job performance
1 CONSTRUCT VALIDATION OF THE JOB PERFORMANCE MEASUREMENT: A CASE OF THE MALAYSIAN PUBLIC SERVICE AGENCIES Johanim Johari, Khulida Kirana Yahya and Abdullah Omar Universiti Utara Malaysia Abstract Many studies have conceptualized job performance as a multidimensional construct, however researchers have thus far provided inconclusive agreement on what factors should be included in measuring job performance. Hence, a re- conceptualization of the job performance is crucial due to the inconsistencies in the measures of this construct. This study attempts to examine the psychometric properties of the job performance constructs by integrating the task performance and organizational citizenship behavior (OCB) items. The latter, which comprised of sportsmanship, civic virtue, courtesy, altruism, conscientiousness, and innovative behavior, is conceptualized as the contextual performance of employees. A priori proposition was made that the job performance measurement could be explained by two factors, which are task and contextual performance. SPSS version 12 and AMOS 4 were used to analyze the data. Findings supported the hypothesis that job performance can be measured two factors. However, four factors of OCB i.e. civic virtue, sportsmanship, conscientiousness, and courtesy, loaded on the task performance factor, while altruism and innovative behavior loaded on the contextual performance construct. Keywords: job performance, contextual performance, organizational citizenship behavior, public service. Introduction The Malaysian Public Service is the single largest employer employing about 10.6 percent of Malaysia’s total workforce (Malaysian Public Service Department, 2007). The sizeable human capital needs to be mobilized and managed effectively and efficiently to unleash their full potential for the benefits of the nation. Energy and resources should be focused on transforming human capital in the public service by strengthening their knowledge, skills, abilities, characteristics, and competencies needed to manage and lead the public administration of the future (Malek, 2006). Reforming the public service and transforming its human capital are efforts precondition to provide excellent and high quality service to the stakeholders and its clients in the 21 st century. Until today, various measures have been taken to alleviate performance-related problems in the Malaysian Public Service. New Public Management (NPM) was introduced in the 1990s to make absolute changes in the public administration. This is to gear up public sector efficiency and performance in service quality and delivery (Siddiquee, 2006). This approach also emphasized the critical role of performance in the public sector. In 2004, Key Performance Indicators (KPIs) was introduced as a basis to benchmark the performance of the public sector and this circular requires all 2 frontline agencies to develop and implement key indicators (Malaysian Public Service Department, 2007). The Government has introduced a number of policies and programs towards reinforcing the need for a high performance workforce. Similarly in 2004, the National Integrity Plan and the Integrity Institute of Malaysia (IIM) was introduced to reduce corruption, mismanagement and abuse of power, to enhance the efficiency level in the public service delivery systems, and to improve corporate governance and business ethics (Malaysian Public Service Department, 2007). Notwithstanding the tremendous reformation being made and great emphasis being placed on performance, the public service has so far failed to bring about the desired results, particularly in the service quality and delivery. In fact, there is a decline in the Malaysia Public Sector competence compared to other Asian countries. To illustrate, Malaysia has been ranked 28 th place in the World Competitiveness Report from the Geneva-based Institute for Management Development (IMD) in 2005 (IMD World Competitiveness Report, 2005). Comparatively, this ranking has dropped from the 18 th place in 2004. In 2006, Malaysia Public Sector has managed to somewhat improve the ranking to the 22 nd place, nonetheless the ranking in terms of government efficiency has deteriorated from 16 th place in 2004 to 26 th in 2006 (“IMD World Competitiveness Report”, 2006). In fact, the overall ranking has dropped slightly to the 23 rd place in 2007 (“IMD World Competitiveness Report”, 2007). Among the perceived weaknesses are lack of communication skills among civil servants, lack of leadership commitment in managing human resources at the agency level, resistance to change among the public sector employees, difficulty in handling and dealing with poor performers due to the lack of an exit policy, over-centralized decision making, lack of systematic succession planning that leads to unclear career paths, and lack of focus on human resource management by the top management in the agencies (Malek, 2006). Apparently, these problems rooted in the poor performance among the government officials. This is supported by Siddiquee (2006), who claimed that public sector in Malaysia has long been criticized for its inflexibility, ineffective accountability, and poor performance among the officials. The bleak picture is evident when the Malaysian Public Service has recorded 58 percent increase in formal complaints from the public (Siddiquee, 2006). In 2005, a total of 2,707 complaints were received with regard to the public service poor performance, such as delays in service provision, unfair actions/decisions of the