Fairness of Performance Appraisal and Organizational Commitment

This study is conducted to examine the perception of public servants towards the fairness of performance appraisal and its effect on organizational commitment. This study also examined the intermediary effects of satisfaction in these two relationships. The data for this study were obtained through a survey among 425 employees of government agencies. The findings show that perceived fairness of performance appraisal has influenced their commitment towards organization through the mediating factor of satisfaction. This finding is consistent with the efforts by the government to establish a more transparent and more accountable decision-making process in an organization. In conclusion, to improve performance evaluation to be more effective in influencing organizational commitment, satisfaction of the civil servants as well as fair performance management within the organization should be given priority.


Introduction
Generally, the structure of the Malaysian government consists of three levels namely the federal government as the central government, the state governments and the local governments.Malaysian public servants serve in various government agencies at these three different levels of government forming the largest population of workforce in the country providing support to economic growth.Hence, emphasis on the efficiency and effectiveness of the civil service would greatly impact the national competitiveness in the global economy.Noting the great impacts by civil servants' performance, the government would definitely require commitment by public officials to his job and the organization.This has in fact been recognized that organizations can realize all the functions and objectives if they have committed workers (Riketta, 2002).
Recently, the relationship between organizational justice and organizational commitment has been widely researched.Several studies support the notion that fairness of performance appraisal is saliently related to employees' commitment, satisfaction and motivation to their organization (Morrow, 2011;Abdul Shukor et.al, 2008 andColquitt et al., 2001).According to Meyer et al. (1997), perception of fairness illustrates that the organization is committed to its employees.If people feel that the decisions are fair, they would respond with commitment, higher satisfaction and would be more willing to involve in 'extra-role behavior' (Colquitt et al., 2001).In fact, researchers such as Cook et al., (2004) argue that the evaluation performance will not be effective if it is perceived as unfair by those involved in the evaluation process.Since employees' perception of justice is important to the performance outcomes, management of the organizations was advised to have the ability to change the employees' perception of the performance evaluation (Tang et al., 1996).However, the current literatures are still unclear in giving explanations of the form of these relationships.For example, Masterson et al. (2000) argues that the evidence on the impact of justice in the workplace is still lacking and more research is needed to highlight the issue.Researchers like Colquit (2001) calls for more researches to further strengthen the study of justice so that important aspects such as the definition, causes and effects of organizational justice can be holistically determined.As the process of assessing employee performance are very important determinants of the organizational justice (Greenberg, 1986;Folger, et al., 1992), the issue relating to the performance appraisal is relevant to be studied.
Consequentially, the main objective of this study is to investigate the effect of fairness of performance appraisal on organizational commitment.As postulated by Jawahar (2007) and findings from previous studies, the effectiveness of the performance appraisal is also influenced significantly by the reactions to the performance appraisal.Therefore, the other objective is to explore the impact of employees' satisfaction of performance appraisal on these relationships.Based on the discussion of the literatures, the theoretical framework that guides the present study is presented in figure 1.This model predicts that the fairness of the performance appraisal has a positive relationship with organizational commitment.It is also predicted that the satisfaction in performance appraisal mediates the relationship between fairness of the performance appraisal with the organizational commitment.
Figure 1.Conceptual framework of the relationship between performance evaluation justice and organizational commitment

Literature Review and Hypotheses
Many of the definitions of organizational commitment focus on commitment-related behaviors and in terms of attitude (Mowday et al., 1979).According to Mowday et al, (1979), the organizational commitment is defined as the relative strength of an individual's identification with and involvement in a particular organization.Employees who have strong organizational commitments are characterized as having strong belief in the values and goals of the organization, willing to step up efforts in the interests of the organization and have a strong desire to remain in the organization.This model of commitment is also known as affective or attitudinal commitment that explains the organizational commitment from the perspective of the employees' attitude towards the organization (Mowday, 1998;Allen et al., 1990, Mathieu et al., 1990).Another popular definition is based on the model of organizational commitment developed by Allen et al., (1990), which consists of three dimensions i.e. affective, continuance and normative commitment.They also used the term affective commitment to characterize Porters' views of the construct (Panggabean, 2001).
Authors such as Cullen, et al., (2003) and Elliot (1994), claim that organizational commitment is related to the organizational phenomenon such as citizenship behavior, achievement and productivity.For instance, Elliot (1994) stresses that organizational commitment will affect productivity because employees are willing to work hard and Riketta (2002) argues that committed employee is assumed to be more determined to work with and has a low turnover intention (Elliot, 1994 andSarminah, 2006).As a result of low turnover, organizations are more capable in reducing operational costs in managing human resources in terms of recruiting, retaining and training.
Committed workers are also more ethical as they are not missing from work (Mathieu et al., 1990) and more supportive of the organizational values (Randall et al., 1997)

