Affective organizational commitment and citizenship behavior: Linear and non-linear moderating effects of organizational tenure

Journal of Vocational Behavior 79 (2011) 528–537

Contents lists available at ScienceDirect

Journal of Vocational Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j v b

Affective organizational commitment and citizenship behavior: Linear and non-linear moderating effects of organizational tenure Thomas W.H. Ng a,⁎, Daniel C. Feldman b a b

The University of Hong Kong, School of Business and Economics, Pok Fu Lam, Hong Kong The University of Georgia, Terry College of Business, Athens, GA 30602, USA

a r t i c l e

i n f o

Article history: Received 12 February 2011 Available online 15 March 2011

Keywords: Organizational commitment Citizenship behavior Tenure

a b s t r a c t Utilizing a meta-analytical approach for testing moderating effects, the current study investigated organizational tenure as a moderator in the relation between affective organizational commitment and organizational citizenship behavior (OCB). We observed that, across 40 studies (N = 11,416 respondents), the effect size for the relation between affective organizational commitment and non-self measures of OCB was .23. However, we also found that organizational tenure moderated the relation in a non-linear way. Before 10 years of tenure, the strength of the commitment–OCB relation increased as organizational tenure increased; after that point, the strength of the commitment–OCB relation decreased as organizational tenure increased. In short, the moderating effect of tenure follows a curvilinear pattern. © 2011 Elsevier Inc. All rights reserved.

In recent years, many organizations have tried to cultivate a workforce that is not only strongly committed to its employers but also strives for high performance (Deery & Iverson, 2005; Rousseau, 1998). In academic research, the relation between organizational commitment and employee performance has also generated a great deal of interest among scholars (Jaramillo, Mulki, & Marshall, 2005; Riketta, 2002). Compared to non-committed workers, employees with high organizational commitment have been found to have more positive job attitudes and to exhibit stronger performance on the job (Harrison, Newman, & Roth, 2006; Meyer, Stanley, Herscovitch, & Topolnytsky, 2002). Wright and Bonett (2002) argue that the commitment–performance relation deserves more attention because there are likely to be moderators of that relation. In particular, they found support for their prediction that tenure would moderate the relation such that committed workers with shorter tenure were likely to perform more strongly than committed workers with longer tenure. That is, as average tenure increases, the strength of the positive commitment–performance relation decreases. The current study extends Wright and Bonett's (2002) work in several ways. First, in the Wright and Bonett (2002) study, the focus was on in-role job performance. However, beyond core job duties, committed workers are also likely to go the extra mile and perform tasks that are not necessarily expected by others, such as helping out coworkers. Indeed, since going beyond the call of duty is largely left to the discretion of workers themselves, we might expect that the impact of commitment on these citizenship behaviors might be even greater than its impact on in-role performance. Organizational citizenship behaviors (OCB) are extremely important for promoting organizational performance (Podsakoff, Blume, Whiting, & Podsakoff, 2009). In fact, Lievens, Conway, and De Corte (2008) found that, in many situations, raters of job performance gave greater weight to OCB than to in-role performance. Relying on the same meta-analytical approach that Wright and Bonett (2002) used, we extend their work by examining the moderating effect of organizational tenure in the relation between commitment and OCB. We are aware of no prior studies which have directly examined the moderating effects of organizational tenure in that relation.

⁎ Corresponding author. E-mail addresses: [email protected] (T.W.H. Ng), [email protected] (D.C. Feldman). 0001-8791/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jvb.2011.03.006

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

529

Second, while Wright and Bonett (2002) combined both self-ratings and others' ratings of job performance in their study, here we focus solely on others' ratings of OCB (supervisory and peer ratings) and objective measures. We chose this strategy because self-ratings of job performance may be positively biased, as employees tend to rate their own job performance significantly higher than others do because of self-serving biases (Harris & Schaubroeck, 1988). Researchers have found that there is a particularly strong self-inflation bias in self-ratings of OCB (Organ & Ryan, 1995), while it can be argued that supervisors and peers may give more objective ratings of OCB. Third, while Wright and Bonett (2002) examined the linear moderating effects of tenure in the commitment–performance relation, in the present study we examine its curvilinear moderating effects, too. As we will discuss in more detail later, there are reasons to expect that the strength of the relation between commitment and performance may weaken over time, including the depletion of personal resources used to contribute extra over an extended period of time (English, Morrison, & Chalon, 2010). Beyond extending Wright and Bonett's (2002) work, the current study also attempts to advance organizational commitment theory by examining curvilinear moderating effects. Most studies of organizational commitment examine linear moderating effects. These linear moderating effects occur when the strength of the commitment–outcome relation changes in one direction (either increases or decreases) along with changes in the value of the moderator. In contrast, a curvilinear moderating effect occurs when the strength of the commitment–outcome relation increases as the value of the moderator increases, but then decreases beyond a certain value of the moderator. (Another form of a curvilinear moderating effect occurs when the strength of the commitment–outcome relation decreases as the value of the moderator increases, but then increases beyond a certain value of the moderator.) Testing for both linear and non-linear moderating effects has theoretical and practical implications for commitment research. Theoretically speaking, we cannot draw conclusions about the linear moderating effects of organizational commitment without first considering potential curvilinear effects (cf. Cortina, 1993). Practically speaking, identifying the curvilinear moderating effects on the commitment–outcome relations could help managers identify the inflection points when employees' behaviors are most likely to change. In the present study, we expect that while organizational commitment may be a strong driver of OCB at the time of entry and the first few years of employment, the strength of that relation is likely to decline as organizational tenure increases. 1. Definitional issues 1.1. Organizational commitment Organizational commitment is a stabilizing force that binds individuals to organizations (Bentein, Vandenberg, Vandenberghe, & Stinglhamber, 2005; Meyer & Herscovitch, 2001; Riketta, 2002). Organizational commitment is one of the most commonly examined attitudes in the organizational sciences literature (see Meyer et al., 2002 for a quantitative review) and has particularly interested researchers since Allen and Meyer (1990) proposed a three-dimension model of the construct. Specifically, Allen and Meyer (1990) proposed that there are three different mind-sets underlying organizational commitment: affective, normative, and continuance. Affective organizational commitment refers to employees' emotional attachment to, involvement in, and identification with their employers (O'Reilly & Chatman, 1986). Normative organizational commitment is defined as the perceived obligation to stay, with some connotations of moral imperatives to do so. Continuance organizational commitment is the extent to which employees perceive that they have to stay with their employers because the costs of leaving are too high (McGee & Ford, 1987). Following Wright and Bonett's (2002) approach, we focus on affective organizational commitment for both theoretical and practical reasons. Theoretically speaking, affective organizational commitment has greater relevance for predicting OCB than normative and continuance commitment do since the performance of OCB is often driven by positive affect and emotions rather than obligations and calculations (Lee & Allen, 2002). Practically speaking, we were able to identify fewer than 10 studies in our article search which examined the relations of normative and continuance commitment with others' ratings (not self-reports) of OCBs. This small number of studies precluded us from conducting additional moderated meta-analyses on these variables (Hunter & Schmidt, 1990). 1.2. Organizational citizenship behavior (OCB) OCB encompasses behavior which promotes the organization through strengthening and maintaining its social system (LePine, Erez, & Johnson, 2002). Researchers have documented the importance of OCB for organizational functioning over the last two decades (Organ, 1988; LePine et al., 2002). OCB (sometimes called pro-social behaviors, extra-role behaviors, or contextual performance) are not specific to an individual's own job duties, but rather support the broader organizational environment in which core performance takes place (Motowidlo & Van Scotter, 1994; Organ, 1988). Examples of OCB are compliance with organizational norms, not complaining about trivial matters, and helping coworkers. Employees' aggregated OCBs have been regularly linked to group, unit, and organizational productivity (Podsakoff et al., 2009). Multiple dimensions of OCB have been proposed over the years (e.g., Organ, 1988; Smith, Organ, & Near, 1983; Williams & Anderson, 1991). LePine et al. (2002) argue that these various dimensions are indeed measuring the same underlying, broad construct of OCB. They found that the various dimensions of OCB are closely related to each other and that their correlations with work outcomes are largely similar. Following LePine et al.'s (2002) recommendation, then, we do not make distinctions among types of OCB in this paper.