administrators, abuse of power, misconduct of the officials, and failure to enforce rules. Even though the number of complaints has dropped slightly from 2,792 in 2004 to 2,707 in 2005, it should be noted that complaints on unsatisfactory quality of service have almost doubled from 2.91 to 4.43 percent (Public Complaints Bureau, 2005). Further, the overall number of complaints has dropped from 3,397 in 2006 to 2,378 until September 2007. However, the number of complaints on unsatisfactory service quality has tripled to15.60 and 15.53 percent in 2006 and 2007 respectively, compared to the preceding years (Public Complaints Bureau, 2007). Various explanations have been given by scholars on factors related to poor performance in the public service sector. For instance, Givan (2005) revealed the use of inappropriate performance indicators as the root cause while Black, Briggs, and Keogh (2001) pointed out the challenge for public servants to fulfill the need of multiple stakeholders as the core problem. Findings by George and Hegde (2004) acknowledged the role of traits and behaviors in employees, which may influence the customer satisfaction at different level. Mwita (2000) diverted the focus to the top management personnel, who are responsible to identify the ‘performance gap’ and provide avenues for the public officials to upgrade their performance accordingly. Pandey and Rainey (2006) raised the similar issue of vague organizational goals as the major hindrance to a better performance among public servants, 3 including those in the management level. In an earlier study, Pandey and Welch (2005) reported that the pervasive effect of red tape in the public sector hampers achievement at both individual and organizational level respectively. It was also noted that organizational politics and unfair reward allocation impede excellent performance of the public officials (Morrison, 2000). Taken together, Kim (2006) and Lee, Nam, Park, and Lee (2006) strongly suggested that research on job performance of public officials is considered very necessary for two main reasons: firstly, research findings in this area are thus far inconclusive; and secondly, job performance has a profound impact on the performance of the public sector as a whole. These bases prompted the need for practitioners and academia to systematically examine antecedents of job performance among the public officials. Objectives of the study Studies on job performance have adopted various different measures in capturing the performance construct. The inconsistencies of these measures may not reflect the actual factors related to the employees’ job performance. Given this, there is a need to develop a more comprehensive instrument that can capture clearly the job performance construct. Therefore, this study examined the construct validity of the task performance and OCB items as contextual performance construct in the Malaysian Government agencies setting. We examined the construct validity of the job performance instrument which is also served as a preliminary investigation of the psychometric properties of the Malay language version of the job performance instrument with a sample of respondents from the Malaysian Public Service agencies. In specific, the objectives of this study were twofold: firstly, to assess the internal consistency reliability of the job performance dimensions and the total score, and secondly, to assess the construct validity of the job performance instrument utilizing exploratory and confirmatory factor analytic procedures. The items and dimensions of job performance scale were developed and also adapted based on the existing instruments that assessed the two dimensions of job performance: task performance and OCB. Literature review Conceptualization of job performance Job performance has become one of the significant indicators in measuring organizational performance in many studies (Wall et al. 2004). Even though performance is oftentimes determined by financial figures, it can also be measured through the combination of expected behavior and task-related aspects (Motowidlo, 2003). In fact, performance that is based on an absolute value or relative judgment may reflect overall organizational performance (Gomez- Mejia, Balkin, and Cardy, 2007; Wall et al. 2004). Additionally, job analysis can also be used in developing performance standard required of each employee (Heneman and Judge, 2005). Job analysis specifies work behaviors and knowledge, skills, abilities, and other characteristics (KSAOs) required of the job incumbents. Most importantly, Wiedower (2001) and Pincus (1986) asserted that performance measure that is based on the performance appraisal items offers higher reliability in evaluating performance. Schmitt and Chan in Motowidlo (2003) categorized job performance into ‘will-do’ and ‘can- do’. The former refers to individuals’ KSAOs required in performing certain job and the latter denotes the motivation level that individuals may have in performing their work. On the 4 same ground, Motowidlo and Van Scotter (1994) pointed out that performance construct should consist of task performance and contextual performance. Both constructs are influenced by different factors, for instance job-related experience determines task performance while individual’s personality type determines contextual performance (Motowidlo and Van Scotter, 1994). In a parallel fashion, Cardy and Dobbins in Williams (2002) conceptualized performance as work outcomes that relates closely to task performance, such as the quantity