Fairness of Performance Appraisal and Organizational Commitment
Process in evaluating the performance of employees is one of the most important determinants of organizational justice (Greenberg, 1986;Folger et al., 1992).Although researchers argue about the category of fairness, there is a general consensus that organizational justice consists of at least two components, namely distributive and procedural justice (Arif et al., 2011).The procedural justice is the perception of workers that procedures used to evaluate their performance is fair while distributive justice means that performance or rewards received from the use of these procedures is fair (Greenberg, 1990;Colquitt et al., 2001).Further review in the literatures indicates that fair practices in human resource management, particularly in terms of performance appraisal has a predictive role in the employees' attitude such as the organization's commitment (Jehad et al., 2011).Performance appraisal is viewed as an important mechanism for changing employees' attitude and behaviors such as affective commitment (Morrow, 2011).Moreover, a study by Tam (1996) indicates that the commitment of government employees is higher in the federal government agencies that implement a fair and transparent assessment.
Therefore, based on the above discussions, the study proposes the following hypothesis: H1: There is a significant relationship between fairness of performance appraisal with the organizational commitment.

The Intermediary Role of the Performance Satisfaction
Several studies had shown that the effect of performance appraisal on employee attitudes and behavior is through the satisfaction of performance evaluation.For example, a study conducted by Tang et al., (1996) found that elements of organizational justice are important in predicting the employees' satisfaction of the performance appraisal and commitment to the organization.They suggest that managers have control over the employees' perception of the organizational justice in order to give effect to their satisfaction, commitment and participation in the organization.Another study conducted by Panggabean (2001) supports the view that satisfaction of the performance appraisal is important in linking the perception of fairness of performance appraisal with the organizational commitment.Similarly, a study by Thurston (2001) found that the relationship between fairness in performance appraisal with organizational citizenship behavior is mediated by the satisfaction of the former.Therefore, based on the above arguments, the following hypothesis is proposed: H2: Satisfaction in performance evaluation serves as a mediator of the relationship between fairness of performance appraisal with organizational commitment.

Population and Sampling
Generally, this study focuses on public servants working in the federal departments appointed by the Public Service Commissions.These departments implement the policies for a ministry at the federal level.Other government services such as armed forces, judicial and legal services; police, education and state public service has their own council or commissions and are not included in this study.
Cluster sampling is one of the probability sampling where in this method, groups (i.e.offices), not individual is selected (Salkind, 2000).This study has identified 48 public departments to form a group or cluster.Out of the 48 departments, 34 have a number less than 2,000 employees (small), 11 departments have number of employees ranging between 2,001 to 5,000 (medium) and 3 departments have more than 5,000 employees (large).A total of 9 departments was randomly selected according to size categories from these groups comprising of 5 small departments, 2 medium-sized departments and 2 large departments.Selection according to the size of the organization was to prevent the sample from being unrepresented.At the second level, 12 states have been clustered to three zones; East, North and South of Peninsular Malaysia.Next, three states were selected to represent their respective zones bringing the total samples to 27 units.
A total of 805 questionnaires were randomly distributed to selected respondents in 27 federal departments (representing 9 major departments in 3 states in Peninsular Malaysia).Out of the 614 (76 percent) questionnaires returned, a total of 425 usable questionnaires representing 25 departments were used for analysis.This amount is in excess of the specified sample size of 400 (Sekaran 2000).

