1992). We expected unit level procedural justice eects to parallel those documented in individuals level studies. More importantly, we anticipated that work unit justice perceptions would explain variance in job satisfaction and organizational commitment beyond that accounted for by members' individual procedural justice perceptions. We thus oer that: Individuals in work units whose justice context re¯ects more fair treatment will report higher levels of job satisfaction and organizational commitment than would be expected on the basis of their individual procedural justice perceptions. Of course, the opposite eect was expected to occur for individuals in work units whose justice context re¯ected less fair treatment.
Method Research setting and subjects Respondents for the study were 323 non-supervisory employees of a large savings and loan corporation located in the southwest United States. The sample represented 53 of the savings and loan's 58 locations in ®ve metropolitan areas. Although branch-wide policies were in place, branch managers made personnel decisions and day-to-day operational decisions for their units. As such, each branch functioned as a self-contained unit that required interactions among its employees. From a multilevel perspective (Rousseau, 1985), these features help de®ne two clear levels of analysis, the individual and branch (unit). Between three and 15 employees from each of the 53 branches completed the surveys, administered by a researcher primarily at sessions held on company time and property. The average within-branch response rate across the branches was 70 per cent. The jobs held by participants were bank teller, customer service representative, clerk, and loan ocer. Fifty-seven per cent of the respondents were between the ages of 18±29 years; 15 per cent were over 40 years old. Twentytwo per cent of the respondents had a college degree and 84 per cent were Caucasian. # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
A MULTILEVEL ANALYSIS
135
Measures The items comprising all of the individual level measures used a 5-point, Likert response format. Responses were coded such that higher scores indicated higher levels of procedural justice, job satisfaction, or organizational commitment. To obtain individuals' scores on these measures, items within each scale were averaged.
Job satisfaction Job satisfaction was measured using 16 items (a 0.90) from the Minnesota Satisfaction Questionnaire (Weiss, Dawis, England and Lofquist, 1967).
Organizational commitment Organizational commitment was assessed with the 9-item short form (a 0.89) of the Organizational Commitment Questionnaire (OCQÐMowday, Steers and Porter, 1979).
Procedural justice Individual evaluations of procedural justice were gauged using six items (a 0.81) that re¯ected aspects of both Leventhal et al.'s (1980) six procedural fairness rules and Greenberg's (1986) procedural justice determinants. Instructions focused respondents on the individual level of analysis and the items themselves referenced respondents' personal experience within the organization. Sample items include: `Gives you the opportunity to express your side'; `Enables you to modify or reverse a decision that has already been made'; and `Gives consideration to what you say when making a decision'.
Procedural justice context Because there were no other studies from which to draw a measure of procedural justice context, we constructed one re¯ecting traditional facets of procedural justice and managerial aspects of branch banking. Four items (a 0.81) comprised the procedural justice scale, focusing respondents on the overall fairness of facets central in the organization's human resource system: performance appraisal, raises, bene®ts, and working conditions. In branch banking, unit managers control daily operations including task and social maintenance duties, and they exert appreciable in¯uence on branch employees' cognitive appraisals of the work context. Thus, respondents were asked to consider their supervisor and organization together when judging the general fairness of pay raise and performance appraisal procedures. These items were as follows: `Overall, how fair are the procedures and policies used by your supervisor and your organization to handle performance appraisal?' and `Overall, how fair are the procedures and policies your supervisor and your organization follow in reaching a decision about the size of pay raises?' Parallel items addressed the overall fairness of organization procedures used in determining work conditions (e.g. workload, assignments, etc.) and employee bene®ts. The four items were averaged to form individuals' scores. Within each unit, members' scale scores were then averaged to create procedural justice context scores for all of the units. # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
136
K. W. MOSSHOLDER, N. BENNETT AND C. L. MARTIN
Analyses Con®rmatory factor analysis Because organizational commitment, job satisfaction, and procedural justice measures were intercorrelated, we conducted a con®rmatory factor analysis to examine their viability as separate scales. We combined respective scale items to increase indicator stability (West, Finch and Curran, 1995), thereby reducing the number of indicators for the organizational commitment and procedural justice factors to three each, and the number of indicators for the job satisfaction factor to four. The LISREL 8 statistical package was used for the analyses (JoÈreskog and SoÈrbom, 1993), and the comparative ®t (CFI) and incremental ®t (IFI) indices were used to assess model ®t. A CFI of 0.98 and an IFI of 0.98 were obtained for a correlated three-factor model, indicating an acceptable level of ®t. A model positing that a single factor underlies the three measures was then tested, but did not ®t the data well (CFI 0.71, IFI 0.72). Finally, a twofactor model in which job satisfaction and organizational commitment indicators were combined was examined to consider whether they formed one aective construct. The CFI was 0.83 and the IFI was 0.83 for the two-factor model. Comparing the two- and three-factor models, the chisquare dierence test was signi®cant (Dw2 301.30, df 2, p 5 0.001). Taken together, these analyses indicate that the measures could be examined separately.
