work and occupations - kocwcontents.kocw.net/kocw/document/2015/sungkyunkwan/kimsanguk… ·...

http://wox.sagepub.com/Work and Occupations

http://wox.sagepub.com/content/38/1/3The online version of this article can be found at:

DOI: 10.1177/0730888410383367 2011 38: 3 originally published online 8 November 2010Work and Occupations

Sang-Wook Kim and Charles W. MuellerOccupational Contexts : The Case of South Korea

Occupational and Organizational Commitment in Different

Published by:

http://www.sagepublications.com

can be found at:Work and OccupationsAdditional services and information for

http://wox.sagepub.com/cgi/alertsEmail Alerts:

http://wox.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://wox.sagepub.com/content/38/1/3.refs.htmlCitations:

What is This?

- Nov 8, 2010 OnlineFirst Version of Record

- Jan 20, 2011Version of Record >>

at SungKyunKwan University on May 26, 2012wox.sagepub.comDownloaded from

http://wox.sagepub.com/

http://wox.sagepub.com/content/38/1/3

http://www.sagepublications.com

http://wox.sagepub.com/cgi/alerts

http://wox.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://wox.sagepub.com/content/38/1/3.refs.html

http://wox.sagepub.com/content/38/1/3.full.pdf

http://wox.sagepub.com/content/early/2010/11/05/0730888410383367.full.pdf

http://online.sagepub.com/site/sphelp/vorhelp.xhtml


Work and Occupations38(1) 3 –36

© The Author(s) 2011Reprints and permission: http://www.sagepub.com/journalsPermissions.nav

DOI: 10.1177/0730888410383367http://wox.sagepub.com

383367WOX38110.1177/0730888410383367Kim and MuellerWork and Occupations© The Author(s) 2011

Reprints and permission: http://www.sagepub.com/journalsPermissions.nav

1Sungkyunkwan University, Seoul, Republic of Korea2University of Iowa, Iowa City, IA, USA

Corresponding Author:Sang-Wook Kim, Department of Sociology, Sungkyunkwan University, 3-53 Myungryun-Dong, Jongno-Ku, Seoul 110-745, Republic of Korea Email: [email protected]

Occupational and Organizational Commitment in Different Occupational Contexts: The Case of South Korea

Sang-Wook Kim1 and Charles W. Mueller2

Abstract

Although the positive relationship between occupational commitment and organizational commitment has been well documented for occupation- specific and organization-specific studies, there is much less research that documents between-occupation differences in this relationship. Drawing from the occupational subculture literature, the authors offer and test hy-potheses about between-occupation differences in levels of occupational and organizational commitment and in the correlation between these two forms of work commitment. Data from the 2005 Korean General Social Survey are used in the analysis. They find that (a) these two forms of commitment, although distinct constructs, are positively related for those in all occupa-tions, with no significant between-occupation differences in the correlations; however (b) occupational commitment is greater for professionals than for managers as well as the nonprofessional cluster of clerical, service and sales, and manual occupations; and (c) managers exhibit greater organizational commitment relative to semiprofessionals and those in the nonprofessional cluster of clerical, service and sales, and manual occupations. In addition,



4 Work and Occupations 38(1)

it is only for professionals that occupational commitment is greater than organizational commitment. The authors do not believe that these results are unique to Korean society but argue that research such as theirs on be-tween-occupation differences in work commitment needs to be undertaken in other cultural contexts. Perhaps more important, the impact on work commitment of the conflict between occupational subcultures and organiza-tional cultures needs to be a focus in future research.

Keywords

organizational commitment, occupational commitment, occupational contexts, occupational subculture, Korean General Social Survey

For close to five decades sociology of work and organization scholars have expended considerable energy on studying various forms of work commit-ment. The most attention has been given to organizational commitment, although work motivation, professional commitment, career commitment, occupational commitment, and job involvement have also been studied extensively (Meyer & Allen, 1997; Morrow, 1993). The total body of research has been massive in large part because of these forms of commitment having been examined as independent, intervening, moderating, and dependent vari-ables in various models offered to understand employee behaviors better. As a byproduct of this research, an interest emerged in the interrelationships among various forms of work commitment. Because these studies typically report correlations among different types of commitment, it was possible to conduct meta-analyses (Lee, Carswell, & Allen, 2000; Mathieu & Zajac, 1990) of the true (corrected) intercorrelations among these forms of commit-ment. As a consequence, we now have solid evidence about how various forms of commitment are interrelated. There still are, however, a number of unanswered questions about the interrelations of different forms of commit-ment. One of these questions concerns occupation differences in employee commitments, that is, for our purposes, occupation differences in organiza-tional commitment and occupational commitment. Concretely, our goal is to use major occupational groupings as contexts that could differentially affect employee occupational and organizational commitment. We develop and test hypotheses (a) about how levels of organizational and occupational commit-ment vary by occupation and (b) about how the correlation of organizational and occupational commitment varies by occupation.



Kim and Mueller 5

The bulk of the research on various forms of commitment has been based on data from Western nations, and the United States in particular. In addition, previous research has relied primarily on occupation-specific samples and/or samples from single organizations. There is very little research on non- Western data from national samples that allows for examining how occupa-tional contexts affect organizational and occupational commitment. We hope to contribute toward remedying this deficiency with our analysis of data from a recent national sample from South Korea.

Conceptual and Definitional IssuesMorrow (1993) has conceptualized work commitment as encompassing job involvement, organizational commitment, occupational commitment, and work ethic endorsement. Our focus is only on organizational commitment and occupational commitment. And although acknowledging the existence of various dimensions or types of organizational commitment (Meyer & Allen, 1997), our concern is only with affective organizational commitment. Following Porter, Steers, Mowday, and Boulian (1974), we define organiza-tional commitment as “the relative strength of an individual’s identification with and involvement in a particular organization” (p. 604).

As discussed by Lee et al. (2000), career commitment, professional com-mitment, and occupational commitment have been used interchangeably in the commitment literature. This is found in meta-analyses (Lee et al., 2000; Mathieu & Zajac, 1990) that examine what is labeled as the relationship between organizational and occupational commitment but that actually relies on data sets that measure career commitment, professional commitment and occupational commitment and subsume them under the label “occupational commitment.”

Even though they are often used interchangeably, career, occupational, and professional commitment are not the same. Career commitment is typi-cally used in two distinct ways: (a) to refer to a lifelong series of related training/vocational decisions and work-related activities (Arther, Hall, & Lawrence 1989; Greenhaus, 1987) and (b) to refer to attachment to an identi-fiable and specific line (type) of work (Blau, 1985; Morrow, 1993). Professional commitment, although often thought about as involving a long-term attachment (a career), specifies loyalty to the profession (professional occupation) one has selected and been trained in. Clerical, sales, service, and manual occupations seldom are identified as providing careers (as defined above in its first use) and are not occupations for which there would be a




professional commitment, but there clearly is some level of occupational commitment for those working in these occupations. Faced with these defini-tional dilemmas, Lee et al. (2000) provide a convincing argument that “occu-pation” is a more global or general concept for one’s work or job than is career or profession and is preferable when studying a broad range of occupa-tions because it includes both professional and nonprofessional occupations. In addition, relying on the second definition of career (see above), we agree with Lee et al. (2000) who emphasize that their definition of occupation spec-ifies an “identifiable and specific line of work” (p. 800). We adopt their defi-nition of occupational commitment as “a psychological link between a person and his or her occupation that is based on an affective reaction to that occupa-tion” (p. 800).

Previous Research on the Relationship Between Organizational and Occupational CommitmentThe vast majority of the research that considers both occupational and orga-nizational commitment is mainly interested in the correlation between the two. Counter to some claims (Kalleberg & Berg, 1987) of a zero-sum relationship—if commitment to one focus increases, commitment to other foci will decrease—there is absolutely no question that the two forms of commitment are positively related. With no moderator analysis for type of occupation, Mathieu and Zajac (1990) found the true (corrected) correlation between occupational commitment and organizational (attitudinal) commit-ment to be .45. Lee et al. (2000) found the affective organizational commit-ment and occupational commitment correlation also to be .45. Most interesting for our goals, however, is their moderator analysis that examines this correlation for professionals and nonprofessionals. They hypothesize a stronger correlation for those working in “profession-consistent” work con-texts but then state “It is not at all clear how occupational and organizational commitment are related among nonprofessional employees” (Lee et al., 2000, p. 801). The correlation for professionals in organizations that corre-spond to the profession (e.g., nurses in hospitals) was .48, for professionals in noncorresponding organizations (e.g., nurses in schools) the correlation was .23, whereas the correlation for nonprofessionals (all studies that did not study professions) was .45. The finding that work context is important for professionals is consistent with the research of Wallace (1995) who found, for lawyers, that the correlation between occupational and organizational commitment was .47 when the lawyers worked in professional organizations and only .28 when they worked in nonprofessional organizations. In a related



Kim and Mueller 7

meta-analysis of samples of only professionals, Wallace (1993) found in a moderator analysis the correlation for professionals who manage to be .47, whereas it was .29 for professionals who did not manage. We found no stud-ies of this correlation across multiple occupation contexts.

Much less is known about between-occupation differences in the mean levels of occupational and organizational commitment. As we argue below for studying occupations as meaningful contexts, we review the limited research that has been conducted.

