JOURNAL OF ORGANIZATlONAL BEHAVIOR, Vol. 9,297-308 (1988)

The impact of interpersonal environment on burnout and organizational commitrnent' MICHAEL P. LEITER Psychology Department, Acadia University, Canada

AND CHRISTINA MASLACH Psychology Department. University of California. Berkeley, U.S.A .

Summary

Organizational commitment and burnout were related to interpersonal relationships of nurses in a small general hospital. Regular communication contacts among personnel were differentiated as supervisor or coworker contact, and these categories were further differentiated into pleasant and unpleasant contacts. The results were consistent with a view of burnout in which emotional exhaustion leads to greater depersonalization which subsequently leads to diminished personal accomplishment. Interpersonal contact with personnel in the organization was related to the development of burnout at each stage. Patterns of pleasant and unpleasant contacts with supervisors and coworkers were related to the three aspects of burnout in a distinct manner. High burnout was related to diminished organizational commitment, which was also related to aspects of the interpersonal environment of the organization. The results are discussed in the context of a comprehensive approach to psychological adjustment to a worksetting.

Introduction In recent years, increasing attention has been paid to the phenomenon of burnout, particularly in human services professions. Burnout appears to be a response to interpersonal stressors on the job, in which an overload of contact with people results in changes in attitudes and behaviours towards them. More specifically, burnout has been defined as a syndrome of emotional exhaustion, depersonalization, and a reduced sense of personal accomplishment, which can occur among individuals who work with people in some capacity (Maslach and Jackson, 1986, 1984a). Emotional exhaustion refers to feelings of being emotionally overextended and drained by one's contact with other people. Depersonalization refers to an unfeeling and callous response toward these people, who are usually the recipients of one's service or care. Reduced personal 'This research was supported by a grant from the Social Sciences and Humanities Research Council of Canada. Please address requests for reprints to Michael Leiter, Psychology Department, Acadia University, Wolfville, Nova Scotia, Canada BOP 1x0.

0894-3796/ 88/040297-12$06.00 1988 by John Wiley & Sons, Ltd.

0

Received 2 December 1986 Revised 9 April 1987

298

M. P. LEITER AND C. MASLACH

accomplishment refers to a decline in one’s feelings of competence and successful achievement in one’s work with people. These three aspects of the burnout syndrome have been the focus of numerous research studies investigating the causes and outcomes of burnout (see Maslach and Jackson, 1986). Virtually all discussions of burnout have proposed that it is a product of both personal and environmental factors, and various studies have looked at one or both of these categories. However, the bulk of the research evidence to date suggests that environmental factors, particularly characteristics of the work setting, are more strongly related to burnout than are such personal factors as demographic and personality variables (e.g. Burke, Shearer and Deszca, 1984; Golembiewski and Scicchitano, 1983; Maslach and Jackson, 1984b). Consistent with the analysis of burnout in interpersonal terms, many of the work characteristics that have been linked to burnout involve contact with other people, whether they be clients or coworkers. Thus, for example, burnout has been correlated with a greater percentage of time in direct care of clients (Lewiston, Conley and Blessing-Moore, 1981; Maslach and Jackson, 1982), more difficult client problems (Meadow, 1981; Pines and Maslach, 1978), caseload (Maslach and Jackson, 1984b; Maslach and Pines, 1977), and a low degree of peer support (Burke et al., 1984; Jackson, Schwab and Schuler, 1986; Maslach and Jackson, 1982). In some cases, interactions with coworkers have bees cited as the most important sources of job stress and burnout (Gaines and Jermier, 1983; Leiter and Maslach, 1986). These findings suggest that contact with people can be a major source of distress, frustration, or conflict in human services professions, and that such a negative experience can be an important element in an employee’s satisfaction with the job and commitment to continue working in it. Such negative interpersonal experiences contrast with the many positive and rewarding aspects of contact with coworkers, which have been discussed most extensively in the research literature on social support in the work setting (Beehr, 1985; House, 1981). Different types of support have been identified (such as emotional versus instrumental), as have different sources of support (such as coworker, supervisor, spouse). Contact with people, whether it be number of people and/or frequency of contacts, has been one primary measure of social support, and consequently such contact has been presumed to be positive in nature. However, recent analyses have pointed out that contact is not always supportive (Barrera, 1985; Brodsky, 1976), an argument that is borne out by the burnout research cited above. Thus, one goal of the present study was to assess both positive and negative contacts on the job, and to investigate their separate contributions to burnout and organizational commitment. In line with the distinction between different sources of social support, the current study distinguished between two different types of job contacts: coworkers and supervisors. It seemed reasonable to assume that the nature of the contact between two fellow employees would not be the same as that between an employee and his or her supervisor. For example, positive contact with a supervisor might involve praise, guidance, and promotions, while positive contact with a coworker might refer more to friendship, help, and comfort. It is conceivable that these different types of positive contact would have different relationships to burnout and commitment, and that a similar argument could be made with regard to negative contacts. Interpersonal contacts are often negative because of various conflicts between the people involved. There may be disagreements about how the job is to be carried out, or there may be multiple demands that cannot be easily met. Conflict can also stem from organizational sources (e.g. institutional policies) or intra-individual sources (e.g. the worker’s personal values) which clash with competing job demands. Work environments that are characterized by high levels of role conflict are more difficult and unpleasant settings in which to work and require much more effort (thus leading to greater exhaustion). Such environments produce a greater level of burnout

