Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Stress, attributions, and coping : predictors of emotional exhaustion in male post-secondary instructors Stephens, Eileen 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1992_fall_stephens_eileen.pdf [ 4.4MB ]
Metadata
JSON: 831-1.0054151.json
JSON-LD: 831-1.0054151-ld.json
RDF/XML (Pretty): 831-1.0054151-rdf.xml
RDF/JSON: 831-1.0054151-rdf.json
Turtle: 831-1.0054151-turtle.txt
N-Triples: 831-1.0054151-rdf-ntriples.txt
Original Record: 831-1.0054151-source.json
Full Text
831-1.0054151-fulltext.txt
Citation
831-1.0054151.ris

Full Text

STRESS, ATTRIBUTIONS, AND COPING: PREDICTORS OF EMOTIONAL EXHAUSTION IN MALE POST-SECONDARY INSTRUCTORS by EILEEN STEPHENS B.A.(HONS.), THE UNIVERSITY OF BRITISH COLUMBIA M.A., QUEEN'S UNIVERSITY A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES (Department of Counselling Psychology) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1992 © Eileen Stephens, 1992 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of C o u n s e l l i n g Psychology The University of British Columbia Vancouver, Canada Date October 14, 1992 DE-6 (2/88) Abstract This study used a learned-helplessness model to examine burnout. Two h i e r a r c h i c a l multiple regression analyses tested the extent to which job stress, a t t r i b u t i o n s t y l e , and coping strategies were associated with burnout ( i . e . , emotional exhaustion) i n male instructors (N = 108), aged 30 and 55 (M = 44.1), employed at a post-secondary i n s t i t u t i o n i n western Canada. I t was expected that those a t t r i b u t i o n s t y l e s and coping strategies associated with greater "personal" and "universal" helplessness would be associated with greater emotional exhaustion, whereas those associated with greater "morale" and "enthusiasm" would be associated with lower emotional exhaustion S p e c i f i c a l l y , greater characterological (internal-stable) and task d i f f i c u l t y (external-stable) a t t r i b u t i o n s were expected to be associated with greater emotional exhaustion under high-stres s conditions. Under low-stress conditions, greater behavioural (internal-unstable) a t t r i b u t i o n s were expected to be associated with greater emotional exhaustion. Greater emotional exhaustion was also expected to be associated with greater escape-cognitive and escape-active coping r a t i o s , but with lower contro l - c o g n i t i v e and control-active coping r a t i o s . Those a t t r i b u t i o n s t y l e s and coping strategies associated with personal helplessness, i . e , characterological ( i n t e r n a l -stable) a t t r i b u t i o n s and escape-cognitive coping, were expected to make stronger contributions to emotional exhaustion than were those associated with universal helplessness, i . e . , task-i i i d i f f i c u l t y (external-stable) a t t r i b u t i o n s and escape-active coping. In addition, control-cognitive coping was expected to be more negatively associated with emotional exhaustion than was cont r o l - a c t i v e coping. Results indicated that greater escape-active coping was associated with greater emotional exhaustion. The contribution of the universal-helplessness product-term (stress by external-stable attributions) approached s i g n i f i c a n c e . Under high-stress conditions, greater task d i f f i c u l t y (external-stable) a t t r i b u t i o n s were associated with greater emotional exhaustion. The enthusiasm product-term (stress by internal-unstable attributions) also contributed s i g n i f i c a n t l y to emotional exhaustion. Under low-stress conditions, greater behavioural (internal-unstable) a t t r i b u t i o n s were associated with greater emotional exhaustion. The negative association between c o n t r o l -cognitive coping and emotional exhaustion approached s i g n i f i c a n c e . Results suggested that burnout may involve universal rather than personal helplessness. This f i n d i n g was not anticipated. Results also supported the theory that internal-unstable (behavioural) a t t r i b u t i o n s may represent a v u l n e r a b i l i t y to burnout. Longitudinal studies would need to be conducted to t e s t any causal implications of the present findings. i v Table of Contents Page Abstract i i Table of Contents i v L i s t of Tables v i i L i s t of Figures v i i i Acknowledgement x Dedication x i Introduction 1 Review of L i t e r a t u r e 4 D e f i n i t i o n s of Burnout 4 Competing Explanations of Burnout 6 A t t r i b u t i o n s as Predictors of Burnout 7 A t t r i b u t i o n s i n Learned-Helplessness Theory 10 Support f o r the Locus-Attribution Hypothesis 12 Longitudinal research 12 Cross-sectional research 14 Theory 17 Summary 19 Support f o r the Control-Attribution Hypothesis 20 Longitudinal research 2 0 Cross-sectional research 21 Locus-of-control research 26 Summary 30 Implications for Emotional Exhaustion 31 Implications for Job S a t i s f a c t i o n 34 Addi t i o n a l Burnout/Learned-Helplessness Links 35 Burnout research 35 V Learned-helplessness research 36 Coping 37 Implications for Personal Accomplishment 41 Summary and Integration 41 Main expectations 42 Composite a t t r i b u t i o n variables 45 Product-terms 46 Corollary expectations 48 Longitudinal implications 51 Sample 52 Hypotheses 53 Main hypothesis 53 Corollary hypothesis 54 Exploratory Questions 55 Method 56 Respondents 56 Procedure 57 Instruments 57 C r i t e r i o n variables 57 Predictor variables 61 Demographic data 68 Data Analysis 68 Results 73 Descriptive S t a t i s t i c s and Preliminary Analyses 73 Pearson Product Moment Correlations between Variables . . 79 Regression Analyses 84 Exploratory Analyses 90 Post-Hoc Analyses 90 v i Discussion 93 Universal Helplessness 93 Personal Helplessness 96 Limitations 98 Suggestions for Future Studies 102 P r a c t i c a l Implications 104 References 107 Appendices 119 A. Informed consent attachments, reminders and i n s t i t u t e endorsement 119 B. Advertising 124 C. Instruments 125 D. A n c i l l i a r y r e s u l t s 151 v i i L i s t of Tables Page Table 1. Glossary of Composite Variables and Product-Terms 47 Table 2. Demographic Characteristics of Sample 73 Table 3. Means and Standard Deviations for Predictor and C r i t e r i o n Variables 76 Table 4. Correlations for Demographic and C r i t e r i o n Variables (N = 108) 80 Table 5. Correlations for Predictor and C r i t e r i o n Variables (N = 108) 81 Table 6. Multiple Regression of Emotional Exhaustion on "Helplessness-Orientation" Predictors (N = 108) 85 Table 7. Multiple Regression of Emotional Exhaustion on "Helplessness-Risk" Predictors (N = 108) 89 Appendix D. Table 1. Correlations for Teacher-Stress Factors and C r i t e r i o n Variables (N = 108) 152 Table 2. Correlations for Stress, Locus, S t a b i l i t y , and C r i t e r i o n Variables (N = 108) 153 Table 3. Correlations for Weighted A t t r i b u t i o n s and C r i t e r i o n Variables (N = 108) 154 Table 4. Multiple Regression of Low Personal Accomplishment on "Helplessness-Orientation" Predictors (N = 108) 156 Table 5. Multiple Regression of Low Personal Accomplishment on "Helplessness-Risk" Predictors (N = 108) 158 Table 6. Multiple Regression of Job ( D i s ) s a t i s f a c t i o n on "Helplessness-Orientation" Predictors (N = 108) 160 Table 7. Multiple Regression of Job ( D i s ) s a t i s f a c t i o n on "Helplessness-Risk" Predictors (N = 108) 161 v i i i Table 8. Correlations for A t t r i b u t i o n s , Coping Factors, and C r i t e r i o n Variables (N = 108) 165 Table 9. Correlations for Predictor and C r i t e r i o n Variables (n =16) 166 ix L i s t of Figures Page Figure 1. Individual Attitudes and Perceptions i n the Burnout Process 8 Figure 2. Model Integrating A t t r i b u t i o n s and Coping . . . . 