apllication methods
Journal of Applied Psychology 1993, Vol.78, No. 4,538-551
Copyright 1993 by the American Psychological Association, Inc. 0021-9010/93/$3.00
Commitment to Organizations and Occupations: Extension and Test of a Three-Component Conceptualization
John P. Meyer, Natalie J. Allen, and Catherine A. Smith
The authors tested the generalizability of J. P. Meyer and N. J. Allen's (1991) 3-component model of organizational commitment to the domain of occupational commitment. Measures of affective, continuance, and normative commitment to occupation were developed and used to test hypothe- ses concerning their differential relations with antecedent and consequence variables. Confirma- tory factor analyses conducted on data collected from samples of student and registered nurses revealed that the 3 component measures of occupational commitment were distinguishable from one another and from measures of the 3 components of organizational commitment. Results of correlation and regression analyses were generally consistent with predictions made on the basis of the 3-component model and demonstrated that occupational and organizational commitment contribute independently to the prediction of professional activity and work behavior.
Although commitment has been a subject of interest for some time, the past decade has seen a steady increase in the attention given to studying it. Not only has the number of stud- ies directly concerned with the development and consequences of commitment increased dramatically, but also commitment is often included as a variable in studies where it is not the pri- mary focus of attention. Indeed, commitment has begun to rival job satisfaction in this regard. Commitment theory and research have also been the subjects of several critical reviews in recent years (e.g., Griffin & Bateman, 1986; Mathieu & Zajac, 1990; Meyer & Allen, 1991; Morrow, 1983; Mowday, Porter, & Steers, 1982; Reichers, 1985).
With the increased attention given to commitment, there have been many important developments in both theory and research. Two of these are of particular importance for the present research. First, it has become increasingly apparent that commitment is a complex and multifaceted construct. For many years theorists and researchers have been defining and operationalizing commitment in different ways, and, as a re- sult, it has been difficult to synthesize the results of the accu- mulating research. It is now acknowledged that commitment
John P. Meyer and Catherine A. Smith, Department of Psychology; Natalie J. Allen, Centre for Administrative and Information Studies; University of Western Ontario, London, Ontario, Canada.
This research was supported by research grants from the Social Sciences and Humanities Research Council of Canada (410-89-370) and from Imperial Oil.
We thank the School of Nursing at Queen's University, the College of Nurses of Ontario, and the many student nurses and registered nurses who participated in this research. Particular appreciation is extended to Rita Maloney and Ena Howse, at Queen's University, and to Marilyn Rewak, at the College of Nurses of Ontario, for their assistance in accessing the research participants. Finally, we gratefully acknowledge the assistance of Annette Chauvin, Alanna Leffley, Tram Nguyen, Joseph Tavares, and Sue Tessier in data collection and analysis.
Correspondence concerning this article should be addressed to John P. Meyer, Department of Psychology, Social Science Centre, University of Western Ontario, London, Ontario, Canada N6A 5C2.
can take different forms, and it is therefore imperative that researchers state clearly what form or forms of commitment they are interested in and that they ensure that the measures they use are appropriate for the intended purpose. Second, there has been a broadening of the domain within which com- mitment is studied. Some of the earliest and most influential work within the organizational behavior literature (e.g., Bu- chanan, 1974; Mowday, Steers, & Porter, 1979; Porter, Steers, Mowday, & Boulian, 1974; Steers, 1977) examined employees' commitment to their employers, commonly referred to as orga- nizational commitment. More recently, there has been an in- crease in research examining commitment to unions (e.g., Full- agar & Barling, 1989; Gordon, Philpot, Burt, Thompson, & Spiller, 1980), employment (e.g., Jackson, Stafford, Banks, & Warr, 1983), professions (e.g., Aranya, Pollock, & Amernic, 1981; Morrow & Wirth, 1989), careers (e.g., Arnold, 1990; Blau, 1985,1988,1989), and so on.
Unfortunately, these two developments have taken place somewhat independently. Consequently, researchers examining commitment to entities other than employing organizations of- ten view commitment as a unidimensional construct (e.g., Ar- anya et al., 1981; Blau, 1985; Jackson et al., 1983). In cases in which commitment is considered multidimensional, the di- mensions tend to be domain specific. For example, the measure of union commitment developed by Gordon et al. (1980) was found to reflect four underlying dimensions: union loyalty, re- sponsibility to the union, willingness to work for the union, and belief in unionism. To date there has not been a systematic attempt to develop a multidimensional conceptualization of commitment that can be applied across domains. The develop- ment of such a conceptualization would help to foster commu- nication among researchers who are studying commitment within these various domains. This is particularly important as attention begins to focus on the impact of dual or multiple commitments (e.g., T. E. Becker, 1992; Darden, Hampton, & Howell, 1989; Gordon & Ladd, 1990; Reichers, 1986; Steffy & Jones, 1988).
We undertook this research to examine the generalizability
538
COMMITMENT TO ORGANIZATIONS AND OCCUPATIONS 539
of a three-component conceptualization of commitment that was developed by Meyer and Allen (1991) in the context of organizational commitment. In this initial effort, we were inter- ested in determining whether the three-component conceptual- ization could contribute to our understanding of occupational commitment. If a contribution could be demonstrated, it would be possible to use similar methods to assess generaliz- ability to still other domains (e.g., unions, employment, and management).
Meyer and Allen's (1991) Three-Component Model of Commitment
In their review of the organizational commitment literature, Meyer and Allen (1991) identified three distinct themes in the definition of commitment: commitment as an affective attach- ment to the organization, commitment as a perceived cost asso- ciated with leaving the organization, and commitment as an obligation to remain in the organization. They referred to these three forms of commitment as affective, continuance, and nor- mative commitment, respectively. Common to the three ap- proaches is the view that commitment is a psychological state that (a) characterizes the employee's relationship with the orga- nization and (b) has implications for the decision to continue or discontinue membership in the organization. Beyond this, how- ever, the nature of the psychological state for each form of com- mitment is quite different. Employees with a strong affective commitment remain with the organization because they want to, those with a strong continuance commitment remain be- cause they need to, and those with a strong normative commit- ment remain because they feel they ought to do so. Meyer and Allen argued that one can achieve a better understanding of an employee's relationship with an organization when all three forms of commitment are considered together. Consequently, they developed a three-component model of organizational commitment. According to the model, employees can experi- ence varying degrees of all three forms of commitment. Meyer and Allen hypothesized that each component develops as the result of different experiences and has different implications for on-the-job behavior.
A variety of antecedents of affective commitment have been identified, including personal characteristics, structural char- acteristics, job-related characteristics, and work experiences (Mowday et al, 1982). Meyer and Allen (1991) noted that by far the strongest and most consistent relationships have been ob- tained with work experiences. Employees whose experiences within the organization are consistent with their expectations and satisfy their basic needs tend to develop a stronger affective attachment to the organization than do those whose experi- ences are less satisfying. Continuance commitment presumably develops as employees recognize that they have accumulated investments or "side bets" (H. S. Becker, 1960) that would be lost if they were to leave the organization, or as they recognize that the availability of comparable alternatives is limited. Fi- nally, normative commitment develops as the result of social- ization experiences that emphasize the appropriateness of re- maining loyal to one's employer (Wiener, 1982) or through the receipt of benefits (e.g., tuition payments or skills training) that
create within the employee a sense of obligation to reciprocate (Scholl,1981).
Although all three forms of commitment should be nega- tively associated with turnover, Meyer and Allen (1991) sug- gested that there might be differences in the on-the-job behav- ior and performance associated with affective, continuance, and normative commitment. Specifically, they suggested that affective commitment and, to a somewhat lesser extent, norma- tive commitment, should be positively related to job perfor- mance and organizational citizenship, whereas continuance commitment might be expected to be unrelated, or negatively related, to these consequence variables. For this reason, they cautioned that not all forms of commitment are alike and that organizations concerned with keeping employees by strength- ening their commitment should carefully consider the nature of the commitment they instill.
