DOES INTERPERSONAL JUSTICE NEGATIVELY PREDICT OVERALL PERFORMANCE APPRAISAL FAIRNESS? A PAKISTANI STUDY
Building on recent research in organizational justice, this study examined the impact of four factors of justice on perceived overall fairness of performance appraisal (PA) systems. The study used a survey questionnaire to collect data from 259 public servants in Pakistan. Based on fundamental ethical principles that employees desire public managers to be unbiased, we developed the hypotheses that were, subsequently, tested using variance-based structural equation modeling in SmartPLS 3.0. The results of the study revealed evidence that, in the context of PA systems, high interpersonal justice is indeed associated with decreased perceptions of overall fairness. The findings of the study provide guidance to public managers that distributive justice, procedural justice, and informational justice perceptions are more conducive for overall fairness perceptions in the PA system than interpersonal justice. Although the negative effect of interpersonal justice seems plausible, surprisingly, empirical research hitherto has not directly investigated this possibility.
INTRODUCTION
Performance appraisal (PA) is common in organizations, yet it is considered a complex process (Asim, 2001; Murphy, 2020; Murphy et al., 2018). Some researchers (e.g., Folger et al., 1992; J. Kim, 2016) have highlighted competing demands of appraisers and appraisees as a problem in PA. On the one hand, appraisers consider PA their legitimate administrative authority to maintain smooth functioning of their administrative unit(s), and then, communicate this to key stakeholders, such as top management and peer administrative units (Aguinis, 2013; Murphy & Cleveland, 1995). On the other hand, appraisees desire higher ratings (Harris & Shaubroeck, 1988), which may get them cherished outcomes, such as pay raises, training and development, and promotions. Under these circumstances, both appraisers and appraisees may prefer pursuing their self-interests at the expense of the organization's interest (Rosen et al., 2016).
While anticipating the above situation, appraisers' motives to maintain harmonious interpersonal relationships with appraisees may come into play (Rosen et al., 2016). That is, appraisers may set aside appraisees' actual job performance and inflate ratings about appraisees (Aguinis, 2013). As a result, appraisers strengthen their interpersonal relationships with appraisees, sometimes at the cost of the overall fairness of the PA system (OFPAS; Cawley et al., 1998; Pichler, 2012). Whether appraisers employ a confrontational or conciliatory approach while rating performance, they face the challenge of demonstrating overall fairness and effectiveness of the PA system (for example, by exhibiting the integrity of the system; Rosen et al., 2016). In line with this, recent literature warrants more studies on the association between justice and leaders' (appraisers') relationships with subordinates (Mumtaz & Rowley, 2020).
The above challenges are essentially the same for PA systems across public sector organizations. Due to broad goals in public sector organizations, appraisers are constrained to use subjective assessment criteria for writing appraisal reports (S. P. Robbins & Mathew, 2009). Subjective judgments are prone to vary among appraisers, increasing the complexity of the PA process (Asim, 2001; J. Kim, 2016). Thus, as purported by the theory of self-serving behavior (Johns, 1999), appraisees in public sector organizations may feel that the overall PA system is ambiguous and uncertain. This state of affairs is likely to increase the tendency to pursue self-interest among appraisees (Rosen et al., 2016).
Following the theory of self-serving behavior (Johns, 1999), we assume that public managers (appraisers) more often rely on subjective assessment criteria resulting in speculative ratings, and then, offset their subjective judgments with rating adjustments to preserve their interpersonal relationship with appraisees. Although these ratings may favor appraisees, it may create an impression among them that such favorable ratings undermine the overall fairness perceptions of a PA system. Substantiating this, the recent PA literature (e.g., Bernardin et al., 2016; Gupta & Singh, 2013; Rosen et al., 2016) suggests that the supervisor's diligent efforts for developing and maintaining good interpersonal relationships with the subordinates may result in paradoxical consequences for fairness perceptions among appraisees. Hence, it can be expected that the supervisor's intent to maintain a good interpersonal relationship with the subordinate—that is, the subordinate's positive perceptions of interpersonal justice (IntJ)—may backfire as reflected in the subordinate's negative perceptions of overall fairness. This line of reasoning is consistent with the justice literature, which specifies that people accord authorities with legitimacy based on the extent to which the authorities have instilled neutrality to manage their conflicts (e.g., Zapata et al., 2016).
