THE INFLUENCE OF COLLECTIVISM AND RATER ERROR ON ORGANIZATIONAL CITIZENSHIP AND IMPRESSION
PHILSUNG KIM Korea Institute of Industrial Technology
JI-HWAN LEE Korea Advanced Institute of Science and Technology
Although the employer may not necessarily recognize this, organizational citizenship behavior (OCB) enhances organizational performance. However, most contemporary appraisal measures fail to distinguish between OCB and other types of employee behaviors such as impression management behavior (IMB) and in-role behavior (IRB), all of which employees could display in order to gain higher performance ratings. In this paper we distinguished empirically between OCB, IMB, and IRB, and examined their selection factors. Our regression results showed that an employee’s collectivism was positively associated with his/her OCBs. Contrary to our prediction, however, collectivism was also positively related to IMBs. In the context of human resource management, an employee’s perception of rater error had a negative effect on OCBs and IRBs. Implications for management and future research are discussed.
Keywords: collectivism, rater error, organizational citizenship behavior, impression management behavior.
Since Bateman and Organ (1983) developed the concept, organizational citizenship behavior (OCB) has emerged as a popular area for organizational behavior study. With the rapid growth of OCB theory and research, there has been a proliferation of similar concepts, such as prosocial organizational
SOCIAL BEHAVIOR AND PERSONALITY, 2012, 40(4), 545-556 © Society for Personality Research http://dx.doi.org/10.2224/sbp.2012.40.4.545
Philsung Kim, Korea Institute of Industrial Technology; Ji-Hwan Lee, Graduate School of Management, Korea Advanced Institute of Science and Technology. The authors wish to acknowledge the comments of the late Professor Jinjoo Lee on this manuscript. Correspondence concerning this article should be addressed to: Ji-Hwan Lee, Graduate School of Management, Korea Advanced Institute of Science and Technology, 85 Hoegiro, Dongdaemoon-gu, Seoul 130-722, Republic of Korea. Email: email@example.com
ORGANIZATIONAL CITIZENSHIP BEHAVIOR546
behavior (Brief & Motowidlo, 1986), civic organizational behavior (Graham, 1991), organizational spontaneity (George & Brief, 1992), intraorganizational volunteerism (Peloza & Hassay, 2006), and contextual performance (Borman & Motowidlo, 1993).
However, research on OCB and other forms of extrarole behaviors has resulted in a lack of accurate assessment of the similarities and differences between these constructs (Podsakoff, MacKenzie, Paine, & Bachrach, 2000). Along with the abovementioned constructs, impression management behavior (IMB) has received considerable attention in the organizational behavior literature (Bolino, 1999; Bolino, Varela, Bande, & Turnley, 2006; Chen, Lin, Tung, & Ko, 2008; Rosenfeld, Giacalone, & Riordan, 1995). IMB researchers have identified the tactics used by individuals to enhance their image as perceived by others (Jones & Pittman, 1982; Schlenker, 1980). Some researchers have commented that IMB may often mistakenly be coded as OCB, and likewise, OCB may often be categorized as IMB (Bolino, 1999; Schnake, 1991). This potential confusion of OCB and IMB becomes more problematic in the context of human resource management (HRM).
In this study our aim was to distinguish empirically between OCB and IMB, and to examine the relative contributions of collectivism and rater error to those behaviors.
Theory and Hypotheses
Individualism-Collectivism and Organizational Behaviors Emerging from classical origins (e.g., Durkheim, 1933; Weber, 1947), the
concept of individualism-collectivism was proposed as a predictor of individual differences by Parsons and Shills (1951), who considered it to be a bipolar construct. Whereas individualism refers to a self-orientation, an emphasis on self-sufficiency and control, and the pursuit of individual goals that may, or may not, be consistent with group goals, collectivism involves subordination of personal goals to those of a larger group and emphasizes sharing, cooperation, group harmony, and concern for group welfare (Wagner, 1995).
Organizational researchers who have studied the relationship between individualism and collectivism and OCB have gained similar results. Paine and Organ (2000) took a cross-cultural view, offering a heuristic model in which the relationship between individualism, collectivism, and OCB is shown and it is suggested that collectivistic cultures could promote OCB. Moorman and Blakely (1995) studied individualism and collectivism within an organization and suggested that collectivistic individuals are more likely than are those who are individualistic to perform OCBs. Because OCB usually requires subordination of self-interest (Organ, 1988) and because the intensity of willingness to cooperate
ORGANIZATIONAL CITIZENSHIP BEHAVIOR 547
(Barnard, 1938) varies widely among individuals, it could be expected that, in a business setting, more collectivistic employees would be more likely to perform OCBs.