44 Figure 3. Regression of Emotional Exhaustions on External-Stable Attributions within High and Low-Stress Subgroups 87 Figure 4. Regression of Emotional Exhaustion on Internal-Unstable Attributions within High and Low-Stress Subgroups 91 X Acknowledgement I wish to thank my supervisor, Dr. Bonnie Long, f o r her long-standing commitment to t h i s study. Dr. Long contributed to t h i s project both her expertise and her i n t e r e s t i n the areas of teacher stress, coping, and job s a t i s f a c t i o n . I thank too my committee members, Drs. Richard Young and Todd Rogers, and also Dr. Walter Boldt, who assisted with the analysis. I am g r a t e f u l to my colleagues who f i l l e d out a lengthy questionnaire at what was a d i f f i c u l t time of the year f o r them and to the management and unions, who supported my research. I wish to thank Dr. Chris Peterson for permission to use The Expanded ASQ, Dr. Janina Latack for permission to use the Latack coping scale, and The Consulting Psychologists Press, Inc. f o r permission to use the Maslach Burnout Inventory. Dedication x i E. George Stephens (1920-1991) This t h e s i s i s dedicated to my father who, i n facing a diagnosis of cancer, demonstrated the wisdom of h i s optimism. 1 Introduction Burnout i s believed to threaten the professional wellbeing of teachers, s o c i a l services workers, nurses, and other health-care professionals, along with the welfare of the students, c l i e n t s , and patients they serve. Since the term "burnout" was f i r s t coined (Freudenberger, 1974), the burnout syndrome has increasingly come under scrutiny of researchers ( K a h i l i , 1988). Yet despite two decades of research, we know l i t t l e more today about how to remedy the syndrome than we did i n the early 1970s. The f a i l u r e of the burnout research to provide remedies for the burnout syndrome indicates a underlying problem i n the l i t e r a t u r e : burnout lacks a clear d e f i n i t i o n and a sound t h e o r e t i c a l model by which research may be guided (Savicki & Cooley, 1983). The burnout l i t e r a t u r e currently o f f e r s no consensus on the etiology of the burnout syndrome. Although an accepted etiology would not necessarily give r i s e to accepted remedies (Savicki & Cooley, 1983), nevertheless an established model of burnout would encourage the r a t i o n a l i z a t i o n of contradictory findings and help focus the search for remedies. Presently, the best operational d e f i n i t i o n of burnout i s the Maslach Burnout Inventory (MBI, Maslach & Jackson, 1981). Maslach and Jackson validated the MBI by comparing scores on the MBI with those on the Job D i s s a t i s f a c t i o n Scale (Hackman & Oldham, 1974) for s o c i a l service and mental health-care workers (N = 91). Because job d i s s a t i s f a c t i o n accounted f o r less than 6% 2 of variance i n any of the MBI subscales, Maslach and Jackson concluded that burnout i s d i s t i n c t from job d i s s a t i s f a c t i o n . A burnout model put forward by Savicki and Cooley (1983) suggests that burnout may bear more s t r i k i n g s i m i l a r i t i e s to learned-helplessness. Burnout and learned-helplessness are both believed to involve dysfunctional responses to f r u s t r a t i o n and passive coping strategies, as well as long-term negative emotional consequences. However, the learned-helplessness l i t e r a t u r e , unlike the burnout l i t e r a t u r e , boasts a w e l l -researched t h e o r e t i c a l model and tested remediation strategies. Were borrowing from the learned-helplessness model j u s t i f i e d , our understanding of the burnout syndrome might thereby be s i g n i f i c a n t l y increased. Learned-helplessness studies have generally taken depression as t h e i r c r i t e r i o n variable (Sweeney, Anderson, & Bailey, 1986). In the face of non-contingency events, depression has been associated with and predicted by a "pessimistic" ( i n t e r n a l -stable-global) a t t r i b u t i o n s t y l e (Peterson & Seligman, 1984). To t e s t whether the learned-helplessness model helps explain burnout, two h i e r a r c h i c a l MR regression analyses were used to examine emotional exhaustion. Teacher stress was entered f i r s t as an in d i c a t o r of non-contingency events; a t t r i b u t i o n s t y l e s and coping strategies consistent with learned-helplessness were tested f o r t h e i r additional contributions to burnout. Those a t t r i b u t i o n s t y l e s and coping strategies consistent with personal and universal helplessness were expected to be associated with greater emotional exhaustion under high-stress conditions. In addition, those consistent with greater e f f o r t were expected to 3 be associated with greater emotional exhaustion under low-stress conditions. The responses of male post-secondary i n s t r u c t o r s were assessed using a cross-sectional survey design. Although post-secondary i n s t r u c t o r s are believed to be highly-stressed, they are under-represented i n the l i t e r a t u r e ( S e i l e r & Pearson, 1984-1985). Literature Review 4 In the two decades since the term burnout was f i r s t coined, researchers have, with varying degrees of success, attempted to define burnout and to i d e n t i f y i t s causes, stages, and appropriate remediation strategies. O r i g i n a l l y , the term burnout meant "to f a i l , wear out, or become exhausted by making excessive demands on energy, strength, or resources" (Freudenberger, 1974, p. 159). Freudenberger noted that when highly-motivated s o c i a l services workers were placed i n f r u s t r a t i n g work s i t u a t i o n s , they began a long and c o s t l y psychological decline. This decline proceeded from enthusiasm to f r u s t r a t i o n , stagnation, and apathy (Edelwich & Brodsky, 1980). Once burned out, workers "often f a i l [ e d ] to see t h e i r s i t u a t i o n as stemming from inside themselves" (Freudenberger, 1977, p. 26). They thus became problems not only to themselves, but often to t h e i r employers, co-workers, and c l i e n t s as well. D e f i n i t i o n s of Burnout Most empirical research has operationally defined burnout as a c l u s t e r of symptoms a r i s i n g i n response to s t r e s s . Teacher burnout, for example, has been defined as a syndrome caused by prolonged teacher stress and characterized by p h y s i c a l , emotional, and a t t i t u d i n a l exhaustion (Cunningham, 1983). Symptoms included i n the burnout c l u s t e r have been diverse. E i n s i e d e l and T u l l y (1981) l i s t e d 84 burnout symptoms (with some overlapping), and C a r r o l l and White (1982) l i s t e d 47 separate symptoms. 