Burnout, job satisfaction, and job performance

zyxw zyxw Burnout, Job Satisfaction, and Job Performance Mary Randall Phillip Health Centre and William A Scott Australian National University Abstract: Job “burnout” is often used in ways not well distinguished from older concepts, such as job dissatisfaction and poor performance. An attempt was made to distinguish the three notions, both theoretically and operationally, and to investigate their presumably distinctive correlates in two samples of employees 248 nurses (professional sample) and 108 service employees (non-professional). In both samples the three measures were intercorrelated, but burnout could be distinguished from subjective incompetence in that the latter state waq found most commonly among employees ofjunior status. In the professional sample, but not in the non-professional sample, burnout could barely be distinguished from job dissatisfaction, in that the latter state was somewhat less associated with the personality characteristics of “striver/ achiever” (a component of “Type A personality”) and commitment to life (a component of “hardiness?. Nevertheless, more theoretical and operational attention are needed if clear distinctions among these three reactions to employment are to be made. ~ zyxw zy zyxwvu In much of the current literature, the concept of burnout, its causes, and prevention are discussed without attempting a clear definition, or distinction from other reactions to the job, such as dissatisfaction and poor performance. In the research to be presented here, an attempt was made to define burnout, both conceptually and operationally, as distinct from job dissatisfaction and poor subjective performance. Then correlates of the three variables were compared to see if they corresponded to theoretically meaningful distinctions. Conceptions Burnout Burnout is defined, Following Maslach and Jackson (1981) and Meier ( 1 983), as “feelings and attitudes of a worker involving a combination of ( 1 ) emotional exhaustion and desire to withdraw from others and work; (2) depersonalisation and negative attitudes to clients and other staff; (3) dissatisfaction with personal accomplishments on the job and negative Requests for reprints should be sent to Professor W. A. Scott, Department of Psychology, Australian National University, PO Box 4, Canberra, A.C.T. 2601. Australian Psychologist Vol. 23 No. 3, November 1988 335 336 zyxwvutsrq zyxwv zyxwv Burnout, Job Satisfaction and Job Performance self-evaluation; and (4) lack of expected reward and recognition on the job.” Some authors (such as Jones, 1982) include behavioural attributes in definitions of burnout, but it appears more reasonable to measure such aspects in a performance inventory rather than a burnout scale, as burnout has traditionally been associated with a state of mind which may contribute to poor performance. Although sources of burnout have not been clearly indicated from research to date, two alternative explanations have been proposed. The first is that burnout arises due to “depletion of personal resourcefulness, flexibility and positive energy” (Carmichael, 1979). This interpretation assumes initial competence of the employee, and gradual decline in morale and performance, with concurrent increase in feelings of burnout. Alternatively, burnout may arise due to an early criris of competence. The latter possibility was proposed by Cherniss (1980), who argued that new employees in the helping professions, while holding the necessary qualifications, lack the coping skills, insight and resources which are gained through experience on the job. This interpretation would predict feelings of burnout over the initial period of work, which diminish with time in the job. zyxwv zyxw Job Dissat isfact ion Dissatisfaction with one’s job, though presumably a consequence or concomitant of burnout, is more focused on the job than on the self. A dissatisfied employee does not like the job, whereas a “burned out”emp1oyee feels incapable of performing it adequately. Sources of job dissatisfaction include a great many jobJrelated characteristics independent of the incumbent’s morale and subjective competence - such as pay, supervisory practices, and congeniality of colleagues. It is expected, therefore, that: Hypothesis I : Job dissatisfaction is associated predominantly with judged negative characteristics of the job, whereas burnout is associated both with these and with reported characteristics of the self. Performance Poor performance on the job refers to an external standard which the incumbent fails to meet. The