Put differently, appraisees' perceptions of IntJ can be more crucial, as well as atypical, toward their overall fairness perceptions of the PA system (Bernardin et al., 2016). Under these circumstances, demonstrating fairness in the PA system of public sector organizations cannot be reckoned as straightforward as it can be for other personnel functions. Acknowledging this void, during the past 2 decades, some public management researchers investigated what factors determine the perceived fairness of a PA system (S. E. Kim & Rubianty, 2011; Nesbit et al., 2012; Rubin & Kellough, 2012).
Although the negative impact of IntJ on overall fairness perceptions in the context of PA systems seems plausible, surprisingly, empirical research hitherto has not directly investigated this possibility. A limited number of studies have explored PA-related interactions between appraisers and appraisees in public sector organizations (e.g., Harrington & Lee, 2015). Furthermore, previous studies paid less attention to overall fairness as compared to four factors of justice (Mohammad et al., 2019). Therefore, in line with the contemporary organizational justice (OJ) research stream (e.g., Ambrose & Arnaud, 2005; Ambrose & Schminke, 2009; Colquitt & Zipay, 2015; Goldman & Cropanzano, 2015; Lind, 2001a, 2001b), the purpose of our article is to investigate the simultaneous effects of all four factors of justice on overall fairness perceptions in the context of PA system in public sector organizations in Pakistan. This step seeks to differentiate the effect of IntJ on overall fairness perceptions from other forms of justice—i.e., distributive, procedural, and InfJ. Thus, our article is expected to motivate PA and OJ researchers to refine the dimensionality of OJ theory, from the perspective of the theory of self-serving behavior (Johns, 1999).
LITERATURE REVIEW
Organizational Justice
The OJ field distinguishes between various conceptualizations of justice that are relevant for understanding the functioning of organizations. A common typology of various conceptualizations pertains to the four factors of justice—that is, distributive, procedural, interpersonal, and informational (Colquitt, 2001). Distributive justice (DJ) refers to perceptions that the outcomes or resources (e.g., salary, promotions, appraisal ratings) that employees derive from their social exchange with the organization are fair. Frequently, such DJ perceptions are fueled by social comparisons (e.g., Adams, 1965; Smith et al., 2012). Procedural justice (PJ) refers to perceptions that the decision-making procedures used to arrive at certain outcomes are fair. For instance, people prefer procedures that give them voice, that consider all relevant information, and that are free of bias (e.g., De Cremer & Tyler, 2005; Tyler & Blader, 2003; Van den Bos & Lind, 2002; van Prooijen et al., 2012).
The third factor, IntJ, then, considers how employees experience their relationship with their direct supervisor. Specifically, experiences of IntJ are characterized by the perceptions that one is treated with respect and dignity, which has a substantial impact on outcomes such as the evaluation of authorities, job satisfaction, organizational commitment, and the like. The fourth factor, informational justice (InfJ), pertains to the explanations that people receive as to why certain procedures are chosen, or why certain decisions are made (Colquitt et al., 2001). In addition, several studies have identified that fairness perceptions are influenced by demographic characteristics like age, gender, and tenure of employees (e.g., Cohen-Charash & Spector, 2001; Kulik et al., 1996).
Organizations use PA results to make various human resource decisions, such as salary increments, promotions, training and development, and documentation for legal purposes (Dubinsky et al., 1989; Taylor et al., 1995). Since these human resource decisions directly affect employees' careers, employees aspire to fair human resource decisions, and therefore also seek fairness of the PA system.
OJ and PA System
PA and OJ literatures suggest the significance of investigating employees' fairness perceptions—that is, distributive, procedural, interpersonal, and informational fairness toward various facets of the PA system (Latham & Wexley, 1994). This is because employees' fairness perceptions of a PA system can be helpful in determining perceptions of the system's success and usefulness—that is, the effectiveness of the PA system (Erdogan et al., 2001). The literature suggests that a careful demonstration of principles of OJ in a PA system may help organizations increase their employees' morale, productivity, job performance, organizational commitment, job satisfaction, and motivation (Blau, 1999; Boyd & Kyle, 2004; Colquitt et al., 2001). In contrast, a poorly managed and implemented PA system may result in employees' perceptions of unfairness, creating lack of trust in the system (Marsden & French, 1998). Therefore, employee fairness perceptions of a PA system can have substantial implications for both employees and the organization.