Hierarchical linear modeling Hierarchical linear modeling (HLM) provides for a more robust examination of cross-level models (Bryk and Raudenbush, 1989, 1992; Bryk, Raudenbush and Congdon, 1994), allowing for the investigation of both within- and between-unit eects on an individual level dependent variable. A within-unit, `level-1' analysis is used to determine two separate parameters (i.e. intercept and slope terms) describing the relationship between the predictors and the focal dependent variable within each unit. The intercept and slope parameters obtained from the level1 analysis then serve as the dependent variables in equations used for a between-unit, `level-2' analysis. In the present study, the level-1 analysis considered relations between individuals' procedural justice perceptions and, each in turn, their organizational commitment and job satisfaction. Our level-2 variable, procedural justice context, was then used to predict hypothesized variance in the intercept terms from these within-unit regressions. If the parameter estimate (gÐgamma coecient) associated with work unit procedural justice in the level-2 model were statistically signi®cant, a contextual eect for procedural justice would be indicated.
Results Preliminary analyses Descriptive statistics for the individual level study variables are reported in Table 1. As would be expected, the intercorrelations among study variables were positive and moderate in magnitude. To examine the homogeneity of procedural justice context perceptions within bank branches, we computed interrater agreement indices (rwg ÐJames, Demaree and Wolf, 1984, 1993). This index indicates the degree to which unit members concur in their individual assessments regarding a # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
A MULTILEVEL ANALYSIS
137
Table 1. Descriptive statistics Variable 1. Organizational commitment 2. Job satisfaction 3. Procedural justice
M
S.D.
1
2
3
3.16 3.17 2.66
0.82 0.67 0.77
Ð 0.55 0.59
Ð 0.53
Ð
n 323. All correlations are signi®cant at p 5 0.01.
contextual variable. A mean rwg of 0.77 was found across the 53 bank branches. We also computed eta coecients as another indicator of between-unit variation in the procedural justice context measure (James, 1982). An eta of 0.60 (F 2.85, p 5 0.001) was obtained, indicating that 36 per cent of the variance in this measure lies between units. These results suggest that there was both within-unit agreement and between-unit variation in the procedural justice context measure.