Measures of the three components of organizational com- mitment were developed and found to be psychometrically sound (Allen & Meyer, 1990). Factor analytic studies of the Affective, Continuance, and Normative Commitment scales have shown that they measure relatively distinct constructs (Al- len & Meyer, 1990; Hackett, Bycio, & Hausdorf, 1992; McGee & Ford, 1987; Meyer, Allen, & Gellatly, 1990; Shore & Tetrick, 1991). The three scales have also been found to correlate differ- entially with variables purported to be antecedents of commit- ment (Allen & Meyer, 1990; Meyer & Allen, 1984; Meyer, Bobo- cel, & Allen, 1991; Shore & Tetrick, 1991). Finally, the various forms of commitment have been found to be differentially re- lated to on-the-job behavior and job performance, as predicted (Allen & Smith, 1987; Hackett et al., 1992; Meyer, Paunonen, Gellatly, Coffin, & Jackson, 1989; Randall, Fedor, & Longen- ecker, 1990; Shore & Barksdale, 1991).
Extension of the Model to Occupational Commitment
Although the three-component model was developed in the context of organizational commitment, it is reasonable to ex- pect that it might also be applied in other domains. Previous research examining commitment to occupations has tended to take a unidimensional perspective (e.g., Aranya et al., 1981; Blau, 1989; Morrow & Wirth, 1989). Like the early work on organizational commitment, commitment to occupations has typically been conceptualized as an affective attachment to the occupation.
It should be noted that the terms occupation, profession, and career have been used somewhat interchangeably in the com- mitment literature. We chose to use the term occupation rather than profession (Aranya et al., 1981; Morrow & Wirth, 1989) because we believe that both professionals and nonprofession- als can experience commitment to the work they do. Because participants in our research were nurses and nursing students who, according to the definition provided by Kerr, Von Glinow, and Schreisheim (1977), would be considered professionals, we used the term profession in the measures we developed for this study. However, the wording of the items is such that someone wanting to use the scales to measure commitment to other oc- cupations could do so simply by substituting the appropriate descriptors. We also chose to avoid using the term career com- mitment (e.g., Arnold, 1990; Blau, 1985) because of ambiguity in
540 J. MEYER, N. ALLEN, AND C. SMITH
the meaning of career. Career can be denned as a planned pat- tern of work from entry into the work force to retirement or as involvement in a particular job, organization, occupation, or profession. Because we were interested in assessing commit- ment to a particular line of work, we felt that the term occupa- tion was more appropriate.
The value of taking a multidimensional approach to the study of occupational commitment is that, as in the case of organizational commitment, it provides a more complete un- derstanding of a person's tie to his or her occupation. Although all three forms of commitment might be related to an individ- ual's likelihood of remaining in an occupation, the nature of the person's involvement in that occupation might be quite dif- ferent depending on which form of commitment is predomi- nant. A person who is affectively committed (i.e., has a strong desire to remain in the occupation) might be more likely than someone who is not so attached to keep up with developments in the occupation (e.g., by subscribing to trade journals or at- tending conferences), to join and participate in relevant associa- tions, and so on. The same might be true of individuals who have a strong normative commitment (i.e., a sense of obligation to remain). In contrast, individuals who have a strong continu- ance commitment (i.e., who recognize high costs associated with leaving the occupation) might be less inclined than those who remain for other reasons to involve themselves in occupa- tional activities besides those required to continue member- ship. As has been found to be the case with organizational com- mitment (e.g., Meyer et al., 1989), continuance commitment to the occupation might be expected to correlate negatively with the tendency to engage in behaviors that are beneficial from the standpoint of the occupation or profession (e.g., promotion of the occupation to the public or compliance with professional standards).
As was true in the case of organizational commitment de- scribed earlier, we also expected the antecedents of the three components of occupational commitment to differ. Affective commitment was expected to develop when involvement in the occupation proved to be a satisfying experience (e.g., provided the opportunity to do satisfying work or afforded the opportu- nity to develop valued skills). Continuance commitment should have developed as the individual made investments (side bets) that would be lost or reduced in value if he or she were to change occupations. Examples of such investments are the sta- tus associated with membership in an occupation and the time and effort put into acquiring occupation-specific skills. Finally, normative commitment was expected to develop as the result of the internalization of normative pressures to pursue a course of action, and the receipt of benefits that created a sense of obliga- tion to reciprocate. For example, being a member of a family with a history of involvement in a particular occupation or receiving financial support to pursue a career could contribute to the development of normative commitment.
Research Objectives and Overview
We conducted this research to develop measures of affective, continuance, and normative commitment to occupations and to use these measures to test the generalizability of Meyer and Allen's (1991) three-component model. To develop measures,
we followed procedures similar to those used by Allen and Meyer (1990) in the development of their organizational com- mitment scales. To test the generalizability of the three-com- ponent model, we correlated the occupational commitment measures with variables identified as potential antecedents and consequences of commitment. Finally, in response to concerns that have been raised in the literature about the proliferation of commitment measures (e.g., Morrow, 1983), we attempted to demonstrate that organizational and occupational commit- ment are relatively independent constructs and that each con- tributes uniquely to the understanding of, and ability to pre- dict, work behavior.
We conducted the research with two samples of participants involved in a single occupation, nursing. The first sample con- sisted of student nurses; the second sample was composed of registered nurses who, as a group, had considerable experience in the nursing profession. We collected data from the student- nurse sample primarily to develop scales, but we measured some additional variables to use in testing the differential-rela- tions hypothesis. For example, we expected satisfaction with the nursing program to correlate positively with affective com- mitment, and possibly with normative commitment, to the nursing profession, but not with continuance commitment. We expected that continuance commitment would increase as stu- dents progressed through the 4-year program because of the increased time and effort they had invested. Finally, we ex- pected that affective and normative commitment would be posi- tively associated with extracurricular nursing-related activities (e.g., having a summer job related to nursing) and with the in- tention to establish a long-term career in nursing. Theoretically, continuance commitment should also tie individuals to their occupation. But because these students were in the very early stages of their nursing careers, it was not clear whether we could expect continuance commitment to correlate with long-term intentions. We did expect, however, that continuance commit- ment, in and of itself, would not be related to students' involve- ment in nursing-related activities beyond those required for their degrees.
The questionnaires administered to registered nurses con- tained a much wider range of variables and afforded a better opportunity to test the differential-relations hypothesis. We ex- pected that affective commitment to the occupation would be greater among those nurses who were satisfied with their current employment situation (e.g., where they worked and the work they were doing). We expected continuance commitment to correlate positively with variables reflecting increased in- vestment in the profession (e.g., age, years in nursing, and full- time as opposed to part-time status). Finally, we expected nor- mative commitment to be greater among those nurses with a strong general sense of obligation to others.
The outcome variables examined with the registered-nurse sample can be divided into two groups: those that are of direct relevance to the profession (e.g., intention to leave nursing and professional involvement) and those that are relevant to the employing organization (e.g., intention to leave the organiza- tion, performance, and citizenship). Although commitment to the nursing profession should have a greater impact on the former, we expected that commitment would also affect the latter. We expected all three components of occupational com-
COMMITMENT TO ORGANIZATIONS AND OCCUPATIONS 541
mitment to be negatively related to intention to leave the nurs- ing profession (and possibly the organization); however, we ex- pected, consistent with previous research (e.g., Meyer et al, 1989), that affective commitment would be positively asso- ciated, and continuance commitment negatively associated, with activities that benefit the profession, the organization, or both (e.g., professional involvement and job performance). There is less basis for prediction concerning links with norma- tive commitment, but, given the nature of the construct, we expected that it would be positively associated with desirable professional and organizational activities.
Method
Participants and Data Collection Procedures
We collected data for this study from two samples, one consisting of student nurses and the other of registered nurses.
Student Nurses
For 2 consecutive academic years (1988-1989 and 1989-1990), we collected data at the beginning and the end of the year from students in a 4-year nursing program at Queen's University, Kingston, Ontario. For purposes of this study, we combined data obtained in the first and second years of the study. Questionnaires were administered during regular class meetings in required courses. This ensured that we would reach the majority of students in each year of the program.
The first questionnaire was administered in September, approxi- mately 2 weeks after the start of classes, and was completed by 366 students. Of these, 123 were in the first year of the program, 90 were in the second year, 79 were in the third year, and 74 were in the final year. The average age of participants at this administration was 21.68 years. Nine of the students were men and 352 were women; 5 students did not indicate their sex. The second questionnaire was administered in March during the second to the last week of classes in the winter term and was completed by 296 students (286 women, 8 men, and 2 who did not report their sex). The number of students in successive years of the program was 98,83,63, and 52, respectively. The average age of respon- dents was 21.98 years.