All four forms of justice—i.e., distributive, procedural, interpersonal, and informational—certainly seem to fit in the context of PA systems. Therefore, all four factors of justice are important to be studied (Malagueño et al., 2019). Indeed, PA systems typically pertain to decision-making processes where certain outcomes are distributed (i.e., appraisal ratings), using a specified decision-making structure that can vary in perceived fairness (i.e., PJ). Pointing to the relevance for IntJ, PA ratings may be influenced by the mutual respect that exists between the appraiser and the appraisee. Moreover, it is likely that employees seek to know the basis for their appraisal ratings, suggesting a prominent role of InfJ. As such, all four factors of justice should be incorporated when investigating how employees determine the overall fairness of the PA system.
Overall Fairness
Studying overall fairness mirrors a theoretical and empirical trend in OJ research—that is, investigating overall fairness perceptions above and beyond any specific factor of justice (see, e.g., Colquitt & Zipay, 2015; Goldman & Cropanzano, 2015). The literature suggests that distinct types of justice may not give a complete picture of employees' fairness perceptions in work settings, but overall fairness provides a broader view of individuals' justice perceptions. Hence, overall fairness perceptions are considered useful for overcoming the limitations of current OJ research (Ambrose & Schminke, 2009). Overall fairness perceptions about the organization are referred to as “systemic justice,” a term that was first used by Sheppard et al. (1992). In work settings, employees utilize overall fairness perceptions as a heuristic to decide whether they should trust their organization (Lind, 2001a). Fairness is “a global impression of fair treatment” (Lind & van den Bos, 2002, p. 196) and appropriateness (Colquitt & Zipay, 2015). In line with this conceptualization, recently defined terms of justice and fairness are that “justice” denotes conduct that is morally required, whereas “fairness” denotes an evaluative judgment as to whether this conduct is morally praiseworthy (Goldman & Cropanzano, 2015, p. 313). Thus, we maintain that overall fairness of the PA system is a global assessment of whether or not the existing PA system in an organization is fair.
The Role of IntJ in OFPAS
IntJ pertains to whether a supervisor has treated their appraisees with respect and dignity during the appraisal period (Greenberg, 1993a). Generally, it is assumed that all four factors of justice contribute positively toward employees' overall fairness perceptions. However, Zapata et al. (2016) propounded that IntJ may have surprising negative consequences. We argue that in the specific context of PA systems, this may be the case. Our main argument is that when employees evaluate a PA system, high perceptions of IntJ may result in low overall fairness perceptions. This inverse effect of IntJ on overall fairness may be attributable to the complex nature and interpersonally sensitive dynamics that are associated with a PA system.
In the literature, scholars have identified various problems related to PA systems: For instance, appraisers reckon PA as an unpleasant and unproductive practice, and thus, they inflate performance ratings (Kikoski, 1999). Highlighting the issues of a PA system, Murphy and Cleveland (1995) contended that lenient ratings by superiors are one of the realities of a PA system, which create various complications within the system. In this regard, PA researchers have reported that appraisers give inflated ratings to avoid various conflicts, as they believe that a good relationship between the supervisor and the subordinate can be damaged due to a negative PA (Murphy & Cleveland, 1991); or in some cases appraisers commit preferential treatment—i.e., their ratings about appraisees are based on personal likes or dislikes (Longenecker et al., 1987). Such preferential treatment, however, violates fundamental norms about justice and ethics.
As a case in point, van Prooijen et al. (2006) argued that authorities can be biased by showing inappropriate preference to one person over the other, which may be due to common group membership, friendship, or other affiliations that lead decision makers to favor one over the other recipient. Their research indeed indicates that, under certain circumstances, biased authorities can elicit exceptionally low ratings of justice. De Cremer (2004) reasoned that a biased authority is not reliable and as a result of this negative impression, any subsequent act of the authority is perceived negatively, even if they have used accurate procedure(s). These considerations suggest that people who are concerned about a fair PA have reason to be suspicious of an overly friendly supervisor. Indeed, research indicates that communal and well-intended emotions such as compassion can promote injustice due to preferential treatment (Batson et al., 1995). Inflated ratings due to the appraiser's intent for managing conflict or preferential treatment might decrease the perceptions that a PA system is fair overall, per se.