and quality of work done, and job relevant behaviors that consist of behavioral aspects useful in achieving task performance (Williams, 2002). In other words, job relevant behaviors provide support in performing task-related matters. Although various ways have been adopted to measure job performance, there is a need to develop a comprehensive instrument that can capture the performance construct. As such, job performance is best measured in terms of task performance and contextual performance or job relevant behaviors needed to enhance performance-related matters. Given the very similar concept of contextual performance and OCB, this study used OCB items to measure contextual performance. Conceptualization of task performance According to Motowidlo (2003), scholars have given limited attention on the most appropriate concept of task performance despite the fact that an accurate definition of task performance or in-role performance is crucial before any interventions are made to improve human performance in organizations. In human resource management studies, task performance has been measured using a range of criterion measures, including supervisory ratings, productivity indexes, promotability ratings, sales total, and turnover rate. Although these indicators might be presumed to reflect performance at various degrees, Gomez-Mejia et al. (2007) stated that task performance should be distinguished into quality of work done, quantity of work performed, and interpersonal effectiveness. Motowidlo (2003) defined task performance or in-role behaviors as the organization’s total expected value on task related proficiency of an employee. In other words, task performance is the behaviors related specifically to performing job-related matters. Task performance can be measured in terms of the absolute value or relative judgment (Gomez-Mejia et al., 2007; Wall et al., 2004). The former is based on the figures or financial indicators, such as productivity and profitability. Relative judgment focuses on the overall performance of an employee or organization, which is based on task-related and behavioral aspects. It has also been reported that performance should be measured broadly to enhance its reliability (Chockalingam, Schmidt, and Viswesvaran, 1996) but the scope of measurement should be most specific. For example, performance measurement should be based on performance appraisal items or job analysis in order to increase both validity and reliability (Pincus, 1986; Ashton, 1998; Wiedower, 2001). According to Wall et al. (2004), most human resource management researches adopted subjective measure of performance in tapping job performance, which is most appropriately measured based on task related and behavioral aspects. Most importantly, subjective measure allows researchers to generalize the findings to a larger performance construct (Wall et al. 2004). This is in accordance to Motowidlo’s (2003) assertion that task performance is best construed as a behavioral construct because it involves psychological process that is related to selection, training, motivation, and facilitating situational processes. 5 Organizational citizenship behavior as a contextual performance The biggest challenge for employers in managing human resources is to get their employees working beyond what is stated in their job descriptions voluntarily. In fact, maximizing efforts from employees is important in sustaining competitive advantage, keeping abreast with changes, and promoting innovation (Organ, 1997). This situation demands for organizational citizenship behavior or OCB to be exhibited by all employees in the organization. Organ (1997) and Podsakoff et al. (2000) introduced organizational citizenship behavior, which is also known as the contextual performance or extra-role performance, as a prominent contributing factor to organizational effectiveness. Organizational citizenship behavior or OCB was first introduced in the early 1980s by Bateman and Organ (Organ et al. 2006). It has been defined by Organ (1988) as: “An individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system and that in aggregate promotes the effective functioning of the organization. By discretionary, we mean that the behavior is not an enforceable requirement of the role or job description that is the clearly specifiable terms of the person’s employment contract with the organization; the behavior is rather a matter of personal choice, such that the omission is not generally understood as punishable” (p. 4). In other words, OCB concerns with the positive behavioral aspects that are neither stated in job description nor enforced by employment contract. Besides contextual performance, OCB has been also coined as the extra-role behaviors or discretionary behaviors (Organ et al., 2006). When first introduced by Bateman and Organ, OCB was distinguished into general compliance that concerns with what employees should do and altruism that focuses on employees’ willingness in helping others (Organ et al., 2006). Later, Organ (1985) expanded OCB into five distinct dimensions namely, altruism, civic virtue, conscientiousness, courtesy, and sportsmanship. Following this, the concept of OCB has gone through several transformations. For instance, Williams and Anderson (1991) divided OCB into OCB-I that focuses on behaviors at individual level and OCB-O that deals with employee behaviors at organizational level. Then, Organ (1997) categorized OCB into three dimensions, which are helping, courtesy, and conscientiousness. According to Koster and