Occupations as Contexts: Arguments and HypothesesAlthough the heyday of interest in how occupational and organizational commitment are interrelated may have passed, there still is considerable research being undertaken (see Blau, 2003; Blau & Holladay, 2006; E. M. Chang, 1999; J. Y. Chang & Choi, 2007; Cohen, 1999, 2000, 2006, 2007; Den Hartog & Belschak, 2007; Hackett, Lapierre, & Hausdorf, 2001; McAculay, Zeitz, & Blau, 2006). And although the earlier as well as these more recent studies have been quite valuable, they have two striking simi-larities that make their generalizability problematic: (a) the vast majority of the studies are of professionals or semi-professionals and (b) the studies typically rely on data from one or only several organizations. As a conse-quence, we have only limited data across a full range of occupations for a total national sample. These limitations, by themselves, are basis enough for looking at a full range of occupations for a representative national sample. However, there are two other reasons for wanting to examine a more com-plete range of occupations when asking how occupational and organizational commitments are related.

A second reason for emphasizing occupational context comes from the relatively recent, and bold, claim that occupations, rather than social classes, are the more important bases for understanding societal stratification and inequality. Coming from a mainstream macrosociological perspective, and citing the classical work of Durkheim (1893), the major proponents (Grusky & Sørensen, 1998) see occupations as a critical basis for social community identity. Specifically, they see three of the basic claims about social classes—that they exhibit closure, they are readily open to exploitation, and they con-tain the seeds for collective action—as actually more characteristic of occupations than of what are usually identified as social economic classes. They conclude that occupations are the more significant unit of analysis for studying and understanding societal stratification and inequality (see Johnson,




Mortimer, Lee, & Stern, 2007, on how work values cluster for employees; Sengupta, Edwards, & Tsai, 2009, for the differences between “good” and “bad” jobs; and Sørensen, 2000, for counter arguments and discussion). This argument is entirely consistent with our final reason for examining the effects of occupational context.

The third reason comes from the “occupational subculture” literature. Although organizational and sociology of work ethnographic studies have for some time identified and studied occupational subcultures (e.g., H. Applebaum, 1981; Becker, 1951; Bryant, 1972; Gamst, 1980; Lipset, Trow, & Coleman, 1956; Pilcher, 1972; Salaman, 1974; Steffens, 1972; Terkel, 1974), arguably the most often–cited theoretical work is by Harrison Trice (1993) and Trice and Beyer (1993), where the argument is made for going beyond just the study of organizational cultures to consider the exis-tence of occupational subcultures within organizations. Trice (1993) con-cluded that, at the time of his writing, the focus in the study of organizational behavior (and work cultures) was almost entirely on organizations, rather than on occupations. Accepting that most organizations usually have multiple cultures rather than a single homogeneous culture (Dalton, 1959; Gregory, 1983; Louis, 1985; Riley, 1983), Trice and Beyer (1993) argue that studying organizational subcultures is critical to the understanding of employee behav-ior in these organizations. Organizational subcultures “consist of distinctive clusters of ideologies, cultural forms, and other practices that identifiable groups of people in an organization exhibit” (Trice & Beyer, 1993, p. 174). Within modern work organizations, occupational cultures are often the most well-organized sources of subcultures (Trice, 1993). This is true for occupa-tions because “(1) they provide the basis for diffuse subcultures in the general society, and (2) their members often form face-to-face subcultures within the organizations that employ them” (Trice, 1993, p. 178). When occupation members’ lives become integrated with occupational identities and ideolo-gies, the social groups that emerge are referred to as occupational communi-ties (Trice & Beyer, 1993). The degree to which an occupational community is strong or weak depends on the existence of the following closely related characteristics:

(1) Esoteric knowledge and expertise, (2) extreme or unusual demands, (3) consciousness of kind, (4) pervasiveness—the occupational culture permeates nonworking life, (5) ideologies that confer favorable self-images and social value to the tasks, (6) the extent to which members of the occupation are members’ primary reference group, and (7) the abundance of consistent cultural forms. (Trice, 1993, p. 26)



Kim and Mueller 9

There are clear implications of this conceptualization of an occupational community for between-occupation differences in degree of organizational commitment and occupational commitment—members of occupations where these characteristics are more likely found will display the greatest occupa-tional commitment, and likely greater occupational commitment than organi-zational commitment.

Merging stratification and organizational perspectives, Ritzer and Trice (1969) argue that those in low-status occupations (e.g., janitors and clerks) are unlikely to be very committed to their occupations, whereas those in high-status occupations (e.g., doctors and lawyers) will be much more highly com-mitted to their occupations. They argue that those in lower status occupations, to give meaning to their work lives, will direct their attention, and commit-ment, toward the organization.

This line of reasoning is expanded when Trice (1993) argues that “those occupations in which becoming a member is an ordeal generate more com-mitment (to the occupation) than those in which becoming a member is easier” (p. 131). He compares the “rites of passage” of university professors and physicians with those of becoming a school teacher or a social worker and argues that role (occupational) commitment will be greater for the two former occupations because of the larger and more complex roles they must learn.

Most of our hypotheses come from the above literature. Because profes-sionals are more likely to be part of an occupational community—journals with peer reviews, professional meetings, part of national or international occupational labor markets (as opposed to local or firm labor markets), more likely to have experienced a more rigorous “ordeal” to enter the occupation, and clearly are of higher status—their occupational commitment will be higher than all other occupational groups. Following Ritzer and Trice (1969), occupational commitment should be lowest for those in unskilled occupa-tions where organizational commitment should be higher. Managers may experience “more difficult rites of passage” than clerical, sales, service, and manual workers and thus have relatively high occupational commitment, but it is much more likely for them to have a stronger commitment (when com-pared with professionals) to their employing organization. Managers have perhaps the greatest investment in the employing organization. Their job involves making and implementing decisions that are designed to make the business successful. If these decisions are not sound, then the manager must take responsibility for any resulting failures. Thus, their commitment must be to the organization itself and not just to “doing the job.” Our first two hypoth-eses come from these arguments.




Hypothesis 1: Occupational commitment will be lowest for clerical, service and sales, and manual occupations and highest for profes-sional occupations.

Hypothesis 2: Organizational commitment will be highest for managers.

Our third hypothesis is about the difference between the degree (or level) of occupational and organizational commitment (we will refer to this as the “commitment gap”) over different occupations. Based on the previous argu-ments we expect the commitment gap to be greatest for those in professional occupations, where occupational commitment will be greater than organiza-tional commitment (see Wang & Armstrong, 2004, for empirical support for higher occupational commitment for professionals; but see Kwon & Banks, 2004, for a study of professionals where organizational commitment is higher than professional commitment).

There are at least three factors operating here. The first involves the deci-sion to undertake training in a profession. This is, in itself, a commitment to the profession because the training in most professions is time-consuming and expensive. The second factor is the socialization (both during training and while working in the profession) professionals experience. An attitudi-nal, and often affective, attachment to the profession is instilled as the person is socialized into the profession. Professionals have been socialized to view their occupation (profession) as their “home” and not the organization they work for. The third factor, already mentioned above, is the structure of the labor market professionals belong to. Professionals operate more in an exter-nal occupational labor market (McDuff & Mueller, 2000; Osnowitz, 2006); professionals are selected and sorted based on national and international job opportunities rather than just local opportunities. All three of these factors should operate to instill a higher level of occupational commitment, relative to organizational commitment, among professionals.

We concur with Lee et al. (2000) that hypotheses about nonprofessional occupations are not as easily formulated, so we limit our hypothesis about the “commitment gap” to professionals. However, based on our early depictions of occupational categories, we would not be surprised if managers are more organizationally committed than occupationally committed and among cleri-cal, service and sales, and manual occupations, organizational commitment is also greater than occupational commitment.

Hypothesis 3: Relative to those in other occupations, professionals will exhibit greater occupational commitment than organizational com-mitment. (We refer to this as the “commitment gap hypothesis.”)



Kim and Mueller 11

Our fourth hypothesis takes us from predicting between-occupation mean differences to predicting between-occupation correlation differences. It is correlations among work commitment forms that have held the interest of researchers much more often than mean differences. Although the meta-analyses we have already referred to clearly show a positive relationship between occupational commitment and organizational commitment, we have yet to find data that disaggregate by more than a few crude occupational categories. As with mean differences, we are able to examine this correlation for six occupational categories: professionals, semiprofessionals, managers, clerical workers, sales and service employees, and manual workers. As we argue in the section on the Korean context, we believe our fourth hypothesis drawn primarily from Western data and research does not need to be altered for Korea (see E. M. Chang, 1999; Kim & Seo, 2003, to support this). Our fourth hypothesis is the following:

Hypothesis 4: Occupational commitment and organizational commit-ment are positively correlated for all occupations.