Hypothesis 1: In the context of PA systems for public servants, all four factors of justice may predict OFPAS such that:
-
H1a: DJ may positively predict OFPAS
-
H1b: PJ may positively predict OFPAS
-
H1c: InfJ may positively predict OFPAS
-
H1d: IntJ may negatively predict OFPAS
METHOD
Procedure
We collected the data from public servants working in two public sector organizations in the Southern district of the province, Khyber Pakhtunkhwa, Pakistan. The Southern District is located far from the provincial headquarter, Peshawar. So, it has fewer facilities and resources available for the public servants. Above this, the working environment was challenging as the district experienced dwindling law and order situations in the past. The PA system is vital for public servants as it guides many important career-related decisions. Concerning these circumstances, investigating fairness perceptions of public servants about the PA system was needed.
Although there were various departments in the district, we chose the departments of education and health. We selected these two departments for various reasons. First, both departments have a primary influence on public service delivery. Second, both departments have setups in the whole district. The education department has schools throughout the district, while the health department has established basic health units and rural health centers in the district. The public servants in Basic Pay Scale 16 through 19, having experience of 2 or more years, were eligible to patriciate in the study. Participants were working as subject specialists, headmistresses and headmasters in the education department, or as medical officers and doctors working in administrative posts in the health department.
For the data collection from eligible employees, the principal author personally administered a questionnaire (English version) with a cover letter explaining the purpose of the study. All the respondents were provided with envelopes to ensure confidentiality of their responses. Participants were encouraged to complete the questionnaire during break time, and they were also requested to immediately return the questionnaire after completion.
Participants
The questionnaire was distributed among 273 eligible employees. The survey resulted in data from 261 employees with response rate of (95%) in which 259 questionnaires were usable for data analysis. Among them, 212 were males and 47 were females. A total of 28 respondents reported to perform duties as an appraiser in their department. The majority of respondents (n = 118) were between the age groups of 36–45, whereas 80 and 50 respondents were between the age group of 25–35 and 46–55 respectively, and the remaining 11 were older than 55 years. For details see Table 1.
Measures
Except for demographic variables, all other items were measured on 5-point Likert response scales ranging from 1 (strongly disagree) to 5 (strongly agree).
Organizational Justice Perceptions
Appraisees' fairness perceptions of the PA system were measured through scales of DJ, PJ, IntJ, and InfJ, which were designed and validated in the specific context of PA systems (Thurston, 2001; Thurston & McNall, 2010; Walsh, 2003). Some items were modified slightly to match the requirements of our purposes.
PJ was measured with three subscales—i.e., “Setting Performance Targets,” “Reporting Officer Confidence,” and “Seeking Appeals,”—comprising a total of 19 items. The sample items (from each subscale) were: “My performance targets are set on the basis of my job description,” “My department assigns me a reporting officer who is fully aware of my work,” “In my department performance ratings are challenged if they are unfair,” and “performance evaluation report influences personnel decisions.” All 19 items were averaged into a single PJ scale (α = .88).
DJ was measured with two subscales—i.e., “Accuracy of Performance Ratings” and “Concern over Ratings”—comprising a total of 13 items. The sample items (from each subscale) were “My reporting officer gives me genuine rating even if it might irritate me” and “In my department performance rating is based on the quantity of my work.” All 13 items were averaged into the DJ scale (α = .95).
IntJ was measured with two subscales—i.e., “Respect by Reporting Officer” and “Sympathy of Reporting Officer”—comprising a total of 8 items. The sample items (from each subscale) were: “My reporting officer is nice to me” and “My reporting officer is responsive to my feelings.” The IntJ scale was also highly reliable (α = .78).
InfJ was measured with three subscales—i.e., “Clarification of Performance Targets,” “Performance Feedback,” and “Explanation of Rating Decisions”—comprising a total of 15 items. The sample items (from each subscale) were: “My reporting officer clearly explains regularly about his/her expectations of my performance,” “My reporting officer regularly lets me know how I am doing my job,” and “My reporting officer always explains the decisions that concern me.” The items from these subscales were averaged into the InfJ scale (α = .96).