employee’s own judgment, though vulnerable to influence from personal and situational factors, is mainly focused on a discrepancy from standard that is discernible to colleagues and supervisors as well. Performance has, thus, a more objective, consensually shared referent than burnout and dissatisfaction. It is therefore expected that: Hypothesis 2: Employees’ negative judgments of their own job performance are less related than job dissatisfaction and burnout to characteristics of the person and job; more related to their own inexperience and low status in the organisation and to their supervisors’ negative appraisals of the employees’job performance. Correlates of Burnout While no previous study has investigated distinctive correlates of burnout (i.e., correlates different from those of dissatisfaction and poor per- zy zyxwvut zyxwv zyxwv Burnout, Job Satisfaction and Job Performance 337 formance), a number of recent findings pertain to employees identified only as “burned out”; not all of them are mutually consistent, however; nor do they uniformly support the predictions made above. Studies by Maslach and Jackson (198 I), Schwab and lwanicki (1982), and Weinberg, Edwards and Garove ( 1983) found higher-than-average rates of burnout among young, single employees. Although consistent with the second interpretation of “burnout” (see above) as reflecting a crisis of competence, this fails to distinguish “burnout” from felt incompetence, for which no new label is required. Berkeley Planning Associates( 1977) reported burnout relatively frequently in high-status positions, such as professional and managerial staff of childcare facilities. Direct service providers and agency directors experienced less burnout. (The scale used in this study was named alienation, although its constituents were almost identical to those of burnout.) In contrast, Weinberg et al. (1983), using a burnout inventory, found no systematic difference in amount of burnout among administrators, professionals and directcare personnel in I4 residential settings for the developmentally disabled. Similarly, Riggar, Harrington Godley, and Hafer( 1984) found no significant differences in burnout between rehabilitation administrators and direct-service providers in 155 respondents. The different clientele or the different location of high status workers in the first study may account for its atypical findings. Maslach and Jackson (1982) reported burnout in conjunction with certain kinds of job stress produced by relations with clients or inadequate feedback on their work, while Berkeley Planning Associates (1977) claimed a connection with felt lack of support on the job. As these characteristics of the job were judged by the respondingemployees, rather than independently assessed, there is some ambiguity about the direction of effect. They would appear even more germane to job dissatisfaction than to burnout. According to Hypothesis I , the major distinctive correlates of burnout should be characteristics of the self, such as personal hardiness (Kobasa, 1979; Kobasa, Maddi & Kahn, 1982). Hardiness consists of several components: (a) internal locus of control; (b) commitment to life, or the ability to feel deeply involved in one’s activities; (c) challenge, that is, the anticipation of challenges as exciting rather than as threats to security; this involves flexibility and adaptability. Following Hypothesis I , burnout would be associated with lack of hardiness. In addition, following Hypothesis I , it was expected that “Type A”personality, reflecting a striverachiever disposition (Caplan, Cobb, French, Van Harrison & Pinneau, 1975)would be distinctively associated with burnout, under the assumption (Carver & Humphries, 1982) that such people become socially isolated through their obsession with work, and are less able to tolerate job-related frustrations. This prediction is most pertinent to a “depletion-of-resources’: view of burnout, whereas the predicted negative correlation with hardiness is consistent with either interpretation presented above. zyxwvutsr zyxwvuts zyxwvutsr zyxwv zyxwvu 338 Burnout, Job Satisfaction and Joh Performance Method Subjects As there is some question whether burnout appears mainly among helping professionals, or is equally prevalent among other service employees (Cherniss, 1980; Edelwich & Brodsky, 1980; Ford, Murphy & Edwards, 1983), it seemed appropriate to investigate the hypothesised relationships in two employment settings. Sample I (n = 248) included 83% of the 300 registered nurses from all areas of the Royal Canberra Hospital; ratings of