Our fifth hypothesis specifies between-occupation correlation differences. Although not specifically studying the issues addressed in our article, Krecker (1994) discusses contexts in which there is a “loose coupling” of individual (occupational) and organizational careers. While previous research is informative about the “bundling” of the two forms of commitment for professionals, we know essentially nothing about this for other occupations. Lee et al. (2000) do examine the relationship between occupational (com-bines occupational, career, and profession commitment measures) and orga-nizational commitment for professionals versus nonprofessionals, but as we have said previously, the authors are unwilling to commit to hypotheses about nonprofessional employees. Lee et al. (2000) seem somewhat sur-prised that the nonprofessional correlation (.448) is so similar to the correla-tion (.484) for professionals in corresponding organizations (we computed a weighted r from the Lee et al. data for all professionals to be .443). They explain this by concluding that the mechanisms operating to produce these similar correlations are quite different. Professionals have developed their career commitment prior to their development of organizational commit-ment. If the organization that employs them is concurrent with their career expectations (i.e., it is a “corresponding” organization), then the two forms of commitment will be more highly correlated. This larger correlation in this situation is consistent with Wallace’s (1995) claim that bureaucratic organizations are not that incompatible with professional values and that




professionals have learned to adapt to bureaucratic work environments. Unfortunately, we do not have access to data on the types of organizations professionals are employed by. However, we still argue that for profession-als, in general, there is a weaker bundling of occupational and organizational commitment. Among nonprofessionals, where there is less antibureaucracy fervor and there is less intense pre-entry occupational socialization, career and organizational commitment are likely to develop concurrently based on workplace experiences. Our fifth hypothesis is the following:

Hypothesis 5: Occupational and organizational commitment will be “bundled” (correlated) the lowest for professionals.

Our final hypothesis is an extension of the argument presented for Hypothesis 2 and comes from arguments and data provided by Wallace (1993). Wallace argues that professionals who also manage have a greater vested interest in the organization they work for, thus more strongly merging their organizational and career commitment. The success of the organization is important in their career advancement, even if they do not continue employment in that company. Wallace found the corrected correlation for professionals who manage to be .469, whereas it was .287 for those who did not manage. Although mean differences do not necessarily translate into cor-relation differences, Wang and Armstrong (2004) found that professionals who are project managers have higher professional commitment than profes-sionals who are not project managers. Our sixth hypothesis is the following:

Hypothesis 6: Among professionals, the correlation between occupa-tional commitment and organizational commitment will be higher for those who supervise others.

The Korean ContextEmployment practices, rather than occupational contexts, are what differ-entiate Korean workers and Western workers. On a continuum ranging from merit- to seniority-based employment practices (Dore, 1973; Lincoln & Kalleberg, 1990), Korean workers are, relative to their Western counter-parts, located more toward the seniority-based end of the continuum (Bae & Chung, 1997; Steinberg, 1989). Despite some objections to this traditional characterization (Bae & Form, 1986), seniority still is highly valued in Korea. Most Korean work organizations are hierarchically struc-tured in terms of seniority rules and regulations, and substantial rewards



Kim and Mueller 13

and benefits accrue to senior positions (Eom, 2006). Specifically, and rela-tive to Western organizations, pay schemes in most Korean organizations are often based on firm tenure and age rather than on individual perfor-mance or achievement. Various kinds of payment, fringe benefits, and allowances are often offered independent of an employee’s performance. Although performance evaluation matters in deciding crucial organiza-tional rewards such as pay and promotion, it is not always the major crite-rion in those decisions. Instead, decisions can be informal and personalistic in nature or can be based on collectivistic (group) rules and regulations. In summary, performance evaluations are less important in Korea than in the West (Kim, 1999).

The question before us is “do these cultural differences require that we qualify our hypotheses?” We argue that they do not. We found no evidence that these cultural workplace features in Korea characterize one occupational category more than another. If this is so, then we believe our hypotheses are appropriate for Korean employees. In fact, in a study to compare the levels of and correlations between organizational and occupational commitments for medical doctors and nurses who were employed in large-scale hospitals of Korea, Kim and Seo (2003) found results consistent with our expectations: Doctors (professionals) are committed more to their occupations than organi-zations, whereas nurses (semiprofessionals) are committed more to their organizations. In addition, correlations between the two commitments are lower for doctors (unadjusted r = .422) than nurses (r = .627).

MethodData

The 2005 Korean General Social Survey (KGSS; Kim et al., 2006) is the source of data we use to test the hypotheses. The KGSS, a Korean version of the General Social Survey in the United States, is a national sample survey conducted each year since 2003. The target population includes the adult population aged 18 years or older who live in households in Korea. A repre-sentative sample is drawn from this population by means of multistage area probability sampling procedures. Structured face-to-face interviews that are administered by a trained group of interviewers are then conducted for the selected sample.1 The 2005 KGSS included two topical modules, the International Social Survey Programme (ISSP) 2005 module for Work Orientations III and the special module on Inequality and Equity of Korea. Data from both modules are used in the analysis.




The survey, in the field from June to August 2005, yielded a valid sample of 1,613 out of 2,500, for a response rate of 64.5%.2 From these 1,613, we conducted the analysis on 535 cases (552 minus 17 cases lost because of listwise missing data). The reduction from 1,613 to 552 requires an explana-tion. Two factors are operating. In Korea, as in most General Social Survey–type surveys conducted around the world, the target population is the total adult population (employed and unemployed) with no restriction on the upper extreme of the age. About 40% of the Korean adult population is unemployed (e.g., students, housewives, and the like). In addition, Korea, relative to developed Western nations, has a smaller percentage of wage workers who are employed. The percent of non-self-employed wage earners among the adult population in 2005 in Korea was 39.6%; whereas in Sweden, it was 66.6%; in Great Britain, it was 52.4%; in the United States, it was 61.2%; and in Germany, it was 50.8% (International Labor Office [ILO], 2006).

MeasurementTable 1 contains descriptive statistics for the variables used in the analysis. Organizational commitment was measured by three items adapted from the Organizational Commitment Questionnaire (Porter et al., 1974), and occupa-tional commitment was measured by three items adapted from Blau’s (1985) measure of what he referred to as career commitment. However, as discussed by Lee et al. (2000), Blau’s (1985) definition and measurement of career emphasizes a person’s “line of work” rather than a series of jobs, training choices, and various work-related activities over one’s lifetime. In capturing “line of work” attachment, the measure we used can legitimately be labeled as occupational commitment. Both measures capture an employee’s affective orientation (loyalty) to the employer and the occupation, respectively. Table 2 provides the results of a confirmatory factor analysis designed to substantiate that the two forms of commitment are positively related but are conceptually distinct forms of commitment. Using LISREL analysis several competing, nested factor models were compared. The two-factor model, as opposed to a one-factor model, clearly has the best fit with the data, meaning that organizational commitment and occupational commitment, although positively correlated, are empirically, as well as conceptually, distinct con-structs. Lee et al. (2000) in their meta-analysis found a corrected correlation of .449 between these two forms of commitment, and Mathieu and Zajac (1990) found the correlation to be .438. Our LISREL-corrected correlation between these two latent constructs is also positive but substantially larger at .73. We are reluctant to claim that this difference is meaningful because it



Kim and Mueller 15

Table 1. Descriptive Statistics for the Variables (N = 535)

VariablesNo. of Items Mean (Range) SD α

Source of the Measures

Commitment Organizational commitment 3 3.342 (1-5) 0.840 .700 Porter et al.

(1974)a

Occupational commitment 3 3.053 (1-5) 1.013 .829 Blau (1985)a

Occupation Occupational categoriesb 1 NA NA NAc ISCO-88

(ILO, 1990)Employee characteristics Genderd 1 0.606 (0-1) 0.489 NAc — Age 1 38.198 (18-72) 10.538 NAc — Years of schooling 1 13.586 (3-19) 3.001 NAc — Firm tenuree 1 5.935 (0.083-35.000) 6.832 NAc — Union membershipf 1 0.168 (0-1) 0.374 NAc — General training 2 2.743 (1-4) 0.700 .636 —Workplace rewards Wageg 1 212.312 (25-1,050) 147.032 NAc —h

LN (Wage) 5.144 (3.219-6.957) 0.677 Advancement 1 2.837 (1-5) 1.066 NAc —h

Job security 1 3.277 (1-5) 1.101 NAc —h

Autonomy 1 3.236 (1-5) 1.153 NAc —h

Skill improvement opportunity

1 3.224 (1-5) 1.099 NAc —h

Service opportunity 2 3.536 (1-5) 0.877 .810 —h

Routinization 1 2.735 (1-5) 1.023 NAc —h

Professional value factors Value-career opportunity 1 4.458 (1-5) 0.660 NAc —h

Value-autonomy 1 3.972 (1-5) 0.831 NAc —h

Value-service orientation 2 4.144 (1-5) 0.704 .842 —h

Note: ISCO = International Standard Classification of Occupations; ILO = International Labor Office.a. The items used in the measure are adapted from the source indicated.b. 1 = Managers (n

1 = 86), 2 = Professionals (n

2 = 62), 3 = Semiprofessionals (n

3 = 77), 4 = Clerical

(n4 = 99), 5 = Service/sales (n

5 = 66), 6 = Manual (n

6 = 145).

c. Coefficient alpha cannot be assessed for a single item (Nunnally, 1978).d. 0 = Female, 1 = Male.e. In cumulative years.f. 0 = Nonmember, 1 = Member.g. Monthly average work income. The unit is 10,000 won in Korean currency, with 10,000 won in 2005 being equivalent to approximately US$9.7.h. Detailed measures are available in the International Social Survey Programme website (http://www.issp.org).



16

Tabl

e 2.

Con

firm

ator

y Fa

ctor

Ana

lyse

s fo

r th

e C

ompe

ting

Mod

els

of O

rgan

izat

iona

l and

Occ

upat

iona

l Com

mitm

ent

(N =

535

)

Fact

or L

oadi

ngsa

Com

petin

g Fa

ctor

Mod

els

O

rgan

izat

iona

l C

omm

itmen

tO

ccup

atio

nal

Com

mitm

ent

(M0)

. N

ull M

odel

(M1)

. O

ne-F

acto

r M

odel

(M2)

. Tw

o-Fa

ctor

Mod

el

Org

aniz

atio

nal c

omm

itmen

t

1. I

am w

illin

g to

wor

k ha

rder

tha

n

I

have

to

in o

rder

to

help

the

firm

or

org

aniz

atio

n I w

ork

for

succ

eed.