Overall Fairness of the PA System
Appraisees' overall fairness of the PA system (OFPAS) was measured by using six items developed by Ambrose and Schminke (2009). A sample items was: “In my department overall performance evaluation system is fair.”
ANALYSIS AND RESULTS
Before selecting a tool for analyzing statistical data, we performed a Shapiro-Wilk W test (p > .05; Shapiro & Wilk, 1965). The results showed that the dependent variable (DV) was not normally distributed for each response group of the independent variables (IVs; see Supplemental Table SI). Based on these results, we employed partial least squares structural equation modeling technique (PLS-SEM), using SmartPLS 3.0 (Hair et al., 2017). Moreover, in the current study we collected the data through a survey at a single time point, and therefore we applied ex ante and ex post approaches to minimize the risk of common method variance (Podsakoff et al., 2003; see Supplemental Material).
Measurement Model
The measurement model of reflective constructs includes internal consistency, convergence of validity, and discrimination (Hair et al., 2017). First, we tested a measurement model for internal consistency. Table 2 shows that the Cronbach's α and composite reliability (ρc) scores for all the reflective constructs are above the acceptable threshold value of ≥ .70 and ≥ .80, respectively. Therefore, the model's internal consistency has been established (Nunnally & Bernstein, 1994). For the convergent validity, those items were retained that had standardized factor loadings above .70 and were significant (Bagozzi & Yi, 1988; Hair et al., 2010). In the social sciences, sometimes researchers obtain low loadings of the constructs (Hulland, 1999). Nevertheless, consistent with the recommendations of Hair et al. (2017), for analysis, we omitted all the items with standardized factor loadings less than .70 (see Supplemental Table SII). Table 2 shows that the values of average variance extracted (AVE) for all constructs are above the threshold (≥ .50). Thus, results show that the criteria for convergent validity were met.
Furthermore, we evaluated the discriminant validity of the scale on the basis of Fornell and Larcker's (1981) criteria (see Supplemental Table SIII). If the square root of the AVE of the latent construct (reported on the diagonal) is higher than correlation coefficients in the corresponding rows and columns, then discriminant validity is established (Felipe et al., 2016). However, in case of low loadings, Fornell and Larcker's criteria do not perform well to assess the discriminant validity (Henseler et al., 2016). Since in the current study, few loadings were low, therefore, we used the correlation between Heterotrait-Monotrait (HTMT). For HTMT distribution statistics, we applied 5,000 bootstrapped samples. The results showed that the confidence interval did not include 1. Table 3 shows that all HTMT values are less than the cut-off value of .85 (Kline, 2011). This demonstrates discriminant validity.
Structural Model
For hypotheses testing, we ran the research model using a 5,000-iteration bootstrapping procedure and obtained t-values and β for the model. We assessed the model fit by using standardized root mean square residual (SRMR) as recommended by Henseler and Sarstedt (2013). Our model achieved SRMR = .09, that is greater than the threshold of .08 (Hair et al., 2017). Per recommendations of Hair et al. (2017), Stone-Geisser Q2 for cross-validated redundancy index should be greater than 0 (Geisser, 1974). The results of blindfolding process reported that the value of Q2 is greater than 0. This implied that the model has good predictive power; for details see Table 4.
Hypotheses Testing
Table 5 shows the Pearson correlation coefficients, while Table 4 shows the results of our hypotheses that four factors of justice predict overall fairness perceptions, with an expected negative impact of IntJ perceptions in the context of PA system. Consistent with the recommendations of Hair et al. (2017), we tested the hypotheses by employing bootstrapping procedure (5,000 iterations, bias-corrected, 95% confidence intervals). Results presented in Table 4 shows significant direct paths between DJ and OFPAS (β = .20, t = 1.97, p < .05, f2 = .03), PJ and OFPAS (β = .20, t = 2.85, p < .05, f2 = .03), InfJ and OFPAS (β = .34, t = 2.15, p < .01, f2 = .06), and IntJ and OFPAS (β = –.27, t = 2.65, p < .01, f2 = .06). Of importance is that these results substantiate our assumption that IntJ has a negative relation with overall fairness in the context of a PA system. Thus, the results confirm our H1d. DJ, PJ, and InfJ displayed a positive relation with OFPAS. Thus, our hypotheses H1a, H1b, and H1c also are confirmed1.