5 The most widely-used measure of burnout, the Maslach Burnout Inventory (Maslach & Jackson, 1981), i d e n t i f i e s three symptom fa c t o r s : emotional exhaustion, low personal accomplishment, and depersonalization. I t measures these symptoms i n terms of t h e i r frequency and i n t e n s i t y . As functional d e f i n i t i o n of burnout, the MBI does l i t t l e , however, to c l a r i f y the process of the syndrome's development (Savicki & Cooley, 1983). I t leaves researchers open to the charge of defining burnout simply as that which burnout instruments assess (Herr & Cramer, 1984) and perpetuates the r i s k of spurious findings. C r i t e r i o n measures of emotional stress, such as the MBI, are l i k e l y to co r r e l a t e highly with other instruments that are also l a r g e l y c h e c k l i s t s of negative emotions ( K a h i l i , 1988). Lee and Ashforth (in press) have preferred to view burnout as a sequence, rather than as a c l u s t e r of symptoms. They recently tested two models of burnout (Golembiewski, Munzenrider, & Stevenson, 1986; L e i t e r & Maslach, 1988) i n an attempt to i d e n t i f y a sequential ordering of MBI factors. Although t h e i r r e s u l t s were inconclusive, a post-hoc analysis allowed them to re v i s e the L e i t e r and Maslach model and to suggest that, among l i n e supervisors and managers i n the human service sector (N = 233), emotional exhaustion leads d i r e c t l y to both depersonalization and diminished personal accomplishment. Lee and Ashforth underscored the c e n t r a l i t y of emotional exhaustion i n the burnout process (cf. Gaines & Jermier, 1983; Maslach, 1982; Shirom, 1989), noting that personal accomplishment may be only weakly re l a t e d to emotional exhaustion (cf. Maslach & Jackson, 1981). 6 Competing Explanations of Burnout In discussing the burnout process, most researchers have focused on the possible causes of the syndrome. With respect to these causes, they have disagreed about the r e l a t i v e importance of the i n d i v i d u a l and the s i t u a t i o n . Freudenberger (1974, 1980) i d e n t i f i e d both i n d i v i d u a l and s i t u a t i o n a l causes of burnout, whereas Maslach (1977) argued that burnout originates mainly i n the s i t u a t i o n . Consistent with Maslach*s analysis, Pines and her colleagues (Pines, Aranson, & Kafrey, 1981) distinguished burnout from depression, on the premise that burnout i s more s i t u a t i o n a l and depression more personal i n etiology. They cautioned against i d e n t i f y i n g personal causes for burnout for fear that blaming a burnout v i c t i m might tr i g g e r depression i n that i n d i v i d u a l . As yet, issues concerning the etiology of burnout have not been adequately tested; most empirical studies of burnout have been cross-sectional i n design. Reviewing the f i r s t decade of burnout research, Perlman and Hartman (1982) noted that, although the majority of a r t i c l e s i d e n t i f i e d both i n d i v i d u a l and organizational causes and f o c i of solutions for the syndrome, those c i t i n g only one cause or solution favoured the organization over the i n d i v i d u a l by a r a t i o of almost 15:3. Among organizational causes of burnout, r o l e ambiguity and r o l e c o n f l i c t (Crane & Iwanicki, 1986), task overload (Berkley Planning Associates, 1977; Pines & Kafry, 1978; Pines & Maslach, 1978), poor s o c i a l support (Berkley Planning Associates, 1977; Pines & Kafry, 1978), and low influence over p o l i c y decisions (Barad, 1979; Maslach & Pines, 1977) have a l l been c i t e d . 7 Popular though s i t u a t i o n a l explanations of burnout have been, they nevertheless beg a compelling question posed by Savicki and Cooley (1983): "Why i s i t that some in d i v i d u a l s i n some work settings f l o u r i s h , and others become exhausted" (p. 235)? Though a variety of personal variables have been associated with burnout, no clear answer to t h i s question has emerged. The association of burnout with seemingly inconsistent personal variables, such as a type A personality (Nagy & Davis, 1985) and an external locus of control (Holt, Fine, & Tollefson, 1987; Mclntyre, 1984) suggests, however, that burnout involves a process of personal change. Savicki and Cooley (1983) looked f o r an answer to t h e i r question i n the very process of a t t r i b u t i n g causes. I f blaming the burnout vi c t i m p o t e n t i a l l y aggravates the syndrome, could i t be that self-blame i s i t s e l f a personal variable contributing to burnout? Self-blame and other a t t r i b u t i o n s of r e s p o n s i b i l i t y involve both the person and the s i t u a t i o n . As s i t u a t i o n a l l y s e n s i t i v e c a t a l y s t s of personal change, a t t r i b u t i o n s have been implicated by Savicki and Cooley i n the burnout process. A t t r i b u t i o n s as Predictors of Burnout Savicki and Cooley (1983) suggested that two of the MBI subscales, depersonalization and personal accomplishment, are confounded with a t t r i b u t i o n s . They placed depersonalization at the high-burnout end of t h e i r " i d e n t i f i c a t i o n with the c l i e n t " (1983, p. 232) a t t r i b u t i o n dimension and low personal accomplishment at the high-burnout end of t h e i r "therapeutic locus of control and personal accomplishment" (1983, p. 232) a t t r i b u t i o n dimension (see Figure 1). Together, these two 8 9 dimensions were taken to r e f l e c t the degree of personal r e s p o n s i b i l i t y a therapist assumes for outcomes (Cooley & Savicki, 1987). According to Savicki and Cooley (1983), extreme a t t r i b u t i o n s for involvement ( i d e n t i f i c a t i o n with c l i e n t s ) and control (personal impact on outcomes) are both undesirable. Such extremes indicate either elements of burnout or a predisposition to the syndrome. Under s t r e s s f u l conditions, a t t r i b u t i o n s r e f l e c t i n g over-involvement may predispose i n d i v i d u a l s to depersonalization; those r e f l e c t i n g over-control may predispose them to low personal accomplishment. Savicki and Cooley (1983) explained the mechanism by which a t t r i b u t i o n s may lead to burnout. Over-involved workers l i k e l y confuse boundaries between themselves and others, a t t r i b u t i n g the causes of change i n t h e i r c l i e n t s to themselves i n such a way that "the c l i e n t ' s problems and successes become the therapist's problems and successes" (1982, p. 417). Although a therapist's over-involvement may seem a l t r u i s t i c i n i t i a l l y , as burnout progresses i t leads to defensive depersonalization of c l i e n t s , i . e . , to a perception that the c l i e n t i s an adversary or merely an object i n the therapeutic process. E n t h u s i a s t i c a l l y over-controlling i n d i v i d u a l s , f o r t h e i r part, i n i t i a l l y aspire to "save the world" (1983, p. 233). As they burnout, however, over-controlling i n d i v i d u a l s come to see themselves as having l i t t l e impact on outcomes. They thus experience the external locus of control that Savicki and Cooley associate with low personal accomplishment (Savicki & Cooley, 1982). 