(R

)

.583

χ2 = 1

231.

880,

df

= 1

5G

FI =

—A

GFI

= —

RM

SR =

—N

FI =

—C

FI =

—

IFI =

—

χ2 = 1

25.0

64 (

p <

.01)

df =

9G

FI =

.920

AG

FI =

.814

RM

SR =

.067

NFI

= .8

98C

FI =

.905

IFI =

.905

χ2 = 1

7.66

7 (p

< .0

5)df

= 8

GFI

= .9

89A

GFI

= .9

71

RM

SR =

.033

NFI

= .9

86

CFI

= .9

92IF

I = .9

92

2.

I am

pro

ud t

o be

wor

king

for

my

firm

or

orga

niza

tion.

(R

).8

70

3.

I w

ould

tur

n do

wn

anot

her

job

that

o

ffere

d qu

ite a

bit

mor

e pa

y in

ord

er

to

stay

with

thi

s or

gani

zatio

n. (

R)

.625

Occ

upat

iona

l com

mitm

ent

1.

I de

finite

ly w

ant

a ca

reer

for

mys

elf

in t

his

occu

patio

n. (

R)

.761

2.

Thi

s is

the

idea

l voc

atio

n fo

r a

life

wor

k. (

R)

.854

3.

If I

coul

d do

it a

ll ov

er a

gain

, I w

ould

c

hoos

e to

wor

k in

thi

s pr

ofes

sion

. (R

).7

52

Not

e: G

FI =

goo

dnes

s-of

-fit

inde

x; A

GFI

= a

djus

ted

good

ness

-of-f

it in

dex;

RM

SR =

roo

t m

ean

squa

re r

esid

ual;

NFI

= n

orm

ed fi

t in

dex;

CFI

= c

ompa

rativ

e fit

in

dex;

IFI =

incr

emen

tal f

it in

dex.

a. V

alue

s ar

e la

mbd

as (

λs)

obta

ined

from

the

com

plet

ely

stan

dard

ized

sol

utio

ns in

LIS

REL

8. A

ll va

lues

are

sig

nific

ant

at t

he .0

01 le

vel.

LISR

EL-c

orre

cted

zer

o-or

der

corr

elat

ion

betw

een

the

two

late

nt v

aria

bles

is .7

33 (

p <

.001

). (R

) in

dica

tes

reve

rse-

code

d.



Kim and Mueller 17

must be remembered that our Korean results are based on only one sample, whereas the meta-analyses rely on multiple samples for which an average correlation over numerous samples is obtained.

The occupational categories are based on the International Standard Classification of Occupations 1988 (ISCO-88; ILO, 1990). The first digit of the ISCO-88 four-digit codes, which consist of 390 unit groups in total, hier-archically distinguishes a total of nine major categories: (1) legislators, senior officials, and managers; (2) professionals; (3) technicians and associate pro-fessionals; (4) clerks; (5) service workers and shop and market sales workers; (6) skilled agricultural and fishery workers; (7) craft and related trades work-ers; (8) plant and machine operators and assemblers; (9) elementary occupa-tions. These are all mutually exclusive categories based on the ISCO-88 (ILO, 1990).3

This categorization differentiates skills acquired through education and training, rather than other differences (e.g., industry or employment status; Ganzeboom & Treiman, 1996), and facilitates grouping into professional and various nonprofessional categories. Categories 4 through 9 are all clearly nonprofessional occupations (ILO, 1990), and categories 6 through 9 are manual occupations and are grouped together accordingly (we label these as MAN).4 We retain the first five categories as distinct: (1) Managers (MG), (2) Professionals (PR), (3) Semiprofessionals (SP), (4) Clerical (CL), and (5) Service and Sales (S&S).

The other variables listed in Table 1 are control variables. In assessing the hypothesized between-occupation “commitment gap” differences (Hypothesis 3), factors known to influence commitment will be controlled. Specifically, we use three different sets of controls: employee characteristics, workplace rewards, and professional value factors. Employee characteristics are the sociodemographic characteristics of an individual employee and include gen-der, age, educational attainment, firm tenure, union membership, and general training. Workplace rewards include the organizational inducements of wages, advancement opportunities, job security, autonomy, skill improve-ment opportunity, service opportunity, and routinization (variety is the reward).5 Following Wallace (1995) and Osinsky and Mueller (2004), profes-sional value factors refer to several of the most important work conditions consistent with professional values and include the importance of career opportunities, autonomy, and service orientation. These value factors are sup-posed to have the strongest effect on the work commitment of professionals. Complete descriptions of these control variables are found on the ISSP web-site (http://www.issp.org).




The appendix presents the correlations among all of the variables in the analysis. There are only four correlations greater than .500. Organizational commitment and occupational commitment are correlated by .585, routiniza-tion is correlated by −.539 with occupational commitment and −.601 with advancement opportunities, and skill improvement opportunities is corre-lated .541 with advancement opportunities. Multicollinearity is not a serious problem for the variables in the analysis.

ResultsTable 3 displays the means by occupation for all variables, the analysis of variance for occupation differences, and the multiple comparison test results for occupation differences among the means. All the F tests are significant except for union membership, the importance (value) of autonomy and the importance (value) of a service orientation. The post hoc tests identify sig-nificant differences but also allow for identifying any clustering of occupa-tions based on similarities (no differences in means) for each of the variables in the table.

With reference to occupation differences in employee characteristics, managers (92%) and manual workers (75%) are overwhelmingly male, and they are also the oldest employees. Not surprisingly, professionals stand alone with the most schooling (16.9), whereas service and sales and manual employees cluster together with the least schooling. Although managers are not different from professionals in their high level of tenure, it is managers who are consistently higher in tenure than all of the other occupations. Korea is unique in that there are no significant differences in between-occupation union membership (average is 17%). Professionals have a greater level of general training relative to clerical, service and sales, and manual workers.

For workplace rewards the post hoc tests indicate that the least number of between-occupation differences is for job security and autonomy where it is the professionals who stand out uniquely as being rewarded the most on these two factors. With regard to wages, it is managers and professionals who clus-ter together with high pay. For advancement opportunities and skill improve-ment opportunities, there are two distinct clusters: managers, professionals and semiprofessionals (high opportunities) and clerical, service and sales, and manual (low opportunities). Professionals are unique with low routiniza-tion (high variety), whereas service opportunities differ by occupation but do not seem to fit any particular pattern. Finally, it is interesting that for the three value variables, it is only for the importance of career opportunities that there are between-occupation differences and, specifically, only for professionals



Kim and Mueller 19

Table 3. Means, ANOVA Results, and Post Hoc Between-Occupation Testsa (N = 535)

Variables Meanb F Ratio MG PR SP CL S&S

Commitment Organizational commitment Managers (n

1 = 86) 3.717 (0.796) — — — — —

Professionals (n2 = 62) 3.522 (0.909) n.s. — — — —

Semiprofessionals (n3 = 77) 3.429 (0.780) 7.223 n.s. n.s. — — —

Clerical (n4 = 99) 3.263 (0.771) (p < .001) * n.s. n.s. — —

Service/Sales (n5 = 66) 3.081 (0.869) * n.s. n.s. n.s. —

Manual (n6 = 145) 3.170 (0.812) * n.s. n.s. n.s. n.s.

Occupational commitment Managers (n

1 = 86) 3.264 (0.792) — — — — —



Clerical (n4 = 99) 2.913 (0.998) (p < .001) n.s. * n.s. — —

Service/sales (n5 = 66) 2.601 (1.090) * * * n.s. —

Manual (n6 = 145) 2.766 (0.916) * * * n.s. n.s.

Employee characteristics Male Managers (n

1 = 86) 0.919 (0.275) — — — — —

Professionals (n2 = 62) 0.484 (0.504) * — — — —

Semiprofessionals (n3 = 77) 0.546 (0.501) 21.202 * n.s. — — —

Clerical (n4 = 99) 0.444 (0.499) (p < .001) * n.s. n.s. — —

Service/sales (n5 = 66) 0.303 (0.463) * n.s. n.s. n.s. —

Manual (n6 = 145) 0.752 (0.434) n.s. * n.s. * *

Age Managers (n

1 = 86) 42.41 (7.397) — — — — —

Professionals (n2 = 62) 35.76 (7.790) * — — — —


Clerical (n4 = 99) 34.39 (9.473) (p < .001) * n.s. n.s. — —

Service/sales (n5 = 66) 34.76 (10.789) * n.s. n.s. n.s. —

Manual (n6 = 145) 43.31 (11.860) n.s. * * * *

Years of schooling Managers (n

1 = 86) 14.442 (2.189) — — — — —

Professionals (n2 = 62) 16.903 (1.512) * — — — —

Semiprofessionals (n3 = 77) 15.000 (2.323) 78.918 n.s. * — — —

Clerical (n4 = 99) 14.343 (1.960) (p < .001) n.s. * n.s. — —

Service/sales (n5 = 66) 12.273 (2.440) * * * * —

Manual (n6 = 145) 10.972 (2.681) * * * * *

Firm tenure Managers (n

1 = 86) 10.689 (8.412) — — — — —



Clerical (n4 = 99) 4.942 (6.339) (p < .001) * n.s. n.s. — —

(continued)




(continued)


Service/sales (n5 = 66) 2.775 (4.916) * * n.s. n.s. —

Manual (n6 = 145) 5.369 (5.884) * n.s. n.s. n.s. n.s.