DISCUSSION
Integrating Johns' (1999) theory of self-serving behavior with OJ theory (Greenberg, 1990, 1993a, 1993b) we set the main objective of this paper to study the influence of four factors of justice on OFPAS in the public sector. The paper furnishes the following key findings:
First, as proposed, IntJ negatively predicted OFPAS. Consistent with John's (1999) theory of self-serving behavior, appraisees' perceptions of OFPAS are contingent upon the personal costs they incur with respect to their performance ratings and subsequent outcomes—e.g., pay and promotion. Therefore, appraisees' perceptions of OFPAS can emasculate IntJ. Moreover, DJ, PJ, and InfJ positively predicted OFPAS. These effects emerged in a Pakistani context, thus extending insights into the dynamics of organizational justice to a non-Western culture (see Supplemental Material).
Although IntJ is necessary, and appraisees do desire respectful treatment from their supervisors (Colquitt et al., 2001; Folger & Bies, 1989), interpersonal affect is a reality in the supervisor-subordinate relationship and is likely to influence the PA rating process (Robbins & DeNisi, 1998). Kallejian et al. (1953) argued that an increased number of positive interactions between the appraiser and the appraisee make the appraiser prone to leniency. Under these circumstances, PA ratings are contingent upon personal relationships instead of actual work performance (Ohemeng et al., 2015), implying that the appraiser gives unfairly inflated PA ratings to the appraisee (Aguinis, 2013). Therefore, sympathetic behavior of the appraiser may be appreciated on an interpersonal level yet at the same time prompt suspicions about the OFPAS in the public sector (J. Kim, 2016). Of importance, this observation does not necessarily mean that appraisees also experience decreased satisfaction. For example, previous research reveals that people consider overpayment as unfair, yet experience positive emotions about such an unfair resource distribution (van den Bos et al., 1998). In a similar vein, our findings reveal that high IntJ is associated with lower overall fairness, but this does not mean that higher IntJ also decreases appraisees' satisfaction. Due to compassionate treatment of the appraiser, the appraisee may be quite satisfied yet perceive the appraiser as biased, thereby violating the neutrality doctrine, and undermining the legitimacy of the PA system in the public organization.
Second, among the three factors of justice (i.e., DJ, PJ, and InfJ), InfJ significantly predicted overall fairness. Interestingly, InfJ emerged as the strongest predictor of appraisees' OFPAS. This finding reveals that, different from previous research that reported PJ as the strongest predictor of appraisees' reaction toward the PA system (e.g., Bies & Moag, 1986; Cohen-Charash & Spector, 2001), sometimes InfJ can be regarded most influential toward PA effectiveness. In addition, appraisees' are more concerned about InfJ in the PA system of the sample organizations, possibly because of the limited flow of PA-related information in the PA system, which is frequently regarded as “confidential.” Furthermore, due to confidentiality of the system, there is no arrangement of appraisal review meetings in which appraisers discuss appraisal ratings with appraisees. Research suggests that successful functioning of a PA system in the public sector is contingent on transparency and openness (Ireland et al., 2003). Research by Folger et al. (1992, p. 142) illustrates that “feedback given on a regularly recurring and timely basis” increases employees' fairness perceptions. Subsequently, employees strive to improve their performance (Iqbal et al., 2019). In line with this, recently, organizations are shifting toward the conversation days to provide regular performance feedback to employees (Dessler, 2020).
Thus, OFPAS in the participating organizations could be improved by increasing the communication between the appraiser and the appraisee in terms of performance feedback; providing greater access to PA-related information; and making the PA system more open than is currently the case.
Finally, DJ and PJ perceptions positively influenced employees' OFPAS. As it turned out, both factors of justice had a similar effect size on overall fairness perceptions. This finding is consistent with prior research that accorded both DJ and PJ as cornerstones for overall fairness (see Leventhal, 1980; Leventhal et al., 1980). Further, these findings are consistent with Lind and Tyler (1988) who asserted that both PJ and DJ are important in ensuing OFPAS. This implies that appraisees are concerned about both types of justice. Pertaining to DJ, various important decisions are made because of appraisal ratings; therefore, employees are always conscious about the fair distribution of PA outcomes (Murphy & Cleveland, 1995). Holbrook (1999) argued that when appraisees receive favorable outcomes—i.e., high PA ratings—they perceive distributive fairness. Similarly, appraisees also care for PJ that reflects the standards followed and processes applied for making these decisions (Folger & Cropanzano, 1998). Hence, both factors of justice, DJ and PJ, could contribute in enhancing the OFPAS.