Hierarchical linear modeling analyses In keeping with HLM procedures recommended for examining contextual eects (see Hofmann, in press; Kidwell, Mossholder and Bennett, in press), we ®rst determined whether there was systematic between-unit variation in our dependent variables. Calculations revealed that 27 per cent of the variance in job satisfaction and 18 per cent of the variance in organizational commitment was between-unit variance. Chi-square statistics indicated that both of these between-unit variance estimates were signi®cantly dierent from zero (satisfaction: w2 159.89, df 52, p 5 0.001; commitment: w2 120.05, df 52, p 5 0.001), suggesting that systematic between-unit variance in each of the dependent variables existed. Next, we estimated a random coecient regression model that included individual level procedural justice as the independent variable and individuals' job satisfaction (or organizational commitment) as the dependent variable. A signi®cant w2 statistic for the t00 estimates produced at this step (see Hofmann, in press) would provide evidence that there is systematic variance in the intercepts across units. Results show that this occurs only for job satisfaction (job satisfaction: w2 96.14, df 52, p 5 0.001; organizational commitment: w2 61.58, df 52, p 4 0.05). Because there was no systematic variance in the intercepts for organizational commitment, only the portion of the hypothesis regarding job satisfaction received further consideration. Given that there was both systematic within- and between-unit variance in job satisfaction and signi®cant variance in the level-1 intercepts, we estimated an intercepts-as-outcomes model for job satisfaction in a level-2 analysis. The t-test for the g01 parameter produced in the level-2 analysis provides the operational test of our hypothesis (Bryk and Raudenbush, 1992). This t-test was signi®cant (g01 2.03, t 6.22, p 0.001), con®rming that procedural justice context explains variance in individuals' job satisfaction beyond that accounted for by individual level procedural justice perceptions. As expected, individuals in branches whose justice context re¯ected higher procedural justice exhibited higher levels of job satisfaction. The between-unit variance explained by procedural justice context can be computed using the t00 coecients produced by this model and the random coecient regression model (see Hofmann, in press). From this information, we determined that procedural justice context explained 75 per cent of the between-unit variance in individuals' job satisfaction. Because 27 per cent of the variance in job satisfaction was found to be between units, this meant that procedural justice context explained 20 per cent of the variance in employees' job satisfaction. # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
138
K. W. MOSSHOLDER, N. BENNETT AND C. L. MARTIN
Discussion Researchers have focused mainly on the individual level of analysis in examining the role of procedural justice in organizations. Thus, the primary contribution of the present study was to investigate the role that the procedural justice context plays in relation to employees' aective work reactions. Speci®cally, we found that individuals belonging to units with higher perceptions of procedural justice registered greater job satisfaction than would have been expected based on their individual justice perceptions alone. By design, policies and procedures aect many individuals simultaneously. The degree of procedural justice extant within any particular bank branch could be expected to elicit similarly favorable eective reactions among employees. This might explain the emergence of a unit level eect for job satisfaction. Such an eect was not found regarding organizational commitment, however, perhaps partly because of its substantive focus. Commitment may be characterized in terms of congruence between individuals and the organization over longer time spans. Where commitment is concerned, procedural justice may provide employees with information concerning their individual ®t with the organization and whether a relationship with it should be continued. Rousseau and Parks (1992) note that perceived fair treatment is critical to the continuation of relational contracts, which often involve issues of commitment to the organization. They also propose that adherence to terms of relational contracts are subject to individuals' interpretations. Thus, it seems that fair treatment may stimulate commitment, but more on an individual basis because only individuals are in a position to interpret what fair treatment means