INTERPERSONAL ENVIRONMENT

299

among the employees (Jackson et al., 1986; Schwab and Iwanicki, 1982), and burnout, in turn, should produce a lower level of organizational commitment. Organizational commitment refers to employee identification with, and involvement in, a particular organization (Mowday, Steers and Porter, 1979). Higher levels of commitment are reflected in lower rates of turnover (Price and Mueller, 1981; Steers, 1977), and are also believed to be related to enhanced productivity and better delivery of services. It seems reasonable to postulate that the experience of burnout would lead employees to be less committed to the organization, and thus more likely to quit their job. If employees are feeling emotionally exhausted by their work, are developing depersonalized views of the people they serve, and are feeling that they are not accomplishing much, then they would probably become less enthusiastic and accepting of the organization’s goals, less dedicated to achieving them, and more prone to withdrawing from work (both psychologically and physically). Thus, organizational commitment may be the mediating variable in the link that has been found between burnout and both turnover and job withdrawal (Jackson et al., 1986; Lazaro, Shinn and Robinson, 1984; Maslach and Jackson, 1984b). Similarly, burnout may be an important mediating variable between interpersonal aspects of the work environment and organizational commitment. Previous research has found that role conflict and interpersonal contacts (as operationalized in terms of communication networks) are predictors of commitment (Eisenberg, Monge and Miller, 1983; Morris and Sherman, 1982), although the underlying dynamic has not really been specified. It could be argued that burnout is the key factor here, in that it is a likely response to interpersonal stressors and conflict, which then instigates withdrawal from the organization. For the most part social support at work has been considered as moderating the relationship between job stressors and stress reactions of workers. In some circumstances, however, it may be useful to conceive of the social environment as a direct contributor to workers’ reactions to a worksetting. To explore the role of social contacts, this study took place in a small hospital which was fairly homogeneous in regard to stressors. While it is recognized that a wide variety of stressors (caseload, access to resources, etc.) contribute to job stress, a comprehensive exploration of burnout can benefit from an exploration of direct relationships between the social environment of a worksetting and burnout.