For PA of public servants in participating organizations, the same appraisal rating form is being used for officials working in the same job grades across all departments. However, few technical job positions are an exception. Hence, there is a general form for jobs that lack objective assessment criteria. This practice is opposite to recommendations of previous research. For example, Lee (1985) suggested that PA rating forms should be in accordance with the nature of appraisees' job, and if this is not in accordance with the nature of appraisees' job, then it would not be possible to make the PA system more effective. Therefore, a valid and reliable PA rating form may improve overall fairness perceptions of appraisees. In so doing, the subsequent PA ratings may help to establish DJ to enhance appraisees' overall fairness perceptions of PA system.
Theoretical and Managerial Implications
Our article has implications for researchers as well as managers. Overall, the most significant theoretical implication of the study is that it reveals a negative impact of IntJ on OFPAS. Generally, it has been assumed that there should be high IntJ, that is, the appraiser should treat the appraisees with respect and dignity (Cohen-Charash & Spector, 2001). Indeed, respectful treatment is inevitable for IntJ. However, the findings presented here suggest that at least in the specific context of PAs, the appraiser should not be tempted to exercise unwarranted lenience for the purpose of maintaining good interpersonal relationships. If the organization's goal is to maximize overall fairness, the appraiser must be judicious in following and implementing established standards of PA system in an organization. Naturally, we did not assess the implications of this for other important variables, such as identification with the organization, positive and negative affect, or organizational citizenship behaviors. Further research is therefore necessary to more precisely establish the positive and negative implications of establishing high IntJ in a PA system.
Another noteworthy implication of the present research is that it has examined the relationship among four factors of justice (i.e., distributive, procedural, interpersonal, and informational) in Pakistani culture. OJ researchers have recommended that OJ dynamics be tested in cultures other than the United States (McFarlin & Sweeney, 2001). More generally, it has been noted that research findings within the social sciences rely too heavily on samples that are “WEIRD” (Western, Educated, Industrialized, Rich, and Democratic; Henrich et al., 2010). As noted by Henrich et al. (2010), these samples are not representative, and possibly quite unusual, in comparison to the global human population. As such, organizational psychology in general, and research on PA in the public sector and OJ theory in particular, may benefit from more research conducted in different cultures than the United States and Western Europe.
Finally, the results of our article may be informative for public servants and top management in the public service of Pakistan. Our article found that InfJ perceptions are more conducive for overall fairness of the PA system than IntJ. Hence, the authorities can design interventions for crafting the appraisal system open in order to ensure InfJ.
Limitations and Directions for Future Research
The contribution of our study could be viewed with some limitations—e.g., its cross-sectional design. Data were collected at a single point in time, which limits conclusions of causality. Moreover, there can be a risk of common method variance, as the independent and the dependent variables were obtained from the same source (Podsakoff et al., 2003). These limitations, however, do not explain or compromise the main contribution of this article, namely that IntJ sometimes is a negative predictor of overall fairness.
To the best of our knowledge, there is hardly any study that has been conducted previously on appraisees' (public servants') fairness perceptions of PA system in Pakistan. Therefore, in the public service of Pakistan, more research should be conducted to determine employees' fairness or justice perceptions about various outcomes/benefits such as pay, promotion, training, and development. In the current research, we have studied the fairness perceptions of a PA system with four factors of OJ. However, future research may be conducted to identify the problems of a PA system by considering the external environment—hat is, the social justice concerns that exist outside the organization.