10 According to Savicki and Cooley (1983) , a moderate degree of involvement and control i s functional. Unless moderation i s maintained, stress i n the form of negative events t r i g g e r s a s h i f t between the extremes on the involvement and control dimensions. A predisposition to burnout then develops into the full-blown syndrome. The more extreme the a t t r i b u t i o n s , the more i n e f f e c t i v e the therapist w i l l be, and the more l i k e l y stress i n the form of negative events i s to preciptate a dysfunctional a t t r i b u t i o n a l s h i f t (Brickman et a l . , 1982; Savicki & Cooley, 1982). A t t r i b u t i o n s i n Learned-Helplessness Theory By implicating a t t r i b u t i o n s i n the burnout process, Savicki and Cooley (1983) created hypotheses s i m i l a r to those found i n the learned-helplessness l i t e r a t u r e (Peterson & Seligman, 1984). According to the reformulated learned-helplessness theory, characterological a t t r i b u t i o n s for uncontrollable negative events i . e . , a t t r i b u t i o n s that are i n t e r n a l , stable, and global, both co r r e l a t e with and predict learned-helplessness depression when negative events occur. A t t r i b u t i n g uncontrollable negative events to i n t e r n a l , stable, and global causes involves blaming one's character for f a i l u r e s . To make such a t t r i b u t i o n s f o r negative events, individuals must characterize causes of f a i l u r e as r e s i d i n g i n themselves (an involvement issue), as being constant ( l i k e l y a control issue), and as generalizing across s i t u a t i o n s , i . e . , " i t ' s me; i t ' s going to l a s t forever; and i t ' s going to a f f e c t everything I do" (Peterson & Seligman, 1984, p. 350) . 11 Although considered t r a i t l i k e , a t t r i b u t i o n s are not considered invariant (Peterson & Seligman, 1984). Peterson and hi s colleagues (1981) explained the possible etiology of characterological a t t r i b u t i o n s . Such a t t r i b u t i o n s may a r i s e as a l o g i c a l response to persistent events. I n i t i a l l y , i n d i v i d u a l s who make i n t e r n a l a t t r i b u t i o n s w i l l l i k e l y a t t r i b u t e the causes of events to t h e i r behaviour, an i n t e r n a l , unstable, s p e c i f i c a t t r i b u t i o n . But as the number of s i m i l a r events increases, they w i l l come to adopt characterological, i . e . , i n t e r n a l , stable, g l o b a l . a t t r i b u t i o n s , because the constant factor i n a se r i e s of s i m i l a r events, s i m i l a r over time or over s i t u a t i o n s , i s one's character. Among female undergraduates (N = 87), characterological explanations for negative events increased with the number of negative events reported f o r the previous year. In the face of negative events, a t t r i b u t i o n s seem to become more i n t e r n a l and stable. In any a t t r i b u t i o n a l s h i f t from behavioural to characterological a t t r i b u t i o n s , responses on the inter n a l / e x t e r n a l or locus a t t r i b u t i o n dimension remain constant. Whether the c r i t i c a l s h i f t occurs on the s t a b i l i t y or g l o b a l i t y dimension i s unclear (Peterson, Schwartz, & Seliman, 1981). In eit h e r case, those who blame t h e i r characters f o r negative events, making internal-stable-global a t t r i b u t i o n s , have been found, a f t e r a series of negative events, to cease t r y i n g to avoid such events and to secumb to the cognitive, motivational, and emotional d e f i c t s of learned-helplessness depression (Sweeney et a l . , 1986). 12 Those making external-stable-global a t t r i b u t i o n s f o r negative events are also expected to s u f f e r some helplessness d e f i c i t s under the stress of non-contingent events (Abramson, Seligman, & Teasdale, 1978). Rather than b e l i e v i n g themselves to be personally involved i n the causes of negative events, however, external a t t r i b u t o r s believe instead that relevant others would also f a i l under s i m i l a r circumstances. They thus secumb to "universal" rather than "personal" helplessness, experiencing cognitive and motivational d e f i c i t s , but not the self-esteem d e f i c i t s of the learned-helplessness depressed. Styles of a t t r i b u t i n g negative events to one's behaviour (an internal-unstable attribution) or to external causes have been found to be inconsistent with personal helplessness or learned-helplessness depression (Peterson et a l . , 1981). Those who make in t e r n a l a t t r i b u t i o n s for f a i l u r e may defend against depression i n the face of uncontrollable negative events, i t seems, either by maintaining behavioural a t t r i b u t i o n s i n the face of stress or by adopting more external ones. Support for the Locus-Attribution Hypothesis Savicki and Cooley's (1983) burnout hypotheses, l i k e those of learned-helplessness theory, indicate a defensive s h i f t from in t e r n a l to external a t t r i b u t i o n s i n the face of s t r e s s . According to Savicki and Cooey's locus hypothesis, over-involved indiv i d u a l s are predisposed to depersonalize, i . e . , to become less personally involved i n a c l i e n t ' s change, as they encounter negative events and begin to burn out. Longitudinal research. Wade, Cooley, and Savicki*s (1986) findings are consistent with the claim that over-involved 13 i n d i v i d u a l s are vulnerable to burnout. In a l o n g i t u d i n a l study, Wade and h i s colleagues tested the r e l a t i o n s h i p between burnout and personal r e s p o n s i b i l i t y , as measured by the A t t r i b u t i o n s of Treatment Outcome measure (ATO, Cooley & Savicki, 1987). Personal r e s p o n s i b i l i t y focuses l a r g e l y on blaming oneself for negative outcomes. According to Cooley and Savicki, i t r e f l e c t s both involvement and control. Over the course of a year, s o c i a l -services workers (N = 46) who were most prone both to accept personal r e s p o n s i b i l i t y for the outcomes of t h e i r c l i e n t s and to blame themselves when t h e i r c l i e n t s did not improve eithe r increased t h e i r l e v e l s of burnout on two or more of the s i x MBI subscales or remained high on two or more of the subscales (p_ < .05) (Wade et a l . , 1986). As r e f l e c t e d i n the personal r e s p o n s i b i l i t y factor, involvement seems to have predicted burnout over the course of a year. No s p e c i f i c reference was made i n t h i s study to associations between personal r e s p o n s i b i l i t y and depersonalization; nor were associations between personal r e s p o n s i b i l i t y and f i n a l burnout scores reported. Wade and h i s colleagues' (1986) findings were l i m i t e d by the nature of the ATO (see Appendix C). Although the personal r e s p o n s i b i l i t y factor was taken