work performances of 163 (66% of the responding nurses) were made by their immediate supervisors (with each responding nurse’s permission). Sample 2 (n = 108) comprised 89% of the 122 workers from all areas of the Motor Vehicle Registry Section of the Commonwealth Public Service, in Canberra; ratings of their work performance were made (with the focal employee‘s permission) by the immediate supervisors of 99 (92% of the responding workers). Demographic differences between the two samples, as well as mean differences on the substantive variables of this study, may be inferred from Table I . Because scores on most of the substantive variables have no absolute meanings, point-biserial correlations (rather than means and standard deviations) are reported, to give an idea of the relative magnitudes of differences between the two samples. These are trivial for the dependent variables, and of occasionally moderate size for reported job characteristics. Nurses tended to have been in similar employment longer, to spend more time with clients, and to report both more prestige and more stress than did Motor Registry employees. Most pronounced are the sex and education differences between the two samples. Measures zyx zyxw Measures used were identical for the two samples, apart from changing the word “patient” to “client” in the Motor Registry questionnaire Burnour was measured in two ways. Firstly, the original 22-item Maslach Burnout Inventory (Maslach & Jackson, 1981) was used. This contains a series of statements about emotional exhaustion, depersonalisation, and personal accomplishment. Respondents were required to rate on scales of 0-6 and 0-7 the frequency and intensity with which they had experienced feelings referred to in the statements. In the present study, alphacoefficients (Cronbach, 195I) for the emotional exhaustion subscale were .86 (frequency) and .86 (intensity); for the depersonalisation subscale they were .71 (frequency) and .77 (intensity); and for the personal accomplishment subscale they were .76 (frequency) and .77 (intensity). Secondly, burnout was measured using 9 items from the Maslach Burnout Inventory combined with 8 other items which were reverse worded. This avoids the major methodological problem in the original version, that pro and con items are not balanced; hence, bias due to acquiescence set may account for correlations with other scales. Further items relating to lack of recognition and reward were also added. The reliability of the composite measure (21 items) was .87. zyxwv zy zyx zyxwv zyx zyxwvu zyxwv Burnout, Job Satisfaction and Job Performance 339 Table 1 Comparison of Nurses and Motor Registry Employees. Demographic variables Female Age Single Number of children Years education Dependent variables Burnout Job dissatisfaction Poor performance (self-reported) (supervisor-rated) Job characteristics Level Time in position Time in similar work Hours with clients Nurses ( n = 248) Motor Registry ( n = 108) I 47% 7.04 * 97% Mz32.I ( S B 8 . 4 ) M=37.4 (SB12.7) 3.96 * 16% 2.95 * 31% W 1 . 2 ( S D 1 . 4 ) M=1.9 ( S D 1 . 2 ) 4.79 * W15.7 ( S B 1 . 5 ) M.ll.2 I S D 1 . 5 ) 56.44 * i" Job characteristics (cont.) r" -. 10 Prestige .25 * .I5 * Stress .39 * Social support .05 .07 Personality characteristics -.O I Hardiness (composite) .I2 * Internality .09 -.21 * Commitment to life .02 .O I Desire for security -.I4 * .44 * .28 * Flexibility Striver/achiever .I2 * .I5 * * p < .05. positive (point-biserial)correlation indicates that the mean score for nurses was higher than that for Motor Registry employees. *A Using the original version, patterns of external correlates for frequency and intensity ratings were virtually identical, implying that these two kinds of ratings were redundant. This contrasts with Maslach and Jackson's (1981) interpretation that frequency and intensity are distinct because they correlated only .52. Moreover, the average similarity in external correlates of the three balanced subscales (emotional exhaustion, depersonalisation, and lack of personal accomplishment) was 0.73. (It is noteworthy that Maslach and Jackson, 1981, also reported similarities in some correlates of the subscales; for instance, in a study of 9 1 social service and health workers, the correlations of job satisfaction with emotional exhaustion, depersonalisation and lack of personal accomplishment were -.23, -.22 and -.17 respectively.) In the present study, a small but definite improvement in predicted correlations with other variables was found with the composite scale of counterbalanced items. This result prompted us to