Registered Nurses
Questionnaires were mailed to 1,000 randomly chosen registered nurses from the membership of the College of Nurses of Ontario. Of the questionnaires delivered (24 were returned undelivered), 603 were returned with usable data, fora response rate of6I.8%. Of those who participated, 98% were women, 79% were married, 56% worked full time, 35% worked part time, 2% were casual employees, and 7% were not currently employed. Average tenure in the nursing profession was approximately 15 years. Of those who were employed, the majority were staff nurses (64%); head nurses composed the next largest group represented (10%). The majority of the participants were working in general hospitals (55%); others were doing home or extended care (9%) or were working in doctors' offices (5%), nursing homes (4%), psychiat- ric hospitals (4%), educational institutions (3%), public health (3%), and other settings (12%). Participants ranged in age from 21 to 70 years, with a mean age of 40 years.
Measures
Student Nurses
Commitment, Questionnaires administered to the student nurses included 30 items, each of which was written to assess affective, contin- uance, or normative commitment to the nursing profession. Responses to these items were made on 7-point scales (1 = strongly disagree and 7 = strongly agree). Because one purpose of this study was to develop measures of three components of occupational commitment, we in- cluded more items than we would ultimately use. This allowed us the flexibility to conduct item analyses and select the best items for inclu- sion in the commitment scales.
Antecedent and outcome variables. Other measures obtained to demonstrate construct validity of the commitment scales included sat- isfaction with the nursing program, involvement in career-related work, and career plans. We assessed satisfaction with the nursing pro- gram by asking students to indicate, on a 7-point scale, how satisfied they were with course variety, course content, quality of instruction, amount of practical experience, and quality of practical experience. We averaged responses to these questions to form a composite index of satisfaction with the nursing program.
We assessed involvement in career-related work in both question- naires. At the beginning of the academic year, we asked students whether they had worked at nursing-related jobs the previous summer (1 = no and 2 = yes); at the end of the year, we asked whether they had a nursing-related summer job lined up or if they planned to look for one (1 = no, and not looking and 2 = yes, or no, but looking).
Finally, we asked students, at both the beginning and end of the year, to indicate whether they planned to continue to work full time in some capacity in the nursing profession until they retired and whether they expected to change careers and do something unrelated to nursing. We also asked them to estimate the number of years they expected to work in the nursing profession and the total number of years they expected to remain in the work force. We standardized the ratio of years in nursing to years in the work force and combined it with standardized scores on the work-full-time and change-career items to yield a compos- ite measure of plans to continue in nursing.
Registered Nurses
The questionnaire administered to our registered-nurse sample was quite extensive and was designed to address a variety of issues. The variables of particular interest for this study were the occupational and organizational commitment scales, as well as several variables identi- fied as potential antecedents and consequences of commitment.
Commitment. The affective, continuance, and normative occupa- tional-commitment scales included on this questionnaire each com- prised six items selected on the basis of item analyses conducted with the student-nurse sample (to be described later). The corresponding organizational commitment scales were modified versions of those reported by Allen and Meyer (1990). Modification in the case of affec- tive commitment included eliminating the two items found to have the weakest loadings on the relevant factor in the confirmatory factor analyses (CFAs) reported by Meyer et al. (1990). These items were ex- cluded more for the sake of brevity than because they were inappropri- ate items. To increase clarity, we removed three items from the Continu- ance Commitment Scale, including two reverse-keyed items, and added a new item modified from the new occupational commitment scale (i.e., "If I had not already put so much of myself into this organiza- tion, I might consider working elsewhere"). Changes to the Normative Commitment Scale were more extensive. Items were rewritten to em- phasize more clearly the sense of obligation to the organization that is characteristic of normative commitment as conceptualized by Meyer and Allen (1991). Responses were made on 7-point scales (1 = strongly
542 J. MEYER, N. ALLEN, AND C. SMITH
disagree and 7 = strongly agree) and were averaged to yield composite commitment scores.
Antecedent and consequence variables. The demographic and em- ployment variables we examined in this study as correlates of commit- ment included age, years in nursing, and employment status (1 = ca- sual 2 = part time, and 3 = full time). We also asked nurses to indicate whether they were working in the clinical area of their choice and in the geographic area of their choice.
We assessed job satisfaction using Quinn and Staines's (1979) five- item Facet-Free measure. The nurses' general sense of obligation to others was assessed using a five-item scale developed for this study. Items were written to reflect a general belief in, and need to fulfill, one's obligations to others (e.g., "There are things in life that one ought to do whether he/she likes it or not" and "I feel obligated to live up to the expectations of those who have helped me along the way"). Re- sponses were made on 7-point scales (1 = strongly disagree and 7 = strongly agree) and were averaged across items to yield composite sense of obligation scores.
Among the outcome variables considered were self-report measures of behavioral intentions, behavior, and performance of relevance to the nursing profession, to the organization, or to both. The outcomes of most direct relevance to the profession included intention to leave the profession and professional involvement. Organization-relevant out- comes included intention to leave the organization as well as several measures of on-the-job activities.
We measured intention to leave the nursing profession with three items asking participants how frequently they thought about getting out of nursing, how likely it was that they would explore other career opportunities, and how likely it was that they would leave the nursing profession within the next year. Responses were made on appropri- ately labeled 7-point scales and were averaged across items to yield composite intention scores.
We assessed professional involvement using a measure that was based on an instrument developed by Kramer (1974). Participants were asked to indicate on 4- to 6-point scales how many professional courses they had taken since becoming a nurse (adjusted for years in nursing), how many journals they currently subscribed to or read on a regular basis, how many professional books they had purchased within the last 5 years, their degree of involvement in professional associations, and the amount of nursing-related volunteer work they had done. Responses to these questions were standardized and com- bined to form a composite professional-activity score.
We measured intention to leave the organization with three items paralleling those in the intention-to-leave-the-occupation measure. Specifically, we asked nurses how frequently they thought about leav- ing their current employer, how likely it was that they would search for a job in another organization, and how likely it was that they would actually leave the organization within the next year. Responses were made on appropriately labeled 7-point scales and were averaged across items to yield intention-to-leave scores.
Among the on-the-job behaviors we measured were three modes of responding to dissatisfaction at work identified by Hirschman (1970) and Farrell (1983): voice, loyalty, and neglect. (Note that exit, the other major response variable identified by Farrell, 1983, and by Hirsch- man, 1970, was measured as intention to leave the organization). Voice, loyalty, and neglect were each assessed with two-item measures that we developed for this study. The voice items reflected a willingness to make suggestions for improvement (e.g., "When I see ways that patient care can be improved I generally try to bring it to the attention of my superiors"). The loyalty items reflected a willingness to accept things as they are (e.g., "I generally trust the judgment of the decision-makers in this organization"). The neglect items reflected passive withdrawal in the face of dissatisfaction (e.g., "I get so frustrated at times with the way things get done in this organization that I just stop caring"). Re-
sponses were made on 7-point scales (1 = strongly disagree and 7 = strongly agree) and were averaged to form composite scores for voice, loyalty, and neglect.
We also obtained self-report measures of absenteeism, tardiness, job performance, and organizational citizenship. We asked participants to indicate how many days in the past year they had been absent from scheduled work. To help distinguish voluntary from involuntary ab- sence, we also asked them to indicate how many of these days were missed because they "didn't feel like going to work." Tardiness was measured by asking participants to indicate on a 7-point rating scale how frequently they arrived for work at least 10 min late.
We assessed job performance with two items. We first asked partici- pants to indicate the overall evaluation they received from their super- visor on their most recent performance appraisal. We then asked them to provide their own assessment of their overall performance. Both ratings were made on 7-point scales (1 = poor and 7 = excellent).
We measured organizational citizenship using scales we developed specifically for this study. Several items were written to describe behav- iors that reflect good or poor (reverse-keyed) examples of organiza- tional citizenship (as described by Organ, 1987) within a nursing con- text. A principal-components analysis led to the identification of two citizenship factors that we labeled Helping Others (four items, e.g., "When my own work is completed, I try to help other nurses who may have a particularly heavy load") and Use of Time (five items, e.g., "When my work load gets particularly heavy, I work extra hours [e.g., through breaks or lunch, overtime] without pay to get things done on time"). Responses were made on 7-point scales (1 = strongly disagree and 7 = strongly agree) and were averaged across items.