to r e f l e c t i n t e r n a l a t t r i b u t i o n s , i n t e r n a l and external attributions on the ATO f a i l e d to covary i n any consistent way among helping professionals (N = 94). The external sources of change factor, designed to r e f l e c t external a t t r i b u t i o n s for p o s i t i v e and negative events, was l a r g e l y unrelated, f o r example, to the personal impact fac t o r , a factor designed to r e f l e c t i n t e r n a l attributions for p o s i t i v e and 14 negative events. Although the personal r e s p o n s i b i l i t y did cor r e l a t e moderately with the personal impact factor, r = .39, p_ < .05, the mean i n t e r - f a c t o r c o r r e l a t i o n among a l l the ATO factors was a nonsignificant .22. Cooley and Savicki (1987) accounted for the lack of covariance among ATO factors by concluding that helpers evaluate i n t e r n a l and external factors for producing change i n c l i e n t s independently. However, the lack of covariance may instead be due to the mixing of att r i b u t i o n s for p o s i t i v e and negative events on the ATO. Attributions for po s i t i v e and negative are not expected to covary i n unison (Peterson & Seligman, 1984). In the i n t e r e s t of gaining stronger r e s u l t s , measures of a t t r i b u t i o n generally separate attributions for good and bad events more c l e a r l y than do either the ATO or locus of control measures. On the ATO, the personal r e s p o n s i b i l i t y factor favours negative events; whereas three items on t h i s factor r e f e r to negative events, only one re f e r s to po s i t i v e events (see Appendix C). Despite the l i m i t a t i o n s of the ATO, Wade and h i s colleagues (Wade et a l . , 1986) nevertheless found support, using the ATO, for t h e i r hypothesis that high involvement predisposes respondents to burnout. Cross-sectional research. The related hypothesis that high involvement s h i f t s to low involvement i n the burnout process cannot be tested adequately by means of a cros s - s e c t i o n a l design. Nevertheless, cross-sectional research can indicate the degree to which burnout i s associated with i n t e r n a l and external a t t r i b u t i o n s . Savicki and Cooley (1983) expected that burnout, and p a r t i c u l a r l y depersonalization, would be negatively 15 associated with involvement. This expectation, however, was not confirmed. Among helping professionals (N = 94), the propensity to take personal r e s p o n s i b i l i t y i n the face of f a i l u r e was p o s i t i v e l y associated with emotional exhaustion, r = .28, p < .003, frequency, and r = .18, p < .046, i n t e n s i t y ; i t f a i l e d to cor r e l a t e s i g n i f i c a n t l y with depersonalization (Cooley & Savicki, 1987). Quigley and her colleagues (Quigley, Slack, & Smith, 1987) s i m i l a r l y found high involvement i n work to be p o s i t i v e l y associated with emotional exhaustion. Structured interviews with secondary-school coaches (N = 75) showed that excessive involvement was a major source of burnout. Here, involvement suggested both time commitment and work overload; i t was not s p e c i f i c a l l y r e l a t e d to a p a r t i c u l a r a t t r i b u t i o n s t y l e . Nevertheless, Quigley and her colleagues drew a connection between p r a c t i c a l and emotional over-involvement when they suggested that the greater emotional exhaustion among single coaches as compared to married coaches (single M = 33.2; married M = 28.5) might be due to the tendency of s i n g l e coaches to over-i d e n t i f y with t h e i r work. The highly-involved teacher coaches i n Quigley and her colleagues' study experienced more emotional exhaustion and greater personal accomplishment, but le s s depersonalization, than did other occupational groups. Maslach and Jackson (1981) also reported a p o s i t i v e association between involvement with emotional exhaustion. They found that the MBI's optional involvement factor correlated p o s i t i v e l y with emotional exhaustion, frequency, r = .40, p_ < .001, and i n t e n s i t y , r = .44, p < .001, among human services 16 professionals (N = 1025)• I t also correlated p o s i t i v e l y with depersonalization, r = .09, p < .05, frequency, and r = .17, g <.001, i n t e n s i t y , and with personal accomplishment, r = .14, p_ > .05, frequency, and r = .21, p < .05, i n t e n s i t y . The ATO 1s external source of change factor, on the other hand, correlated negatively with depersonalization, r = -.18, p < .046, and non s i g n i f i c a n t l y with emotional exhaustion among helping professionals (N = 94). This finding could be taken to suggest that external attributions for negative events are unrelated to emotional exhaustion and negatively associated with depersonalization. However, the nature of the external a t t r i b u t i o n s tapped must be taken into account. Unlike the ASQ (Peterson et a l . , 1982) frequently used i n learned-helplessness research, the external source of change factor mixes p o s i t i v e and negative events (see Appendix C). Thus external a t t r i b u t i o n s as measures by the ATO may involve empowering rather than to depersonalizing others. Using the ASQ (Peterson et a l . , 1982) to measure of pessimism, McMullen and Krantz (1988) tested the association between burnout and at t r i b u t i o n s . Their pessimism v a r i a b l e measures the difference between composite i n t e r n a l - s t a b l e - g l o b a l a t t r i b u t i o n s for negative events and composite external-unstable-s p e c i f i c a t t r i b u t i o n s for p o s i t i v e events. Among female day-care workers (N = 67), pessimism correlated p o s i t i v e l y with both emotional exhaustion, r = .23, p < .05, and depersonalization, r = .24, p < .05. Among women, both emotional exhaustion and depersonalization seem to be associated with i n t e r n a l a t t r i b u t i o n s i n the face of negative event. 17 Again, however, r e s u l t s must be interpreted cautiously. Because McMullen and Krantz's (1988) pessimism v a r i a b l e includes two dimensions besides the internal/external (locus) a t t r i b u t i o n dimension, i t r e f l e c t s not only i n t e r n a l , but also stable and global a t t r i b u t i o n s for negative events, and not only external, but also unstable and s p e c i f i c a t t r i b u t i o n s for p o s i t i v e events. A summed score across three a t t r i b u t i o n dimensions confounds r e s u l t s with respect to any one dimension's c o r r e l a t i o n with e i t h e r emotional exhaustion or depersonalization. Nevertheless, the sort of pessimism variable used by McMullen and Krantz i s generally taken i n learned-helplessness research to r e f l e c t i n t e r n a l a t t r i b u t i o n s for negative events (Sweeney et a l . , 1986). Theory. Although not supported i n burnout research, Savicki and Cooley 1s (1983) claim that depersonalization represents a defensive strategy for dealing with d i s t r e s s i n g self-blame finds support i n burnout theory ( K a h i l i , 1988; Lee & Ashforth, i n press). Lee and Ashforth suggested that depersonalization may be a