use the single balanced scale as a measure of burnout in subsequent analyses. Results utilising the original, separate scales were virtually the same as those to be reported here. Job dissatisfaction was measured by nine statements pertaining to the position, use of respondent's ability, perceived confidence of management in the respondent'sjudgment, pay and promotion opportunities, and attitudes to supervisor and other workers. The statements were derived from a measure of job satisfaction and motivation used by Fiorgionne and Peeters zyxwvutsrqp zyxwv zyxwvutsr 340 Burtiout. Job Satisfaction and Job Performance zyx zyx zy ( 1982).Together they cover, in short form, all aspects of the Job Description Index (Smith, Kendall & Hullin, 1969). Such abbreviated forms - for example, Kunin’s (1955) “Faces” Scale of job satisfaction, have been found to correlate highly with the larger Job Description Index. Coefficient alpha for the job dissatisfaction scale was .72 in the present study. Poor work performance was measured from two sources: subjective opinion of the focal employee and the opinion of his or her supervisor. Both scales referred to amount of work done, accuracy and quality, use of information, initiative, time management, and interaction with staff. These aspects of performance included items from Scott and Stumpfs (1984) Rating of Workers, following theoretical guidelines as to the behavioural concomitants of burnout (Booth, 1984; Edelwich & Brodsky, 1980 Jones, 1982; Maslach, 1982). Pro and con items were balanced, and dichotomous answers (“generallytrue”or “generally false”)were used for most of the scale. ’The subjective performance rating consisted of six items, with an alpha coefficient of .59. The supervisor’s rating consisted of nine items, with an alpha coefficient of .77. The correlation over 262 employees (both samples combined) between supervisor’s rating and the employee’s own rating ofjob performance was .25 (p < .001). In addition, three specific aspects of the job were appraised by the employee: presrige (including recognition, desirability, and autonomy); stress (upsetting situations and clients who are difficult to deal with); and sociul supporl (team spirit, availability of support from family and other staff). Respondents replied “generally true” or “generally false” to I8 questions, which had been balanced for positive and negative content. The statements were prepared by the first author on the basis of reports from Jones (1982) and Maslach (1982). Alpha coefficients for the scales were: prestige (4 items), .49; stress (6 items), 0.59; and social support (8 items), .52. Mean job prestige scales from employees within each of 10 occupational groups of the Motor Registry were significantly correlated with mean prestige ranking of the lojobs provided by an independent sample of Motor Registry employees (r = .65, p < .Ol). Hence, job prestige perceived by incumbents was related to the consensually judged prestige as seen by outsiders. Incumbent-perceived level of stress in thejob was also related to independent rankings made by other workers of stress levels in the lojobs ( r = .57, p < .05). However, the mean views of incumbents as to levels of support on the job were not significantly related to the mean views of others in the Motor Registry, when asked which occupational groups were most likely to encourage support and teamwork. Hurdiness was a combined measure of three components: (internal) locus of control, (high) commitment to life, and (high) challenge (Kobasa, 1979). Locus oJcontrol was measured with a modified version of Levenson’s scale (Lefcourt, I98 I). This scale is similar to Rotter’s (1966) Internal-External Control Scale, but it is directed at the respondent’s behaviour rather than abstract beliefs. Modifications included reducing the number of items from 24 to 15, and making the response categories dichotomous (rather than zyxwv zy zyx Burnout. Job Satisfaction and Job Performance 341 zyxw zy zy &point scales), to reduce respondent bias toward extreme or neutral replies (Scott, 1968). Coefficient alpha for this scale was .54. Challenge was measured using a modified version of the Flexibility scale of Caplan et al. (1975) and inverse ratings of statements from the Security Scale of the California Life Goals Evaluation Schedules (Hahn, 1966).Coefficient alpha for this scale (I4 items) was .47. The Commilmenl to Life scale (9 items, a = .55) was constructed from the Alienation from Life scale (Maddi, Kobasa & Hoover, 1979). measuring the respondent’s sense of purpose in life, involvement with