in particular for their future relations with the organization. Our purpose was not to pit individual and contextual explanations of procedural justice eects against one another, but to demonstrate that contextual eects do occur. Just as instrumental and noninstrumental aspects of procedural justice may concurrently in¯uence individuals' reactions to treatment by their organization (Shapiro and Brett, 1993; Tyler and Lind, 1992), so too may individual and unit level justice components. Although the research base is not yet sucient to establish a cross-level theory of procedural justice, we speculate that individual and unit in¯uences regarding procedural justice could resemble what Wegner (1982) describes as the tacitfocal function. In brief, an individual's tacit awareness of one social entity, such as their work unit, can serve in a secondary capacity for evaluative purposes. Focal awareness orients one toward a target social entity (e.g. other individuals, events, etc.) that is evaluated in terms of standards of the particular tacit system that has been engaged. Some researchers (e.g. Tyler, 1990) propose that individuals have personal tacit standards of fairness deriving from the general culture. Naturally, these standards come into play when individuals assess the fairness of organizational procedures. However, given the salience of the collective work unit, we oer that such standards can come to serve as an active referent for procedural justice appraisals. In the present study, the average rwg for procedural justice perceptions was 0.77, suggesting a collective quality to members' assessments of fair treatment by the organization. When this happens, unitbased eects may emerge in addition to reactions based on individuals' personal understanding of procedural justice. Some limitations of the study should be acknowledged to put them in perspective. Common method variance artifacts are always possible in survey research designs where data are collected from one source. While such artifacts could have aected our individual level results, they are not a likely explanation for the obtained procedural justice context eects because these eects stem from aggregate responses. Also, because of their cross-sectional nature, the data could be used only to demonstrate contextual eects rather than explain how they occur. Longitudinal research # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
A MULTILEVEL ANALYSIS
139
should be conducted to begin to understand the processes that underlie contextual procedural justice. Viewed broadly, this study suggests some new directions for organizational research on procedural justice. With regard to the primary ®ndings, an implication is that researchers should also consider the procedural justice context in which people work, as knowing about their individual perceptions may not reveal the full scope of potential repercussions for work attitudes and reactions. For example, people who feel that they personally have been treated fairly may not react with a commensurate level of job satisfaction if they are in a work unit that has, on average, perceptions to the contrary. In this scenario, the justice context could act to attenuate the level of job satisfaction expected based solely on individual levels of perceived procedural justice. In a reverse type of scenario, the justice context could have an accentuating eect. Over time, employees may have critical interactions with organizational decision makers that strongly aect their views of organizational fairness. We would not argue that contextual factors outweigh direct experience in determining procedural justice assessments and eects. But where employees' experience or information is insucient, they may also draw upon information gleaned from others' critical interactions to make more informed inferences about procedural fairness. Tyler and Lind (1992) reason that across situations, individuals' understanding of fairness in authority relations develops, in part, through such an inductive process. The present study is consistent with this rationale, only within the speci®c context of the work unit. Moreover, our results imply that work unit in¯uences may not only aid employees' understanding of organizational fairness, but also aect how they react to this understanding. Decision makers should recognize that whereas organizational procedures may be designed to be fair to all organization members, the interpretation, execution, and impact of procedures at dierent hierarchical levels can vary (Sheppard et al., 1992). Organizations should insure that their agents (e.g. managers, department heads, ®rst-line supervisors) are attuned to general guidelines for fair behavior. Moreover, these guidelines should be made appropriate and meaningful to the particular level of personnel aected. Whether such guidelines are drawn from formal lists such as Leventhal's (Leventhal et al., 1980) or other sources, those in decisionmaking positions should be familiar with procedural fairness issues. Organizations could also educate decision makers about social processes, such as making adequate social accounts (Bies, Shapiro and Cummings, 1988) or impression management tactics (Greenberg, 1990b), that may be useful in applying justice guidelines. If information exchanges among employees play a role in the formation of procedural justice perceptions, it becomes all the more important for organizations to not only be fair with their employees, but to cultivate an image of fairness and avoid the predicament of injustice (Bies, 1987; Greenberg, 1990b). Focusing on this idea with respect to work units, perceptions of procedural justice can be proactively in¯uenced to the degree that an organization acts with fairness towards aggregate units as well as individuals. Sheppard et al. (1992) discuss this as part of the challenge of systemic justice. The present ®ndings suggest that it may be useful to begin studying antecedents of unit level perceptions of procedural justice. Organization-wide processes like budgeting and resource allocation, performance appraisal, downsizing, and strategy implementation may be intertwined with the procedural justice context. Certainly, further research is needed to consider whether there is an isomorphism of sorts between procedural justice processes occurring at the individual and unit levels. One conjecture does seem reasonable at this point, however: both antecedents and eects of procedural justice at the unit level of analysis are likely to be more complex. Hopefully, this study will encourage researchers to further investigate procedural justice phenomena from a multilevel perspective. # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
140
K. W. MOSSHOLDER, N. BENNETT AND C. L. MARTIN
References Alexander, S. and Ruderman, M. (1987). `The role of procedural and distributive justice in organizational behavior', Social Justice Research, 1, 177±198. Ambrose, M. L., Harland, L. K. and Kulik, C. T. (1991). `In¯uence of social comparisons on perceptions of organizational fairness', Journal of Applied Psychology, 76, 239±246. Bies, R. J. (1987). `The predicament of injustice: The management of moral outrage', Research in Organizational Behavior, 9, 289±319. Bies, R. J., Shapiro, D. L. and Cummings, L. L. (1988). `Causal accounts and managing organizational con¯ict: Is it enough to say it's not my fault?' Communication Research, 15, 381±399. Bryk, A. S. and Raudenbush, S. W. (1989). `Methodology for cross level organizational research', Research in the Sociology of Organizations, 7, 233±273. Bryk, A. S. and Raudenbush, S. W. (1992). Hierarchical Linear Models: Applications and Data Analysis Methods, Sage, Newbury Park, CA. Bryk, A. S., Raudenbush, S. W. and Congdon, R. J. (1994). Hierarchical Linear Modeling with the HLM/ 2L and HLM/3L Programs, Scienti®c Software International, Chicago. Cropanzano, R. and Greenberg, J. `Progress in organizational justice: Tunneling through the maze'. In: Cooper, C. L. and Robertson, I. T. (Eds) International Review of Industrial and Organizational Psychology, John Wiley & Sons, New York, in press. Erickson, B. H. (1988). `The relational basis of attitudes'. In: Wellman, B. and Berkowitz, S. (Eds) Social Structure: A Network Approach, Cambridge University Press, New York, pp. 99±121. Folger, R. and Konovsky, M. A. (1989). `Eects of procedural and distributive justice on reactions to pay raise decisions', Academy of Management Journal, 32, 115±130. Folger, R., Rosen®eld, D., Grove, J. and Corkran, L. (1979). `Eects of ``voice'' and peer opinions on responses to inequity', Journal of Personality and Social Psychology, 37, 2253±2261. Greenberg, J. (1986). `Determinants of perceived fairness of performance evaluations', Journal of Applied Psychology, 71, 340±342. Greenberg, J. (1990a). `Organizational justice: Yesterday, today, and tomorrow', Journal of Management, 16, 399±432. Greenberg, J. (1990b). `Looking fair versus being fair: Managing impressions of organizational justice', Research in Organizational Behavior, 12, 111±157. Hofmann, D. A. `The logic and rationale of HLM', Journal of Management, in press. James, K. and Cropanzano, R. (1990). `Focus of attention and