530

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

1.3. Organizational tenure In this study, we define organizational tenure as the length of employment in an organization (McEnrue, 1988; Shirom & Mazeh, 1988). One of the most frequent uses of organizational tenure in past research has been as a proxy for work experience or amount of job-related knowledge (Quinones, Ford, & Teachout, 1995; Sturman, 2003). Organizational tenure has been considered as a quantitative indicator of work experience by many researchers because, by remaining with a firm for additional years, employees develop a wider set of work skills and become more knowledgeable about the organization as a whole (Bird, 1996). Our focus in this paper is on organizational tenure rather than other types of tenures because, as we argue later, there are theoretical reasons why long organizational tenure could either increase or decrease committed employees' willingness and capacity to engage in OCB. 2. Theoretical background 2.1. Affective organizational commitment and OCB While researchers have identified a wide range of predictors of OCB, attitudinal predictors have been found to have the most consistent effects (Organ & Ryan, 1995). The theoretical arguments that are used to support such a relation are largely predicated on social exchange theory (Blau, 1964) and the inducements-contributions model (March & Simon, 1958). In formalizing social exchange theory, Blau (1964) suggests that reciprocity is an important characteristic of social relationships. Individuals who perceive that they are valued and respected are likely to reciprocate with trust and emotional engagement in their exchanges with others. In putting forth their inducements–contributions model, March and Simon (1958) suggest that organizations offer inducements for employees to join and stay with their firms, while employees make contributions to organizational effectiveness through high levels of performance on their jobs. Taken together, these two theories suggest that individuals who have a high level of affective organizational commitment will have greater intentions to reciprocate to the organization through engaging in OCB (Cropanzano, Rupp, & Byrne, 2003). Affective organizational commitment has been observed in meta-analyses to be positively related to OCB as well (LePine et al., 2002; Meyer et al., 2002; Organ & Ryan, 1995). Specifically, Organ and Ryan (1995) found that, across four empirical studies, the correlations of affective organizational commitment with the altruism and compliance dimensions of OCB were .23 and .30, respectively, although these studies combined both self- and others' ratings of altruism. Across 17 empirical studies, LePine et al. (2002) observed that the correlation between affective organizational commitment and OCB was .21; here, too, both self- and others' ratings were included in that estimate. Finally, based on eight empirical studies, Meyer et al. (2002) observed that the correlation between affective organizational commitment and supervisor-rated OCB was .27. None of these three previous quantitative reviews has addressed the potential moderating effects of organizational tenure. 2.2. Linear moderating effects of organizational tenure Although organizational tenure is frequently studied and measured (Quinones et al., 1995; Sturman, 2003), the theoretical process by which organizational tenure affects job behavior is not fully understood (Ng & Feldman, 2010). Wright and Bonett (2002) argue that it is particularly important to examine the effects of tenure in the relation between affective organizational commitment and job performance because higher tenure is likely to be positively related to both commitment and performance, thereby inflating or creating a spurious correlation between commitment and performance. Like Wright and Bonett (2002), we argue that the correlation between commitment and OCB might be spurious when organizational tenure is not considered because organizational tenure is related to both affective organizational commitment (Meyer et al., 2002) and OCB (Ng & Feldman, 2010). Unlike Wright and Bonett (2002), though, we argue that two alternative predictions could be made regarding the moderating effect of organizational tenure. On one hand, there are theoretical reasons to believe that the commitment–OCB relation is stronger for individuals with higher organizational tenure. This reasoning is based on both human capital theory (Becker, 1964) and social capital theory (Burt, 1992; Granovetter, 1973). Specifically, individuals who accrue longer years of service are likely to acquire greater amounts of human and social capital as well. Human capital consists of the skills and knowledge that individuals acquire to enhance their potential productivity and success in the labor market (Becker, 1964). By staying longer with a particular organization, workers gain more extensive knowledge about their specific organizational environment, which in turn might facilitate the performance of OCBs (Ng & Feldman, 2010). For instance, by staying longer with an organization, employees gain greater knowledge about organizational goals and therefore can make more constructive suggestions accordingly. Similarly, Slaughter, Ang, and Boh (2007) argue that firm-specific human capital (e.g., knowledge about business processes and strategies) will increase with organizational tenure. Social capital is also likely to increase with organizational tenure. Social capital is the sum of the actual and potential resources embedded within, available through, and derived from the network of relationships developed by an individual (Nahapiet & Ghoshal, 1998). Individuals who have worked longer in an organization are likely to have established greater numbers of social links and stronger links with colleagues both inside and outside their departments (Slaughter et al., 2007). In turn, these links encourage greater people-directed OCBs like courtesy and altruism. Extensive links in an organization's social network also assist employees in spreading innovation and ideas throughout a unit. Last here, individuals with longer organizational tenure, who have