Relationships can also be identified between individualism-collectivism and other types of work behaviors. Impression management is a behavior that individuals employ to protect their self-image or influence the way they are perceived by others (Schlenker, 1980). Because IMB arises from, and is primarily motivated by, self-interest (Lester & Brower, 2003), it is instinctively appealing to believe that individuals who have greater individualistic tendencies would be more likely to perform IMBs. In-role behavior (IRB) at work is another important type. IRB involves fulfilling the responsibilities specified in the employee’s job description within the organization. However, in our research we have not yet found any evidence to support a prediction of differences between individualists and collectivists in their tendency to perform in-role behaviors. Thus, we have only two hypotheses in our first set, as follows: Hypothesis 1: An individual’s collectivism will be positively associated, and individualism will be negatively associated, with organizational citizenship behavior. Hypothesis 2: An individual’s collectivism will be negatively associated, and individualism will be positively associated, with impression management behavior.
Perceived Rater Errors and Organizational Behaviors HRM practices that promote OCB have also been of interest in organization
studies (Werner, 2000). In the performance evaluation context, a rater is supposed to base decisions on information regarding a subordinate’s IRB and OCB (Werner, 1994). However, as an employee’s IMB can also affect a rater’s judgment (Kipnis & Schmidt, 1988; Wayne & Ferris, 1990), individuals may use IRB, OCB, and IMB strategically to achieve a higher performance rating. Given the limited amount of resources and capabilities available, an individual must divide and allocate them into these three types of behaviors (Bolino et al., 2006). When a performance appraisal is biased, distorted, or inaccurate, the chances of increasing the productivity of an employee are largely limited. As the cognitive components of satisfaction stem from one’s perception of fairness, people who perceive unfairness are likely to withhold or reduce their contribution to the organization, and, thus, have lower OCB (Borman & Motowidlo, 1993; Organ, 1988, 1990).
Rater error might also affect IMB. Given that IMB is intended to influence the perception of others about an individual’s perception of his or her own image (Goffman, 1959), that individual may attempt to affect the rater’s judgment when there is rater error during the appraisal process. If the rater makes errors
ORGANIZATIONAL CITIZENSHIP BEHAVIOR548
while observing and evaluating an employee, IMB can also have an effect on the evaluation results (Wayne & Ferris, 1990). In such a situation, employees may well attempt to manage the rater’s evaluation by using IMB. IRB may also be influenced by rater error. The primary purpose of appraisal is to evaluate IRB among various types of organizational behaviors. If employees perceive that the performance appraisal process is impaired by the rater’s cognitive distortion, they would decrease IRB and find another way to enhance their performance ratings. Thus, we proposed the following hypotheses: Hypothesis 3: Perceived rater error will be negatively associated with organizational citizenship behavior. Hypothesis 4: Perceived rater error will be positively associated with impression management behavior. Hypothesis 5: Perceived rater error will be negatively associated with in-role behavior.
Sample We collected data using a questionnaire survey conducted as part of a research
project at LG Information and Communication’s Research and Development Center. This organization is one of the largest and most advanced information technology research centers in Korea. Self-administered questionnaires were sent to all the qualified researchers in the center (N = 452) of which 353 complete and usable questionnaires were returned, yielding a response rate of 78.1%. The sample comprised 93.4% male (n = 330) and 6.6% female (n = 23) respondents. The ages of the respondents ranged between 19 and 41 years, with an average of 29.4 (SD = 3.47) years.
Measures Organizational citizenship behavior and in-role behavior. We assessed the
respondents’ OCB and IRB using a 21-item scale developed by Williams and Anderson (1991). The OCB items included OCBs targeting individuals (OCBI) and those benefiting the organization as a whole (OCBO). The IRB items were focused on behaviors that are recognized not only by formal reward systems but are also part of the requirements of the job descriptions. The respondents were asked to indicate the extent to which they engaged in each behavior on a 7-point Likert scale (ranging from 1 = never to 7 = always).
Impression management behavior. Seven items from Wayne and Ferris’s (1990) scale were used to measure IMB. Results of the principal components analysis indicated there were three types of impression management: job-focused, supervisor-focused, and self-focused. As some researchers (e.g., Ferris, Judge, Rowland, & Fitzgibbons, 1994) have reported that job-focused and self-focused
ORGANIZATIONAL CITIZENSHIP BEHAVIOR 549
IMBs give rise to methodological problems, we used a shortened version of the Wayne and Ferris scale to assess IMB, considering only 1 of the 3 types of behaviors: supervisor-focused impression management. Respondents were asked to rate how often they engaged in seven impression management behaviors on a 7-point scale ranging from 1 = never to 7 = always.