possessions, skills, goals and achievements. The srriver/achiever scale (5 items, a = .58) was adapted from the Tq’peA scale of Caplan et al. (1975) and Vickers ( 1976). This scale measures expressed involvement in work, persistence, achievement orientation, and competitiveness. Dichotomous response categories (“generally true” and “generally false”) were supplied. Procedure After obtaining management approval in both settings, the principal investigator explained the purpose of the study to employees seen individually or in small groups. Each person was assured of confidentiality, and of obtaining feedback on the general results. Participation was voluntary and respondents gave permission for their supervisor to rate their performance in confidence. Numbers, rather than names, were used on the supervisors’ questionnaire to enhance confidentiality. Questionnaires were collected personally by the principal investigator three or four days afterdistribution. Results All analyses were performed within each sample independently and results compared in two ways: by tests of the difference between corresponding correlation coefficients and by overall measures of similarity between two correlation matrices (which consisted of product-moment correlations computed over the corresponding pairs). In general, results for the two samples were very close, except as noted below. zyxw Intercorrelations Among Dependent Variables Table 2 shows intercorrelations among the three dependent variables, along with supervisor-judged job performance. In the interest of simplifying comparisons over external correlates, all dependent variables are named and scored in the “undesirable” direction. Burnout and job dissatisfaction were substantially correlated, but at lower levels than their respective scale reliabilities, suggesting that distinct, though related, traits were being measured. Both were associated with poor self-rated job performance. Supervisor-judged performance correlated significantly with self-rated performance and with job dissatisfaction among nurses. Comparing the six pairs of correlations between the two samples. the coefficient of similarity represented in Table 2 is 0.86, indicating very similar results, in spite of the fact that one group consisted mainly of professionals, the other mainly of 342 zyxwvutsrqpo zyx Burnout, Job Satisfaction and Job Performance service employees. The only significant difference between corresponding correlation coefficients was between job dissatisfaction and burnout, which was higher among Motor Registry employees than among nurses. zyxwvut zyxwvu zyx Table2 IntercorrelationsAmong Burnout, Job Dissatisfaction and Job Performance. I 2 3 4 I . Burnout (37)" (.72)" 2. .lob dissatisfaction Sample I .45* Sample 2 .65* 3. Poor performance (self-rated) ( .59)a .30* .19* Sample I .26* .37* Sample 2 4. Poor performance (.77)" (supervisor-rated) .07 .13* .25* Sample I Sample 2 .04 .08 .I4 NOIP.Diagonal values in brackets are scale reliabilities estimated, for the combined sample (N=356), from Cronbach's (1951)alpha. n for Sample 1 (nurses) = 248; for Sample 2 (Motor Registry employees), 108; supervisors' ratings were obtained on 163 and 99, respectively. * p < .05. Both samples combined. zyxwvutsr zyxwv Cornpararive Correlates of the Three Job Orientations Hypotheses 1 and 2 propose differences in the correlates of burnout, job dissatisfaction, and poor subjective performance, corresponding to differences in the intended meanings of the three constructs: dissatisfaction associated with judged inadequacies of the job, burnout reflecting perceived inadequacies of the self, as well as the job, and subjective incompetence associated with low status, inexperience, and inadequate job performance. Table 3 presents the relevant correlations for the two samples, organised by the three classes of predictor variables. Results are complicated by the fact that, contrary to expectation, high striving correlated negatively, rather than positively, with all three outcomes. Considering the signed values of correlations, the mean correlations of burnout with the four personality characteristics (excluding the hardiness composite in favour of its components) were .06 and .05 in the two samples. But since the direction of correlation is not specified in Hypothesis I , it seemed equally appropriate to compare the absolute values of correlations. The mean absolute correlations of burnout with the four personality characteristics were .I8 and .21 among nurses and Motor Registry employees, respectively, while the mean correlations of burnout with