Pilot Study
A pilot survey was conducted involving a total of 45 respondents in a public institution.Scores of alpha reliability coefficient (α) for each dimension of fairness in performance appraisal are in the range of 0.738 to 0.961 and 0.695 to 0.918 for variables of satisfaction in performance evaluation.The score of the alpha reliability coefficient (α) for organizational commitment is 0.928.

Research Instruments
The study involved three research instruments i.e the fairness of performance appraisal, the satisfaction of performance evaluation, and the organizational commitment.Instruments under fairness of performance appraisal comprises of two components of justice i.e. procedural justice and distributive justice.The instrument contained 56 questions that were originally built by Thurston (2001) and later used by Walsh (2003).It is measured using a 5 point Likert scale from 1 to 5 where 1 = strongly agree, 2 = agree, 3 = not sure, 4 = disagree and 5 = strongly disagree.
Instrument for satisfaction of performance appraisal consists of 16 items that measure the performance appraisal system satisfaction, satisfaction rating received and the satisfaction of the supervisor.The empirical study was done by Thurston (2001) and Walsh (2003) to prove its validity.It is also measured using a 5 point Likert scale from 1 to 5 where 1 = strongly agree, 2 = agree, 3 = not sure, 4 = disagree and 5 = strongly disagree.Likert scale is 'continuous' scale which has been adopted by most researchers in the field of management in the United States and Europe and is also recognized in the leading journal publications.
This study also uses scaling of 'Organizational Commitment Questionnaire' or better known as the OCQ.The measurement of OCQ is widely used by other researchers and is highly reliable when used in Asia (Zakaria, 1987 and1988).The questionnaires consist of 15 questions which was initially developed by Mowday et al. (1979) and is a unidimensional measurement to assess employees' attitudes towards the organization in three main aspects namely; (i) receiving strong beliefs and values and goals of the organization; (ii) willingness to meet their responsibilities in the interest of the organization; and (iii) strive to continue to serve the organization.
A study by Tam (1998) shows that OCQ Scale has a 0.88 of Alpha reliability coefficients and 1.05 of the standard deviation.

Characteristics of Respondents
Generally, the proportion of respondents in terms of ethnics background and category of positions (i.e.Support groups) almost matching the characteristics of the population.Most of the respondents are Malays (96.9 percent), occupies the position in the support category (86.4 percent), aged less than the 45 years (66.3 percent), and were married (82.1percent).Most of them also hold High School Certificate of Malaysia (STPM) (68.9 percent).The proportion of respondents by gender is almost equal in which the male respondent is 214 people (50.4 percent) and the number of women is 211 people (49.6 percent).Most respondents had also served more than 5 years (60 percent).A total of 134 respondents (32.5 percent) involved in supervision tasks including conducting the performance evaluation.

Statistical Analysis
Path analysis was used to test hypotheses about the direct effects of performance evaluation justice on organizational commitment.Path analysis was also used to determine the role of performance appraisal satisfaction as mediator.To see the effects of satisfaction as mediator, the analysis was conducted using multiple regression methods and Sobel Test as suggested by Baron and Kenny (1986), Kenny (2012a) and Kenny (2012b).
Through the use of multiple regression method, at the first stage, the analysis should indicate that the independent variable X affects a dependent variable Y (path c).Similarly, the second stage of analysis needs to demonstrate the effect of independent variable X on the mediator M (path a).A third level of analysis should also indicate that the mediator M variable affects the dependent variable Y, when the effect of X on Y is controlled (path b).Furthermore, in the fourth stage of statistical analysis, the effect of independent variable X onto the dependent variable Y will disappear when the mediator M is controlled (path c ').In brief, the relationship is shown graphically as in figure 2: Subsequently, the Sober Statistical Test was performed to determine whether the association between independent and dependent variables were significantly reduced when a mediator variable is included.
Comparison is done to the relationships between the independent variable with a dependent variable before and after the inclusion of a mediator.In short, this test will assess whether the effect of a mediator is significant by treating the ratio value as Z test (i.e., larger than 1.96 in absolute value is significant at the level of .05)based on the following formula: (1) Wherein a is the regression coefficients (unstandardized) for the relationship between the independent variables X with the mediator variables M, b is the regression coefficient (unstandardized) for the association between a mediator and the criterion variables, Sa is the standard error for the relationship between the independent variable with mediator variables, and Sb is the standard error for the association between mediator variables and the criterion variable.