Results
Scale Development
Our first objective in analyzing the student-nurse data was to identify items to be included in measures of affective, continu- ance, and normative commitment to the nursing profession. We conducted principal-components analyses on the 30 commit- ment items. Analyses of data obtained at both the beginning and end of the year revealed that items generally loaded on the factor representing the construct they were intended to mea- sure. In selecting items, we examined the pattern of loadings over the two analyses, looking for items with consistently high loadings on the intended factor and low loadings on the other two factors. We also tried to minimize item redundancy. The final scales each included 6 items.
To determine whether the three commitment scales indeed measured distinct constructs, we conducted CFAs of the covari- ance matrices derived from the data obtained from the student- and registered-nurse samples. We obtained maximum likeli- hood solutions using LISREL 7 (Joreskog & Sorbom, 1989). Only the analysis conducted on the registered-nurse sample was truly confirmatory; the results of the analyses conducted with the student-nurse data were biased by the fact that items were selected on the basis of earlier analyses. Nevertheless, these additional analyses allowed us to determine whether an oblique three-factor structure fit the data better than an alternative structure. The alternative structures chosen for comparison were a one-factor solution (all commitment items loading on one factor) and an oblique two-factor solution (affective and normative commitment items defining one factor and continu- ance commitment defining the other). Inclusion of the two-fac- tor model was based on the previous finding that affective and
COMMITMENT TO ORGANIZATIONS AND OCCUPATIONS 543
Table 1 Overall Fit Indexes for Occupational Commitment Scales
Model df RNI PNFI
Nursing students: beginning of year (N = 312)
Null model 1 factor 2 factors 3 factors
7,871.98 1,005.51
597.66 347.88
171 135 134 132
— .887 .940 .972
— .782 .823 .845
Nursing students: end of year (N = 275)
Null model 1 factor 2 factors 3 factors
6,455.65 1,165.08
679.85 353.42
171 135 134 132
— .836 .913 .965
— .738 .800 .833
Registered nurses (N = 530)
Null model 1 factor 2 factors 3 factors
16,644.64 1,602.87 1,037.79
475.72
171 145 134 132
— .911 .945 .979
— .804 .828 .845
Note. Dashes indicate that the fit index was not applicable to the null model. RNI = relative noncentrality index; PNFI = parsimonious normed-fit index.
normative commitment, although distinct, are positively corre- lated (Allen & Meyer, 1990).
We assessed the fit of the various models by computing two indexes: the relative noncentrality index (RNI; McDonald & Marsh, 1990) and the parsimonious normed-fit index (PNFI; Mulaik et al, 1989). For both indexes, higher values indicate better fit to the data; the PNFI corrects for the number of pa- rameters estimated (see Mulaik et al., 1989). The RNI and PNFI values for the various models are reported, along with the chi-square values on which they were based, in Table 1. As can be seen, the three-factor solution provided the best fit to the data, even when a correction was made for the greater number of estimated parameters.
Examination of the correlations among the factors in the three-factor solution revealed that, as expected, affective and normative commitment to the nursing profession correlated significantly in all three data sets (M = .403). The correlation between affective and continuance commitment was somewhat smaller and negative, but was also significant (M = -.227). There was also a low, but significant, positive correlation be- tween continuance and normative commitment (M = .234).
To determine whether occupational commitment could be distinguished from organizational commitment, we conducted a further CFA using the registered-nurse data. Here we com- pared the fit of a six-factor solution (three occupational- and three organizational-commitment factors) with the fit of a three-factor solution in which all affective commitment items defined one factor, all continuance commitment items a second factor, and all normative commitment items a third factor. The fit indexes are reported in Table 2. As can be seen, the six-factor solution provided the better fit. The scale items and their load- ings on the six factors are reported in Table 3.
Correlations among the commitment factors are presented in
Table 4. The largest correlations across domains (i.e., occupa- tional and organizational commitment) tended to be between corresponding forms of commitment. The largest of these correlations was between continuance commitment to the oc- cupation and to the organization. Those nurses who indicated that it would be costly to leave their current employers also indicated that it would be costly to leave the nursing profession. Correlations between constructs within domain were similar to those obtained in the analysis of the occupational commitment items reported above, and in previous CFAs of the organiza- tional commitment scales. The one exception to this is a larger than expected correlation between affective and normative commitment to the organization.
Correlates of Occupational Commitment Among Student Nurses
The means, standard deviations, and reliability estimates (co- efficient alpha) for variables included in the correlation analy- ses conducted on the student-nurse data are reported in Table 5. The correlations are reported in Table 6.
Age, in spite of the restriction in range for this sample, corre- lated negatively with affective commitment and positively with continuance commitment both at the beginning and end of the year. Year in the program correlated positively with continu- ance commitment and negatively with affective and normative commitment on both occasions of measurement. Thus, as stu- dents progress through the nursing program, their recognition of the cost associated with leaving the nursing profession in- creases but their desire and sense of obligation to remain de- creases. As expected, affective commitment to the nursing pro- fession correlated significantly (positively) with ratings of satis- faction with the nursing program when both are measured early in the academic year. However, the correlation between these measures obtained near the end of the year was not significant.
By examining the behavior and behavior-intention measures, one can see that continuance commitment measured at the beginning of the year correlated positively with reports of hav- ing had a nursing-related job the previous summer. Affective commitment measured at the end of the year correlated posi- tively with having secured, or intending to look for, a nursing- related job during the upcoming break. Thus, whereas the per- ceived cost associated with leaving the nursing profession is greater among those who have already had nursing experience,
Table 2 Overall Fit Indexes for Occupational- and Organizational-Commitment Scales
Model
Null model 3 factors 6 factors
x2
36,838.97 2,851.37 1,588.15
df
666 591 579
RNI
_ .938 .972
PNFI
.880
.893
Note. N = 530 for all models. Dashes indicate that the fit index was not applicable to the null model. RNI = relative noncentrality index; PNFI = parsimonious normed-fit index.
544 J. MEYER, N. ALLEN, AND C. SMITH
Table 3 Standardized Parameter Estimates for Six-Factor Solution
Item Parameter estimate Item
Parameter estimate
Factor 1
Nursing is important to my self-image. .406 I regret having entered the nursing profession. (R) .736 I am proud to be in the nursing profession. .709 I dislike being a nurse. (R) .819 I do not identify with the nursing profession. (R) .639 I am enthusiastic about nursing. .733
Factor 2
I have put too much into the nursing profession .469 to consider changing now.
Changing professions now would be difficult for .693 me to do.
Too much of my life would be disrupted if I were .754 to change my profession.
It would be costly for me to change my profession .590 now.
There are no pressures to keep me from changing .347 professions. (R)
Changing professions now would require .631 considerable personal sacrifice.
Factor 3
I believe people who have been trained in a .561 profession have a responsibility to stay in that profession for a reasonable period of time.
I do not feel any obligation to remain in the .596 nursing profession. (R)
I feel a responsibility to the nursing profession to .792 continue in it.
Even if it were to my advantage, I do not feel that .647 it would be right to leave nursing now.
I would feel guilty if I left nursing. .580 I am in nursing because of a sense of loyalty to it. .645
Factor 4
I would be very happy to spend the rest of my .645 career with this organization.
I really feel as if this organization's problems are .410 my own.
I do not feel a strong sense of "belonging" to my .735 organization. (R)
I do not feel "emotionally attached" to this .680 organization. (R)
I do not feel like "part of the family" at my .735 organization. (R)
This organization has a great deal of personal .749 meaning for me.
Factor 5
Right now, staying with my organization is a .504 matter of necessity as much as desire.
It would be very hard for me to leave my .592 organization right now, even if I wanted to.
Too much of my life would be disrupted if I .678 decided I wanted to leave my organization now.
I feel that I have too few options to consider .700 leaving this organization.
If I had not already put so much of myself into .454 this organization, I might consider working elsewhere.
One of the few negative consequences of leaving .483 this organization would be the scarcity of available alternatives.
Factor 6
I do not feel any obligation to remain with my .580 current employer. (R)
Even if it were to my advantage, I do not feel it .638 would be right to leave my organization now.
I would feel guilty if I left my organization now. .658 This organization deserves my loyalty. .718 I would not leave my organization right now .735
because I have a sense of obligation to the people in it.
I owe a great deal to my organization. .691
Note. Factors 1-3 reflect affective, continuance, and normative commitment to the occupation, respectively; Factors 4-6 reflect affective, continuance, and normative commitment to the organization, respectively. Only estimated parameters are reported here; all other values were fixed at zero. All parameter estimates reported here are significant (p < .05). (R) = reverse keyed.
desire to remain in the profession is associated with efforts to obtain such experience.
