Exploring applicant reactions to selection processes from an organizational justice perspective

David Hart Oliver


Applicants' reactions to employment testing processes may result in many negative consequences for organizations. This study investigated the role of 10 procedural rules on job applicants' perceptions of procedural justice (or fairness). Relationships among applicants' perceptions of procedural justice and several outcome variables were also assessed. Two-hundred-and-nine (209) applicants at a large telecommunications company completed a written survey and face-to-face interview immediately after employment testing. The written survey measured perceptions of the opportunity to perform, ease of faking, job relatedness, consistency of administration, knowledge of results, selection information, honesty, two-way communication, invasiveness of questions, and interpersonal effectiveness of the administrator. The oral survey measured perceptions of procedural justice and four outcome variables (i.e., affect, organizational attractiveness, intentions to recommend the organization to others, and intentions to accept a job offer). The following hypotheses were examined using standard multiple regression and structural equation modeling techniques. Perceptions of the procedural rules were divided into three classifications and hypothesized to predict procedural justice. Procedural justice was hypothesized to predict the four outcome variables. This proposed model is very similar to a section of Gilliland's (1993) applicant reaction model. Multiple regression results indicated that the three procedural rule classifications predicted procedural justice $(R\sp2$ ranged from.22 to.33), and the variables of honesty, opportunity to perform, consistency of administration, job relatedness, and two-way communication made unique contributions to procedural justice. Procedural justice predicted the four outcome variables (bivariate correlations ranged from.37 to.57). Structural equation results indicated that two of the three classifications nearly fit the proposed model. Variables with significant path coefficients to procedural justice included honesty, opportunity to perform, and consistency of administration. Overall, the results of this study support Gilliland's (1993) model. Applicants' perceptions of the selection process influence perceptions of fairness, and perceptions of procedural justice are related to several important outcome variables. Inconsistent results between the two statistical procedures are believed to be due to measurement error in the regression procedures, which highlights the importance of accounting for error in measurement when assessing applicant reactions.