بررسی مفهوم سازی جنبه های رضایت شغلی کارگرن یقه آبی در مقابل مفهوم سازی جنبه های رضایت شغلی کارگران یقه سفید

This study examined the degree to which blue- versus white-collar workers differentially conceptualize various job facets, namely the work itself, co-workers, supervisors, and pay. To examine these potential differences, we conducted a series of analyses on job satisfaction ratings from two samples of university workers. Consistent with the study hypothesis, results revealed that blue- and white-collar workers held different conceptualizations regarding the nature of co-workers, pay, and the work itself, but not of supervisors. In general, more dimensions for each facet emerged for the white-collar workers, suggesting that these individuals possess more differentiated and multidimensional evaluations of these job facets than do blue-collar workers. Discussion focuses on the meaning and implications of the findings.

Among the most robust findings in organizational psychology is that workers’ satisfaction with various facets (i.e., aspects or features) of their jobs has significant individual and organizational consequences (for reviews, see Brief, 1998, Hulin and Judge, 2003, Spector, 1997 and Warr, 2007). While workers’ satisfaction with myriad aspects of their jobs may be consequential, scholarly research has largely converged on particular job features that are of greatest import. Some of those features, the ones under investigation here, are the degree to which individuals are satisfied with their supervisor, co-workers, pay and benefits, and the nature of their work (Smith, Kendall, & Hulin, 1969). Almost universally, both satisfaction with regard to these job facets is assessed using self-report measures (see Brief, 1998). An often unstated assumption underlying this approach is that workers view these aspects in the same manner. By implication, being (dis)satisfied with these various job features “means the same thing” for different groups of employees. While some workers are more satisfied with their co-workers than are others, for instance, the assumption is that all respondents evaluate their co-workers using a similar frame or along the same dimensions; the difference is thought to be one of degree, not kind. However, some evidence, discussed below, suggests that different groups of workers may evaluate these job aspects using dissimilar conceptualizations. Such dissimilarities are significant for at least two reasons. First, they are meaningful because researchers often infer that mean differences in satisfaction ratings among various groups provide evidence regarding the nature of those characteristics that predict more or less satisfaction. For instance, a finding that white-collar workers are more satisfied with their work duties than are their blue-collar counterparts is usually interpreted to mean that the objective characteristics of the two job types (e.g., level of control, autonomy) result in greater satisfaction with the work itself (e.g., Hackman and Oldham, 1980 and Humphrey et al., 2007). However, to the degree that qualitative differences exist with respect to how these different groups think about such facets, interpreting findings across groups or comparing groups’ mean satisfaction levels on these facets becomes a questionable practice (Steenkamp & Baumgartner, 1998). Second, and perhaps more significant, is that these dissimilar conceptualizations would be indicative of qualitatively different experiences of work. A finding that white- versus blue-collar employees rate their satisfaction with co-workers using more or fewer dimensions, for instance, suggests that these two groups actually experience and form judgments about their co-workers in different ways. As noted by Logan, O’Reilly, and Roberts (1973), job satisfaction researchers should not simply assume that different groups of workers have identical perceptions of the satisfactions they derive from their job. Rather, scholars should examine different satisfaction patterns for various employee groups as an important strategy for understanding the nature of job satisfaction and, even more broadly, the nature of subjective work experience. The purpose of the current paper is to explore the notion that blue- versus white-collar workers evaluate these facets in qualitatively different manners. We chose these particular groups because they report mean difference in facet job satisfaction (e.g., Herzberg et al., 1957, Pearson, 1998 and Wan and Leightley, 2006) and because their distinct work experiences (Morgeson & Humphrey, 2006) have the potential to lead them to think differentially about what contributes to or represents satisfaction with these job facets. To these ends, the remainder of the paper unfolds as follows. First, using existing research and our own theoretical intuition, we broadly suggest that these two groups of workers possess somewhat dissimilar conceptualizations of these facets. Next, we conduct a series of configural equivalence (i.e., invariance) analyses