reported job characteristics were .27 and .33 for the two samples; mean absolute correlations of job dissatisfaction with the four personality zyxwv zyx zyxw zy zyxwvut zyx zyxwvutsrqponmlk Burnout, Job Satisfaction and Job Performance 343 Table 3 Correlates of Burnout, Job Dissatisfaction, and Subjective Performance Among Nurses (Sample I ) and Motor Registry Employees (Sample 2). Burnout I Poor per- Dissatisfaction 2 Characteristics of self (low) Hardiness .22* .23* (External) Locus of control" .02 .08 (low) Commitment to life .38* .41* (low) Challenge" .02 -.Ol (high) Striver/ Achiever -.24* -.30* Respondent-judged job characteristics (low) Prestige .25* .43* (high) Stress .28* .22* .35* (low) Social support .29* Incumbent's status Age (young) .25* .27* -.03 .08 Education (low) Level (low) .07 . I8* (little) l i m e in this position -.19* (little) Time in this kind of work .24*b -.02 (poor) Supervisorjudged performance .I0 .04 I forrnance 2 I 2 .02 .I I .27* .19* -.03 .07 . I8* .I7 .26* .23* -.a -.05 .36* -.02 -.04 -.07 -.21* -.27* -.24* .39* .26* .46* .54* .I0 .I0 .I0 .35* .07 .25* .24* .05 .19* .27* . I 3f .23* .oo . I8* -.07 .21* .08 -.I7 .27* .04 .I I -.07 .32* .24* .I7* .08 .25* .14* .18* -.05 .42* .34* .37* Nore. Sample I , n = 248; Sample 2, n = 108. * p < .05. ' This sub-scale was included in the Hardiness composite. differs from that for Sample 2 at p * Correlation for Sample I < .05. characteristics were . I I for nurses and . I3 for Motor Registry employees, while the correlations with job characteristics were .37 and .33 in the two samples. Thus, the job correlates of burnout and job dissatisfaction were similar, whereas the personality correlates of burnout tended to be higher than the personality correlates ofjob dissatisfaction ( t for combined samples 2.32; d= ' 7; p < .05). Hypothesis I , therefore, received some support. The main departure from expectations pertains to the hypothesised positive correlation between burnout and the striver/ achiever disposition, which actually turned out in the opposite direction. Among the three components of (low) hardiness, only (low) commitment to life was a significant correlate of burnout, while (low) sense of challenge and external locus of control were not. The overall similarity in pattern of the 13 external correlates (excluding the hardiness composite) of burnout and job dissatisfaction was .90(p < 344 zyxwvutsrqp zyxwvuts zyxwv zyxwvu zyxwv zyxwvu zyxw zyxwvu Burnnut. .lob Satisfaction and .luh Performance .001) among Motor Registry employees, and the only significant difference between them pertained to the correlation with the Commitment to Life scale, for which the correlation with burnout was higher than that with job dissatisfaction. Among nurses, however, the similarity in external correlates was only .66 0, < .05) and several significant differences appeared. Correlations with the hardiness composite and its component, commitment to life, the striver-achiever scale, and time in this kind of work were all higher for burnout, while correlations with job prestige and social support were higher for job dissatisfaction. Thus, burnout and job dissatisfaction could be fairly well distinguished among nurses, but not among Motor Registry employees, on the basis of their external correlates. (This difference may be partly attributable to the larger sample of nurses.) Hypothesis 2, that subjective incompetence reflects low status in the organisation more than does burnout or job dissatisfaction, received some support. The mean correlations of the six status variables with subjective incompetence were .25 among nurses and .I9 among Motor Registry employees, while the mean correlations of these job status variables with burnout were just .I3 and .Oh, and their mean correlations with job dissatisfaction were . I I and .03, respectively, in the two samples. (The significance of the differences was reflected in r-ratios for paired elements of 5.04 (p < .01) and 2.49 (p < .05) respectively.) Mean correlations of job characteristics with subjective incompetence werejust .09and .I8 forthe two samples - significantly lower, on average, than the mean correlations of job characteristics with burnout and dissatisfaction. Mean absolute correlations of personality characteristics with subjective incompetence were .21 and .I2 respectively; these did not differ significantly from the mean absolute characteristics. Correlations of burnout and job dissatisfaction with personaliry l h e similarity in overall pattern of external correlates between subjective incompetence and burnout was .35 (ns) in both samples; the similarity in external correlates of subjective incompetence and job dissatisfaction was -.24 and .37 (ns) in the two samples. Thus, subjective performance could be fairly well distinguished from both burnout and job dissatisfaction. The significant differences between correlates of subjective incompetence and burnout in both samples were that respondent-judged (1ow)job prestige and social support were more strongly associated with burnout, while (low) position level was more strongly associated with subjective incompetence. The significant differences between correlates of subjective incompetence and job satisfaction in both samples were that respondent-judged (low)job prestige and social support were more strongly associated with dissatisfaction, while (little) time in present position and (little) time in similar work were more strongly associated with subjective incompetence. Contparison qf the Two Samples The mean levels of burnout and subjective incompetence did not differ significantly between the two samples, and nurses were only slightly less zyxwv zyx zy Burnout. Joh Satisfaction and Joh Performance 345 satisfied with their jobs than were Motor Registry employees (Table I). Moreover, there were few significant differences between the samples in correlations with external variables (Table 3). Time in present position and time in similar job were more negatively related to burnout among nurses than among Motor Registry employees. Also, time in position was more strongly correlated with (good) subjective performance among nurses than among Motor Registry employees. zy zyxwvu Summary and Discussion An attempt was made to distinguish the concept of burnout from related concepts ofjob dissatisfaction and poor self-judged job performance, and to confirm the distinctions by a comparison of the major correlates of these appraisals in two samples: nurses and non-professional employees of a government agency. Generally, it was found that job dissatisfaction was associated mainly with employee-judged characteristics of the job. Poor self-rated performance was associated mainly with the incumbent's low status and, to some extent, with the personality characteristics of low hardiness and a negative orientation toward striving and achievement. Burnout was associated about equally with these same personality characteristics and with employee-judged characteristics of the job. Thus, there is some basis for distinguishing empirically among burnout, job dissatisfaction, and subjective incompetence. even though these attitudes are interrelated. The fact that similar results appeared for the two, rather different, samples indicates that burnout, as well as the other two responses to work, has similar meanings - that is, similar antecedents and consequences - among non-professional, as well as professional, employees of large organisations. Central to both burnout and poor subjective performance are one aspect of low hardiness - that is, a low "commitment to life" - and a low striver/ achiever orientation. The fact that it was low, rather than high, striving which was associated with burnout, as well as with subjective incompetence, suggests that the "Type A" personality syndrome does not lead to burnout, but rather to continuing job commitment. A longitudinal study would be required, however, to clarify the relationship; as both striving/ achieving orientation and burnout were assessed at the same time rather than with a considerable time lag between them, it cannot be known for certain whether the long-term effect of the striver/achiever orientation on job attitudes is positive or negative. The fact that, among nurses, burnout is associated with brief tenure in the present position and in similar work suggests that, in this occupation, the reaction is more likely to appear as insecurity of newcomers than as alienation among experienced employees. Burnout may not, therefore, be the most apt term for the state; perhaps it can be characterised, instead, as relatively undeveloped commitment to life and work. In the non-professional sample, however, burnout was more common among employees who, 346 zyxwvutsrqp zy zyxwvu zyxwvu zyxwvut Burnout, Job Satisfaction and Job Performance although relatively young, had spent considerable time in the present position, thereby warranting use of the term in its colloquial meaning. Thus, the very meaning of “burnout” may depend on the kind of employee and work setting considered. The need to develop more distinctive measures of burnout and job dissatisfaction is apparent. References Berkeley