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

531

developed more links with others in the work place, might be more frequently asked for help from coworkers and might be more effective in providing such assistance. Therefore, while affective organizational commitment may spark employees' initial willingness to engage in OCB, organizational tenure enhances that willingness and increases employees' ability to engage in OCB effectively. In short, this line of reasoning suggests that organizational tenure helps increase individuals' stock of human capital and social capital, which in turn facilitates committed workers' performance on extra-role tasks. Thus, the positive relation between affective organizational commitment and OCB will be stronger for those individuals with longer organizational tenure. Hypothesis 1. Organizational tenure positively moderates the effect of affective organizational commitment on OCB such that the relation is stronger for individuals with higher organizational tenure. An opposite prediction can also be made regarding the linear moderating effect of tenure on the commitment–OCB relation. Specifically, longer-tenured employees might accumulate stress over time and that accumulated stress might dissipate longer-tenured employees' energies for engaging in OCB. Super's (1957, 1980) life-span, life-space model suggests that individuals go through multiple stages of career development over the course of their lives: exploration and establishment (early career), maintenance (mid-career), and eventual disengagement (late career). In the early-career stage, the main tasks of individuals are learning a wide assortment of skills, establishing relationships with others at work and in the industry, and advancing to new levels of responsibilities (Super, Savickas, & Super, 1996). However, as individuals enter the mid-career (or maintenance) stage, their major concerns become preserving their self-concepts, holding on to career achievements already attained (Gibson, 2003), and achieving better work-nonwork balance (Williams & Savickas, 1990). Indeed, as individuals enter the late-career (or disengagement) stage, their career trajectory is clearly towards less involvement in work activities and more involvement in family and personal life activities (Adler & Aranya, 1984). Consequently, we propose that employees with very long organizational tenure might have less interest or willingness to engage in tasks beyond their required job duties. In addition, employees with very long organizational tenure might have less physical and emotional energy to engage in OCB (Luthans, Avolio, Avey, & Norman, 2007). Engaging in OCB places greater demands on employees and, over time, their psychological, emotional, and physical resources to engage in OCB might diminish. Previous research has shown that longer tenure is associated with greater job boredom (Drory, 1982; Kass, Vodanovich, & Callender, 2001) and less favorable perceptions of workplace environments (English et al., 2010; Huang, Shi, Zhang, & Cheung, 2006). Thus, the link between affective organizational commitment and OCB may become weaker as tenure increases, since engaging in OCB takes an increasingly heavy toll on employees over time (Bolino & Turnley, 2005). Indeed, Wright and Bonett (2002) use this argument to explain the moderating effects of tenure observed in their study. Specifically, they proposed that employees with shorter tenure are usually more motivated to work and usually derive greater satisfaction from doing so. As such, for these employees, there is a positive relation between affective organizational commitment and in-role job performance. In contrast, Wright and Bonett (2002) argue that longer tenured workers are more likely to experience burnout and, as a result, feel less motivated to work as time passes. In Hypothesis 2, we propose that the same dynamics (a weaker relation between affective organizational commitment and performance as organizational tenure increases) will hold for OCB as well. Hypothesis 2, then, represents a competing hypothesis to Hypothesis 1; that is, it proposes a negative linear moderating effect of organizational tenure. Hypothesis 2. Organizational tenure negatively moderates the effect of affective organizational commitment on OCB such that the relation is weaker for individuals with higher tenure.

2.3. Curvilinear moderating effects of organizational tenure There is yet a third plausible hypothesis regarding how organizational tenure moderates the relation between affective organizational commitment and OCB. In this study, we also explore whether organizational tenure has a curvilinear moderating effect. That is, we examine whether the strength of the commitment–OCB relation increases as organizational tenure increases up to a certain number of years, but then decreases after that inflection point has been reached. To date, the curvilinear moderating effects of organizational tenure on the commitment–OCB relation have not been tested. There is some related evidence, though, which suggests that organizational tenure might have a curvilinear moderating effect on this relation. For example, Ng and Feldman (2010) observed that organizational tenure is curvilinearly related to OCB. Specifically, the corrected correlations for the tenure–OCB relation were .15 for employees with less than 3 years of tenure, .09 for employees with 3–6 years of tenure, .05 for employees with 7–10 years of tenure, −.01 for employees with 11–14 years of tenure, and −.00 for employees with more than 14 years of tenure. Ng and Feldman's (2010) study, then, raises the possibility that the moderating effect of organizational tenure in the commitment–OCB relation might be curvilinear in nature as well. Peter and Hull (1969) also argued that organizational tenure and performance are not linearly related. They suggest that employees get promoted up organizational hierarchies until they land in positions for which they do not have adequate skills and from which they can no longer get promoted. As a result, organizations have large numbers of long-tenured employees in positions which they cannot perform well. Peter and Hull's (1969) thesis, then, is that the relation between organizational tenure and performance will weaken over time.