to examine the fit of the same, generally accepted single, undimensional factor model for each facet across the groups. Results indicating less than excellent fit would provide preliminary evidence that the factor structures differ across groups and, by implication, that the two groups evaluate the facets in somewhat discrepant ways. To explore these potential differences, we then conduct exploratory factor analyses (EFA) for each facet to generate alternative configurations. Finally, based on these preliminary findings, we develop theroetically-based alternative configurations which we evaluate on a separate subsample using confirmatory factor analysis (CFA). The paper closes with a discussion of the meaning and implication of the results. 1.1. Differences in job satisfaction between blue- and white-collar workers For this paper, we follow others in defining blue-collar workers as those who perform primarily physical work and whose career paths are relatively restricted (Gibson and Papa, 2000 and Lederer, 1987) and white-collar workers as professional and semi-professional employees (Hammer & Ferrari, 2002). A particularly robust finding from studies comparing job satisfaction across these groups is that blue-collar workers tend to be less satisfied than their white-collar counterparts with various aspects of their jobs, such as pay (e.g., Lee et al., 1981 and Weaver, 1975), their supervisors (e.g., O’farrell and Harlan, 1982), and the work itself (e.g., Lee et al., 1981 and O’farrell and Harlan, 1982). While such differences almost certainly do largely reflect objective qualities of the jobs, such comparisons may also be somewhat erroneous to the degree that the two groups evaluate these facets in discrepant manners. Indirect support for this differential conceptualization derives from findings indicating that these groups think about their jobs and job experiences in somewhat discrepant ways. Hennequin (2007), for example, discusses how blue- and white-collar workers may hold distinct ideas about the meaning of career success, given the dissimilar indicators of success and differing paths to such success that characterize the two groups. Related research documents that these two groups often attach unequal importance to the role of work in their lives and identity (see Dubin, Champoux, & Porter, 1977), develop different meanings of work (Harpaz, 1986 and Yuchtman-Yaar and Gottlieb, 1985), and have different levels of job involvement (Kaufman, 1982). The two groups also perceive and react to some aspects of the job differently, such as role strain (Mathieu & Hamel, 1989) and upward mobility (Prince, 2003). In addition, different job characteristics may be of more or less significance in determining overall job satisfaction across these two groups (Berger, 1986 and Ronen and Sadan, 1984). Here, we attempt to extend these findings by examining whether these differences manifest in dissimilar ratings of satisfaction with job facets. Specifically, we assess the degree to which these employees differentially conceptualize four particularly meaningful and significant features of work, namely the work itself, the supervisor, co-workers, and pay (Warr, 2007). As discussed below, we do so by comparing the two groups’ patterns of responses to the corresponding four facet scales of the Job Descriptive Index (JDI; Smith et al., 1969), which is the most widely used and validated facet measure of job satisfaction (Kinicki et al., 2002 and Smith and Stanton, 1998). Our general prediction is that white-collar workers will have more differentiated views of at least some of these facets than will blue-collar workers. We base this proposition on the notion that white-collar workers’ jobs generally are more complex than are those of their blue-collar counterparts (Morgeson & Humphrey, 2006). Owing to this great complexity, white-collar employees potentially have more information and/or dimensions upon which to evaluate the facets of their jobs. For instance, the greater variety in their tasks (Morgeson & Humphrey, 2006) may lead white-collar workers to evaluate their work in a more differentiated or nuanced way. Similarly, their greater potential for multiple types of interactions with co-workers (e.g., interactions with co-workers of differing job status and positions) could also lead to more differentiated or refined judgments about satisfaction with co-workers. Conversely, for blue-collar workers, whose jobs tend to be more routinized and monotonous (Dubin & Champoux, 1977), the factor structures should be simpler and less differentiated, as these employees should have more consistent information upon which to rely in evaluating each facet. Based on this rationale, we offer the following broad hypothesis: Hypothesis. There is greater factor complexity (i.e., dimensionality) for white-collar workers than for blue-collar workers with respect to the two groups’ ratings of satisfaction with the work itself, one’s co-workers, one’s supervisor, and one’s pay.