Both affective and normative commitment correlated signifi- cantly in a positive direction with intention to remain in the nursing profession. Continuance commitment correlated nega- tively with intention to continue in nursing, although only the correlation for data obtained at the beginning of the year was significant.
To determine whether the three components of occupational commitment contributed independently to the prediction of intention to continue in nursing, we regressed the intention measure on the three commitment scales. For data obtained at the beginning of the year, commitment accounted for 10% of
the variance in intention to continue, and both continuance commitment (0 = —.20) and normative commitment (0 = .20) made significant independent contributions to prediction. Commitment measured at the end of the year also accounted for 10% of the variance in intention, with affective commitment (ft = .24) and normative commitment (0 = . 18) making signifi- cant independent contributions to prediction.
Correlates of Occupational and Organizational Commitment Among Registered Nurses
Means, standard deviations, and reliability estimates for the variables included in the correlational analyses are reported in
COMMITMENT TO ORGANIZATIONS AND OCCUPATIONS 545
Table 4 Correlations Among Latent Variables in Six-Factor Solution
1. 2. 3. 4. 5. 6.
Latent variable
ACS-OCC CCS-OCC NCS-OCC ACS-ORG CCS-ORG NCS-ORG
1
_ -.117*
.485*
.431* -.191*
.355*
2
— .215* .027 .743* .119*
3
— .333* .152* .618*
4
— -.062
.737*
5 6
— .151* —
Note. ACS-OCC = affective commitment to the occupation; CCS- OCC = continuance commitment to the occupation; NCS-OCC = normative commitment to the occupation; ACS-ORG = affective commitment to the organization; CCS-ORG = continuance commitment to the organization; NCS-ORG = normative commit- ment to the organization. * p < .05.
Table 7. The correlations are reported in Table 8. With one exception (age and continuance commitment to the occupa- tion), all three forms of commitment to the profession and to the organization correlated significantly and positively with age and years in nursing. Affective commitment to both the organi- zation and occupation were positively correlated with ratings of whether the individual was currently working in the clinical area, and in the geographical location, of his or her choice.
Ratings of job satisfaction correlated positively with affective and normative commitment and negatively with continuance commitment to both the occupation and organization. As ex- pected, normative commitment both to the organization and the occupation correlated positively with the sense of obligation to others. Continuance commitment to the organization and to the occupation also correlated positively with sense of obliga- tion.
As expected, all three forms of commitment to the occupa- tion correlated negatively with intention to leave the nursing profession. Affective and normative commitment also corre- lated positively with involvement in professional activities; con-
tinuance commitment did not correlate significantly. The orga- nizational commitment measures also correlated significantly with intention to leave the occupation and with professional activity. Specifically, affective and normative commitment were negatively related to intention to leave. Normative commitment was positively related and continuance commitment was nega- tively related to professional activity.
Both organizational and occupational commitment corre- lated significantly with several of the organization-relevant out- come measures. For example, affective and normative commit- ment to both the organization and the occupation correlated negatively with intention to leave the organization. These same commitment variables correlated positively with two of the re- sponses to dissatisfaction, voice and loyalty, and negatively with a third, neglect. They also correlated positively with effective use of time and negatively with voluntary absence. Both contin- uance commitment to the organization and continuance com- mitment to the occupation correlated positively with the ne- glect measure. Only the organizational commitment measures correlated significantly with supervisor performance-appraisal estimates. Consistent with previous findings (Meyer et al., 1989), affective commitment correlated positively and continu- ance commitment correlated negatively with reported supervi- sor evaluation of performance.
It is clear from these correlations that both commitment to the organization and commitment to the occupation are related to occupation- and organization-relevant activities. To deter- mine whether occupational commitment made a unique con- tribution to the prediction of these outcome variables, we con- ducted hierarchical regression analyses in which we entered the organizational commitment variables into the equations first, followed by the occupational commitment variables. The re- sults of these regression analyses are reported in Table 9.
Of particular interest in Table 9 is the change in the squared multiple correlation attributable to the inclusion of the occupa- tional commitment measures. As can be seen, including occu- pational commitment did increase the ability to predict impor- tant outcome variables. Most notable was the increase in ability
Table 5 Means, Standard Deviations, and Reliabilities for Measures Obtained From Student Nurses
Beginning of year End of year
Variable
Affective commitment Continuance commitment Normative commitment Age Year in program Satisfaction with nursing
program Had nursing-related
summer job Have or seek nursing-related
summer job Intent to continue in
nursing
M
5.57 3.84 3.37
21.19 2.22
4.03
1.49
—
0.01
SD
1.16 1.41 1.19 3.59 1.12
1.22
0.50
—
0.77
No. of items
6 6 6 1 1
5
1
—
3
a
.87
.79
.73 NA NA
.77
NA
NA
.67
M
5.38 3.83 3.30
21.51 2.18
4.29
—
1.85
0.07
SD
1.14 1.43 1.14 3.24 1.10
1.18
0.36
2.36
No. of items
6 6 6 1 1
5
1
3
a
.85
.83
.77 NA NA
NA
NA
NA
.71
Note. Dashes indicate that these measures were not obtained. NA = not applicable.
546 J. MEYER, N. ALLEN, AND C. SMITH
Table 6 Correlations Among Variables and the Three Components of Occupational Commitment at Beginning and End of Nursing Students' Academic Year
Beginning of year
Variable
Age Year in program Satisfaction with nursing
program Had nursing-related
summer job Have or seek a nursing-related
summer job Intent to continue in
nursing
ACS
-.14* -.40**
.37**
-.06
—
.59**
CCS
.33**
.49**
-.09
.25**
—
-.22**
NCS
-.10 -.20**
.07
.07
—
.32**
End of year
ACS
-.14* -.39**
.05
—
.28**
.59**
CCS
.23**
.43**
-.06
—
.00
-.11
NCS
-.06 -.14*
.04
—
.05
.33**
Note. Dashes indicate that these measures were not obtained. ACS = affective commitment scale; CCS = continuance commitment scale; NCS = normative commitment scale. *p<.05. **p<.01.
to predict intention to leave the nursing profession. However, occupational commitment also added significantly to the pre- diction of professional activity, intention to leave the organiza- tion, loyalty, neglect, voluntary absence, tardiness, helping others, and use of time. It should also be noted that, of the occupational commitment measures, affective commitment contributed most frequently to prediction, particularly in the case of the organization-relevant outcomes.
Discussion
The results of this research provide preliminary evidence for the generalizability of Meyer and Allen's (1991) three-compo- nent model of commitment. Specific contributions made by the research include (a) the development of reliable measures of affective, continuance, and normative commitment to occupa- tions; (b) evidence that the three components of occupational commitment are differentially related to variables considered to be antecedents or consequences of commitment; and (c) evi- dence that organizational and occupational commitment con- tribute independently to the prediction of important organiza- tion-relevant outcome variables (e.g., turnover intention, perfor- mance, and citizenship).
Our CFAs of the occupational and organizational commit- ment measures revealed that it is indeed possible to distinguish among three aspects of commitment, both within and across domains. It was also apparent, however, that the components of commitment to occupations and organizations are not com- pletely independent. In fact, some fairly strong correlations were identified. The positive correlation, within domain, be- tween affective and normative commitment is consistent with previous research (e.g., Allen & Meyer, 1990; Hackett et al., 1992; Randall et al., 1990). The correlation between affective and normative commitment to the organization obtained in this study, however, was stronger than that found in the past. Unfortunately, we cannot determine from the data available whether this difference is due to something unique about the
sample we used or to the modifications made to the normative commitment scale in this study.
That a positive correlation emerged consistently between af- fective and normative commitment might be explained, in part, by the fact that the two have many common antecedents. For example, affective and normative commitment were found in this study to be associated with positive work experiences. It appears that having such experiences can lead one to develop an affective attachment, a sense of obligation, or both, to the entity associated with those experiences (e.g., an occupation or an organization). There has not been sufficient attention given to date to the investigation of the processes by which affective and normative commitment develop; therefore, it is not possi- ble to predict in advance whether the dominant response will be attachment or obligation. Moreover, although unique ante- cedents of normative commitment may well exist, little atten- tion has been paid to identifying them. Given that affective and normative commitment both tend to be associated with desir- able outcome variables, it might be reasonable to question whether these issues matter. Meyer and Allen (1991) suggested, however, that there may be important differences in the conse- quences of affective and normative commitment. Among other things, they hypothesized that the beneficial effects of norma- tive commitment might be more short-lived (i.e., lasting only until a debt has been repaid) than those resulting from affective commitment. The antecedents and processes involved in the development of affective and normative commitment, there- fore, warrant further investigation.