Hypotheses This analysis of interpersonal aspects of the work environment rests on the general assumption that role conflict and contacts with other employees influence the level of burnout, which, in turn, influences level of organizational commitment. More specifically, this analysis leads to the following set of predictions about the relationships among these variables. First of all, it is postulated that contacts with other employees in the organization will be predictive of the three different aspects of burnout. Contacts are defined in terms of a mutual choice by both of the relevant employees, and not simply in terms of the self-report of one of them. Negative contacts (that is, contacts that are unpleasant or stressful in some way) are potentially stressful and will be directly related to higher levels of emotional exhaustion and indirectly with depersonalization, but unrelated to levels of personal accomplishment. Depersonalization is expected to be negatively related to pleasant coworker and supervisor contact, because the absence of such contact diminishes a worker’s capacity to cope with exhaustion and stress. In contrast, personal accomplishment will be related only to the presence of positive contacts (contacts that are pleasant or supportive). Because supervisors are in a position to provide instrumental as well as emotional support (Beehr, 1985), contacts with supervisors may

300

M. P.LEITER AND C. MASLACH

show different relationships to burnout than do contacts with coworkers. In addition to interpersonal contacts, the presence of role conflict is expected to be related to burnout. Based on previous research findings, role conflict should be most strongly predictive of the emotional exhaustion aspect of burnout. A second set of hypotheses has to do with the interrelationships among the three aspects of the burnout syndrome. According to earlier theorizing (Maslach and Jackson, 1982), emotional exhaustion is likely to be the first phase of burnout, given that it is a response to the emotional stressors of the job. Once emotional exhaustion occurs, an individual may attempt to cope with it by detaching him or herself from others and developing a depersonalized response to them. Thus, emotional exhaustion should be a predictor of depersonalization. Once depersonalization occurs, the individual should begin to feel less successful on the job and should evaluate him or herself less positively in terms of actual accomplishments. Thus, depersonalization should be a predictor of level of personal accomplishment. In other words, the presence of emotional exhaustion will only lead to a sense of reduced personal accomplishment if depersonalization occurs as a mediating variable. It should be noted that these hypotheses about the interrelationships of the three aspects of burnout stand in contrast to those postulated by Golembiewski and Munzenrider (1984), in which depersonalization is predicted to lead to decreases in personal accomplishment, which subsequently lead to higher levels of emotional exhaustion. Finally, it is hypothesized that scores on the three aspects of the burnout syndrome will predict levels of organizational commitment. That is, higher levels of emotional exhaustion and depersonalization, and lower levels of personal accomplishment, will each be related to lower levels of commitment. It is also expected that organizational commitment will be positively correlated with pleasant supervisor and coworker contact, and negatively correlated with both types of unpleasant contact. It is unclear whether the relationships of organizational commitment with the communication variables will be independent of organizational commitment with the burnout scales.

Method Subjects Subjects were 52 of 74 nurses and support staff (49 females and three males) of a small private hospital in an urban area of Northern California. The hospital comprised an emergency room and an inpatient facility which was divided into a seven-bed cardiac carelintensive care unit and a general ward. Inpatients were predominantly persons recuperating from surgery in a nearby general hospital. Emergency room patients were largely members of the health plan with which the hospital was associated, as well as persons admitted as police emergencies. Support staff were full-time workers of the hospital from other clinical areas: radiology and physical therapy. The hospital was one unit of a large health care service with hospitals and clinics throughout the western United States. It was closely associated with an outpatient facility a mile away. Doctors and some of the support staff divided their time between the two locations, but nurses and the support staff participating in the study worked only in the hospital. In addition to direct service staff, persons working in billing, personnel, dietary, and general administration also worked at the hospital, but did not take part in the study. Participants were differentiated as supervisors, who included the director of nursing, the assistant director and six supervisors who regularly covered a given shift and as coworkers who comprised the remaining subjects.