Union member Managers (n

1 = 86) 0.198 (0.401) — — — — —



Clerical (n4 = 99) 0.232 (0.424) (p > .05) n.s. n.s. n.s. — —

Service/sales (n5 = 66) 0.061 (0.240) n.s. n.s. n.s. n.s. —

Manual (n6 = 145) 0.172 (0.379) n.s. n.s. n.s. n.s. n.s.

General training Managers (n

1 = 86) 2.924 (0.644) — — — — —



Clerical (n4 = 99) 2.712 (0.650) (p < .001) n.s. * n.s. — —


Manual (n6 = 145) 2.535 (0.728) * * * n.s. n.s.

Workplace rewards Wage Managers (n

1 = 86) 359.59 (178.19) — — — — —


Semiprofessionals (n3 = 77) 199.03 (138.03) 46.123 * * — — —

Clerical (n4 = 99) 173.96 (101.56) (p < .001) * * n.s. — —


Manual (n6 = 145) 159.48 (83.44) * * n.s. n.s. n.s.

Advancement Managers (n

1 = 86) 3.233 (0.966) — — — — —



Clerical (n4 = 99) 2.707 (0.872) (p < .001) * * * — —


Manual (n6 = 145) 2.338 (1.036) * * * n.s. n.s.

Job security Managers (n

1 = 86) 3.523 (1.026) — — — — —


Semiprofessionals (n3 = 77) 3.143 (0.983) 9.024 n.s. * — — —

Clerical (n4 = 99) 3.364 (0.974) (p < .001) n.s. n.s. n.s. — —

Service/sales (n5 = 66) 3.106 (1.139) n.s. * n.s. n.s. —

Manual (n6 = 145) 2.945 (1.085) * * n.s. n.s. n.s.

Autonomy Managers (n

1 = 86) 3.488 (1.026) — — — — —



Table 3. (continued)



Kim and Mueller 21


Clerical (n4 = 99) 3.030 (1.092) (p < .001) n.s. * n.s. — —

Service/sales (n5 = 66) 2.939 (1.201) n.s. * n.s. n.s. —

Manual (n6 = 145) 3.048 (1.271) n.s. * n.s. n.s. n.s.

Skill improvement opportunity Managers (n

1 = 86) 3.558 (0.965) — — — — —



Clerical (n4 = 99) 3.020 (1.000) (p < .001) * * * — —


Manual (n6 = 145) 2.897 (1.052) * * * n.s. n.s.

Service opportunity Managers (n

1 = 86) 3.593 (0.814) — — — — —



Clerical (n4 = 99) 3.455 (0.776) (p < .001) n.s. * n.s. — —

Service/sales (n5 = 66) 3.189 (1.037) n.s. * * n.s. —

Manual (n6 = 145) 3.369 (0.868) n.s. * * n.s. n.s.

Routinization Managers (n

1 = 86) 2.663 (0.835) — — — — —

Professionals (n2 = 62) 1.936 (0.698) * — — — —


Clerical (n4 = 99) 2.879 (0.860) (p < .001) n.s. * n.s. — —

Service/sales (n5 = 66) 2.995 (1.029) n.s. * * n.s. —

Manual (n6 = 145) 3.097 (1.095) n.s. * * n.s. n.s.

Professional value factors Value career opportunity Managers (n

1 = 86) 4.523 (0.793) — — — — —



Clerical (n4 = 99) 4.556 (0.499) (p < .01) n.s. n.s. n.s. — —


Manual (n6 = 145) 4.262 (0.736) n.s. * n.s. * n.s.

Value autonomy Managers (n

1 = 86) 4.140 (0.883) — — — — —






Value service orientation Managers (n

1 = 86) 4.169 (0.784) — — — — —



(continued)





Semiprofessionals (n3 = 77) 4.046 (0.713) .420 n.s. n.s. — — —




Note: ANOVA = analysis of variance; MG = managers; PR = professionals; SP = semiprofes-sionals; CL = clerical; S&S = service and sales.a. Scheffé multiple-comparison procedure.b. Standard deviations are given in parentheses.*p < .05, two-tailed test.


versus manual workers where service opportunities are valued more by professionals.

With the exception of the results for the value items, the data in Table 3 are consistent with the claim that the work context varies across occupations or clusters of occupations—there are between-occupation differences in worker characteristics and workplace rewards—and these differences, as we have argued, could produce (or may be the consequence of) different occupational subcultures and occupational communities.

Our first two hypotheses are about between-occupation mean differences in the two forms of commitment. We rely on conservative Scheffé post hoc tests because there are so many possible comparisons. Hypothesis 1 is sup-ported. As hypothesized, clerical, service and sales, and manual occupations are the lowest in occupational commitment and are significantly lower than professionals who have the highest mean (see Table 3). In addition, service and sales and manual employees are lower in occupational commitment than semiprofessionals and managers. Professionals are also higher in occupational commitment than managers when an a priori two-tailed t test is conducted.

There is partial support for Hypothesis 2, which specifies organizational com-mitment differences (see Table 3). Managers do display greater organizational commitment when compared with those in the three nonprofessional categories (clerical, service and sales, and manual occupations) and semiprofessionals (a priori two-tailed t test). The fact that professionals and managers do not differ is consistent with Wallace’s (1995) argument that professionals have adapted to the bureaucratic organizations they are now often employed by.

Hypothesis 3 is concerned with the difference in mean levels of the two commitments and stated that, relative to those in the various nonprofessional occupations, professionals will exhibit greater occupational commitment than



Kim and Mueller 23

organizational commitment. The difference between occupational and organi-zational commitment means has been computed and given as the “deviation” for each comparison (see Table 4). A positive deviation indicates that organi-zational commitment is greater than occupational commitment, whereas a negative deviation indicates that occupational commitment is greater than organizational commitment. Because the means we are comparing do not come from independent samples and the measures do not use the same word-ings, we cannot use significance tests for either independent or correlated samples. We must rely on patterns in the data in drawing conclusions, and we need to emphasize that our statements about mean differences are always “rel-ative to” the other occupation(s) used in the comparison.

Without any controls (Table 4) the professionals are committed more to their occupations than to their organizations (deviation = −.220), whereas employees in the other five occupational categories are committed more to their organizations than occupations (deviations: MG = .453, SP = .057, CL = .350, S&S = .480, MAN = .405). Moreover, this hypothesized pattern of higher occupational commitment for professionals holds after introducing each of the three sets of controls in Table 4.6 We conclude that Hypothesis 3 is supported; relative to all other occupational categories, those in profes-sional occupations are more committed to their occupations than to their organizations.

We learn something about why the commitment gaps are different across occupations by examining the adjusted gap when different sets of variables are controlled. Controlling employee characteristics results in only small changes for all six occupational categories when the adjusted commitment gap is compared with the unadjusted gap. However, when the workplace rewards are controlled, the commitment gap for professionals, although still having occupational commitment higher, is reduced more than for any other occupational category. Several of the other gaps even increase in magnitude. This tells us that workplace rewards, which are consistently higher for pro-fessionals (see Table 3), are actually operating to boost occupational commit-ment relative to organizational commitment. On the surface, this may seem surprising because the rewards are likely provided by the employer (organi-zation) rather than by the profession. However, if we accept the argument about professionals adapting to bureaucratic organizations that hire them (Wallace, 1995), rewarding work conditions provided by the employer actu-ally enhance their occupational commitment.

Hypothesis 4 states that the correlation between organizational and occu-pational commitment will be positive for all occupational groups. The data in Table 5 support this hypothesis. Hypothesis 5, which predicts the smallest



24

Tabl

e 4.

Com

mitm

ent

Gap

s (O

rgan

izat

iona

l Com

mitm

ent

− O

ccup

atio

nal C

omm

itmen

t) b

y O

ccup

atio

nal G

roup

s (N

= 5

35)

(MG

) M

anag

ers

(n1 =

86)

(PR

) Pr

ofes

sion

als

(n

2 = 6

2)

(SP)

Se

mip

rofe

ssio

nals

(n

3 = 7

7)

(CL)

C

leri

cal

(n4 =

99)

(S&

S)

Serv

ice/

Sale

s

(n5 =

66)

(MA

N)

M

anua

l (n

6 = 1

45)

Mea

ns

Who

Do

Not

Su

perv

ise

(n

21 =

41)

Who

Do

Supe

rvis

e

(n22

= 2

1)

Una

djus

ted

Org

aniz

atio

nal c

omm

itmen

t3.

717

3.52

23.

429

3.26

33.

081

3.17

0

(3.5

37)

(3.4

92)

Occ

upat

iona

l com

mitm

ent

3.26

43.

742

3.37

22.

913

2.60

12.

766

(3

.813

)(3

.603

)

[D

evia

tion]

a[.4

53]

[−.2

20]

[.057

][.3

50]

[.480

][.4

05]

Adj

uste

d by

em

ploy

ee c

hara

cter

istic

sb

O

rgan

izat

iona

l com

mitm

ent

3.52

1 <

↓.19

6>3.