532

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

Certainly, Ng and Feldman's (2010) findings and Peter and Hull's (1969) arguments focus on the curvilinear nature of the relation between organizational tenure and job performance rather than on the moderating effect of organizational tenure on the commitment–performance relation. Nonetheless, their work suggests that such a non-linear moderating effect is indeed plausible. Examining curvilinear moderating effects might also help reconcile the two competing perspectives we have proposed earlier (Hypothesis 1 vs. Hypothesis 2). Specifically, the positive moderating effect of tenure (predicted in Hypothesis 1) might be evident in early years of service, as relative newcomers strive to enhance their human and social capital in the firm. However, as veteran employees accumulate more and more stress, longer-tenured employees' willingness or ability to engage in OCB might decline (as predicted in Hypothesis 2). 3. Method 3.1. Literature search We performed a comprehensive search for those field studies published in or before 2010 which examined the relation between affective organizational commitment and OCB. Multiple keywords were used in the electronic database search, such as organizational commitment, affective commitment, attitudinal commitment, citizenship behavior, prosocial behavior, extra-role behavior, and contextual performance. As recommended by Rosenthal (1979), we also searched for unpublished studies and dissertations and utilized numerous research databases, including Dissertation Abstracts International, EBSCOHost, Emerald, Factiva, JSTOR, Oxford Journals, Proquest, PsycINFO, ScienceDirect, Sage Full-Text Collections, and Wiley InterScience databases. In addition, the reference lists of all the articles initially identified were examined to locate additional relevant articles. Several inclusion/exclusion criteria were used. First, we only included studies that were conducted in natural or field settings; experimental studies in which the extent of OCB was influenced by experimenters' manipulation were excluded. Second, we included studies that examined both affective organizational commitment and OCB at the individual level of analysis; studies that examined these constructs at the group or organizational level were excluded. Third, we included articles that reported correlations between affective organizational commitment and OCB or provided sufficient information for conversion of effect sizes into correlation figures. Fourth, we excluded studies that examined OCB in classroom settings. Finally, because self-ratings of OCB might be biased by individuals' tendency to inflate the level of their performance (Organ & Ryan, 1995), we included only those studies which used others' ratings of OCB (ratings provided by supervisors and peers) or objective measures of OCB. These criteria led to a total of 57 published articles, containing a total of 64 independent studies with a total of 17,509 respondents. Two researchers (the first author and a research assistant) were responsible for coding each study for sample size, reliability information, and sample characteristics. Inter-rater agreement was 95%. In cases where coding differences occurred, discussion was used to reach consensus. A complete list of all the articles included in the meta-analysis is available from the authors. A majority (64%) of the studies were conducted in the United States, 14% in other Western cultures (e.g., Canada, Netherlands, Portugal, and Spain), and the remaining 22% in Asian cultures (e.g., China, Hong Kong, India, Singapore, and Taiwan). Twenty-five percent (25%) of the studies were published between 1990 and 1999, and the remaining 75% were published in 2000 or after. The average age across samples was 35.5 years old (ranging from 28.4 to 45 years old) and the average tenure was 7.5 years (ranging from 1.5 to 18.6 years). The sample was close to evenly split in terms of gender (52% male, 48% female). 3.2. Measures of key constructs 3.2.1. Affective organizational commitment Across those studies included, we observed that a variety of scales of affective organizational commitment were utilized. More than half (52%) utilized the scales created by Meyer and his colleagues (Allen & Meyer, 1990; Meyer, Allen, & Smith, 1993). Thirty-eight percent (38%) utilized the scales created by Mowday, Porter, and their colleagues (Mowday, Porter, & Steers, 1982; Porter, Steers, Mowday, & Boulian, 1974). The remaining articles measured affective organizational commitment either with other validated scales (6%) or with new items created specifically for those studies (4%). All the studies included used self-report measures; the average scale reliability was .84. 3.2.2. OCB A wide variety of OCB were identified in our article search, such as OCB directed at coworkers and supervisors, OCB directed at organization, boosterism, civic virtue, courtesy, sportsmanship, compliance, altruism, and social participation. Because LePine et al.'s (2002) meta-analysis showed that these different dimensions are highly correlated with each other and that OCB is the underlying latent construct which drives these different behaviors, we do not differentiate among sub-types of OCB. Rather, in those studies where multiple types of OCB were measured, we combined the dimensions to create a composite variable. We did so to ensure that each empirical study only contributed one effect size to the subsequent meta-analysis, which in turn reduces potential data dependency problems (Hunter & Schmidt, 1990). As with the case of affective organizational commitment, a variety of measures have been used to measure OCB. Scales which have been frequently used include those created by Podsakoff, MacKenzie, Moorman, and Fetter (1990) (22%), Smith et al. (1983) (14%), Farh, Earley, and Lin (1997) (13%), and Williams and Anderson (1991) (11%). Like LePine et al.'s (2002) meta-analysis, we treated different scales of OCB as measuring the same underlying OCB construct. As noted before, we excluded self-ratings of OCB