INTERPERSONAL ENVIRONMENT

301

Instruments Burnout was measured by the Maslach Burnout Inventory (Maslach and Jackson, 1981, 1986). This is a 22-item measure which produces three scores: emotional exhaustion, depersonalization and personal accomplishment. In this sample, Cronbach’s coefficient alpha for emotional exhaustion was 0.91, for depersonalization, 0.63, and for personal accomplishment, 0.73. Role conflict was measured by a scale developed by Rizzo, House and Lirtzman (1970). This measure has been widely used in organizational research, and appears to be the most valid measure of role conflict available (Van Sell, Brief and Schuler, 1981). In this sample, the coefficient alpha was 0.76 for role conflict. Organizational commitment was measured by a short version of the Organizational Commitment Questionnaire (Mowday et al., 1979). Following a suggestion by these authors, the current study used only the items which were scored in a positive direction. This scale produces one score, organizational commitment (alpha = 0.83), which is indicative of the extent to which a subject values the organization for which he or she works. Interpersonal contacts were measured using a procedure developed by Leiter and Meechan (1986). Subjects reviewed a list of all organizational employees and indicated those with whom they had ongoing, regular interactions concerning work, administrative, or personal issues. For two subjects to be considered as being in regular contact with one another, both of them had to report independently that they had such contacts. The restriction to reciprocal citations introduced an element of objectivity into the measure of interpersonal contact, in that a subject’s perception of a relationship had to match the perception of the other member of the dyad. The total number of contacts was differentiated into coworker contacts and supervisor contacts. There was generally a high degree of reciprocity in the reporting of communication contacts. On the average 77 per cent of reported contacts were reciprocated; reciprocation rates ranged from 50 per cent to 100 per cent with most falling between 75 per cent and 85 per cent. After identifying their regular contacts, subjects were then asked to rate each of these relationships as being generally pleasant, neutral or unpleasant. Thus, within the category of coworker contacts, a distinction was made between number of pleasant contacts and number of unpleasant contacts. A similar distincti-on was made for supervisor contacts as well.

Procedure The experimenters first discussed the study with the person in charge of staff development at the hospital, and subsequently with the director of nursing. Next, the experimenters attended a series of nursing staff meetings at which they described the goals and methods of the study and demonstrated the computer procedure. The experimenters informed subjects that the study was an outside research project examining relationships between organizational design and job stress, and that participation was voluntary. The hospital was promised a report at the end of the study summarizing general findings. Individuals were to be given an opportunity at the end of the study to see their scores on the questionnaires. The experimenters assured subjects that individual scores would be kept confidential. An Epson QX-10 microcomputer provided by the experimenter was used to administer the questionnaires. The first screen presented a brief description of the study and provided instructions for entering answers into the computer. The subjects were then asked to identify themselves in a list of participants’ names. Subsequently they completed the Maslach Burnout Inventory, the Role Conflict Scale, the Organizational Commitment Questionnaire, and the interpersonal contacts measure. Subjects completed the questionnaires during working hours

302

M. P. LEITER AND C. MASLACH

whenever their work schedules permitted. The programme required that the subject complete all forms in one session, which lasted from 20 to 55 minutes.

The analysis of the relationship between communication involvement and burnout involved examining correlations. This procedure was supplemented by multiple regressions analyses when to test the extent to which correlations are independent of one another. In these analyses only the data from the 44 nurses and support staff were used, because the definition of a supervisor or a coworker contact was not consistent across the supervisors and coworkers. That is, references by supervisors to supervisors had a different meaning than did references by nurses to the supervisors: The supervisors were referring to contacts with administrative colleagues, while the nurses were referring to contact with people who were in charge of their work. Table 1 displays the Pearson Product Moment Correlation Coefficients among the variables in the study, as well as their means and standard deviations. Table 1. Zero-order correlations Mean 1 Emotional exhaustion 21.63 7.52 2 Depersonalization 3 Personal accomplish. 41.32 4 Role conflict 22.47 5 Organizationalcommit. 45.00 22.98 6 Pleasant coworkers 1.20 7 Unpleasant coworker 7.41 8 Pleasant supervisor 9 Undeasant suDervisor 0.57 * p < 0.05, N

S.D. 1 13.41 5.65 0.49* 6.64 -0.15 8.35 0.65* 9.26 -0.52* 12.67 0.17 2.11 0.24 5.63 -0.11 0.98 0.58*

2

3

4

5

6

-0.35* 0.26 -0.11 -0.37* 0.36* -0.34* -0.27 0.37* 0.19 0.15 0.05 -0.28* 0.19 0.09 -0.12 0.29* 0.32* -0.15 0.19 -0.18 0.47* -0.60* 0.20 0.32* 4.10

7

8

0.17 0.21

-0.20

44.

