ASAC 2003Halifax, Nova ScotiaIain L. DenstenSchool of Economics and ManagementUniversity of NSW at ADFA, Canberra, AustraliaTHE BURNOUT PROCESS RE-EXAMINEDThis study examined the burnout process among 480 senior officers in an Australianlaw enforcement organization. Structural equation models (LISREL) were used toinvestigate three possible burnout processes and the interrelationships among themodified five-factors of the Maslach Burnout Inventory (Maslach and Jackson,1981). Results demonstrated that the Leiter and Maslach (1988) sequential modelapproach was superior to Golembiewski and Munzenrider (1988) phase and basemodel approaches. The study supported emotional exhaustion (psychological strain)as central to, or the trigger of, the other burnout factors. This study also providesadditional support for a five-factor burnout framework and advances the debateregarding the interrelationships among the burnout factors. Theoretical and practicalimplications are discussed.Burnout is generally viewed by researchers as a process and not a reaction to aspecific stressful event (Burke and Greenglass, 1991; Capel, 1991; Greenglass and Burke,1990; Lee and Ashforth, 1993; Leiter, 1990; Wade, Cooley et al., 1986). However, there isno consensus among researchers about the burnout process and how the syndrome proceedsfrom one factor to another (Cordes and Dougherty, 1993; Lee and Ashforth, 1993; Leiter,1993; Toppinen-Tanner, Kalimo et al., 2000), although most burnout models have the basicassumption that the process is sequential (Schaufeli and Enzmann, 1998). Lee and Ashforth(1990) and more recently, Hellesøy, Grønhaug, and Kvitastein (2000) have highlighted thedeficiency in understanding the burnout process and have called for more theoreticaldevelopment. This study aims to address this deficiency in the literature while furtheradvancing our limited understanding of burnout among senior managers.BackgroundThe concept of burnout was a product of intense investigations conducted in the late1970s and early 1980s, and was not defined as a separate entity until 1974 by Freudenberger.According to Farber (1983:14), “burnout is more often the result not of stress per se. . . but ofunmediated stress - of being stressed and having no ‘out’, no buffers, no support system.”Rarely, would a single disturbing action episode be sufficient to cause burnout (Burisch,1993). Maslach and Jackson (1981) pioneered the empirical study of burnout and establishedthe widely-cited definition for this debilitating syndrome. Their definition defines burnout interms of three dimensions of the syndrome, namely emotional exhaustion, depersonalizationof others, and perceived lack of personal accomplishment in working with others. Theimplications of burnout for organizations are well documented (see Burke and Richards,1993; Cordes and Dougherty, 1993). For example, Leiter and Maslach (1988) found that highlevels of burnout were related to diminished organizational commitment and were detrimentalto key aspects of the interpersonal environment within organizations. Burnout has beendescribed as “a sensitive indicator of organizational healthiness” (Cox and Leiter, 1992: 222).While burnout studies have focused on the human services professions, health workers andteachers have been the predominant occupational groups studied (Schaufeli and Enzmann,1998).Law enforcement has been recognized as one of the most stressful human serviceprofessions in modern society (Selye, 1979) and linked to burnout (Maslach and Jackson,20
1991) and individual psychological strain (Kaufmann and Beehr, 1989). Studies of burnoutamong law enforcement officers represents only two percent of 473 studies published injournal articles, books and 538 dissertations between 1978 and 1996 (Schaufeli and Enzmann,1998). However, several burnout studies of law enforcement organizations, have confirmedthe negative impact that job demands have on the home and family lives of officers as a resultof high burnout levels (e.g., Burke and Deszca, 1986). Law enforcement is associated withpsychological and emotional stressors that are often not counterbalanced by positive,emotionally pleasing interactions (Cannizzo and Liu, 1995). The majority of burnout studieson law enforcement have focused at the lowest organizational levels (i.e., street level), whichis