locus of control as moderators of fraternal justice eects', Social Justice Research, 4, 169±185. James, L. A. and James, L. R. (1989). `Integrating work perceptions: Explorations into the measurement of meaning', Journal of Applied Psychology, 74, 739±751. James, L. R. (1982). `Aggregation bias in estimates of perceptual agreement', Journal of Applied Psychology, 67, 219±229. James, L. R., Demaree, R. G. and Wolf, G. (1984). `Estimating within-group interrater reliability with and without response bias', Journal of Applied Psychology, 69, 85±98. James, L. R., Demaree, R. G. and Wolf, G. (1993). `rwg : An assessment of within-group interrater agreement', Journal of Applied Psychology, 78, 306±309. JoÈreskog, K. G. and SoÈrbom, D. (1993). LISREL 8 User's Reference Guide, Scienti®c Software International, Chicago. Kidwell, R. E., Mossholder, K. W. and Bennett, N. `Cohesiveness and organizational citizenship behavior: A multi-level analysis using work groups and individuals', Journal of Management, in press. Klein, K. J., Dansereau, F. and Hall, R. J. (1994). `Levels issues in theory development, data collection, and analysis', Academy of Management Review, 19, 195±229. Konovsky, M. A. and Cropanzano, R. (1991). `Perceived fairness of employee drug testing as a predictor of employee attitudes and job performance', Journal of Applied Psychology, 76, 698±707. Leventhal, G. S., Karuza, J. and Fry, W. R. (1980). `Beyond fairness: A theory of allocation preferences'. In: Mikula, G. (Ed.) Justice and Social Interaction, Springer-Verlag, New York, pp. 167±218. Lind, E. A. and Tyler, T. (1988). The Social Psychology of Procedural Justice, Plenum, New York. McFarlin, D. B. and Sweeney, P. D. (1992). `Distributive and procedural justice as predictors of satisfaction with personal and organizational outcomes', Academy of Management Journal, 35, 626±637. # 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
A MULTILEVEL ANALYSIS
141
Miller, C. E., Jackson, P., Mueller, J. and Schersching, C. (1987). `Some social psychological eects of group decision rules', Journal of Personality and Social Psychology, 52, 325±332. Mowday, R. T., Steers, R. M. and Porter, L. W. (1979). `The measurement of organizational commitment', Journal of Vocational Behavior, 14, 224±247. Rousseau, D. M. (1985). `Issues of level in organizational research: Multilevel and cross level perspectives', Research in Organizational Behavior, 7, 1±37. Rousseau, D. M. and Parks, J. M. (1992). `The contracts of individuals and organizations', Research in Organizational Behavior, 15, 1±43. Shapiro, D. L. and Brett, J. M. (1993). `Comparing three processes underlying judgments of procedural justice: A ®eld study of mediation and arbitration', Journal of Personality and Social Psychology, 65, 1167±1177. Sheppard, B. H., Lewicki, R. J. and Minton, J. W. (1992). Organizational Justice: The Search for Fairness in the Workplace, Lexington Books, New York. Steil, J. M. (1983). `The response to injustice: Eects of varying levels of social support and position of advantage or disadvantage', Journal of Experimental Social Psychology, 19, 239±253. Thibaut, J. and Walker, L. (1975). Procedural Justice: A Psychological Analysis, Erlbaum, Hillsdale, NJ. Tyler, T. R. (1990). Why People Obey the Law, Yale University Press, New Haven, CT. Tyler, T. R. and Bies, R. J. (1990). `Beyond formal procedures: The interpersonal context of procedural justice'. In: Carroll, J. (Ed.) Advances in Applied Social Psychology: Business Settings, Lawrence Erlbaum, Hillsdale, NJ, pp. 77±98. Tyler, T. R. and Dawes, R. M. (1993). `Fairness in groups: Comparing the self-interest and social identity perspectives'. In: Mellers, B. A. and Baron, J. (Eds) Psychological Perspectives on Justice: Theory and Applications, Cambridge University Press, New York, pp. 87±108. Tyler, T. R. and Lind, E. A. (1992). `A relational model of authority in groups', Advances in Experimental Social Psychology, 25, 115±191. Wegner, D. M. (1982). `Justice and the awareness of social entities'. In: Greenberg, J. and Cohen, R. (Eds) Equity and Justice in Social Behavior, Academic Press, New York, pp. 77±117. Weiss, D. J., Dawis, R. V., England, G. W. and Lofquist, L. H. (1967). Manual for the Minnesota Satisfaction Questionnaire, Industrial Relations Center, University of Minnesota, Minneapolis. West, S. G., Finch, J. F. and Curran, P. J. (1995). `Structural equation models with nonnormal variables: Problems and remedies'. In: Hoyle, R. H. (Ed.) Structural Equation Modeling, Sage, London, pp. 56±75.
# 1998 John Wiley & Sons, Ltd.
J. Organiz. Behav. 19: 131±141 (1998)