365

<↓.

157>

3.46

6 <

↑.03

7>3.

317

<↑.

054>

3.26

5 <

↑.18

4>3.

207

<↑.

037>

(3

.523

)(3

.518

)

O

ccup

atio

nal c

omm

itmen

t3.

034

<↓ .

230>

3.52

9 <

↓.21

3>3.

399

<↑.

027>

2.96

1 <

↑.04

8>2.

832

<↑.

231>

2.83

3 <

↑.06

7>

(3.7

98)

(3.6

32)

[Dev

iatio

n]a

[.487

][−

.164

][.0

67]

[.356

][.4

33]

[.374

]A

djus

ted

by w

orkp

lace

rew

ards

c

O

rgan

izat

iona

l com

mitm

ent

3.60

2 <

↓.11

5>3.

088

<↓.

434>

3.26

3 <

↓.16

6>3.

335

<↑.

072>

3.28

5 <

↑.20

4>3.

379

<↑.

209>

(3

.513

)(3

.538

)

O

ccup

atio

nal c

omm

itmen

t3.

058

<↓ .

206>

3.16

5 <

↓.57

7>3.

169

<↓.

203>

3.01

3 <

↑.10

0>2.

884

<↑.

283>

3.05

4 <

↑.28

8>

(3.8

42)

(3.5

46)

[Dev

iatio

n]a

[.544

][−

.077

][.0

94]

[.322

][.4

01]

[.325

]A

djus

ted

by p

rofe

ssio

nal v

alue

fact

orsd

Org

aniz

atio

nal c

omm

itmen

t3.

713

<↓.

004>

3.51

0 <

↓.01

2>3.

450

<↑.

021>

3.25

7 <

↓.00

6>3.

068

<↓.

013>

3.17

6 <

↑.00

6>

(3.5

77)

(3.4

14)

Occ

upat

iona

l com

mitm

ent

3.27

1 <

↑.00

7>3.

752

<↑.

010>

3.39

9 <

↑.02

7>2.

919

<↑.

006>

2.58

4 <

↓.01

7>2.

746

<↓.

020>

(3

.832

)(3

.566

)

[D

evia

tion]

a[.4

42]

[−.2

42]

[.051

][.3

38]

[.484

][.4

30]

a. D

evia

tion

= M

ean

orga

niza

tiona

l com

mitm

ent

− M

ean

occu

patio

nal c

omm

itmen

t.b.

Con

trol

var

iabl

es in

clud

e ge

nder

, age

, yea

rs o

f sch

oolin

g, fir

m t

enur

e, u

nion

mem

bers

hip,

and

gen

eral

tra

inin

g.c.

Con

trol

var

iabl

es in

clud

e ln

_wag

e, a

dvan

cem

ent,

job

secu

rity

, aut

onom

y, sk

ill im

prov

emen

t op

port

unity

, ser

vice

opp

ortu

nity

, and

rou

tiniz

atio

n.d.

Con

trol

var

iabl

es in

clud

e va

lues

in c

aree

r op

port

unity

, aut

onom

y, an

d se

rvic

e or

ient

atio

n.



Kim and Mueller 25

correlation for professions, is not supported, because there are no significant differences for any of the pairs of correlations.7

Our final hypothesis (Hypothesis 6) specifies a correlation difference between professionals who manage and professionals who do not manage. Consistent with the hypothesis, professionals who supervise others (.680) “bundle” organizational and occupational commitment more closely than professionals who do not supervise others (.425; see Table 5). The difference is not statistically significant, although the difference we do observe is con-sistent with that found by Wallace (1993), who observed that professionals who manage (supervise) exhibit a stronger correlation between organiza-tional and occupational commitment than professionals who do not manage.

DiscussionOur hypotheses were of three types: (1) mean between-occupation differ-ences in both occupational commitment and organizational commitment, (2) between-occupation differences in the “commitment gap” between occu-pational commitment and organizational commitment, and (3) between-occupation differences in the correlation between occupational commitment and organizational commitment. The hypotheses of the first and second types were generally supported, whereas there was no support at all for between-occupation correlation differences.

Table 5. Zero-Order Correlations Between Organizational and Occupational Commitment by Occupational Groupsa

Sample Correlation

Total Sample (N = 535) .585*** (MG) Managers (n

1 = 86) .420***

(PR) Professionals (n2 = 62) .507***

Who Do Not Supervise (n21

= 41) .425** Who Do Supervise (n

22 = 21) .680***

(SP) Semiprofessionals (n3 = 77) .627***

(CL) Clerical (n4 = 99) .625***

(S&S) Service/Sales (n5 = 66) .699***

(MAN) Manual (n6 = 145) .530***

a. All tests for between-occupation differences in the relationship between organizational and occupational commitment are not significant. (See Note 7 for a description of how the tests were conducted.)

**p < .01, two-tailed.***p < .001, two-tailed.




Hypothesis 1, about between-occupation differences in occupational com-mitment, received the most support. Those in professional occupations seem to be unique in their inflated occupational commitment. Their extensive pro-fession-specific training and professional socialization as well as the external occupational labor markets they operate in make them highly committed to their occupation, which they view as a career. This strong occupational com-mitment for professionals is also seen when we examined the commitment gap in testing (Hypothesis 3). Only for professionals was occupational com-mitment greater than organizational commitment. Those in nonprofessional occupations (clerical, service and sales, manual) have occupations that they identify with as jobs and not as lifelong occupational careers.

For Hypothesis 2, about between-occupation differences in organizational commitment, we found, as predicted, that managers have the highest mean relative to semiprofessionals and the three nonprofessional categories; orga-nizational commitment is not significantly different for managers and profes-sionals. Finding no difference for managers and professionals is not entirely unexpected, however. In fact, it is consistent with Wallace’s (1995) argument that professionals have largely adapted to the bureaucratic organizations they are now often employed by.

Although we found moderate size positive correlations between occupa-tional and organizational commitment for all six occupational categories (as predicted in Hypothesis 4), there was no support for Hypotheses 5 and 6, about between-occupation differences in the magnitude of the correlation. We believe that larger subsample sizes likely would produce some significant differences, but we are unwilling to claim patterns in the data given the non-significance of the between-occupation comparisons.

ConclusionsOur major conclusion is that occupational context is important in understand-ing how employees commit to their work. This is consistent with classical sociological theory (Durkheim, 1893) as well as with claims by contemporary scholars of stratification (Grusky & Sørensen, 1998), who have argued for the importance of occupations, rather than classes, in understanding human behavior within a social context. Our arguments and hypotheses about occu-pational contexts, however, relied not on stratification arguments but primar-ily on claims made about organizations as presented in the work of Trice (1993) and Trice and Beyer (1993). Although our data allow for only a weak test of what they claim about occupational subcultures, our results are consis-tent with much of what they argue about these subcultures. Their arguments



Kim and Mueller 27

go beyond just the study of occupational subcultures, of course, and we believe that it is their more general examination of multiple organizational cultures and subcultures that will prove the most fruitful in understanding how various work contexts affect employee work commitment. Their goal is not to study occupational subcultures in isolation but, instead, to understand occupational subcultures within organizations. Occupational subcultures can come in conflict with the more general organizational culture in the employ-ing organization. Where this conflict is the greatest, the difference between organizational commitment and occupational commitment should be greatest and the correlation between the two should be smaller. We did not have access to data on the organizations the employees worked for and could not examine this potential conflict. However, we believe this is the direction future research should take.

To address potential conflict created by organizational cultures and occu-pational subcultures Trice (1993) and Trice and Beyer (1993) offer a four-cell typology with Degree of Control by Formal Administrative Principle (pre-dominant vs. subordinate) and Control by Occupational Principle (predomi-nant vs. subordinate). For two of the categories (one where control is prominent, i.e., strong, for both administrative principle and occupational principle and the other where both administrative and occupational principles are subordinate, i.e., weak), there exists either mutual tolerance and accom-modation (the strong–strong context) or egalitarian relations (the weak–weak context). Examples given for the strong–strong category are lawyers and doc-tors in corporations and skilled craft workers, whereas examples for the weak–weak category are food cooperatives, alternative schools and health collectives. The category of prominent occupational principle with subordi-nate administrative principle is labeled “assimilation of management by occupation,” with examples of universities, social work agencies, hospitals, and police departments. Finally, the category of prominent administrative principle and subordinate occupational principle is labeled as “assimilation of occupation by management,” with examples of accountants, engineers, pharmacists and deskilled occupations. Our data did not allow us to under-take the analysis implied by this typology, but it should be used in future hypothesis formulation, in making research design decisions, and in the inter-pretation of results about how various forms of work commitment are related.

Although admitting to providing only a weak test of Trice and Beyer’s (1993) arguments about occupational subcultures, we do argue that most of our results are consistent with their claims and that our analysis by occupa-tional categories goes beyond what has been studied previously when the relationship between occupational and organizational commitment has been




examined. There is, however, one unique feature of Korean society that must also be addressed. As we report above, when the data are described, there is a very large proportion of Korean workers who are self-employed. As a con-sequence, the survey-filtering format resulted in these individuals not being asked about their occupational or organizational commitment. We would be hard pressed to claim that we are missing important data on occupational subcultures for these individuals (no one or only a few to interact with), but this situation does raise a question that, to our knowledge, has not been addressed. These self-employed individuals do, when loosely defined, form an organization. It would be interesting to assess whether their occupa-tional commitment would be highly correlated with their “organizational” commitment.