The purpose of this study was to investigate the degree to which blue- and white-collar workers conceptualize important facets of their jobs differently. We examined this question by examining the two groups’ patterns of responses to a well-established measure of facet job satisfaction. Consistent with our general prediction, the current results indicated that the blue- and white-collar groups appear to perceive individual job facets in qualitatively different ways. These results appear to have several significant theoretical and practical implications. First, consistent with Logan and colleagues’ remarks (1973), the results of this study suggest that job satisfaction research should go beyond only focusing on mean level differences. Studies comparing different groups’ conceptualizations of job satisfaction (i.e., satisfaction patterns) can lead to additional insights into the nature of the satisfaction construct. To date, several studies have revealed distinct job satisfaction patterns for different populations of workers, such as full-time and part-time workers (Logan, O’Reilly, & Roberts, 1973), parent and non-parent workers (Rothausen, 1994), and different specialty groups of physicians (Duffy & Richard, 2006). The present study contributes to and extends this literature by documenting different conceptualizations of job satisfaction facets held by blue- and white-collar workers. When we talk about satisfaction with these facets, we need to consider to whose satisfaction we are referring. Perhaps even more fundamentally, results like the current ones call into question the almost axiomatic notion that job features like “co-workers” or the nature of job characteristics are phenomenologically the same for different workers. With regard to this issue, we strongly call upon researchers to further pursue this topic by, for example, conducting qualitative studies (e.g., interviews) to assess these potentially dissimilar conceptualizations. Also, future studies examining whether these differences hold for other job aspects (e.g., opportunities for promotion, physical aspects of the work environment) would be useful. Finally, queries investigating if the differences seen here manifest with regard to other types of evaluations also would be enlightening. The current results indicate that there are differences in the judgments that contribute to or represent satisfaction with these facets. They do not speak to other evaluative judgments such as perceived support (e.g., from one’s co-workers) or affective reactions (e.g. to one’s co-workers). Findings replicating the current results with other types of judgment or those showing that these present findings do not generalize to other judgments both would be interesting. A second implication of these findings, following from the one above, is that the relationship between these facets and other variables of interest (e.g., contextual performance, team performance) may vary across these groups. That is, to the extent that the facets are not invariant across groups, the relationships with other variables should differ as well. For instance, the present results suggest that the relationship between co-worker satisfaction and variables like individual and team performance may be attenuated or obscured in white-collar samples if one fails to account for the two-dimensional nature of that satisfaction. Presumably, satisfaction with regard to co-workers’ competence is more strongly related to performance than is satisfaction with regard to co-workers’ likability. Conversely, for blue-collar workers, the obtained correlation between (unidimensional) co-worker satisfaction and each criterion seems to be an accurate representation of the relationship. This notion of differential relationships partially may explain Shirom et al., 1999, Smith, 1997 and Warr, 2007 conclusion that older studies examining facet satisfaction–performance correlations have yielded variable effect sizes. Such discrepancies may partially be due to the fact that different studies have included different groups of workers. This possibility also implies that organizations may need to tailor practices or interventions designed to increase facet satisfaction to the particular recipient group. As an example, these results suggest that increased training should result in greater satisfaction with co-workers for blue-collar workers insofar as such training yields greater performance. However, for white-collar workers, such training may lead to more competent workgroups, but not necessarily to greater liking among co-workers. Studies examining this and similar possibilities seem relatively straightforward to conduct and would appear to have important practical implications. Furthermore, these findings seem particularly notable when considered in light of changing trends in the workforce. As the proportion of managerial and professional jobs has grown and continues to grow (Bamundo and Kopelman, 1980 and Mitchell et al., 2003), a more comprehensive and precise understanding of how the job satisfaction of white-collar workers might differ from that of their blue-collar counterparts is increasingly valuable. Much of what we know about the facets, and consequently the facet measures, of job satisfaction derives from research conducted in the 1960s (e.g., the JDI, Smith et al., 1969; the Minnesota Satisfaction Questionnaire, Weiss, Dawis, & England, 1967). Perhaps a critical reconsideration of these facets, especially for white-collar employees, is warranted. Our hope is that the current findings are a first step toward such an examination. A final implication of this study is in regard to the use of measurement equivalence analyses. Traditionally, studies investigating ME have focused almost exclusively on the question of whether the measure is an equally good representation of the underlying construct across time, cultures, languages, administration medium, and so forth (see Vandenberg & Lance, 2000). For example, Candell and Hulin (1986) assessed the ME of the JDI across language and culture. As another example, Donovan, Drasgow, and Probst (2000) tested the ME of two subscales of the JDI across computerized and paper-and-pencil administrations. The current study adopted a different approach by using measurement equivalence (ME) analyses as a means of understanding the underlying construct and different groups’ conceptualizations of that construct. This approach highlights the recognition that ME analyses are useful not only as a check on a measure’s generalizability or psychometric properties, but also as a way to gain insight into the meaning of the theoretical constructs (e.g., for different populations). This latter perspective is consistent with the well-established principle that results of any correlational study speak both to the measure and to the nature of the underlying construct (e.g., Binning and Barrett, 1989 and Hubley and Zumbo, 1996). We believe that also employing ME to learn about constructs and their functioning across different groups holds significant promise. Although ME cannot always be used to directly address theoretical propositions, it can serve as one method of doing so. In sum, this study suggests that blue- and white-collar workers possess different conceptualizations of various aspects of their jobs. In particular, the results indicate that researchers may need to rethink their understanding and treatment of these and perhaps other job facets. We hope the current findings represent a place to start in helping researchers to further explore this idea.