Planning Associates. ( 1977). Evaluarion o j child abuse and neglect demons~ration projecu, 1974-1977: Volume I X . Prvjecr managemenr and worker burnout. Washington, DC: US Department of Commerce. zyxwvuts Booth. R. (1984). Burnour among human service workers. Paper presented at the Annual General Meeting of the Canberra Marriage Counselling Service. Caplan, R. D., Cobb. S.,French, J . P. R.. Jr., Van Harrison, R. & Pinneau, S. R., Jr. (1975). Job drniands and worker health. Main e~ect.sanilocrupationa1d~ferencr.s. Washington, DC: US Department of Health, Education and Welfare. Carmichael, C. (1979, July 18). &>ston Globe. p. 43. J. H. Meyer, Burnout: Ilevelopmental inlluences. Canadian Counsellor, 16. 67-73. Carver. C. S. & Humphries, C. (1982). Social psychology of the Type A coronary-prone behavior pattern. In G.S. Sanders & J. Suls (Eds.), Socia/p.s.vchology ofhealrh and illness. Hillsdale, NJ: Lawrence Erlbaum Associates. zyxwv Cherniss, C. ( 1980). Profe.~.sionalhurnourin humanserviceorganizorions. New York: Praeger. Chronbach, L. J . (1951). Coefficient alphaand theinternalstimulusoftests. Psychomerrika.16, 297-334. Edelwich, J. E. & Brodsky, A. ( 19w)). Burnour. Stares qf di.si1lusivnmenr in rhe helping priJfe.s.sion.s.. New York: Human Sciences Press. Fiorgionne, G. A. & Peeters, V. E. ( 1982). Differences in job motivation and satisfaction among female and male managers. Human Relarions, 35. 10 I - I 18. Ford, D. L., Jr.. Murphy, C. J. & Edwards, K. L. (1953). Exploratory development and validation of a perceptual job burnout inventory: Comparison of corporate sector and human service professionals. Ps.vcho/ogicalReporrs, 52, 993- 1006. Hahn, M. E. (1966). Calfvrnia Lfe Goa1.s Evaluation Schedule. Palo Alto: Western Psychological Services. Jones, J. W. (Ed.). ( 1982). The burnour s,vndrome. Current resrorch, theory. inrerventiom. Park Ridge, IL: London House Press. Kobasa. S. C. (1979). Stressful life events. personality, and health: An inquiry into hardiness. Journal of Persona1it.y and Social Ps.vchologv. 37, I -I I . Kobasa, S. C., Maddi. S. R. & Kahn, S. (1982). Hardiness and health: A prospective study. Journal of Personaliry and Social Psychology, 42, 168-1 77. Kunin, T. ( 1955). The construction of a new type of attitude measure. Persvnnel P~~vrholog.~, 8. 65-77. Lefcourt, H. M. (Ed.). (1981). Research wiih thelocusof conrrolconstrurt:Vol. 1. Assesssmenr methoh. New York: Academic Press. Maddi. S. R.. Kobaw, S. C. & Hoover, M. (1979). An alienation test. Journalof Humanistic P~j’~hology. 19. 73-76. Maslach, C. (1982). Burnout: 7he cosr of c,aring.Englewood Cliffs. NJ: Prentice-Hall zyx zyx zyxwvutsr zyxwvutsr zyxwvutsrqpo zyxwvuts Burnout, Job Satisfaction and Job Performance 347 Maslach, C. &Jackson, S. E. (1981). The measurement of experienced burnout. Journalo/ Oo~o~upa~ioinal Behavior. 2. 99- 1 13. Maslach, C. & Jackson. S. E. (1982). Burnout in health professions: A social psychological analysis. In G . Sanders & J. Suls (Eds.), Siio~iulP.s,vchologyifhealth andillness. Hillsdale. NJ: Lawrence Erlbaum. Meier, S.T. (1983). Toward a theory of burnout. Human Relurivns, 36, 899-910. Peters. C. C. & Van Voorhis. W. R. ( 1940). Staristicalpr~~~eduresandtheir marhematicalhases. New York: McGraw-Hill. Riggar, T. F.. Harrington Godley. S.& Hafer, M. (1984). Burnout and job satisfaction in rehabilitation administrators and direct service providers. Rehahilitation Ciiun.~ellingBullerin. 27, 151-160. Rotter. J. B. (1966). Generalised expectancies for internal versus external control of reinforcement. Psycholvgical Monographs, 80 (609). Schwab, R. & Iwanicki, E. F. (1982). Perceived role conflict. role ambiguity, and teacher burnout. Wuo~arivnalAdministration Quarterly. 18. 60-74. Scott, W. A. (1968). Attitude measurement. In G . Lindsay& E.Aronson (Eds.), Handbook o/ social p.v.vcholvgy; Vol. 2. Reading, MA: Addison-Wesley, 204-273. Scott, W. A. & Stumpf, J. (1984). Personal satisfaction and role performance: Subjective and social aspects of adaptation. Jvurnal of Personality and Social Psycholvgy.47, 8 12-827. Smith. P. C.. Kendall, L. M.& Hullen. C. L. (1969). ThemeasurementoJ’satis/bcrionin work und rerirement. Chicago: Rand McNally. Vickers, R. (1976). Job stress. ego acriviry. andri.sk of coriinary heart disease. Unpublished doctoral dissertation, University of Michigan. Vroom, V. (1964). Work andmorivarion. New York: Wiley. Weinberg. S., Edwards, G. & Garove, W. E. (1983). Burnout among employees of state residential facilities serving developmentally disabled persons. Children and Youth Services Review. S. 239-253.