Normative commitment was also found to correlate posi- tively, within both the occupational and organizational do- mains, with continuance commitment. These correlations, al- though significant, were modest and do not challenge the prop- osition that they are different forms of commitment. Again, this positive association might be explained, in part, by the fact that normative and continuance commitment were found to have common correlates. Although initially predicted to be an antecedent of normative commitment, the measure of general sense of obligation to others was also found to correlate posi-
COMMITMENT TO ORGANIZATIONS AND OCCUPATIONS 547
Table 7 Means, Standard Deviations, and Reliabilities for Measures Obtained From Registered Nurses
Variable
ACS-ORG CCS-ORG NCS-ORG ACS-OCC CCS-OCC NCS-OCC Age Years in nursing Employment status Clinical area of choice Geographic area of choice Job satisfaction Sense of obligation Leave occupation Professional activity Leave organization Voice Loyalty Neglect Total absence Voluntary absence Tardiness Supervisor evaluation of
performance Own evaluation of
performance Helping Others Use of Time
M
3.91 4.03 3.04 5.38 4.73 3.04
39.85 15.39
1.57 0.92 0.88 3.44 5.13 2.75 0.00 3.00 5.77 4.86 2.81 4.60 0.48 1.48
5.92
5.89 6.41 5.65
SD
1.47 1.39 1.41 1.26 1.37 1.44 9.70 8.78 0.55 0.27 0.32 1.14 0.68 1.69 0.58 1.82 1.07 1.33 1.50
19.36 1.30 1.02
1.10
0.79 0.56 1.05
No. of items
6 6 6 6 6 6 1 1 1 1 1 5 6 3 5 3 2 2 2 1 1 1
1
1 4 5
a
.82
.74
.83
.82
.76
.80 NA NA NA NA NA .85 .61 .83 .50 .83 .48 .41 .40 NA NA NA
NA
NA .58 .53
Note. ACS = affective commitment scale; CCS = continuance commitment scale; NCS = normative commitment scale; ORG indicates an organizational focus; OCC indicates an occupational focus. NA = not applicable.
lively with continuance commitment. This finding is consistent with H. S. Becker's (1960) suggestion that "the existence of generalized cultural expectations provides penalties for those who violate them," and therefore serves as a type of "side bet" (p. 36). Thus, feeling a need to live up to others' expectations can contribute to either an obligation-based or cost-based com- mitment, or both.
Correlations between affective commitment and continu- ance commitment to both the occupation and the organization were consistently negative. Again, however, the correlations tended to be relatively small. The link between affective and continuance commitment has received considerable attention elsewhere (e.g., Hackett et al, 1992; McGee & Ford, 1987; Meyer et al., 1990) and will not be discussed further here, other than to reiterate the call for further research to address the extent to which the two forms of commitment might be related over time.
We also found significant correlations between correspond- ing forms of commitment across domains. This was most evi- dent in the case of continuance commitment. One might expect a strong correlation between continuance commitment to the occupation and to the organization in cases in which (a) contin- ued employment in an organization requires continued involve-
ment in the occupation, or (b) there are relatively few organiza- tions that employ members of a particular occupation. The former is very likely to be the case for members of the nursing profession; the latter might also be true, particularly for nurses working in small communities. It remains to be determined whether the correlation between continuance commitment to the occupation and continuance commitment to the organiza- tion will be as large for other occupations.
As noted above, the pattern of correlations obtained with occupational commitment measures and variables identified as potential antecedents or consequences was generally consistent with predictions generalized from Meyer and Allen's (1991) three-component model. Affective commitment was related to positive experiences (e.g., satisfaction with the job or training experience), continuance commitment was related to variables reflecting increased investment (e.g., year in the nursing pro- gram, years in nursing, and employment status), and normative commitment was related to positive work experiences and to a general sense of obligation to others. We found that both affec- tive and normative commitment to the occupation correlated positively with desirable behaviors or behavioral intentions (e.g., professional activity, voice, and citizenship) and negatively with undesirable behaviors (e.g., intention to leave the occupation and the organization, absenteeism, and neglect). On the other hand, although negatively related to intention to leave the nurs- ing profession, continuance commitment was found to be gen- erally unrelated to professional involvement and on-the-job be- havior among registered nurses. Exceptions included a positive correlation with neglect and, unexpectedly, a positive correla- tion with effective use of time. These findings, then, suggest that individuals can become bound to occupations in different ways but that the implications for occupation-relevant behavior other than intention to remain, and for organization-relevant behavior, can be quite different.
Some of the correlations obtained in these analyses are of particular interest because of their theoretical and practical implications. For example, the finding that continuance com- mitment was positively correlated with year in the program for nursing students is consistent with the expectation that continu- ance commitment will increase with the accumulation of in- vestments. Presumably, as nursing students advance in the pro- gram they recognize that they have contributed a great deal of time, effort, and money toward obtaining a degree in nursing and, therefore, acknowledge that it becomes increasingly costly to leave the program. What is also interesting to note, however, is that as continuance commitment increases, affective and nor- mative commitment decrease. Thus, compared with when they enter the program, nursing students' commitment to the pro- fession is based more on cost considerations and less on desire or obligation as they continue in the program. This might be considered a warning for educators and future employers of these students, when combined with the following evidence: (a) Among student nurses, affective and normative commitment correlated positively and continuance commitment correlated negatively1 with intention to establish a long-term career in
1 Although we did not expect continuance commitment to the nurs- ing profession to be positively related to intention to remain in the profession, we did not anticipate the significant negative correlation
548 J. MEYER, N. ALLEN, AND C. SMITH
Table 8 Correlations Among Variables and the Three Components of Organizational and Occupational Commitment for Registered Nurses
Organizational commitment
Variable
Age Years in nursing Employment status Clinical area of
choice Geographic area of
choice Job satisfaction Sense of obligation Leave occupation Professional activity Leave organization Voice Loyalty Neglect Total absence Voluntary absence Tardiness Supervisor evaluation of
performance Own evaluation of
performance Helping Others Use of Time
ACS
.20**
.18**
.03
.13**
.10*
.49**
.08 -.37**
.07 -.45**
.23**
.41** -.38**
.03 -.13** -.01
.16**
.04
.10*
.13**
CCS
.11*
.21**
.05
-.07
-.04 -.22**
.14**
.00 -.15** -.02 -.04 -.03
.25**
.03 -.05
.01
-.12**
-.03 -.06
.08
NCS
.18**
.18**
.06
.08
.04
.37**
.15** -.30**
.14** -.34**
.17**
.40** -.23**
.08 -.15** -.00
.06
-.00 .00 .11**
Occupational commitment
ACS
.18**
.17** -.01
.15**
.18**
.53**
.05 -.52**
.18** -.35**
.15**
.26** -.29**
.04 -.21** -.11*
.07
.05
.23**
.10*
CCS
.07
.16**
.11**
-.02
-.06 -.14**
.16** -.15** -.05 -.05
.00 -.02
.15**
.04 -.03 -.04
-.07
-.01 .03 .12**
NCS
.22**
.24**
.01
.07
.07
.25**
.14** -.35**
.12** -.19**
.10*
.29** -.11*
.11** -.16** -.06
-.05
-.02 .05 .14**
Note. ACS = affective commitment scale; CCS = continuance commitment scale; NCS = normative commitment scale. */><.05. **/?<.01.
nursing; and (b) among working nurses, affective and norma- tive commitment were more likely than continuance commit- ment to be associated with desirable occupation- and organiza- tion-relevant behavior. Additional research is necessary to de- termine whether the changing profile of commitment observed here is unique to the students in our sample or to the nursing profession. If our findings do generalize to other nursing pro- grams or professions, training institutions might be encouraged to explore the reasons behind the decline in affective and nor- mative commitment and the increase in continuance commit- ment among their students.