To conclude, we believe that understanding the relationship between an employee’s organizational commitment and occupational commitment will come from studying the interaction of organizational context (culture) and occupational context (subculture). Although not argued from an “occupa-tional subculture within an organizational culture” perspective, the work of Wallace (1993, 1995) and the meta-analysis of Lee et al. (2000) shows the value of considering organizational context as well as occupational context. Similarly, May, Korczynski, and Frenkel (2002) address how organizational form (context) intersects with occupation (knowledge workers in their study) to affect occupational and organizational commitment. These studies are on the right track but have generally studied single occupations or relied on extremely crude occupational categories. This interest in multiple occupa-tional contexts seems almost to have been an afterthought or just an investi-gation of a potentially interesting moderator variable. Occupational context was at the heart of our study, but our analysis only scratches the surface of the types of issues raised by Trice and Beyer (1993). Future research needs to go beyond our analysis to allow for the study of how the intersection of organi-zational culture and occupational subculture influences employee work commitment.



29App

endi

x. Z

ero-

Ord

er C

orre

latio

ns A

mon

g th

e Va

riab

les

(N =

535

)

Vari

able

s1

23

45

67

89

1011

1213

1415

1617

18

1. O

rgan

izat

iona

l com

mitm

ent

1.0

00

2. O

ccup

atio

nal c

omm

itmen

t.5

85**

* 1.

000

3.

Mal

e.0

83.0

50 1

.000

4.

Age

.174

***

.108

*.1

44**

1.0

00

5. Y

ears

of s

choo

ling

.093

*.1

88**

*.1

02*

−.3

89**

* 1.0

00

6. F

irm

ten

ure

.274

***

.259

***

.239

***

.431

***

.059

1.0

00

7. U

nion

mem

ber

.099

*.1

25**

.107

*.0

98*

.062

.334

***

1.00

0

8. G

ener

al t

rain

ing

.275

***

.345

***

.053

.000

.174

***

.037

.010

1.0

00

9. L

N_W

age

.212

***

.282

***

.447

***

.149

***

.391

***

.480

***

.203

***

.196

***

1.0

00

10. A

dvan

cem

ent

.414

***

.498

***

.078

−.2

09**

*.3

46**

*.0

68.0

36.3

61**

*.3

16**

* 1

.000

11

. Job

sec

urity

.314

***

.324

***−

.013

−.0

90*

.218

***

.253

***

.123

**.0

76.2

60**

*.3

62**

* 1

.000

12

. Aut

onom

y.2

68**

*.2

40**

*.0

92*

−.0

07.1

74**

*.0

58−

.092

*.2

26**

*.1

43**

*.3

65**

*.1

62**

* 1

.000

13

. Ski

ll im

prov

emen

t op

port

unity

.377

***

.440

***

.116

**−

.037

.238

***

.101

*.0

36.4

50**

*.2

67**

*.5

41**

*.2

04**

*.3

72**

* 1

.000

14

. Ser

vice

opp

ortu

nity

.390

***

.351

***

.048

.097

*.1

24**

.151

***

.053

.297

***

.199

***

.284

***

.249

***

.290

***

.476

***

1.0

00

15. R

outin

izat

ion

−.4

14**

*−.5

39**

*.0

19.1

22**

−.2

64**

*−.0

70.0

53−

.320

***

−.1

95**

*−

.601

***

−.2

32**

*−

.368

***

−.4

90**

*−

.356

***

1.0

00

16. V

alue

car

eer

oppo

rtun

ity.1

30**

.063

−.0

83−

.176

***

.175

***−

.089

*−

.085

*.1

19**

.021

.175

***

.013

.126

**.1

32**

.163

***

−.1

80**

*1.

000

17

. Val

ue a

uton

omy

.079

.034

−.0

50.0

08.0

80.0

25.0

27.0

84.0

80.1

05*

.056

.185

***

.079

.141

***

−.0

53.3

10**

*1.

000

18. V

alue

ser

vice

ori

enta

tion

.157

***

.108

*−

.082

.088

*−

.077

.018

−.0

25.0

34−

.035

.057

.075

.060

.118

**.2

34**

*−

.098

*.3

34**

*.4

03**

*1.0

00

*p <

.05,

tw

o-ta

iled

test

. **p

< .0

1, t

wo-

taile

d te

st. *

**p

< .0

01, t

wo-

taile

d te

st.




Acknowledgments

The authors would like to thank the anonymous reviewers for their comments and suggestions to help improve the article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research and/or authorship of this article:

This study was supported by the Samsung Research Fund (2009) at the Sungkyunkwan University.

Notes

1. Further details on the KGSS, plus the related internationally coordinated surveys of ISSP and East Asian Social Survey in Korea, are available in Kim (2004).

2. Two criteria are available to assess the representativeness of the sample obtained from the survey. The first is the response rate of 64.5%, a conservatively esti-mated figure, which suggests a proper representation of the target population. The second is the employment status, one of the most decisive demographic characteristics of the selected sample: employees (552, 34.2%), self-employed (349, 21.6%), and the unemployed (712, 44.1%). The corresponding percentages in the population are 39.6, 19.8, and 40.6, respectively (Korea Ministry of Labor, 2006). A nonparametric chi-square test for the distribution of employment status between the sample and population reveals statistical insignificance, again sug-gesting that the sample provides an adequate representation of the population. The two criteria, taken together, allow us to rule out the possibility that the findings in this study are because of the characteristics unique to the sample analyzed.

3. Total population survey studies such as the International Social Survey Pro-gramme (ISSP) rely on standardized measures of concepts. The Korean Gen-eral Social Survey is a part of the ISSP and must use the International Standard Classification of Occupations (ISCO-88) to measure occupation. To become a member country in the ISSP, careful and thorough training of the interviewers is mandated. Occupation is measured with the following:

What kind of work do you normally do? What do you actually do in that job? If currently not working, tell me about your last job. (WRITE A DETAILED DESCRIPTION ABOUT THE RESPONDENT’S JOB AND



Kim and Mueller 31

THE CODE NUMBER FROM THE INDUSTRY AND OCCUPATIONAL CLASSIFICATION TABLE).

The instructions in the ISCO documentation address issues raised by one of the reviewers of the article. For example, for the general category “Corporate Manag-ers,” the instructions include the following statement: “In some cases where spe-cific professional, technical or operational skills and knowledge may be required of workers at the managerial level, it may be difficult to decide whether a particu-lar job belongs to this or another subgroup.” Concrete instructions (based on what the respondent actually does on the job) are then given regarding how to decide if the respondent belongs to the manager or professional category. We are confident in the validity of the occupational coding in our data, and we would suggest that the large number of data sets used in the various meta-analyses differ substantially in how occupation is measured, with many of them not relying on the rigorous coding required in the ISCO.

4. The analysis of variance results for these four occupational categories indi-cated no significant mean differences for either organizational or occupational commitment.

5. Many of these factors are, in fact, the properties that Lincoln and Kalleberg (1990) call the “corporatist control model” of employee commitment. Such properties are consistent with what are commonly referred to as “high perfor-mance” organizations (E. Applebaum, Bailey, Berg, & Kalleberg, 2000). The model suggests that such organizations today not only strive to secure labor con-trol through normative and symbolic means, instead of coercive and utilitarian means, but they also try to deliberately maximize employee commitment to the workplace by enhancing employee integration, upward mobility, participation in decision making, and the legitimacy of the authority system.

6. These patterns are not changed by considering whether the professionals manage or do not manage (see Table 4).

7. Because traditional significance testing procedures do not exist for differences between correlations, we relied on testing the between-occupation relationship differences using regression analysis that provides t tests for an unstandardized coefficient when occupational commitment is regressed on organizational com-mitment. This involves a series of equations with occupations coded as dummy variables and interaction terms created for occupations and organizational com-mitment (Kutner, Nachtsheim, & Neter, 2004). By changing the occupation in the omitted category across equations, it is possible to conduct all of the tests for between-occupation differences in the relationship between occupational and organization commitment. These are not conservative post hoc tests, yet they indicated no significant differences across all comparisons.




References

Applebaum, H. (1981). Royal blue: The culture of construction workers. New York, NY: Holt.

Applebaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. (2000). Manufacturing advan-tage. Ithaca, NY: Cornell University Press.

Arthur, M. B., Hall, D. T., & Lawrence, B. S. (1989). Generating new directions in career theory: The case for a transdisciplinary approach. In M. B. Arthur, D. T. Hall & B. S. Lawrence (Eds.), Handbook of career theory (pp. 7-25). Cambridge, England: Cambridge University Press.

Bae, K. H., & Chung, C. S. (1997). Cultural values and work attitudes of Korean industrial workers in comparison with those of the United States and Japan. Work and Occupations, 24, 80-96.

Bae, K. H., & Form, W. H. (1986). Payment strategy in South Korea’s advanced eco-nomic sector. American Sociological Review, 51, 120-131.

Becker, H. S. (1951). The professional dance musician and his audience. American Journal of Sociology, 57, 136-144.

Blau, G. (1985). The measurement and prediction of career commitment. Journal of Occupational Psychology, 58, 277-288.

Blau, G. (2003). Testing for a four-dimensional structure of occupational commit-ment. Journal of Occupational and Organizational Psychology, 76, 469-488.

Blau, G., & Holladay, E. B. (2006). Testing the discriminant validity of a four- dimensional occupational commitment measure. Journal of Occupational and Organizational Psychology, 79, 691-704.