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

533

from the current meta-analysis. Therefore, all the studies included here either used supervisor ratings of OCB (88%), peer ratings (11%), or objective measures (1%). Across studies that provided supervisory and peer ratings of OCB, the average scale reliability was .87. 3.3. Meta-analytical procedures Raju, Burke, Normand, and Langlois's (1991) meta-analysis techniques (which include corrections for measurement error variance and sampling error variance) were used in this study. The Raju et al. (1991) procedures are optimally designed to estimate standard errors for corrected correlations when sample-based and assumed (fixed) artifact values (e.g., sample-based reliability estimates) are incorporated into the corrections. A mean corrected correlation was judged to be significant at α = .05 when its 95% confidence interval did not include the value of zero. In Monte Carlo studies, researchers (Mendoza & Reinhardt, 1991; Raju et al., 1991) found that these proposed meta-analytical procedures were more accurate in estimating the mean and variance of true validities than the meta-analytical procedures proposed by other researchers. 3.3.1. Correction for measurement error variance The observed correlations between affective organizational commitment and OCB required disattenuation so that the interpretation of effect sizes would not be confounded by random measurement error variance. Because different studies measure the same construct using different scales, this disattenuation process is needed to adjust for different levels of measurement error contained in different scales. This type of correction utilizes alpha coefficients (i.e., internal consistency estimates) reported in individual studies rather than inter-rater reliability estimates because inter-rater reliability estimates contain systematic rater effects (Murphy & De Shon, 2000). As such, the variance associated with raters would not be a source of random measurement error variance. 3.3.2. Corrections for sampling error variance The second artifact that we corrected for was sampling error due to sample size differences (Raju et al., 1991). 3.3.3. Testing for the moderating effects of tenure In order to test for moderating effects of organizational tenure, we adopted a regression procedure, recommended by Steel and Kammeyer-Mueller (2002), which has been found to be more reliable and robust than other moderation testing methods. An additional advantage of the Steel and Kammeyer-Mueller (2002) method is that continuous moderator variables, such as organizational tenure, can be included. This regression-based moderator testing technique has been used successfully by Wright and Bonett (2002), who examined the moderating effects of tenure in the relation between organizational commitment and in-role job performance. Following this procedure, we used the average tenure of the sample as an independent variable, in a weighted least squares multiple regression, to predict the Fisher-z-transformed corrected correlation coefficients for the commitment–OCB relation. That is, first the independent variable was calculated as the natural logarithm of average tenure of a sample. Next, we applied the Fisher-z transformation to the corrected correlations. This transformed effect size was treated as the dependent variable. The transformed correlations are heteroscedastic with known variance equal to 1/(nj − 3), where nj is the sample size for Study j. Following Wright and Bonett (2002), we used weighted linear squares estimation, where the weights were set equal to nj − 3. If the average tenure of the sample was a significant predictor of the effect sizes, then it would suggest that organizational tenure moderated that relation. This approach can be extended to examine higher-order moderating effects of tenure as well. That is, if the squared value of the natural logarithm of average tenure of a sample was a significant predictor of the effect sizes above and beyond the linear term, it suggested that tenure had a non-linear moderating effect in the relation between affective organizational commitment and OCB. Finally, it should be noted that, following Wright and Bonett (2002), average age of the sample was included as a control variable in the foregoing analyses. 4. Results The meta-analytical results are presented in Table 1. Across the 64 studies (N = 17,509 respondents) we identified in our article search, the effect size for the affective organizational commitment–OCB relation was .22. For the subset of 40 samples which provided complete information on average tenure and age (N = 11,416 respondents), the effect size was .23. To identify initial patterns of the moderating effects of organizational tenure, we separated these 40 studies into three groups of roughly equal size based on average tenure: the low-tenure group (average tenure less than 5 years), medium-tenure group (average tenure between 5 and 10 years), and high-tenure group (average tenure more than 10 years). (Wright and Bonett (2002) also used 5-year intervals of tenure in their data interpretation.) We then conducted meta-analyses for these three groups separately. We observed that the corrected correlation for the commitment–OCB relation was .20 in the low-tenure group, .27 in the medium-tenure group, and .22 in the high-tenure group. In other words, while higher levels of affective organizational commitment were associated with higher levels of OCB in general (as evident by the positive effect sizes), the strength of the relation varied depending on organizational tenure. The relation strengthens from the start of employment up through roughly 10 years of tenure. After that point, the relation between affective organizational commitment and OCB continues to be positive in

534

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

Table 1 Meta-analytical results.

Across all studies identified in the article search Across studies which provided tenure and age information Average tenure b 5 years Average tenure 5–10 years Average tenure N 10 years

N

k

rc

SD

95% CI

17,509 11,416 3627 3860 3929

64 40 14 16 10

.22 .23 .20 .27 .22

.08 .08 .08 .08 .03

(.20, (.21, (.16, (.23, (.20,

.24) .25) .24) .31) .24)

Note. N = cumulative sample size; k = number of studies cumulated; rc = sample-size weighted corrected correlation; SD = standard deviation of rc; CI = confidence interval for rc.

direction but weaker in strength. Thus, the initial examination of the data revealed a non-linear moderating effect of organizational tenure. However, this analysis has not controlled for the confounding effects of age. In fact, previous research has shown that age and organizational tenure are strongly correlated at .70 (Ng & Feldman, 2010). To more formally and rigorously test for the non-linear moderating effect of tenure and include age as a control variable, we again adopted the regression procedure outlined earlier. The regression analysis showed that the average tenure of a sample was a significant and positive predictor (R2 = 2%, p b .05) of the effect sizes for the commitment–OCB relation above and beyond the effect of average age (which was found to be a non-significant predictor in the regression model). These results suggest that organizational tenure was a significant moderator. Furthermore, the squared term for average tenure was a significant and negative predictor (R2 = 12%, p b .05) above and beyond average age and the first-order effect of tenure. This result suggests that the moderating effect of average tenure was non-linear in nature, as identified in the earlier preliminary analysis. We observed that the values for the constant term and regression coefficients for age, tenure, and squared tenure were −06, .01, .85, and −.52, respectively. Therefore, holding the average age of a sample as a constant (we used 35.5 years old, which was the mean age across all samples included in the analyses), the predicted true correlations for the affective organizational commitment–OCB relation are .30 for one year of average tenure, .64 for 5 years of average tenure, .63 for 10 years of average tenure, .58 for 15 years of average tenure, .53 for 20 years of average tenure, .47 for 25 years of average tenure, and .42 for 30 years of average tenure. That is, the relation between affective organizational commitment and OCB strengthened from initial hiring up to 10 years of tenure. However, we also observed that, after 10 years of tenure, the strength of the commitment–OCB relation did decrease as organizational tenure continued to increase, although the relation remained statistically significant. 5. Discussion The findings here provide partial support for our prediction in Hypothesis 1 that the moderating effect of organizational tenure in the commitment–OCB relation is positive in nature, with the relation being stronger for individuals with higher organizational tenure. Before roughly 10 years of organizational tenure, increasing years of tenure strengthened the positive relation between affective organizational commitment and OCB. However, the results also provide partial support for Hypothesis 2 as well. Specifically, after roughly 10 years of service, accruing more years of tenure decreased the strength of the positive relation between affective organizational commitment and OCB. In short, the moderating effect of tenure was curvilinear. In this final section, we discuss the implications of these findings for future theory development and for empirical research. 5.1. Implications for future theory development Over the past few decades, researchers have documented the positive effects of affective organizational commitment on OCB (LePine et al., 2002; Meyer et al., 2002). The present study contributes to this literature in three ways. First, while the current meta-analysis of the commitment–OCB relation is based on cross-sectional studies, the results here regarding organizational tenure are suggestive of changes in the nature of the relation over time (Hofmann, Jacobs, & Gerras, 1992). Also, there is little evidence that the commitment–OCB relation is stable over time. In fact, researchers have found evidence that both affective organizational commitment (Beck & Wilson, 2000) and OCB (Hui, Lam, & Law, 2000) change over time, raising the question of whether the positive relation observed between affective organizational commitment and OCB is generally stable. Second, while some previous researchers have demonstrated the curvilinear main effect of organizational tenure on job attitudes and behaviors (e.g., Ng & Feldman, 2010; Wagner, Ferris, Fandt, & Wayne, 1987), this study is the first to demonstrate the curvilinear moderating effect of organizational tenure. That is, not only does the strength of the commitment–OCB relation change across levels of organizational tenure, but the direction of the change (increases and decreases) varies across levels of tenure as well. This finding highlights the importance of including several additional research streams into future investigations of the relation between affective organizational commitment and OCB because no single theory can fully explain the non-linear moderating effects of tenure we observed here. For example, human capital theory (Becker, 1964) leads us to predict that individuals with longer tenure will have greater willingness and ability to perform OCB, but we observed that willingness and ability to perform declines after the first decade of services. One possibility to consider here is whether the bulk of human capital is acquired early in one's organizational career,