Burnout The findings confirmed two of the predictions concerning emotional exhaustion. It was positively related to role conflict ( r = 0.65, p < 0.05) and unpleasant supervisor contact (r 0.58, p < 0.05), but the expected negative correlation with pleasant supervisor contact was absent (r = -0.1 1, n.s.). A multiple regression analysis confirmed the expectation that role conflict and unpleasant supervisor contacts were each independently related to emotional exhaustion as predicted (see Table 2). Together the two predictor variables accounted for 50 per cent of the variance in emotional exhaustion (adjusted R2 = 0.502, F(2,41) = 22.67, p < 0.001). In addition to the positive relationship with emotional exhaustion, depersonalization was expected to have negative relationships with both types of pleasant contact. Depersonalization was related to pleasant coworker contact (r = -0.27, p < 0.05) as expected, but not pleasant supervisor contact ( r = -0.15, n.s.). It was positively correlated with unpleasant supervisor contact ( r = 0.32, p < 0.05). In a multiple regression analysis only positive coworker contact was significantly related with depersonalization in combination with emotional exhaustion (adjusted R2 = 0.334, 4 2 , 4 1 1 = 1 1 . 8 1 , ~< 0.001). In addition to its relationship with depersonalization, personal accomplishment was expected to be correlated with pleasant supervisor contact and pleasant coworker contacts. As Table 1

INTERPERSONAL ENVIRONMENT

Table 2. Multiple regression analyses of burnout Variable Beta T Emotional exhaustion so1 4.145 Role conflict Unpleasant supervisor .315 2.555 Unpleasant coworker .I45 1.332 Depersonalization Emotional exhaustion .556 4.285 -.373 -2.721 Positive coworker Positive supervisor .030 0.225 Personal accomplishment -.394 -2.91 1 Depersonalization Positive coworker .29 1 2.149

303

Significance pO.O5 p

p < 0.001 p < 0.01 p>o.o5 p < 0.05 p 0.05

304

M. P. LEITER AND C. MASLACH

did not make a significant contribution to predicting organizational commitment after the other two variables were entered (see Table 3). This analysis also indicated that unpleasant supervisor contacts and pleasant coworker contact would account for a substantial portion of the variance in organizational commitment after the two burnout subscales were entered (see Table 3). The four variables account for 49 per cent of the variance in organizational commitment (adjusted R2 = 0.491, F (4.39) = 11.364, p < 0.001).

Unpleasant Coworker Contact

.28

I

-.39

-.30

Depersonalization

I

-.35

f

.13

L

-SO

Accomplishment

T Figure 1. Burnout model

Figure 1 displays a revised version of the model developed in the Introduction. Significant correlations are displayed in bold print; nonsignificant relationships are indicated with dotted connecting lines. In this figure three lines are added: The significant direct relationships of unpleasant supervisor contact with role conflict and organizational commitment, and that of pleasant coworker contact with organizational commitment.

INTERPERSONAL ENVIRONMENT

305

A correlational analysis was performed to provide a more specific perspective on the pleasant and unpleasant aspects of coworker relationships. All 52 subjects were used in this analysis because the supervisor and coworkers contacts were not differentiated. The average commitment and burnout scores of each subject’s contacts were computed; these average scores were then correlated with the subjects’ scores. The analysis indicates that subjects who were committed to the organization tended to interact with coworkers who were also high on organizational commitment (r = 0 . 5 0 , < ~ 0.01) as well as high on personal accomplishment ( r = 0.43, p < 0.01). Further, subjects who were high on personal accomplishment were interacting more often with coworkers who were committed (r = 0.22,p < 0.05). Subjects who depersonalized clients tended to maintain more contacts with coworkers who were high in emotional exhaustion ( r 0.23, p < 0.05) and lower in personal accomplishment ( r = - 0 . 2 6 , < ~ 0.05). The emotional exhaustion level of subjects was unrelated to the scores of their contacts on any of the four measures.