Individualism-Collectivism. Eight items from Wagner and Moch’s (1986) scale were used to measure two components of individualism-collectivism. Individualism-collectivism values (I-C values) pertain to the respondent’s general preferences about working in a more collectivistic environment as compared to a more individualistic environment. Individualism-collectivism norms (I-C norms) are focused on the respondent’s prescriptions for the other members’ behaviors. Respondents were asked to rate the extent to which they agreed with each item on a 7-point scale (1 = disagree to 7 = agree), where higher scores reflected the respondent’s orientation towards collectivistic tendencies.
Rater error. We developed subjective measures to assess an employee’s perception of rater error. The items representing perceived rater error were adopted from the literature on evaluation errors (e.g., Murphy & Balzer, 1989). Through semistructured interviews with a number of managers, we counted the frequency with which they had engaged in the various types of rater error. Using the data, we developed items describing situations for five types of rater error: first-impression error, halo error, personal-bias error, central-tendency error, and leniency error. Respondents were asked to indicate the extent to which they agreed with each item on a 7-point scale (1 = disagree to 7 = agree).
The means, standard deviations, and reliabilities of the scales, along with the correlations among the variables, are reported in Table 1. The reliability of every variable exceeds the minimum level of .70 recommended by Nunnally (1978).
A drawback of the methodology employed in this study is that the participating employees assessed both dependent and independent variables using self- administered questionnaires, which can potentially cause distortion because of self-serving bias and common method variance (CMV). In an attempt to control for possible CMV, ex ante, we used a questionnaire in which items for independent and dependent variables were randomly mixed. Ex post, as every correlation coefficient in Table 1 indicated a small to medium degree of correlation, our concerns about CMV could be alleviated to some extent. Furthermore, we ran Harman’s single-factor test and found that no single factor showed absolutely conspicuous explanatory power judging by the criteria suggested in the literature (see e.g., Podsakoff & Organ, 1986).
ORGANIZATIONAL CITIZENSHIP BEHAVIOR550
Table 1. Descriptive Statistics and Correlations
Variable M SD 1 2 3 4 5 6 7
1. IMB 3.68 .88 .85 2. OCBO 5.21 .72 .09 .70 3. OCBI 4.50 .65 .35*** .28*** .73 4. IRB 5.14 .76 .08 .52*** .29*** .88 5. I-C values 4.48 .95 .12* .27*** .20*** .21*** .74 6. I-C norms 4.71 .81 .22*** .29*** .29*** .20*** .11* .77 7. Rater error 4.02 .69 -.04 -.16** .01 -.15** -.21*** .00 .73
Notes: n = 353; * p < .05, ** p < .01, *** p < .001. IMB = impression management behavior; OCBO = organizational citizenship behaviors targeting organization; OCBI = organizational citizenship behaviors targeting individual; IRB = in-role behavior; I-C values = value dimension of individualism-collectivism; I-C norms = norm dimension of individualism-collectivism. The main diagonal figures in bold contain Cronbach’s internal consistency reliability estimates.
Factor analyses were conducted for all the items measuring the independent variables. Three varimax-rotated factors emerged, and the three-factor solution explained 55.4% of the total variance. Every item loaded predominantly on the predicted factor, with factor loadings ranging from .59 to .87. Thus, all 13 items were used to measure I-C values, I-C norms, and rater error, which showed reliabilities of .74, .77, and .73, respectively. We performed another factor analysis with all the items measuring organizational behaviors at work (OCB, IMB, and IRB). Four varimax-rotated factors emerged, and the four-factor solution explained 51.3% of the total variance. All except two items loaded predominantly on the predicted factor, with factor loadings ranging between .41 and .82. One of the OCBI items, despite loading on the predicted factor (.48), also loaded on the OCBO factor (.54). One of the OCBO items, despite primarily loading on the predicted factor (.47), also loaded on the OCBI factor (.47). Finally, a total of 26 items (two items excluded) were used to determine the OCBI, OCBO, IRB, and IMB scores, which had reliabilities of .73, .70, .88, and .85, respectively. Overall, the results suggest that all the scales used in the study were suitable to measure the proposed constructs distinctly.
Next, hierarchical regression analysis was conducted to test the main research hypotheses relating individualism-collectivism and rater error to the different types of organizational behaviors. Two control variables, age (in years) and gender (dummy coded), were included in the first regression analyses (Step 1). The impact of individualism-collectivism on organizational behaviors was then tested (Step 2). Finally, rater error was added to the regression equations (Step 3). The results from the hierarchical regression analyses are shown in Table 2.
ORGANIZATIONAL CITIZENSHIP BEHAVIOR 551