strategy f o r coping with emotional exhaustion. Ogus and her colleagues (Ogus, Greenglass, & Burke, 1990) suggested that t h i s strategy may be p a r t i c u l a r l y s a l i e n t among men. Relative to women, men have tended report less burnout (Etzion & Pines, 1986), less emotional exhaustion (Maslach & Jackson, 1981), and less depression (Greenglass & Burke, 1988), as well as less involvement, and greater personal accomplishment (Maslach & Jackson, 1981). They have also tended, however, to report higher depersonalization (Anderson & Iwanicki, 1984; Greenglass & Burke, 1988; Maslach & Jackson, 1981; Schwab & Iwanicki, 1982). Although men seem to s u f f e r fewer emotional-18 exhaustion and personal-accomplishment d e f i c i t s than do women, they nevertheless seem to secumb to greater depersonalization. As a means of coping with emotional exhaustion, depersonalization may have d i f f e r e n t implications for men and for women. In a study of teachers (N = 470), men who depersonalized tended to maintain internal-control coping and to cope by attempting to maintain a sense of meaning more than d i d t h e i r female counterparts (Ogus et a l . , 1990). Nevertheless, they showed s i g n i f i c a n t l y greater general d e f i c i t s , reporting poorer o v e r a l l l i v e s t y l e s and greater dependency on medication. In response to stress, men may be both more r e s i s t a n t and more reactive to the d e b i l i t a t i o n of personal helplessness than are women. Unlike g i r l s , boys seem to be conditioned by t h e i r teachers to believe t h e i r behaviour can change outcomes (Dweck & Li c h t , 1980). Teachers tend to c r i t i c i z e boys with behavioural a t t r i b u t i o n s and g i r l s with the more damaging characterological ones. When presented with unsolvable problems, fourth-grade boys gave explanations for f a i l u r e that were more external, unstable, and s p e c i f i c than were those given by g i r l s . Their responses, "I wasn't t r y i n g hard", "I wasn't paying attention", and "I don't care about your t e s t " (Peterson & Seligman, 1984, p.370) seem to r e f l e c t not only behavioural a t t r i b u t i o n s , but also a defensive loss of t r u s t or concern. In the context of locus-of-control studies, males are suspected of reacting defensively to f a i l u r e by using blame-pr o j e c t i n g strategies. Rotter (1975) i d e n t i f i e d men who were highly involved i n c o n t r o l l i n g outcomes, yet reported an external locus of contro l . He suggested that such men were "defensive" 19 externals, i . e . , achievement-oriented in t e r n a l s who adopt external-control attitudes " f a l s e l y " , using blame-projection to save face when f a i l u r e occurs or seems imminent. Hochreich (1974) suggested that such male "defensive externals" might be distinguished from true externals by means t h e i r low-trust scores. Low-trust externals, she suggested, would be more defensive and more "extremely external" or blaming than more congruent externals. Like low t r u s t , depersonalization may also d i s t i n g u i s h defensive from congruent males. I f depersonalization i s a defensive coping strategy, men who depersonalize may be found to show greater blaming tendencies i n helplessness conditions than ei t h e r i n t e r n a l s or more congruent externals. Summary. Although Savicki and Cooley's l o n g i t u d i n a l research (Wade et a l . , 1986) supported t h e i r hypothesis that high involvement leads to burnout (Savicki & Cooley, 1983), cross-s e c t i o n a l research did not support the rela t e d hypothesis that burnout i s associated with low involvement. Instead, cross-s e c t i o n a l research indicated that involvement i s p o s i t i v e l y associated with emotional exhaustion, and perhaps, to a lesser degree, with depersonalization as well (Cooley & Sav i c k i , 1987; McMullen & Krantz, 1988; Maslach & Jackson, 1981; Quigley et a l . , 1987) . Burnout theory suggests that depersonalization may be a defensive coping mechanism for dealing with the s t r a i n of emotional exhaustion (Lee & Ashforth, i n press). More than women, men seem conditioned to both r e s i s t and react to the threat of helplessness (Hochreich, 1974; Rotter, 1975). The 20 higher depersonalization scores among men may r e f l e c t defensiveness i n coping with the d i s t r e s s of emotional exhaustion. Support for the Control-Attribution Hypothesis According to learned-helplessness theory, helplessness d e f i c i t s occur only i n the context of control issues. For helplessness to follow, respondents must believe not only that the causes of negative events reside within themselves, but also that they are unable to influence these negative events (Peterson & Seligman, 1984). In t h e i r control hypothesis, Savicki and Cooley (1983) suggested that high i n t e r n a l control both predicts burnout and s h i f t s to low control, s p e c i f i c a l l y to low personal accomplishment, under high-stress conditions. Longitudinal research. Through longitudinal research, Savicki and h i s colleagues (Wade et a l . , 1986) found support for the f i r s t aspect of t h e i r control hypothesis. As indicated, they found that personal r e s p o n s i b i l i t y predicted burnout over the course of a year (Wade et a l . , 1986). Because the ATO personal-r e s p o n s i b i l i t y factor r e f l e c t s self-blame, Cooley and Savicki (1987) assumed that the scale also r e f l e c t s control over outcomes. On the basis of t h i s assumption, they concluded that high experienced control predicts burnout. Although the conclusion that high experienced control p r e d i c t s burnout may be sound, i t i s poorly based i f i t i s based on the assumption that personal r e s p o n s i b i l i t y implies high experienced c o n t r o l . In the face of negative events, only behavioural (internal-unstable) a t t r i b u t i o n s need imply personal c o n t r o l . Because the ATO does not di s t i n g u i s h between stable and 21 unstable causes, i t cannot d i s t i n g u i s h between behavioural and ch a r a c t e r o l o g i c a l a t t r i b u t i o n s . The reported association between personal r e s p o n s i b i l i t y and emotional exhaustion (Cooley & Sa v i c k i , 1987) cannot therefore, i n i t s e l f , be taken to demonstrate that high control predicts burnout. According to learned-helplessness theory (Peterson et a l . , 1981), however, behavioural a t t r i b u t i o n s for negative events are more l i k e l y to be operative than are characterological a t t r i b u t i o n s i n the early phase of learned-helplessness. By drawing on learned-helplessness theory, the r e s u l t s reported by Wade and h i s colleagues (1986) can be taken to support the conclusion that high control predicts burnout. Cross-sectional research. To some extent, Cooley and Savicki*s (1987) cross-sectional research also supports t h e i r r e l a t e d hypothesis that burnout i s associated with low c o n t r o l . Although personal r e s p o n s i b i l i t y f a i l e d to correlate s i g n i f i c a n t l y with personal accomplishment among helping professionals (N = 94), i t correlated s i g n i f i c a n t l y with emotional exhaustion, r = .28, p <.003, frequency, and r = .18, p < .046, i n t e n s i t y . I f personal r e s p o n s i b i l i t y were assumed to imply high co n t r o l , Cooley and Savicki's (1987) fi n d i n g would lead to the conclusion that high control i s associated with emotional exhaustion. Peterson and his colleagues (1981), however, tested and disproved such an assumption. They showed that self-blame need not imply high control. Among freshman females (N = 87), g u i l t was more strongly associated with characterological a t t r i b u t i o n s than with behavioural a t t r i b u t i o n s f o r negative 22 events, yet characterological a t t r i b u t i o n s were also associated with greater depression. Characterological a t t r i b u t i o n s i n helplessness conditions seem to be associated with both a sense of high r e s p o n s i b i l i t y and response d e f i c i t s such as might be expected among those reporting an external locus of c o n t r o l . To assume, as Cooley and Savicki (1987) do, that high personal r e s p o n s i b i l i t y i n the face of burnout implies high control i s to confuse of causal-attribution and locus-of-control issues. Although both a t t r i b u t i o n and locus-of-control constructs deal with control issues, the f o c i of t h e i r respective measures d i f f e r s (Brickman et a l . , 1982; Marsh, 1984; Wong & Sproule, 1984). Measures of causal-attribution focus on control over the causes of events; measures of locus of co n t r o l focus on con t r o l over events themselves. Attributions tap the causes of a person's control (Seligman, Abramson, Semmel, & von Baeyer, 1979) . Weiner's (1972, 1974) two-dimensional taxonomy f o r causal a t t r i b u t i o n s helps c l a r i f y the d i s t i n c t i o n between causal a t t r i b u t i o n s and locus of control. Weiner c l a s s i f i e d causes of events into four types, representing the four combinations of the endpoints of the locus and s t a b i l i t y dimensions. According to t h i s taxonomy, a b i l i t y , an internal-stable cause, i s le s s i n t e r n a l l y c o n t r o l l a b l e than i s e f f o r t , an internal-unstable cause. Task d i f f i c u l t y , an external-stable cause, and luck, an external-unstable cause, are both taken to imply low i n t e r n a l c o n t r o l . As Weiner's (1972, 1974) a t t r i b u t i o n factors, i . e . , a b i l i t y , e f f o r t , task d i f f i c u l t y , and luck, demonstrate, co n t r o l over the 23 causes of events need not imply control over the events themselves. One might, for example, experience low control over one's l e v e l of a b i l i t y and thus low control over the causes of events, yet through one's a b i l i t y nevertheless experience high co n t r o l over events themselves. Two considerations appear to influence the experience of control over events: the nature of the event i t s e l f and the nature of a t t r i b u t i o n s for the event. Locus-of-control responses seem to be determined (a) by whether the event i s p o s i t i v e or negative and (b) by whether the cause of the event i s deemed c o n t r o l l a b l e . I f a p o s i t i v e event i s attributed to one's l e v e l of a b i l i t y , a high sense of internal control over events l i k e l y r e s u l t s , along with an attitude akin to pride. I f a negative event i s att r i b u t e d to one's l e v e l of a b i l i t y , a low sense of i n t e r n a l control over events l i k e l y r e s u l t s , along with an a t t i t u d e akin to shame. Simi l a r l y , i f a p o s i t i v e event i s a t t r i b u t e d to one's e f f o r t , a greater sense of i n t e r n a l control over events w i l l l i k e l y r e s u l t than i f a negative event i s a t t r i b u t e d to one's lack of e f f o r t . These l a t t e r "behavioural" a t t r i b u t i o n s l i k e l y give r i s e to attitudes akin to self-esteem and to g u i l t . The expected association between Weiner's a t t r i b u t i o n s t y l e s and self-concept was demonstrated by Marsh (1984) . Among f i f t h -grade students (N = 559), greater s u c c e s s / a b i l i t y and success/effort a t t r i b u t i o n s were associated with greater academic self-esteem, both rs = .49. Greater f a i l u r e / a b i l i t y and f a i l u r e / e f f o r t a t t r i b u t i o n s , on the other hand, were associated with lower academic self-esteem, r = -.38 and r = -.26, 24 r e s p e c t i v e l y ( a l l ps <.05). F a i l i n g through lack of a b i l i t y appeared more damaging to self-esteem than d i d f a i l i n g through lack of e f f o r t . Consistent with learned-helplessness theory, f a i l i n g f o r external reasons was nearly unassociated with academic self-concept. Like self-esteem, depression has also been predicted by two-dimensional factor a t t r i b u t i o n s . Sweeney and h i s colleagues (Sweeney et a l . , 1986) demonstrated, through a meta-analytic review of the learned-helplessness l i t e r a t u r e , that a b i l i t y a t t r i b u t i o n s for f a i l u r e have been associated with depression. When two-dimensional factors were used to pre d i c t depression, the e f f e c t s i z e for a b i l i t y (.32) was greater than was the e f f e c t s i z e f o r e f f o r t (.08); s i m i l a r l y the e f f e c t s i z e f o r luck a t t r i b u t i o n s (-.17) was greater than was that f o r task d i f f i c u l t y (-.07). A b i l i t y , l i k e characterological a t t r i b u t i o n s , were those most strongly associated with depression. As tested by the learned-helplessness-depression hypothesis, characterological (internal-stable) a t t r i b u t i o n s f o r negative events imply both high involvement i n the causes of events and low c o n t r o l over events themselves. Maslach's (1982a) d e s c r i p t i o n of burned-out health-care workers suggests that such workers, l i k e helplessness victims, make characterological a t t r i b u t i o n s for negative events. The workers Maslach describes seem to blame t h e i r characters for negative events that l i e beyond t h e i r control: The perception that "I am not cut out for t h i s job"; "Something i s wrong with me"; or "I have become a cold, unfeeling person" i s not uncommon and appeared to be one 25 factor that propelled several professionals into some form of i n d i v i d u a l therapy or led them to quit t h e i r job. Even when they recognized the s p e c i a l s i t u a t i o n a l stresses of t h e i r work [ i t a l i c s mine], people were s t i l l prone to lay blame on some flaw within themselves ("I should have been able to handle i t " ) . Consequently, they experienced a sense of f a i l u r e and a loss of self-esteem, and a state of depression would often set i n . (p. 239) . Should burned-out workers be found to make characterological a t t r i b u t i o n s f o r f a i l u r e , they would l i k e l y f i t Maslach's des c r i p t i o n better than Freudenberger's (1977) de s c r i p t i o n of workers who " f i n d f a u l t with everything and everyone around them, complaining about the organization and reacting c y