Discussion The results of this study provide general support for the hypotheses proposed above, although with some exceptions. Emotional exhaustion was more prevalent for subjects with a negative interpersonal work environment (i.e. more role conflict and more unpleasant contacts with supervisors). Personal accomplishment and depersonalization were each related to pleasant interpersonal contact with coworkers, although in different directions. The data were consistent with the predicted relationships among the three aspects of burnout. The results support the expected relationships between burnout and organizational commitment; in addition, two aspects of burnout (emotional exhaustion and personal accomplishment) each made an independent contribution to the prediction of commitment. A posf hoc analysis suggests that there are direct negative relationships between unpleasant supervisor contacts and organizational commitment and a positive relationship with pleasant coworker contacts and organizational commitment, in addition to the relationship with burnout. Together, the results suggest that interpersonal relationships at work may be instrumental in explaining the variation in stress reactions within a worksetting which is homogeneous in regard to job stressors.

Burnout The findings for emotional exhaustion indicate that it is related to both role conflict and unpleasant contacts with supervisors. These two predictors are correlated with one another, suggesting that role conflict is one aspect of unpleasant supervisor contacts. Role conflict is more concerned with formal aspects of a job, in that it measures the extent to which workers perceive supervisors to be making irreconcilable demands on their time and resources. This relationship is consistent with Beehr (1985), who suggested that supervisor contact is more likely to comprise instrumental support, affecting the conditions which lead to job stress. In contrast, coworker contact provides more emotional support which helps a worker cope with existing stressors. This analysis extends this concept by indicating that unpleasant contact with supervisors is associated with at least one organizational stress, role conflict. Role conflict may be the occasion of unpleasant interactions with persons to whom a worker is accountable, but a worker may also have unpleasant encounters with supervisors which have little to do with that worker’s role definition. The independent contribution of both role conflict and unpleasant supervisor contact indicates that each of these sources of interpersonal stress may be important for the development of emotional exhaustion.

306

M. P. LEITER AND C. MASLACH

In the current study, participants rated most of their relationships positively. The mean number of pleasant contacts was 39.51, while the mean number of unpleasant contacts was 1.77. This difference may indicate that the interpersonal environment of this organization was predominantly pleasant, or that participants were disinclined to note any unpleasant relationships unless the relationships were strongly salient. The difference may also indicate that members of this organization had a large degree of choice concerning their interpersonal relationships. As one subject remarked while completing the interpersonal contacts measure, ‘Why would I spend my time talking to someone I don’t like?’ In any case, the results indicate that the presence of only a few ongoing unpleasant contacts may be sufficient to influence the level of emotional exhaustion. Depersonalization was higher for employees who had higher levels of emotional exhaustion and who had more unpleasant contact with supervisors and less pleasant contact with coworkers. The relationship between depersonalization and unpleasant supervisor contact may be circular, in that workers who depersonalize may receive poor evaluations from their supervisors due to their impersonal treatment of patients. The unpleasant encounters in turn may increase the stress on the worker. The multiple regression analysis suggests that unpleasant supervisor contact is partially mediated through emotional exhaustion. Some aspects of unpleasant supervisor contact may be especially pertinent to depersonalization. For instance, supervisors may be partially responsible for creating a more dehumanized, impersonal work environment through their interpretation of hospital policy. The analysis of contact scores suggests that the relationship of depersonalization with coworker contacts may be more complex than indicated by the absence of a correlation between the two variables. Although workers who depersonalize did not rate more of their coworker contacts as unpleasant, their interpersonal contacts were more likely to be experiencing emotional exhaustion and diminished personal accomplishment. These relationships indicate that workers were not simply aggregating according to their tendency to depersonalize, but that their depersonalization was consistent with the exhaustion and discouragement experienced by the workers and those around them. As expected, personal accomplishment was negatively correlated with depersonalization. The results are consistent with the following sequence: (1) Stressful interactions with supervisors increase the workers’ feelings of emotional exhaustion; (2) high levels of exhaustion lead to depersonalization, unless the workers have frequent supportive contact with their coworkers; (3) as depersonalization persists, the workers’ feelings of accomplishment in their work diminish, although supportive interpersonal contact with coworkers may help to decelerate this process.