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

535

when opportunities for formal training, diverse job assignments, and mentoring are greatest. Thereafter, though, as the accumulation of new human capital slows down, willingness and ability to engage in additional OCB may slow down as well. Another possibility, derived from social capital theory (Burt, 1992; Granovetter, 1973), is that workers in the early years of their careers focus on developing a larger number of weak ties but, as time passes, they focus on developing stronger ties with a more limited set of workers. This shift in social capital building patterns might also explain a subsequent decline in OCB as the size of the core social network shrinks. Yet another possibility, suggested by career stage theory (Super, 1957, 1980), is that individuals in early career are more invested in building their careers (and therefore perform more OCB), but individuals in mid-career and late-career shift more energy into maintaining work-family balance. A related perspective is taken by Wagner et al. (1987), who consider the attitude–outcome relation from a sense-making perspective. They suggest that tenure may have a curvilinear effect on job involvement because newcomers have to spend a great deal of time trying to make sense of their new environments. However, as time passes and individuals become more familiar with those environments, additional tasks which newcomers readily took on early in their careers now seem more boring or less enticing. Last here, the present study suggests that greater attention be paid to the variety of commitment variables used in this research stream. Because of limited data, both Wright and Bonett (2002) and we were able to consider only affective organizational commitment in our research. However, it is possible that organizational tenure might moderate the relations of other forms of organizational commitment with job performance and OCB. For instance, across the eight studies that we were able to identify which examined the relation between continuance commitment and others' ratings of OCB, the corrected effect size was only .02 (N = 1934 respondents). Further, across the seven studies that we found which examined the relation between normative commitment and others' ratings of OCB, the corrected effect size was .09 (N = 1399 respondents). The weak effect size observed here might suggest that there are a variety of moderators (including organizational tenure) which could potentially change the magnitude of these relations.

5.2. Implications for future empirical research Much of our understanding about the relation of affective organizational commitment to OCB has been informed by the use of self-ratings of OCB. Organ and Ryan (1995) found evidence, though, that self-ratings of OCB were consistently higher than those ratings provided by supervisors, peers, and objective counts. We conducted a post hoc analysis comparing the strength of the affective organizational commitment–OCB relation in studies using self-ratings of OCB (N = 60) with studies utilizing others' measures of OCB (N = 64). We found that, in the 60 studies which used self-ratings of OCB, the average corrected correlation between affective organizational commitment and OCB was .39. This effect size was significantly larger than the average corrected correlation between affective organizational commitment and others' ratings of OCB (.22). One approach, taken by Wright and Bonett (2002), is to include source of ratings as a control variable in data analyses. An alternative approach, which we use here, is to conduct separate analyses of self-reports and others' reports of OCB. Along the same lines, we encourage future researchers to consider difference between self-reports and others' ratings of affective organizational commitment. For example, some previous research has found that supervisors make inferences about employees' levels of commitment (Shore, Barksdale, & Shore, 1995). Here, too, self-serving biases might lead employees to rate themselves more highly on commitment than supervisors do. Thus, it is possible that the shape of the commitment–OCB relation might vary even further if we consider self-ratings versus others' ratings of affective organizational commitment as well. The present results also suggest that more attention be paid to the tenure distribution of the sample being investigated. For instance, if the research sample was largely filled with low-tenure employees, then the moderating effect of organizational tenure is likely to be positive in nature. On the other hand, if researchers are studying a sample largely filled with high-tenure employees, the moderating effect of organizational tenure is likely to be negative in nature. Imprecise or incorrect conclusions, then, might be drawn from samples with restricted ranges on the organizational tenure variable. Third, the influence of chronological age in the commitment–OCB relation needs further attention as well. Wright and Bonett (2002) controlled for age in their investigation of the commitment–performance relation because they reasoned that age was highly correlated with organizational tenure. While that argument makes sense theoretically, the effect of age was found to be statistically non-significant in their study. Similarly, in the present study, we observed that age (as a control variable) did not have the same effect size as organizational tenure did. While the weak effect of age in a regression model might be due to multicollinearity with tenure, it is also possible that age, in and itself, does not have a substantial impact on the commitment–OCB relation. For example, while there are theoretical reasons to predict that human capital would increase with organizational tenure (Ng & Feldman, 2010), there are no strong theoretical reasons to expect that aging is associated with material increases in human capital. Finally, the current study has several limitations that should be addressed in future research. The first limitation is that very few studies on OCB are longitudinal in nature and, as a result, it is not possible to infer causality. Second, due to the data contained in individual studies, the average tenure and age of samples were used as proxies for individual-level tenure and age. Last here, the meta-analyses were heavily based on studies conducted in the U.S. and Europe. This shortcoming is partially counteracted by the diversity of samples, with different backgrounds and characteristics, which were included in the meta-analysis to increase the generalizability of its findings. Analyses of cross-cultural differences in how organizational tenure moderates the commitment–OCB relation, though, should be addressed in greater detail as well.