Bryant, C. D. (1972). Sawdust in their shoes: The carnival as a neglected complex organization and work culture. In C. D. Bryant (Ed.), The social dimensions of work (pp. 180-205). Englewood Cliffs, NJ: Prentice Hall.

Chang, E. M. (1999). Career commitment as a complex moderator of organizational commitment and turnover intention. Human Relations, 52, 1257-1278.

Chang, J. Y., & Choi, J. N. (2007). The dynamic relation between organizational and professional commitment of highly educated research and development (R&D) professionals. Journal of Social Psychology, 147, 299-315.

Cohen, A. (1999). The relation between commitment forms and work outcomes in Jewish and Arab culture. Journal of Vocational Behavior, 54, 371-391.

Cohen, A. (2000). The relationship between commitment forms and work outcomes: A comparison of three models. Human Relations, 53, 387-417.

Cohen, A. (2006). The relationship between multiple commitments and organizational citizenship behavior in Arab and Jewish culture. Journal of Vocational Behavior, 69, 105-118.



Kim and Mueller 33

Cohen, A. (2007). One nation, many cultures: A cross-cultural study of the relation-ship between personal culture values and commitment in the workplace to in-role performance and organizational citizenship behavior. Cross-Cultural Research, 41, 273-300.

Dalton, M. (1959). Men who manage. New York, NY: Wiley.Den Hartog, D. N., & Belschak, F. D. (2007). Personal initiative, commitment and affect

at work. Journal of Occupational and Organizational Psychology, 80, 601-622.Dore, R. (1973). British factory, Japanese factory: The origins of diversity in indus-

trial relations. Berkeley: University of California Press.Durkheim, E. (1893). Division of labor in society. New York, NY: Macmillan.Eom, D. W. (2006). The effect of performance-based pay scheme on wage determina-

tion in Korea. Korean Labor Economics Review, 29(2), 29-66. (In Korean)Ganzeboom, H. B. G., & Treiman, D. J. (1996). Internationally comparable measures

of occupational status for the 1988 International Standard Classification of Occu-pations. Social Science Research, 25, 201-239.

Gamst, F. C. (1980). The hoghead: An industrial ethnology of the locomotive engi-neer. New York, NY: Holt.

Greenhaus, J. H. (1987). Career management. Chicago, IL: Dryden.Gregory, K. L. (1983). Native view paradigms: Multiple cultures and culture conflict

in organizations. Administrative Science Quarterly, 28, 359-376.Grusky, D., & Sørensen, J. (1998). Can class analysis be salvaged? American Journal

of Sociology, 103, 1187-1233.Hackett, R. D., Lapierre, L. M., & Hausdorf, P. A. (2001). Understanding the links

between work commitment constructs. Journal of Vocational Behavior, 58, 392-413.

International Labor Office. (1990). International standard classification of occupa-tions: ISCO-88. Geneva, Switzerland: Author.

International Labor Office. (2006). Yearbook of labor statistics. Geneva, Switzerland: Author.

Johnson, M., Mortimer, J., Lee, J., & Stern, M. (2007). Judgements about work: Dimensionality revisited. Work and Occupations, 34, 290-317.

Kalleberg, A. L., & Berg, I. (1987). Work and industry: Structures, markets, and pro-cesses. New York, NY: Plenum.

Kim, S. W. (1999). Behavioral commitment among the automobile workers in South Korea. Human Resource Management Review, 9, 419-451.

Kim, S. W. (2004). A comparison of the characteristics of general social surveys in East Asia. Sungkyun Journal of East Asian Studies, 4, 137-154.




Kim, S. W., Mueller, C. W., Koh, C. Y., Koo, H. R., Kim, W. S., Park, B. J., & . . . .Jun, S. P. (2006). Korean General Social Survey 2005. Seoul, Korea: Sungkyunkwan University Press. (In Korean)

Kim, S. W., & Seo, Y. J. (2003). Organizational and occupational mobilities in firm and occupational internal labor markets: The case of medical profession. Korean Journal of Sociology, 37, 1-29. (In Korean)

Korea Ministry of Labor. (2006). Yearbook of labor statistics. Seoul, Korea. (In Korean)

Krecker, M. L. (1994). Work careers and organizational careers: The effects of age and tenure on work attachment to the employment relationship. Work and Occu-pations, 21, 251-283.

Kutner, M., Nachtsheim, C., & Neter, J. (2004). Applied linear regression models (4th ed.). Boston, MA: McGraw-Hill Irwin.

Kwon, I. W., & Banks, D. W. (2004). Factors related to the organizational and profes-sional commitment of internal auditors. Management Auditing Journal, 19, 606-622.

Lee, K. B., Carswell, J. J., & Allen, N. J. (2000). A meta-analytic review of occupa-tional commitment: Relations with person- and work-related variables. Journal of Applied Psychology, 85, 799-811.

Lincoln, J. R., & Kalleberg, A. L. (1990). Culture, control, and commitment: A study of work organization and work attitudes in the United States and Japan. Cam-bridge, England: Cambridge University Press.

Lipset, S. M., Trow, M. A., & Coleman, J. S. (1956). Union democracy. New York, NY: Free Press.

Louis, M. R. (1985). An investigator’s guide to workplace culture. In P. Frost, L. F. Moore, M. R. Louis, & C. C. Lundberg (Eds.), Organizational culture (pp. 73-94). Beverly Hills, CA: Sage.

Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bul-letin, 108, 171-194.

May, T. Y., Korczynski, M., & Frenkel, S. J. (2002). Organizational and occupational commitment: Knowledge workers in large corporations. Journal of Management Studies, 39, 775-801.

McAculay, B. J., Zeitz, G., & Blau, G. (2006). Testing a “push-pull” theory of work commitment among organizational professionals. Social Science Journal, 43, 571-596.

McDuff, E. M., & Mueller, C. W. (2000). The ministry as an occupational labor mar-ket: Intentions to leave an employer (church) versus intentions to leave a profes-sion (ministry). Work and Occupations, 27, 89-116.

Meyer, J. P., & Allen, N. J. (1997). Commitment in the workplace: Theory, research, and application. Thousand Oaks, CA: Sage.



Kim and Mueller 35

Morrow, P. (1993). The theory and measurement of work commitment. Greenwich, CT: JAI Press.

Nunnally, J. C. (1978). Psychometric theory. New York, NY: McGraw-Hill.Osinsky, P., & Mueller, C. W. (2004). Professional commitment of Russian provincial

specialists. Work and Occupations, 31, 193-224.Osnowitz, D. (2006). Occupational networking as normative control: Collegial

exchange among contract professionals. Work and Occupations, 33, 12-41.Pilcher, W. W. (1972). The Portland longshoremen. New York, NY: Holt.Porter, L. W., Steers, R. M., Mowday, R. T., & Boulian, P. V. (1974). Organizational

commitment, job satisfaction, and turnover among psychiatric technicians. Jour-nal of Applied Psychology, 59, 603-609.

Riley, P. (1983). A structurationist account of political culture. Administrative Science Quarterly, 28, 414-437.

Ritzer, G., & Trice, H. (1969). An occupation in conflict: A study of conflict. Ithaca, NY: Cornell University Press.

Salaman, G. (1974). Community and occupation: An exploration of the work/leisure relationship. London, England: Cambridge University Press.

Sørensen, A. B. (2000). Symposium on class analysis. American Journal of Sociol-ogy, 105, 1523-1558.

Sengupta, S., Edwards, P., & Tsai, C. (2009). The good, the bad, and the ordinary. Work and Occupations, 36, 26-55.

Steffens, R. A. (1972). Class, status, and power among jazz and commercial musi-cians. In C. D. Bryant (Ed.), The social dimensions of work (pp. 206-221). Engle-wood Cliffs, NJ: Prentice Hall.

Steinberg, D. I. (1989). The Republic of Korea: Economic transformation and social change. Boulder, CO: Westview.

Terkel, S. (1974). Working: People talk about what they do all day and how they feel about what they do. New York, NY: Pantheon.

Trice, H. M. (1993). Occupational subcultures in the workplace. Ithaca, NY: Cornell University Press.

Trice, H. M., & Beyer, J. M. (1993). The cultures of work organizations. Englewood Cliffs, NJ: Prentice Hall.

Wallace, J. E. (1993). Professional and organizational commitment: Compatible or incompatible? Journal of Vocational Behavior, 42, 333-349.

Wallace, J. E. (1995). Organizational and professional commitment in profes-sional and nonprofessional organizations. Administrative Science Quarterly, 40, 228-255.

Wang, X., & Armstrong, A. (2004). An empirical study of PM professionals’ commit-ment to their profession and employing organizations. International Journal of Project Management, 22, 377-386.




Bios

Sang-Wook Kim is a professor in the Department of Sociology at the Sungkyunkwan University, Seoul, Korea. His specialty area is sociology of work, and he is interested in employee work orientations, workplace justice, occupational professionalization, and survey research methodology. Recent work of his appeared in Advances in Group Processes (2008), coauthored with Charles W. Mueller.

Charles W. Mueller is Professor Emeritus in the Department of Sociology at the University of Iowa. He continues to conduct research in the area of the sociology of work and in topics related to gender differences in justice evaluations.



work and occupations - kocwcontents.kocw.net/kocw/document/2015/sungkyunkwan/kimsanguk… ·...

Documents