Organizational commitment The results are consistent with the expectation that burnout leads to reduced organizational commitment. Although each of the three aspects of burnout were significantly correlated with organizational commitment, they did not all display a unique relationship with organizational commitment when entered into a multiple regression. The relationship between commitment and depersonalization appears to be redundant with that of the other two aspects of burnout. It may be, as suggested earlier, that the role of depersonalization is in mediating the development of reduced personal accomplishment, and thus it does not have an independent contribution to make to the level of commitment. A post hoc analysis suggests that there is an important direct relationship between unpleasant supervisor contacts and organizational commitment, in addition to the predicted indirect influence mediated through emotional exhaustion. That is, ongoing negative interactions with

INTERPERSONAL ENVIRONMENT

307

supervisors may adversely affect organizational commitment in ways other than through increased burnout. For example, a worker may find hassles with supervisors to be sufficient cause to dislike a worksetting without experiencing burnout in the process. According to this perspective, burnout is one path leading from unpleasant supervisor contact to organizational commitment, but not the only one. Similarly, supportive contact with coworkers may enhance commitment to an organization in ways independent of its relationships with burnout. In addition, the analysis of contact scores indicates that committed workers interacted more often with other workers who were committed, and workers who were indifferent towards the organization interacted more often with like-minded coworkers. Further, it appears that the level of personal accomplishment experienced by both subjects and their contacts was related to organizational commitment. This is another way in which aspects of coworker contact affect organizational commitment directly, in addition to the effects mediated through a subject’s burnout level. In conclusion, this study provides a more complete analysis of workers’ adjustment to a worksetting and provides direction for developing a comprehensive model of burnout and organizational commitment. First, it clarifies relationships of interpersonal contacts with burnout by differentiating between pleasant and unpleasant contacts. Secondly, the examination of contact scores allows a more precise delineation of what is occurring in coworker relationships which are pertinent to the development of burnout and organizational commitment. Thirdly, it provides a framework for explaining the variation of stress reactions in a worksetting which is homogeneous in regard to job stressors. Finally, the study provides a viable perspective on the relationship between burnout and organizational commitment, two concepts which are receiving attention in current organizational research. Research indicates that both of these measures are related to aspects of the interpersonal environment of an organization, as well as to an important organizational outcome, turnover. To avoid conceptual confusion, it is essential to specify the relationship of burnout and organizational commitment to one another and with these other variables. The research reported here furthers this process, and provides direction for subsequent analyses of the contribution of coworker and supervisor relationships to burnout and commitment.

References Barrera, M. (1985). ‘Distinctions between social support concepts, measures, and models’. Unpublished manuscript, Arizona State University. Beehr, T. A. (1985). ‘The role of social support in coping with organizational stress’. In: Beehr, T. A. and Bhagat, R. S. (Eds) Human Stress and Cognition in Organizations, Wiley, New York. Brodsky, C. M. (1976). The Harassed Worker, D. C . Heath, Lexington, Mass. Burke, R. J., Shearer, J. and Deszca, G . (1984). ‘Burnout among men and women in police work: An examination of the Cherniss model’, Journal of Health and Human Resources Administration, 7 , 162-1 88.

Eisenberg, E. M., Monge, P. R. and Miller, K. I. (1983). ‘Involvement in communication networks as a predictor of organizational commitment’, Human Communication Research, 10, 179-201. Gaines, J. and Jermier, J. M. (1983). ‘Emotional exhaustion in a high stress organization’, Academy of Management Journal, 26,567-586.

Golembiewski, R. T. and Munzenrider, R. (1984). ‘Phases of psychological burnout and organizational covariants: A replication from a large population’, Journal of Health and Human Resources Administration, 6, 290-323.