consistent with most burnout studies conducted in various occupations. According to Leeand Ashforth (1993: 370), “very few studies have focused on managers, despite the apparentprevalence of burnout at the managerial level” (e.g., Harvey and Raider, 1984) and thedetrimental effects managers can have on the attitudes and behaviors of individuals they serveand lead (Golembiewski, Munzenrider et al., 1986; Gryskiewicz and Buttner, 1992;Kadushin, 1985; Seltzer and Numerof, 1988). Consequently, the current study aims toaddress this deficiency by investigating burnout among senior law enforcement officers byfocusing on the burnout process within this occupation.The Maslach Burnout Inventory (MBI, Maslach and Jackson, 1981) has been themost widely used instrument for investigating burnout and is recognized to be robust.However, the structural validity of the MBI is not beyond question (Schaufeli, Enzmann etal., 1993), even though many attempts have be made to clarify the number of factors (e.g.,Byrne, 1993; Schaufeli and Van Dierendonck, 1993). The original three factor structuredeveloped by Maslach and Jackson (1981) has dominated all investigations of the burnoutprocess which to date, have been unable to provide conclusive empirical evidence of theburnout process (Hellesøy et al., 2000). A recent study of the MBI factor structure identifiedfive burnout factors (Densten, 2001). This new factor structure incorporated keydevelopments in the conceptualization and measurement of burnout that have occurred sincethe conception of the MBI. Specifically, the expanded five-factor structure increased thecapacity of the MBI to measure more clearly the dimensions of emotional exhaustion andpersonal accomplishment. The emotional exhaustion dimensions relate to ‘feelings of beingemotionally extended and exhausted by one’s work’ (Maslach and Jackson, 1981: 100) andhave both a psychological and somatic (i.e., physical) aspect which are measured by twofactors, namely emotional exhaustion (psychological strain) and emotional exhaustion(somatic strain). According to Enzmann, Schaufeli, Janssen, & Rozeman (1998), being ableto distinguish clearly between psychological and physical forms of exhaustion has soundtheoretical support. Emotional exhaustion, as a single factor has a discriminant validity that isconsidered ‘relatively poor’ because it is strongly related to other concepts, such aspsychosomatic symptoms (Schaufeli and Enzmann, 1998). Consequently, these findingsprovide support for emotional exhaustion having two aspects.The personal accomplishment or professional efficacy dimensions relate to a lack offeelings regarding both job competence and successful achievement in one’s work and aremeasured by two factors (items were reverse scored), namely personal accomplishment (self)and personal accomplishment (others). Personal accomplishment (self) focuses on internallydriven feelings related to self-inefficacy or a lack of ability (Bandura, 1977), while personalaccomplishment (others) focuses on externally driven (e.g., views from others) feelingsrelated to learned helplessness (Seligman, 1975) and the expectation that successfulachievement is unlikely. The depersonalization, cynicism, or dehumanization dimensionrelates to an individual’s attempts to gain emotional distance (e.g., treating a client as anumber) as a means to cope with exhaustion, and is measured by a factor, namelydepersonalization. In summary, the emotional exhaustion factors represent forms of strainwhile the personal accomplishment factors and the depersonalization factor are copingbehaviors.Each MBI burnout factor has a different progression or mechanism but is linked by acommon association or root (Buunk and Schaufeli, 1993). This multidimensional view issupported by empirical evidence (Maslach, 1993). Several models have attempted to explain21
- Page 1 and 2: Volume 24, No. 5, 2003Edited by/Éd
- Page 4 and 5: Pursey HeugensGreg IrvingRoderick I
- Page 6 and 7: TABLE OF CONTENTS - TABLE DES MATI
- Page 8 and 9: study of power displays in mixed-ge
- Page 10 and 11: on expectations and behaviour (Berg
- Page 12 and 13: TaskThis study required the use of
- Page 14 and 15: Task-Gender (male-stereotyped or fe
- Page 16 and 17: It is important to understand the s
- Page 18 and 19: Crown, C.L. & Cummins, D.A. (1998).