Results
Table 1 below shows a significant correlation between the fairness of performance appraisal with performance appraisal satisfaction (r = 696, p<0.01) and with organizational commitment (r = 0.331, p <0.01).Further, a third stage of regression analysis showed that there was a direct effect on performance appraisal satisfaction on organizational commitment (B = .341;p <0.01).Subsequently the fourth stage of analysis confirmed the role of satisfaction in the performance appraisal as a full mediator factor when the effects of fairness of performance appraisal on organizational commitment is no longer significant after performance appraisal satisfaction is controlled in which the path c 'is almost zero (B = .031;p> .01;)as shown in a figure 3: Based on Sober statistical tests, statistical values obtained from the relationship between performance evaluation justice with organizational commitment is Z = 6.68835374, p = 0.000.Hence this mediator analysis confirms that satisfaction with performance appraisal serves as a mediator variable for fairness of performance appraisal in relations to organizational commitment.
Based on the results of regression analysis in table 2, the path diagram showing the relationships between independent variables and the organizational commitment to provide an insight into the role of performance appraisal satisfaction is as in figure 4: Figure 4. Path relationship between fairness of performance appraisal and organizational commitment

Conclusions and Implications
In particular, the objective is to focus on the issues of fairness of performance appraisal in enhancing the organizational commitment among public servants.Satisfaction of the performance appraisal was also included in the discussion as to confirm its implications to the fairness of the performance appraisal as indicated by the studies by Tang et al., (1996), Panggabean (2001), Thurston (2001) and Walsh (2003).
The findings of the path analysis support the hypothesis HI that the fairness of performance appraisal affects the organizational commitment and hypothesis H2 that the satisfaction of performance evaluation serves as a mediator variable in the relationship between fairness of performance appraisal with organizational commitment.
These findings confirm the study done by Panggabean (2001), Thurston (2003) and Walsh (2003) that performance appraisal is a useful mechanism for influencing attitudes towards organizations namely organizational commitment among employees when the employee accepts or is satisfied with the system and its implementation.
The concept of performance evaluation should be strengthened to achieve its objectives.As suggested by Tang et al., (1996), managers need to have control over employees' perception with a fair performance appraisal practices to give effect to satisfaction, commitment and involvement in the organization.As evidenced by this study, civil servants will be satisfied with their performance when they feel that their performance appraisal is fair and this will consequently affect their commitment to the organization.Thus, before any performance appraisal system is introduced, it is pertinent to determine the acceptance or satisfaction of the employees of such system that the system will be fairly conducted.This means, apart from the technical aspect, acceptance and employee reactions should be taken as the basis for the formulation and implementation of performance appraisal system: such action could improve its effectiveness as it can influence the employees' commitment to the organization.
This finding is consistent with the efforts by the government to make a decision-making process as more transparent and has accountability in the organization.In conclusion, to make performance evaluation more effective in influencing organizational commitment, satisfaction of the civil servants should be given priority, as well as a fair performance management within the organization.However, as noted by the low value of model fit, where the value R² is .205, the conceptual framework has to be improved by analyzing other variables that can moderate the relationship between independence variable (IV) and dependence variable (DV).Therefore, in line with previous suggestions by Masterson et al. (2000) and Colquit (2001), more researches are needed into the concept of organizational justice, especially in the area of performance appraisal.

Figure 2 .
Figure 2. Stages of analysis M acts as a full mediator if the first stage of analysis, second and third are met and where the regression coefficient c' is equal to zero.Partial Mediator occurs if the first, second and third level of analysis were met, but the regression coefficient c ' is smaller than the actual value of c.

Figure 3 .
Figure 3.The fairness of performance appraisal on organizational commitment and performance appraisal satisfaction

Table 1 .
Mean, standard deviation and correlationBased on the results of regression analysis in table 2, there was a significant relationship between fairness of performance appraisal with the organizational commitment (B = .329,p <0.01).Fairness of Performance appraisal also has a significant impact on satisfaction of performance evaluation in which, (B = .873;p <0.01).

Table 2 .
The effect of justice factor on satisfaction in performance appraisal and organizational commitment