Also of interest is the pattern of correlations obtained with the summer job-activity measures administered to student nurses. At the beginning of the academic year, continuance commitment correlated positively with reports of having worked the previous summer in a nursing-related job, whereas at the end of the year affective commitment correlated posi- tively with reports of looking for, or having secured, a nursing- related summer job. Thus, it appears that those who indicate an affective attachment to the profession are interested in seeking
obtained. It is possible that the high continuance commitment some students experienced was directed more to the completion of their degree requirements than to the pursuit of a career in nursing. Those who recognize that their intent to continue in the program is largely a function of avoiding loss of investment (e.g., time and effort) might also realize that they have little interest in nursing per se and therefore do not expect a Jong career in the profession.
out relevant work experience, but that, once obtained, this expe- rience contributes to a feeling of greater cost associated with leaving nursing. Although not predicted in advance, this find- ing can be easily interpreted within the framework of the three- component model. Affective commitment is expected to be positively related to prospective behavior (e.g., experience seek- ing), whereas continuance commitment is expected to be sensi- tive to retrospective assessment of accumulated investments (e.g., time and effort spent in acquiring unique skills).
The results of the correlational analyses provide preliminary evidence for the utility of taking a multidimensional approach to the study of occupational commitment. The fact that the three components of commitment were differentially related to occupation-relevant behaviors in both the student- and regis- tered-nurse samples supports the argument advanced earlier that taking a multidimensional perspective provides a better understanding of individuals' involvement in their occupa- tions. This research included only very general measures of occupation-relevant activities, however. Additional research is needed to examine in greater depth the links between the com- ponents of commitment and various types of occupational in- volvement.
We predicted that, in addition to its link to occupation-rele- vant activity, occupational commitment would be related to organization-relevant behavior. The correlations obtained in this study were consistent with this prediction. It is also appar- ent, however, that the correlations observed for occupational commitment were similar to, and often lower in magnitude
COMMITMENT TO ORGANIZATIONS AND OCCUPATIONS 549
Table 9 Regression Analyses: Predicting Outcome Variables From the Three Components of Occupational Commitment for Registered Nurses
Step 1: organizational commitment Step 2: occupational commitment
Full equation
Variable ACS CCS NCS ACS CCS NCS
Note. ACS = affective commitment scale; CCS = continuance commitment scale; NCS = normative commitment scale. */?<.05. **p<.01.
dfo
Leave occupation Professional activity Leave organization Voice Loyalty Neglect Total absence Voluntary absence Tardiness Supervisor evaluation of
performance Own evaluation of
performance Helping Others Use of Time
-.30** -.06 -.39**
.20**
.25** -.34** -.02 -.07 -.01
.18**
.06
.15**
.12*
-.01 -.17** -.03 -.04 -.04
.23**
.02 -.04
.01
-.10*
-.02 -.04
.08
-.11* .20**
-.10* .05 .25**
-.04 .09
-.11 .00
-.04
-.04 -.09
.03
.14**
.05**
.21**
.06**
.20**
.20**
.01
.03**
.00
.04**
.00
.02*
.02**
-.42** .13**
-.22** .06 .06
-.15** .00
-.17** -.11
.03
.04
.23**
.04
-.18** -.02 -.04
.03 -.06
.05
.02 -.01 -.07
-.02
.00
.10
.10
-.10* .02 .06 .01 .10* .03 .09
-.05 -.04
-.10
-.02 -.01
.07
.20**
.02*
.04**
.00
.02*
.02**
.01
.03**
.02*
.01
.00
.05*
.01*
.34
.07
.24
.06
.22
.22
.01
.06
.02
.05
.00
.07
.04
47.63** 6.57**
29.77** 5.99**
25.68** 25.48**
1.27 5.61** 1.68
3.25**
0.40 6.49** 3.72**
6,558 6,542 6,556 6,550 6,550 6,556 6,553 6,543 6,556
6,413
6, 549 6,549 6,542
than, the correlations obtained for organizational commit- ment. To justify including the occupational commitment mea- sures in future studies of organizational behavior it was neces- sary to demonstrate that they contributed to prediction inde- pendently of the organizational commitment measures. This was the purpose of the hierarchical regression analyses.
The results of the regression analyses indicated that occupa- tional commitment did make a significant, albeit small, contri- bution to the prediction of organization-relevant outcomes even when organizational commitment was controlled. Thus, it appears that how an employee behaves on the job may be in- fluenced jointly by commitment to the organization and to the occupation. It might be interesting in future research to exam- ine more carefully how commitments to these domains com- bine to influence behavior. For example, it is possible that the relative influence of occupational and organizational commit- ment is determined by an individual's perception of how rele- vant the behavior is to the occupation compared with the orga- nization. If helping another nurse, for instance, is seen as a professional responsibility, it might be influenced more by oc- cupational commitment than by organizational commitment. The reverse might be true if the behavior is perceived as helpful to the organization.
Our findings are generally consistent with, and extend, the results of earlier research that has taken a unidimensional ap- proach to the study of organizational and occupational (career) commitment (e.g., Aranya et al, 1981; Arnold, 1990; Blau, 1985; Morrow & Wirth, 1989). There are, of course, limitations to the present research that must be acknowledged. Among these is the fact that we studied members of only one occupation. We would like to see these findings replicated with other occupa- tions. The occupational commitment scales we developed can be easily adapted for use with other occupations or professions by inserting the appropriate descriptors.
Other limitations include the absence of causal analyses, the
use of self-report measures of work behavior, and the modest reliability of some measures. With regard to the first of these, the data we relied on most in testing the differential-relations hypothesis, the registered-nurse data, were obtained on a single occasion. Although the findings invite speculation concerning causal relations, further testing will be required for verifica- tion. With regard to the measurement issues, at this stage of the research our primary objective was to develop reliable mea- sures of occupational commitment that could be shown to be conceptually distinct from existing measures of organizational commitment. Our inclusion of other measures was in the ser- vice of this objective. For example, although there are obvious problems associated with the use of self-report measures (e.g., confounding because of common method variance), their use in this case was convenient in that it allowed us to measure a number of relevant variables quite efficiently. Similarly, for the sake of brevity, we developed short measures of several con- structs (e.g., exit, voice, loyalty, and citizenship) and used single- item measures of complex constructs (e.g., voluntary absence and job performance), recognizing that we would sacrifice reli- ability. The fact that these measures correlated with our pri- mary measures in a manner generally consistent with our ex- pectations, in spite of the attenuating effect of unreliability, should perhaps increase confidence in the meaningfulness of these findings. At any rate, we hope that these findings will encourage others to test specific relations examined here using more reliable measures and procedures that are less prone to the problems associated with the use of self-report measures.
Conclusion
These findings contribute to a growing body of research that illustrates the need to take a multidimensional approach to the study of commitment. In this case, we have demonstrated the importance of considering not only different forms of commit-
550 J. MEYER, N. ALLEN, AND C. SMITH
ment to the same entity (e.g., organization), but also commit- ment to different entities that might be relevant to the behavior of interest. Specifically, we were able to show that the predic- tion of various behaviors (e.g., turnover intention and responses to dissatisfaction) can be improved by considering commit- ment to both the organization and the occupation. It is conceiv- able that prediction could be improved further by including measures of commitment to the union, to the work group, to the manager, and to other entities in the regression equation.
The procedures used here to develop occupational commit- ment measures could be easily applied to develop multifaceted measures of commitment to other entities. The underlying con- structs of affective, continuance, and normative commitment appear to be generalizable across domains. This is not in- tended, however, to encourage the proliferation of commit- ment measures (cf. Morrow, 1983). Research directed at further extensions of the model should focus on commitment to entities that are theoretically important for understanding the behavior of interest.
References
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance and normative commitment. Journal of Occu- pational Psychology, 63, 1-18.
Allen, N. J., & Smith, J. (1987, June). An investigation of "extra-role" behaviours within organizations. Paper presented at the annual meet- ing of the Canadian Psychological Association, Vancouver, British Columbia.
Aranya, N., Pollock, J., & Amernic, J. (1981). An examination of profes- sional commitment in public accounting. Accounting Organizations and Society, 6, 271-280.
Arnold, J. (1990). Predictors of career commitment: A test of three theoretical models. Journal of Vocational Behavior, 37, 285-302.
Becker, H. S. (1960). Notes on the concept of commitment. American Journal of Sociology, 66, 32-42.
Becker, T. E. (1992). Foci and bases of commitment: Are they distinc- tions worth making? Academy of Management Journal, 35,232-244.
Blau, G. J. (1985). The measurement and prediction of career commit- ment. Journal of Occupational Psychology, 58, 277-288.
Blau, G. J. (1988). Further exploring the meaning and measurement of career commitment. Journal of Vocational Behavior, 32, 284-297.