CONCLUSION
To conclude, the results of our study suggest that the four factors of justice in a Pakistani sample predict OFPAS. Therefore, our study fits into the OJ research tradition by stipulating that the four justice factors work in concert to predict OFPAS. Moreover, in our study we observed that high IntJ negatively influences the OFPAS. To increase appraisees' OFPAS, appraisers should be instructed to avoid lenient and generous ratings and make assessment on actual job performance. In the absence of objective assessment criteria in the public sector, appraisers evaluate appraisees' performance based on their personal judgments. In such circumstances, appraisers usually assign high ratings, which is against the fairness of the system. Therefore, such rating could not be justified by appraiser's efforts to maintain a good relationship with appraisees. Hence, biases in thoughts or actions of authorities can undermine the legitimacy of the PA system, and decrease ratings of overall fairness. Being mindful of the multifariousness of both PA and OJ theories, however, we call for vigilance that our argument (and the results of this article) may not be employed to deduce that employees should be mistreated, humiliated, or ridiculed. Instead, we suggest that employees should not be given preferential treatment but instead be treated normally, and evaluated based on clearly defined performance criteria. High IntJ in the context of PA system—i.e., influenced by interpersonal affect—can be critical as well as atypical in the specific context of evaluating the fairness of an organization's PA system.
Contributor Notes
MALIK IKRAMULLAH is an associate professor in the Department of Management Sciences, COMSATS University Islamabad, Islamabad, Pakistan. He received his PhD in Management Studies from Gomal University Dera Ismail Khan in June 2012. He served as a guest researcher in the Department of Industrial Organizational Psychology Vrije University Amsterdam, the Netherlands. Dr. Malik is also a Higher Education Commission Indigenous Scholarship Holder and an International Research Support Initiative Program Fellow. Dr. Malik's research is focused on performance management system, organizational justice, and organizational effectiveness. His work can be seen in Personnel Review, Social Justice Research, International Journal of Productivity and Performance Management. He can be contacted at: malik.ikram@comsats.edu.pk
JAN-WILLEM VAN PROOIJEN is an associate professor of psychology at Vrije University Amsterdam, senior researcher at the NSCR, and endowed professor of radicalization, extremism, and conspiracy thinking at Maastricht University. He received his PhD in 2002 from the Department of Social and Organizational Psychology at Leiden University. His research interests include the social origins of injustice, belief in conspiracy theories, and the roots of political extremism. Dr. van Prooijen's earlier work can be seen in renowned journals, e.g., Journal of Personality and Social Psychology, Personality and Social Psychology Bulletin, Journal of Experimental Social Psychology, Group Processes & Intergroup Relations, Personality and Individual Differences, Social Psychological and Personality Science, European Journal of Social Psychology, British Journal of Social Psychology, European Review of Social Psychology, Social Justice Research, and European Journal of Work and Organizational Psychology. Dr. van Prooijen has also authored a book entitled Power, Politics, and Paranoia: Why People Are Suspicious of Their Leaders. He can be contacted at: j.w.van.prooijen@vu.nl
MUHAMMAD ZAHID IQBAL is a professor in the Department of Management Sciences, National University of Modern Languages, Islamabad, Pakistan. He has a PhD in Human Resource Development from National University of Modern Languages, Islamabad, Pakistan, and post doctorate from University of Liverpool Management School, Liverpool, United Kingdom. He teaches a number of graduate courses in the area of human resource development. He is actively involved in research in the areas of training and development, performance appraisal, and conflict management. Dr. Iqbal's earlier work can be found in Journal of European Industrial Training, Total Quality Management & Business Excellence, Personnel Review, Career Development International, and International Journal of Management Reviews. Dr. Iqbal is the corresponding author and can be contacted at: mzahidiqbal@numl.edu.pk
BAHADAR SHAH received his PhD from the University of Warsaw, Warsaw, Poland. He is a retired professor and during his service he has been an initiator of PhD programs (Management Sciences) in different universities of Pakistan. He served as acting Vice Chancellor at Hazara University, Mansehra. Moreover, his tenure of service has revolved around holding other key positions like dean, registrar, principal, and chairman in different universities of Pakistan. His research interests lie in administrative sciences and he has authored over 50 peer reviewed articles. He can be contacted at: bahadarshah@gmail.com
FAQIR SAJJAD UL HASSAN is an assistant professor at the Department of Management Sciences, Khushal Khan Khattak University Karak, Pakistan. He has received his PhD in Management Studies from Gomal University Dera Ismail Khan in June 2012. Dr. Faqir is also an HEC Indigenous Scholarship Holder. Dr. Faqir teaches courses of human resources management and quantitative research methods at graduate level. Dr. Faqir's research interest is in organizational behavior and his research work has been published in Journal of Health Organization and Management, Human Service Organizations: Management, Leadership & Governance and Personnel Review. He can be contacted at: faqirphd@yahoo.com