Golembiewski, R. T. and Scicchitano, M. (1983). ‘Testing for demographic covariants: Three sources of data rejecting robust and regular association’, International Journal of Public Administration, 5,435-447. House, J. S. (1981). Work Stress and Social Support, Addison-Wesley, Reading, MA.

308

M. P. LEITER AND C. MASLACH

Jackson, S. E., Schwab, R. L. and Schuler, R. S. (1986). ‘Understanding the burnout phenomenon’, Journal of Applied Psychology, in press. Lazaro, C., Shinn, M. and Robinson, P. E. (1984). ‘Burnout, job performance, and job withdrawal behaviors’, Journal of Health and Human Resources Administration, 7 , 213-234. Leiter, M. P. (1985). ‘Burnout as a function of communication involvement in multidisciplinary health teams’. Annual Convention of the Canadian Psychological Association, Halifax, Nova Scotia. Leiter, M. P. and Maslach, C. (1986). ‘Job stress and social involvement among nurses’. Paper presented at the Annual Conference of the International Network for Social Network Analysis. Santa Barbara, CA. Leiter, M. P. and Meechan, K. (1986). ‘Role structure and burnout in human services’, Journal of Applied Behavioral Science, in press. Lewiston, N. J., Conley, J. and Blessing-Moore, J. (1981). ‘Measurement of hypothetical burnout in cystic fibrosis caregivers’, Acta Pediatrica Scandinavica, 70, 935-939. Maslach, C. and Jackson, S. E. (1981). ‘The measurement of experienced burnout’, Journal of Occupational Behavior, 2, 99-113. Maslach, C. and Jackson, S. E. (1982). ‘Burnout in health professions: A social psychological analysis’. In: Sanders, G. and Suls, J. (Eds) Social Psychology of Health and Illness, Erlbaum, Hillsdale, NJ. Maslach, C. and Jackson, S. E. (1984a). ‘Burnout in organizational settings’, Applied Social Psychology Annual, 5, 133-153. Maslach, C. and Jackson, S. E. (1984b). ‘Patterns of burnout among a national sample of public contact workers’, Journal of Health and Human Resources Administration, 7 , 189-212. Maslach, C. and Jackson, S. E. (1986). Maslach Burnout Inventory Manual, 2nd edn, Consulting Psychologists Press, Palo Alto, CA. Maslach, C. and Pines, A. (1977). ‘The burn-out syndrome in the day care setting’, Child Care Quarterly, 6, 100-1 13. Meadow, K. P. (198 1). ‘Burnout in professionals working with deaf children’, American Annals of the DeaL 126 ( I ) , 13-22. Mettlin, C. and Woelfel, J. (1974). ‘Interpersonal influence and symptoms of stress’, Journal of Health and Social Behavior, 15, 31 1-319. Morris, J. H. and Sherman, J. D. (1981). ‘Generalizability of an organizational commitment model’, Academy of Management Journal, 24, 5 12-526. Mowday, R. T., Steers, R. M. and Porter, L. W. (1979). ‘The measurement of organizational commitment’, Journal of Vocational Behavior, 14,224-247. Pines, A. and Maslach, C. (1978). ‘Characteristics of staff burn-out in mental health settings’, Hospital and Community Psychiatry, 29,233-237. Price, J. L. and Mueller, C. W. (1981). ‘A causal model of turnover for registered nurses’, Academy of Management Journal, 24, 543-565. Rizzo, J. R., House, R. J. and Lirtzman, S. I. (1970). ‘Role conflict and ambiguity in complex organizations’, Administration Sciences Quarterly, 15, 150-163. Schwab, R. L. and Iwanicki, E. F. (1982). ‘Perceived role conflict, role ambiguity, and teacher burnout’, Educational Administration Quarterly, 18, 60-74. Steers, R. M. (1977). ‘Antecedents and outcomes of organizational commitment’, Administrative Science Quarterly, 22, 46-56. Van Sell, M., Brief, A. P. and Schuler, R. S. (1981). ‘Role conflict and role ambiguity: Integration of the literature and directions for future research’, Human Relations, 34,43-71.