Sanders (2006), OCB has also been defined as customer-service behavior or pro social behavior. However, Chiaburu and Baker (2006) stated that OCB and pro-social behavior or customer-service behavior differ markedly based on the context of the behaviors being performed by the employees. This is because OCB is about reciprocity whereby employees would engage in OCB if they perceive that their supervisors or colleagues exhibit OCB whereas pro-social behavior is the type of behaviors that should be exhibited by employees who are attending to the customers’ needs (Chiaburu and Baker, 2006). Despite numerous conceptualizations of OCB, the most scrutinized concept of OCB is based on the five dimensions by Organ (1985) namely, altruism, civic virtue, conscientiousness, courtesy, and sportsmanship. A more recent concept of OCB includes innovative behavior as one of its dimension. Moon, Van Dyne and Wrobel (2005) noted that this dimension is somewhat different from the classic definition of innovation and creativity because innovative behavior in OCB relates to the engagement level of employees in giving and adopting constructive ideas for the functional improvement of the department or organization. It has been suggested by Moon et al. (2005) that innovative behavior is a crucial important to be included and examined in the OCB construct given the need for organizations nowadays to 6 have employees that can participate actively in delivering ideas for organizational effectiveness and performance. Methodology Procedures Self-administered questionnaires were distributed to the respondents in four government agencies in the state of Kedah, Malaysia. The researchers went to each agency and gave the questionnaires personally to the chief clerk of each department, who were contacted prior to the researchers’ visit. They were briefed on the research objectives and guidelines in answering the questionnaires. A letter from the researcher stating that the research is purely academic and research results will be used for academic purposes only was also enclosed. In the study, respondents were required to evaluate one of their immediate subordinates on task performance and OCB items. Supervisory rating method is chosen over self-rating to avoid common method variance as suggested by many performance-related researchers, e.g. Yousef (1998), Jabroun and Balakrishnan (2001), Crossman and Abou-Zaki (2003), Castro, Armario, and Ruiz (2004), Kim (2006), and Koster and Sanders (2006). Most importantly, Wall et al. (2004), Tubre (2000), Organ et al. (2006), and Organ and Ryan (2001) strongly contended that self-rating performance-related construct will lead to a spuriously high or low correlation, confounded by the common method variance. Further, Bohlander and Snell (2007) and Moideenkutty, Blau, Kumar and Nalakath (2005) asserted that immediate supervisors are the most appropriate source of information with regard to job performance. As such, immediate supervisors were chosen to evaluate their subordinates on task performance and OCB items. Respondents Respondents were 100 public servants from four government agencies in the state of Kedah. A total of 61 of the respondents are male and 39 are female. Because the government agencies in Kedah are predominantly Malay-populated, all of the respondents participated in this study were Malays. Most of the respondents were the exempt employees, which made up 63.0 percent (63) while 37.0 percent (37) of the respondents were non-exempt employees. Questionnaires were given out to the respondents to answer 37 items on task performance and OCB. A total of 200 questionnaires were distributed and 103 were returned. However, only 100 questionnaires were usable for data analysis. Back translation of the items The translation of the original English version of the questionnaire into Malay language was accomplished through back translation procedure. Based on guidelines by Brislin (1970), four different bilingual language experts were employed; two for translating from the source to Malay language, and the other two experts re-translated the items into English. Decentralization was also conducted indicating that the source and the target language version are equally important during the translation procedure (Brislin, 1970). Revision indicated no discrepancies between the English and Bahasa Malaysia version of the questionnaire. We also discussed and verified the items with officers and clerical staff from the Malaysian public service agencies. 7 Measurement of job performance The job performance scale was assessed with seven subscales: task performance, sportsmanship, altruism, courtesy, civic virtue, conscientiousness, and innovative behavior. All items were rated on a seven-point Likert scale, namely 1=very disagree, 2=disagree, 3= slightly disagree, 4=moderate, 5= slightly agree, 6= agree, 7= very agree. To determine the score of this scale, ratings within each scale are summed and divided by the total number of items in that particular scale. Negative statement items on the instrument were reverse-coded so that a high score on the instrument indicates a high degree of job performance for the public servants. Task performance is based on supervisor’s relative judgment that focuses on the overall performance of an employee, which is based on task-related and behavioral aspects (Gomez- Mejia et al., 2007). All of the items were adapted from William and Anderson (1990). Sportsmanship assessed employees behaviors in which one do not complain because he or she has a positive