Blau, G. J. (1989). Testing the generalizability of a career commitment measure and its impact on employee turnover. Journal of Vocational Behavior, 35, 88-103.
Buchanan, B. (1974). Building organizational commitment: The socia- lization of managers in work organizations. Administrative Science Quarterly, 19, 533-546.
Darden, W R., Hampton, R., & Howell, R. D. (1989). Career versus organizational commitment: Antecedents and consequences of re- tail salespeoples' commitment. Journal of Retailing, 65, 80-106.
Farrell, D. (1983). Exit, voice, loyalty, and neglect as responses to job dissatisfaction: A multidimensional scaling study. Academy of Man- agement Journal, 26, 596-607.
Fullagar, C, & Barling, J. (1989). A longitudinal test of a model of the antecedents and consequences of union loyalty. Journal of Applied Psychology, 74, 213-227.
Gordon, M. E., & Ladd, R. T. (1990). Dual allegiance: Renewal, recon- sideration, and recantation. Personnel Psychology, 43, 37-69.
Gordon, M. E., Philpot, J. W, Burt, R. E., Thompson, C. A., & Spiller, W E. (1980). Commitment to the union: Development of a measure and an examination of its correlates [Monograph ]. Journal of Applied Psychology, 65, 479-499.
Griffin, R. W, & Bateman, T. S. (1986). Job satisfaction and organiza- tional commitment. In C. L. Cooper & I. Robertson (Eds.), Interna- tional review of industrial and organizational psychology (pp. 157- 188). New York: Wiley.
Hackett, R. D, Bycio, R, & Hausdorf, P. (1992, August). Further assess- ment of a three-component model of organizational commitment. Academy of Management Best Paper Proceedings, 212-216.
Hirschman, A. O. (1970). Exit, voice and loyalty: Responses to decline in firms, organizations, and states. Cambridge, MA: Harvard Univer- sity Press.
Jackson, P. R., Stafford, E. M., Banks, M. H., & Warr, R B. (1983). Unemployment and psychological distress in young people: The moderating role of employment commitment. Journal of Applied Psychology, 68, 525-535.
Joreskog, K. G, & Sorbom, D. (1989). LISREL 7user's guide. Moores- ville, IN: Scientific Software.
Kerr, S., Von Glinow, M., & Schreisheim, J. (1977). Issues in the studies of "professionals" in organizations: The case of scientists and engi- neers. Organizational Behavior and Human Performance, 18, 329- 345.
Kramer, M. (1974). Reality shock: Why nurses leave nursing. St. Louis, MO: Mosby.
Mathieu, J. E., & Zajac, D. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational com- mitment. Psychological Bulletin, 108,171-194.
McDonald, R. P., & Marsh, H. W (1990). Choosing a multivariate model: Noncentrality and goodness of fit. Psychological Bulletin, 107, 247-255.
McGee, G. W, & Ford, R. C. (1987). Two (or more?) dimensions of organizational commitment: Reexamination of the Affective and Continuance Commitment Scales. Journal of Applied Psychology, 72, 638-641.
Meyer, J. P., & Allen, N. J. (1984). Testing the "side-bet theory" of organizational commitment: Some methodological considerations. Journal of Applied Psychology, 69, 372-378.
Meyer, J. P., & Allen, N. J. (1991). A three-component conceptualiza- tion of organizational commitment. Human Resource Management Review, 1, 61-98.
Meyer, J. P., Allen, N. J., & Gellatly, I. R. (1990). Affective and continu- ance commitment to the organization: Evaluation of measures and analysis of concurrent and time-lagged relations. Journal of Applied Psychology, 75, 710-720.
Meyer, J. P., Bobocel, D. R., & Allen, N. J. (1991). Development of organizational commitment during the first year of employment: A longitudinal study of pre- and post-entry influences. Journal of Man- agement, 17, 717-733.
Meyer, J. P., Paunonen, S. V, Gellatly, I. R., Goffin, R. D, & Jackson, D. N. (1989). Organizational commitment and job performance: It's the nature of the commitment that counts. Journal of Applied Psy- chology, 74,152-156.
Morrow, P. C. (1983). Concept redundancy in organizational research: The case of work commitment. Academy of Management Review, 8, 486-500.
Morrow, P. C., & Wirth, R. E. (1989). Work commitment among sala- ried professionals. Journal ofVocational Behavior, 34, 40-56.
Mowday, R. T, Porter, L. W, & Steers, R. M. (1982). Employee-organi- zation linkages: The psychology of commitment, absenteeism, and turnover. San Diego, CA: Academic Press.
Mowday, R. T., Steers, R. M., & Porter, L. W (1979). The measurement of organizational commitment. Journal of Vocational Behavior, 14, 224-247.
Mulaik, S. A., James, L. R., Van Alstine, J., Bennett, N., Lind, S., & Stilwell, C. D. (1989). Evaluation of goodness-of-fit indices for struc- tural equation models. Psychological Bulletin, 105, 430-445.
COMMITMENT TO ORGANIZATIONS AND OCCUPATIONS 551
Organ, D. W (1987). Organizational citizenship behavior. Lexington, MA: Lexington Books.
Porter, L. W, Steers, R. M, Mowday, R. T, & Boulian, P. V (1974). Organizational commitment, job satisfaction, and turnover among psychiatric technicians. Journal of Applied Psychology, 59,603-609.
Quinn, R. P., & Staines, G. L. (1979). The 1977 quality of employment survey. Ann Arbor: University of Michigan, Institute for Social Re- search.
Randall, D. M., Fedor, D. B., & Longenecker, C. O. (1990). The behav- ioral expression of organizational commitment. Journal of Voca- tional Behavior, 36, 210-224.
Reichers, A. E. (1985). A review and reconceptualization of organiza- tional commitment. Academy of Management Review, 10, 465-476.
Reichers, A. E. (1986). Conflict and organizational commitments. Journal of Applied Psychology, 71, 508-514.
Scholl, R. W (1981). Differentiating commitment from expectancy as a motivating force. Academy of Management Review, 6, 589-599.
Shore, L. M., & Barksdale, K. (1991, August). A longitudinal assess-
ment of the antecedents of affective and continuance commitment. Paper presented at the annual meeting of the Academy of Manage- ment, Miami Beach, FL.
Shore, L. M., & Tetrick, L. E. (1991). A construct validity study of the survey of perceived organizational support. Journal of Applied Psy- chology, 76, 637-643.
Steers, R. M. (1977). Antecedents of outcomes of organizational com- mitment. Administrative Science Quarterly, 22, 46-56.
Steffy, B. D., & Jones, J. W (1988). The impact of family and career planning variables on the organizational, career, and community commitment of professional women. Journal of Vocational Behav- ior, 32,196-212.
Wiener, Y (1982). Commitment in organizations: A normative view. Academy of Management Review, 7, 418-428.
Received April 16,1992 Revision received October 8,1992
Accepted October 8,1992 •
AMERICAN PSYCHOLOGICAL ASSOCIATION SUBSCRIPTION CLAIMS INFORMATION Today's Date:_
We provide this form to assist members, institutions, and nonmember individuals with any subscription problems. With the appropriate information we can begin a resolution. If you use the services of an agent, please do NOT duplicate claims through them and directly to us. PLEASE PRINT CLEARLY AND IN INK IF POSSIBLE.
PRINT FULL NAME OR KEY NAME OF INSTITUTION MEMBER OR CUSTOMER NUMBER (MAY BE FOUND ON ANY PAST ISSUE LABEL)
DATE YOUR ORDER WAS MAILED (OR PHONED)
.PREPAID CHECK ^__CHARGE CHECK/CARD CLEARED DATE:_
STATE/COUNTRY
YOUR NAME AND PHONE NUMBER
TITLE
(If possible, send a copy, front and back, of your cancelled check to help us in our research of your claim.)
ISSUES: MISSING DAMAGED
VOLUME OR YEAR NUMBER OR MONTH
Thank you. Once a claim is received and resolved, delivery of replacement issues routinely takes 4-6 weeks.
———————— (TO BE FILLED OUT BY APA STAFF) ——————
DATE RECEIVED: ACTION TAKEN: _ STAFF NAME:
DATE OF ACTION: _ INV. NO. & DATE: LABEL NO. & DATE:
Send this form to APA Subscription Claims, 750 First Street, NE, Washington, DC 20002-4242
PLEASE DO NOT REMOVE. A PHOTOCOPY MAY BE USED.