536

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

5.3. Conclusion In this study, notable differences in the strength and direction of the commitment–OCB relation are observed across different levels of tenure. Moreover, the association between affective organizational commitment and OCB is both linearly and non-linearly moderated by organizational tenure. Taken together, we hope that this article provides new insights into the complicated effects of organizational tenure on the relation between organizational commitment and work outcomes. References Adler, S., & Aranya, N. (1984). A comparison of the work needs, attitudes, and preferences of professional accountants at different career stages. Journal of Vocational Behavior, 25, 45−57. Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance, and normative commitment to the organization. Journal of Occupational Psychology, 63, 1−18. Beck, K., & Wilson, C. (2000). Development of affective organizational commitment: A cross-sequential examination of change with tenure. Journal of Vocational Behavior, 56, 114−136. Becker, G. (1964). Human capital: A theoretical and empirical analysis with special reference to education. New York: Columbia University Press. Bentein, K., Vandenberg, R. J., Vandenberghe, C., & Stinglhamber, F. (2005). The role of change in the relationship between commitment and turnover: A latent growth modeling approach. The Journal of Applied Psychology, 90, 468−482. Bird, A. (1996). Careers as repositories of knowledge: Considerations for boundaryless careers. In M. B. Arthur, & D. M. Rousseau (Eds.), The boundaryless career: A new employment principle for a new organizational era (pp. 150−168). New York: Oxford University Press. Blau, P. (1964). Exchange and power in social life. New York: John Wiley and Sons. Bolino, M. C., & Turnley, W. H. (2005). The personal costs of citizenship behavior: The relationship between individual initiative and role overload, job stress, and work-family conflict. The Journal of Applied Psychology, 90, 740−748. Burt, R. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press. Cortina, J. M. (1993). Interaction, nonlinearity, and multicollinearity: Implications for multiple regression. Journal of Management, 19, 915−922. Cropanzano, R., Rupp, D. E., & Byrne, Z. S. (2003). The relationship of emotional exhaustion to work attitudes, job performance, and organizational citizenship behaviors. The Journal of Applied Psychology, 88, 160−1690. Deery, S. J., & Iverson, R. D. (2005). Labor-management cooperation: Antecedents and impact on organizational performance. Industrial & Labor Relations Review, 58, 588−609. Drory, A. (1982). Individual differences in boredom proneness and task effectiveness at work. Personnel Psychology, 35, 141−151. English, B., Morrison, D., & Chalon, C. (2010). Moderator effects of organizational tenure on the relationship between psychological climate and affective commitment. Journal of Managerial Development, 29, 394−408. Farh, J., Earley, P. C., & Lin, S. (1997). Impetus for action: A cultural analysis of justice and organizational citizenship behavior in Chinese society. Administrative Science Quarterly, 42, 421−444. Gibson, D. E. (2003). Developing the professional self-concept: Role model construals in early, middle, and late career stages. Organization Science, 14, 591−610. Granovetter, M. (1973). The strength of weak ties. The American Journal of Sociology, 78, 1360−1380. Harris, M., & Schaubroeck, J. (1988). A meta-analysis of self-supervisory, self-peer, and peer-supervisory ratings. Personnel Psychology, 41, 43−62. Harrison, D. A., Newman, D. A., & Roth, P. L. (2006). How important are job attitudes? Meta-analytical comparisons of integrative behavioral outcomes and time sequences. Academy of Management Journal, 49, 305−325. Hofmann, D. A., Jacobs, R., & Gerras, S. J. (1992). Mapping individual performance over time. The Journal of Applied Psychology, 77, 185−195. Huang, X., Shi, K., Zhang, Z., & Cheung, Y. L. (2006). The impact of participative leadership behavior on psychological empowerment and organizational commitment in Chinese state-owned enterprises: The moderating role of organizational tenure. Asia Pacific Journal of Management, 23, 345−367. Hui, C., Lam, S. S. K., & Law, K. K. S. (2000). Instrumental values of organizational citizenship behavior for promotion: A field quasi-experiment. The Journal of Applied Psychology, 85, 822−828. Hunter, J. E., & Schmidt, H. L. (1990). Methods of meta-analysis: Correcting for sources of error and bias in research findings. Newbury Park, CA: Sage. Jaramillo, F., Mulki, J. P., & Marshall, G. W. (2005). A meta-analysis of the relationship between organizational commitment and salesperson job performance: 25 years of research. Journal of Business Research, 58, 705−714. Kass, S. J., Vodanovich, S. J., & Callender, A. (2001). State-trait boredom: Relationship to absenteeism, tenure, and job satisfaction. Journal of Business and Psychology, 16, 317−327. Lee, K., & Allen, N. J. (2002). Organizational citizenship behavior and workplace deviance: The role of affect and cognitions. The Journal of Applied Psychology, 87, 131−142. LePine, J. A., Erez, A., & Johnson, D. E. (2002). The nature and dimensionality of organizational citizenship behavior: A critical review and meta-analysis. The Journal of Applied Psychology, 87, 52−65. Lievens, F., Conway, J. M., & De Corte, W. (2008). The relative importance of task, citizenship and counterproductive performance to job performance ratings: Do rater source and team-based culture matter? Journal of Occupational and Organizational Psychology, 81, 11−27. Luthans, F., Avolio, B. J., Avey, J. B., & Norman, S. M. (2007). Positive psychological capital: Measurement and relationship with performance and satisfaction. Personnel Psychology, 60, 541−572. March, J. G., & Simon, H. A. (1958). Organizations. New York: Wiley. McEnrue, M. P. (1988). Length of experience and the performance of managers in the establishment phase of their careers. Academy of Management Journal, 31, 175−185. McGee, G. W., & Ford, R. C. (1987). Two (or more?) dimensions of organizational commitment: Re-examination of the affective and continuance commitment scales. The Journal of Applied Psychology, 72, 638−642. Mendoza, J. L., & Reinhardt, R. N. (1991). Validity generalization procedures using sample-based estimates: A comparison of six procedures. Psychological Bulletin, 110, 596−610. Meyer, J. P., Allen, N. J., & Smith, C. A. (1993). Commitment to organizations and occupations: Extension and test of a three-component conceptualization. The Journal of Applied Psychology, 78, 538−551. Meyer, J. P., & Herscovitch, L. (2001). Commitment in the workplace: Toward a general model. Human Resource Management Review, 11, 299−326. Meyer, J. P., Stanley, D. J., Herscovitch, L., & Topolnytsky, L. (2002). Affective, continuance, and normative commitment to the organization: A meta-analysis of antecedents, correlates, and consequences. Journal of Vocational Behavior, 61, 20−52. Motowidlo, S. J., & Van Scotter, J. R. (1994). Evidence that task performance should be distinguished from contextual performance. The Journal of Applied Psychology, 79, 475−480. Mowday, R., Porter, L., & Steers, R. (1982). Employee-organization linkages. New York: Academic Press. Murphy, K. R., & De Shon, R. (2000). Interrater correlations do not estimate the reliability of job performance ratings. Personnel Psychology, 53, 873−900. Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital and organizational advantage. Academy of Management Review, 23, 242−266. Ng, T. W. H., & Feldman, D. C. (2010). Organizational tenure and job performance. Journal of Management, 36, 1220−1250. O'Reilly, C. A., & Chatman, J. (1986). Organizational commitment and psychological attachment: The effects of compliance, identification and internalization on prosocial behavior. The Journal of Applied Psychology, 71, 492−499.