attitude (e.g. “He/she always find fault with that the organization is doing.”), altruism measured behaviors in which employees help their colleagues to perform tasks (e.g. “He/she willingly help others who have heavy workload.”), and courtesy assessed behaviors where employees take step to prevent problems with others (e.g. “He/she does not abuse the rights of others.”). Civic virtues examined behaviors that contribute to the political aspect in the organization (e.g. “He/she keeps up with changes in the organization”) and conscientiousness measured behaviors in which employees perform tasks well beyond the minimum required level (e.g. “He/she does not take extra time for breaks.”). All items were adapted from Podsakoff and MacKenzie (1990). Finally, innovative behavior assessed the frequency of contributing ideas and engagement level among employees. The six items measuring innovative behavior were adapted from Moon et al. (2007). A sample item is “He/she tries to make constructive suggestions for improving how things operate in this department.” Results Exploratory and confirmatory factor analyses and internal reliability consistencies and mean were employed to examine the factor structure of the job performance measurement scale. In this study, items for each observed variable was used to identify each of its latent variables. Exploratory factor analysis An exploratory factor analysis was carried out to examine the factorial validity of the Bahasa Malaysia adaptation of the job performance items. Exploratory factor analysis was conducted using principal axis factoring extraction with oblique rotation and a priori criteria of two factors were extracted based on the previous studies. However, four dimensions of OCB, namely sportsmanship, courtesy, civic virtue, and conscientiousness, loaded on the task performance factor. Given this, job performance construct were labeled as task performance, (i.e. items on task performance, sportsmanship, courtesy, civic virtue, and conscientiousness) and contextual performance (i.e. items on altruism and innovative behavior). A total of 12 items were dropped from further analysis because of cross or low factor loading, i.e. less than 0.50 (Hair, Anderson, Tatham, and Black, 1998). 8 Items Factor Loading Task 1 He/she performs tasks that are expected of him/her. 0.694 Task 2 He/she meets formal performance requirements of the job. 0.681 Task 3 He/she is involved in activities that are relevant to his/her yearly performance assessment. 0.783 Task 4 He/she fails to perform essential duties. ® 0.755 Task 5 He/she adequately completes assigned duties. 0.643 Task 6 He/she does not take extra time for breaks. 0.614 Task 7 He/she often work beyond office hours even though not being asked to. 0.762 Task 8 He/she always complains about things that are not important. ® 0.756 Task 9 He/she always finds fault with what the organization is doing. ® 0.638 Task 10 He/she always pays attention to matters that are negative rather than on matters that are positive. ® 0.664 Task 11 He/she is always complaining about work. ® 0.618 Task 12 He/she tries to prevent himself/herself from creating problems for his/her coworkers. 0.874 Task 13 He/she is aware of how his/her behavior affects other people’s jobs. 0.693 Task 14 He/she reads and follows all announcements, memos, and others given out by the organization. 0.782 Task 15 He/she keeps up to date with changes in the organization. 0.527 He/she attend meetings that are not compulsory but are considered important. 0.564 Contextual 1 He/she helps others who have problems with their work. 0.701 Contextual 2 He/she helps others who have heavy workload. 0.620 Contextual 3 He/she always ready to offer help to those around him or her. 0.672 Contextual 4 He/she tries to make innovative suggestions to improve the department/organization. 0.873 Contextual 5 He/she tries to adopt improved procedures for the department/organization. 0.562 Contextual 6 He/she tries to institute new more effective work methods for the department/organization. 0.914 Contextual 7 He/she tries to make constructive suggestions for improving how things operate in this department/organization. 0.989 Contextual 8 He/she makes recommendations on issues that affect the department/organization. 0.931 Contextual 9 He/she speaks up for new changes in this department/organization. 0.716 Eigen Value 10.135 4.011 Variance Explained 40.539 16.045 Alpha 0.921 0.936 Table 1: Factor Loadings for Job Performance Reliability Job Performance # of item Mean α Task Performance 16 5.906 0.921 Contextual Performance 9 5.538 0.936 Overall Job Performance 25 5.773 0.927 Table 2 Summary Statistics for Job Performance and Cronbach’s Alpha Table 2 presents the results of the internal consistency reliability and mean, and standard deviation for the total score and each subscale. Cronbach’s alpha values are 0.921 for ‘task 9 performance’ subscale and 0.936 for ‘contextual performance’ subscale. The internal consistency reliability for the Job Performance scale was 0.927. Confirmatory factor analysis A confirmatory factor analysis was conducted to identify the validity of the job performance construct. The standardized estimates are reported to interpret parameters in the measurement model. Both absolute and incremental fit statistics were used to establish the model fit. Traditional chi-square test and the root mean square error of approximation (RMSEA) were chosen to indicate the absolute fit statistics. Figure 1: A first order measurement model for job performance. Latent constructs