- Page 20 and 21: Smith-Lovin, L., & Brody, C. (1989)
- Page 22 and 23: Table 2Summary of Descriptive Stati
- Page 24 and 25: Table 4Summary of Interaction of Ge
- Page 28 and 29: the interrelationships among the MB
- Page 30 and 31: studies (Friedman and Sarros, 1989;
- Page 32 and 33: This study demonstrated that emotio
- Page 34 and 35: Anxiety, Stress and Coping, in pres
- Page 36 and 37: Washington, DC: Taylor and Francis,
- Page 38 and 39: Table 2Overall Goodness-of-Fit Indi
- Page 40 and 41: ζ 3Depersonalizationη 3β 3,1 (+)
- Page 42 and 43: ASAC 2003Halifax, Nova ScotiaWendy
- Page 44 and 45: views independently, gave rise to S
- Page 46 and 47: track assistant, associate, and ful
- Page 48 and 49: ole conflict are positively related
- Page 50 and 51: Table 3. Regression of climate and
- Page 52 and 53: interests.We speculate the null eff
- Page 54 and 55: DOOP and short for DOOP scales,”
- Page 56 and 57: Pelled, L. H., “Demographic diver
- Page 58 and 59: esulting increase in discretion ove
- Page 60 and 61: al., 1995). However, the distinctio
- Page 62 and 63: ise to strong pressures for conform
- Page 64 and 65: presented with a complete list of a
- Page 66 and 67: Table 3QAP Regression Coefficients
- Page 68 and 69: ConclusionResearchers of social inf
- Page 70 and 71: suggestions for further research.
- Page 72 and 73: performing on the job.” Presumabl
- Page 74 and 75: positively related to performance i
- Page 76 and 77:
proposed hypotheses were supported.
- Page 78 and 79:
ConclusionThis research has made se
- Page 80 and 81:
Be more enthusiastic and exert extr
- Page 82 and 83:
Table 2Correlations Among the Laten
- Page 84 and 85:
performance expectations on feedbac
- Page 86 and 87:
changements qui s’opèrent dans l
- Page 88 and 89:
structures, politiques, systèmes,
- Page 90 and 91:
comportements liés à la communica
- Page 92 and 93:
Tableau 2Échelles de mesure de la
- Page 94 and 95:
prédiction des compétences clés
- Page 96 and 97:
organisation n’est pas uniforme e
- Page 98 and 99:
complètement ce construit. Ces cha
- Page 100 and 101:
Wagner, R.K. and Sternberg, R.J. (1
- Page 102 and 103:
What is a toxin handler?In two arti
- Page 104 and 105:
potentially limited scope and conte
- Page 106 and 107:
work role demands. In this survey,
- Page 108 and 109:
The three factors in this rotated f
- Page 110 and 111:
and service orientation, the abilit
- Page 112 and 113:
ASAC 2003Halifax, Nova ScotiaLisa M
- Page 114 and 115:
self-appraisal group reacted more n
- Page 116 and 117:
effect of voice are the value-expre
- Page 118 and 119:
“Strongly Disagree” to “Stron
- Page 120 and 121:
esults suggest that incorporating s
- Page 122 and 123:
Personality and Social Psychology,
- Page 124 and 125:
Table 1Means, Standard Deviations,
- Page 126 and 127:
Table 3Test of the Mediating Role o
- Page 128 and 129:
ASAC 2003Halifax, Nova ScotiaAnn Fr
- Page 130 and 131:
individual level, polychronicity is
- Page 132 and 133:
construct.Drawing on computer-media
- Page 134 and 135:
Consequence: Impact on Work Overloa
- Page 136 and 137:
Conversation complexity may also mo
- Page 138:
Implications for practiceFuture res
- Page 141 and 142:
ReferencesAncona, D.G., Goodman, P.
- Page 143 and 144:
no. 3 (1994): 381-391.Macan, T.H.,
- Page 145 and 146:
ASAC 2003Halifax, Nova ScotiaIan R.
- Page 147 and 148:
Assessing Measures: Affective Commi
- Page 149 and 150:
implications of psychological contr
- Page 151 and 152:
commitment, affective commitment, c
- Page 153 and 154:
Motivational Process Variables. Amo
- Page 155 and 156:
DiscussionThe main purpose of this
- Page 157 and 158:
approaches zero. In the present stu
- Page 159 and 160:
Extension and test of a three-compo
- Page 162 and 163:
Table 1Descriptive Statistics and Z
- Page 164 and 165:
Table 3Standardized Factor Loadings
- Page 166 and 167:
Table 5Hierarchical Regression Anal
- Page 168 and 169:
ASAC 2003Halifax, Nova ScotiaJoan F
- Page 170 and 171:
ASAC 2003Halifax, Nova ScotiaArla D
- Page 172 and 173:
ASAC 2003Halifax, Nova ScotiaIvy Ky
- Page 174:
ASAC 2003Halifax, Nova ScotiaNina D