T.W.H. Ng, D.C. Feldman / Journal of Vocational Behavior 79 (2011) 528–537

537

Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome. Lexington, MA: Lexington Books. Organ, D. W., & Ryan, K. (1995). A meta-analytical review of attitudinal and dispositional predictors of organizational citizenship behavior. Personnel Psychology, 48, 775−802. Peter, L., & Hull, R. (1969). The Peter principle. New York: William Morrow & Co. Podsakoff, N. P., Blume, B. D., Whiting, S. W., & Podsakoff, P. M. (2009). Individual-and organization-level consequences of organizational citizenship behaviors: A meta-analysis. The Journal of Applied Psychology, 94, 122−141. Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly, 1, 107−142. Porter, L. W., Steers, R. M., Mowday, R. T., & Boulian, P. V. (1974). Organizational commitment, job satisfaction and turnover among psychiatric technicians. The Journal of Applied Psychology, 59, 603−609. Quinones, M. A., Ford, J. K., & Teachout, M. S. (1995). The relationship between work experience and job performance: A conceptual and meta-analytic review. Personnel Psychology, 48, 887−910. Raju, N. S., Burke, M. J., Normand, J., & Langlois, G. M. (1991). A new meta-analytic approach. The Journal of Applied Psychology, 76, 432−446. Riketta, M. (2002). Attitudinal organizational commitment and job performance: A meta-analysis. Journal of Organizational Behavior, 23, 257−266. Rosenthal, R. (1979). The “file drawer problem” and tolerance for null results. Psychological Bulletin, 86, 638−641. Rousseau, D. M. (1998). Why workers still identify with organizations. Journal of Organizational Behavior, 19, 217−233. Shirom, A., & Mazeh, T. (1988). Periodicity in organizational tenure — Job satisfaction relationship. Journal of Vocational Behavior, 33, 38−49. Shore, L., Barksdale, K., & Shore, T. (1995). Managerial perceptions of employee commitment to the organization. Academy of Management Journal, 38, 1593−1615. Slaughter, S. A., Ang, S., & Boh, W. F. (2007). Firm-specific human capital and compensation-organizational tenure profiles: An archival analysis of salary data for IT professionals. Human Resource Management, 46, 373−394. Smith, C. A., Organ, D. W., & Near, J. P. (1983). Organizational citizenship behavior: Its nature and antecedents. The Journal of Applied Psychology, 68, 653−663. Steel, P. D., & Kammeyer-Mueller, J. D. (2002). Comparing meta-analytic moderator estimation techniques under realistic conditions. The Journal of Applied Psychology, 87, 96−111. Sturman, M. C. (2003). Searching for the inverted U-shaped relationship between time and performance: Meta-analyses of the experience/performance, tenure/ performance, and age/performance relationships. Journal of Management, 29, 609−640. Super, D. E. (1957). The psychology of careers. New York: Harper & Row. Super, D. E. (1980). A life-span, life-space approach to career development. Journal of Vocational Behavior, 16, 282−298. Super, D. E., Savickas, M. L., & Super, C. M. (1996). The life-span, life-space approach to careers. In D. Brown, & L. Brooks (Eds.), Career choice and development (pp. 121−178). (3rd ed.). San Francisco: Jossey-Bass. Wagner, J. A., Ferris, G. R., Fandt, P. M., & Wayne, S. J. (1987). The organizational tenure–job involvement relationship: A job-career experience explanation. Journal of Occupational Behavior, 8, 63−70. Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management, 17, 601−617. Williams, C. P., & Savickas, M. L. (1990). Development tasks of career maintenance. Journal of Vocational Behavior, 36, 166−175. Wright, T. A., & Bonett, D. G. (2002). The moderating effects of employee tenure on the relation between organizational commitment and job performance: A metaanalysis. The Journal of Applied Psychology, 87, 1183−1190.