are shown in ellipses and observed variables are shown in rectangles. Two incremental statistics used were Tucker Lewis Index and the Comparative Fit Index. The values of Normed fit Index (NFI), Tucker-Lewis Index (TLI), and Comparative fit Index (CFI) can be between 0 to 1 and values greater than 0.90 and 0.95 reflected an acceptable and excellent fits to the data and the RMSEA values at or less than 0.05 and 0.08 indicated a close and reasonable fit. The Chi-Square Goodness-of-fit statistics with 11 degree of freedom was 12.902. The value of NFI is 0.971, GFI is 0.965, and TLI is 0.992. Figure 1 shows that all loadings of items on their targeted factors were high, statistically significant, and above 0.55 the cutoff point used in the exploratory factor analysis. Both factors are correlated with the correlation value of 0.51 suggesting that these factors are interrelated but relatively are orthogonal of one another. Discussions Findings of the study indicate that only two factors of job performance, namely altruism and innovative behaviors, fall into the OCB construct while the rest loaded on the task performance factor. Such finding is possible because of the supervisory rating approach adopted in this study. In most instances, supervisors perceive OCB item as task performance Task perf .98 TASK e1 .99 .51 CON e2 .72 .62 SPORTS e3 .79 .70 COURTESY e4 .84 .37 CIVIC e5 .60 Contextual .91 ALTRUISM e6 .96 .55 INNOVATI e7 .74 .51 Standardized estimates Chi-square= 12.902 df= 11 ratio= 1.173 p-value= .300 GFI= .965 RMSEA= .042 TLI= .992 CFI= .996 NFI= .971 -.37 .30 10 instead of contextual performance. For example, Wilson (2005) found that 94 percent of supervisors considered OCB as part of their subordinates’ task performance. Borman and Motowidlo (1997) echoed that supervisors viewed and evaluated task and contextual performance as a similar construct in appraising employees’ job performance. Vey and Campbell (2004) conducted a study among respondents with and without supervisory experience to evaluate the in-role and extra role items. It was found that 17 of the 30 OCB items were categorized by 85 percent of the respondents as in-role performance. In fact, respondents without supervisory experience tend to categorize dimensions on altruism, courtesy, and sportsmanship as in-role compared to the other group of respondents. As such, it is interesting to raise several issues pertaining to the ambiguity in understanding in-role and extra role concepts among supervisors. Prior to evaluating subordinates’ performance, it is crucial for supervisors to undergo training for the appraisal task. Bohlander and Snell (2007) noted that supervisors must be adequately trained in appraising subordinates’ performance so that their appraisals are more meaningful and directive. By understanding subordinates tasks and performance standards and objectives or purposes of the appraisal, supervisors can provide useful feedbacks to subordinates and eventually help improve performance. In other words, training appraisers can ensure better performance appraisal process which benefits both subordinates and supervisors. Another plausible explanation for the cross loadings of the OCB items is because of dimensions used in evaluating employees. Malaysian public servants are being evaluated based on several aspects i.e. task and other attributes that are similar to OCB items. This might have inadvertently influenced respondents’ perception of the items. The respondents have been trained to evaluate employees based on the items and therefore OCB items in the performance appraisal were considered as task instead of contextual performance. This is in agreement with Organ’s (1997) and Vey and Campbell’s (2004) assertions that in assessing job performance, OCB items should be included because they are integrally measuring task performance as well. Additionally, the error covariance existed in innovative behaviors and civic virtue due to the fact that both items measured employees’ contribution. The former examined contribution in terms of the political aspect while the latter measured the contribution of ideas to promote general change and improvement. Likewise, cross loading of civic virtues and sportsmanship constructs. Such finding is possible because according to Moon et al. (2005), both factors are considered as the ‘protective’ type of OCB, which occurs when employees take necessary preventive measure to protect the organization from unethical behaviors. Conclusions and recommendations Based on the findings that showed the evidence of construct validity of the job performance items, this instrument can be used in the Malaysian studies. Results of this study also reported coefficient alphas were more than 0.90 for both task and contextual performance and exploratory factor analysis indicated support for the factorial validity of the job performance scale. Such findings suggest acceptable reliability and validity of the instrument. Further, confirmatory factor analysis provided the evidence of construct validity based on tests of significance and assessment of the measurement model fit. Thus, two subscales of job performance can be useful instruments in examining job performance in the Malaysian setting. . 1 CONSTRUCT VALIDATION OF THE JOB PERFORMANCE MEASUREMENT: A CASE OF THE MALAYSIAN PUBLIC SERVICE AGENCIES Johanim Johari, Khulida Kirana Yahya and Abdullah. factorial validity of the Bahasa Malaysia adaptation of the job performance items. Exploratory factor analysis was conducted using principal axis factoring