n i c a l l y to whatever i s suggested or attempted by others (p. 26)." Through reference to learned-helplessness theory, the reported association between personal r e s p o n s i b i l i t y and emotional exhaustion (Cooley & Savicki, 1987) can be taken to support Savicki and Cooley's (1983) c o n t r o l - a t t r i b u t i o n hypothesis. However, Cooley and Savicki f a i l e d to f i n d a s i g n i f i c a n t c o r r e l a t i o n between personal r e s p o n s i b i l i t y and personal-accomplishment. Only the personal impact fac t o r on the ATO correlated s i g n i f i c a n t l y with personal accomplishment, i n t e n s i t y , r = .24, p < .05 (Cooley & Savicki, 1987). Although the ATO does not adequately d i s t i n g u i s h between stable and unstable a t t r i b u t i o n s , locus-of-control measures more d i r e c t l y assess the experience of control, a l b e i t mixing control f o r p o s i t i v e and negative events. An external locus of control, as measured by the Adult Nowicki-Strickland Internal-External 26 Control Scale (ANS-IE, Nowicki & Strickland, 1973), correlated p o s i t i v e l y with emotional exhaustion, frequency, r = .17, p < .001 and with low personal accomplishment, frequency, r = .11, g < .05, among special-education teachers (N = 469) (Mclntyre, 1984). This f i n d i n g supports Savicki and Cooley's (1983) hypothesis that burnout involves low control. Using Rotter's (1966) I-E Locus of Control scale, Capel (1987) also found an association between burnout and low cont r o l . Among a number of predictors, i . e . , years at present p o s i t i o n , hours of e x t r a - c u r r i c u l a r a c t i v i t i e s , frequency with which work i s taken home, years of teaching experience, number of d i f f e r e n t classes, r o l e ambiguity, role c o n f l i c t , and external locus of co n t r o l , an external locus of control best predicted emotional exhaustion and low personal accomplishment among secondary teachers (N = 78). Canonical loadings showed that locus of control contributed most to the o v e r a l l r e l a t i o n s h i p between the predictors and burnout. Depersonalization did not make a s i g n i f i c a n t contribution to the multivariate r e l a t i o n s h i p . Among the low to moderately stressed teachers assessed i n t h i s study, depersonalization may have been moderately e f f e c t i v e as a coping measure fo r minimizing burnout d e f i c i t s . Holt, Fine, and Toffelson (1987) found, among female elementary teachers (N = 192), that those who reported high emotional exhaustion reported a more external locus of control as measured by Rotter's (1966) I-E Locus of Control scale. However, these findings f a i l e d to reach s t a t i s t i c a l s i g n i f i c a n c e . Locus-of-control research. Locus-of-control studies have supported not only the hypothesis that burnout i s associated with 27 low control, but also the related hypothesis that high experienced control predisposes respondents to low c o n t r o l i n the face of negative events. In experimental studies, a high i n t e r n a l locus of control has been found, a l b e i t somewhat inco n s i s t e n t l y , to predispose respondents to helplessness. Hiroto (1974) found, among introductory psychology students (N = 96), that those reporting an external locus of control were more vulnerable than were internals to behavioural helplessness outcomes. However, Pittman and Pittman (1979) showed that, among the same population (N = 90), high internals under high-helplessness conditions were more prone to behavioural and emotional helplessness d e f i c i t s than were high externals. Pittman and Pittman (1979) subjected groups measuring high and low, i . e . , below 8 or above 15, on Rotter's Internal-External Locus of Control scale for good and bad events, to high, low, or no-helplessness t r a i n i n g . Those with a high i n t e r n a l locus of control had s i g n i f i c a n t l y improved performance under low-helplessness t r a i n i n g , but worse performance under high-helplessness t r a i n i n g . Those with a high external locus of control showed a more moderate and l i n e a r pattern of decreasing performance effectiveness as a function of helplessness t r a i n i n g . Both groups i n the Pittman and Pittman (1979) study showed increased h o s t i l i t y under low-helplessness conditions, but high-helplessness respondents showed less h o s t i l i t y than di d low-helplessness respondents. Both groups also showed increased depression under high-helplessness conditions, although high i n t e r n a l s showing the most marked depressive responses. 28 High-helplessness respondents tended to believe that, although they could not solve the problems, others could. They thus seemed to a t t r i b u t e the cause of t h e i r f a i l u r e or lack of c o n t r o l to themselves and therefore to be s u f f e r i n g from personal helplessness. High internals showed the greatest helplessness. However, no d i s t i n c t i o n between high internals and high externals was made with reference to the assessment of others' a b i l i t y to solve problems. In the context of experimental and l o n g i t u d i n a l designs, a t t r i b u t i o n s may be assumed to interact with events to determine locus of control réponses (cf. Weiner 1972, 1974). Under low-stress conditions, an i n t e r n a l locus of control l i k e l y suggests behavioural (internal-unstable) a t t r i b u t i o n s f o r negative events. An external locus of control l i k e l y suggests luck (external-unstable) a t t r i b u t i o n s . Under high-stress conditions, an i n t e r n a l locus of control l i k e l y suggests again behavioural (internal-unstable) a t t r i b u t i o n s for negative events. An external locus of control, on the other hand, l i k e l y suggests ei t h e r task d i f f i c u l t y (external-stable) or the more damaging characterological (internal-stable) a t t r i b u t i o n s f o r negative events. In a longitudinal study of middle-aged male respondents (N = 1262), Krause and Stryker (1984) found that extreme i n t e r n a l locus-of-control respondents were more prone to psycho-p h y s i o l o g i c a l d i s t r e s s than were moderate i n t e r n a l s . They found v u l n e r a b i l i t i e s among extreme and moderate external locus-of-c o n t r o l respondents as well. Only moderate in t e r n a l s showed r e s i l i e n c e to stress ( a l l ps < .05). Krause and Stryker measured 29 stress as the occurence of s t r e s s f u l job and economic-related events. They divided the in t e r n a l and external locus-of-control scores at plus and minus one standard deviation from the mean, but they d i d not separate high and low-stress groups. On the basis of t h e i r findings, Krause and Stryker concluded that e f f e c t i v e coping actions were not undertaken by extreme i n t e r n a l locus-of-control respondents because of the paralyzing g u i l t f e e l i n g s produced by the b e l i e f that one's own actions (sic) are responsible for the occurence of the i n i t i a l event. In a cross-sectional analysis, Krause (1986) found s i m i l a r r e s u l t s for the extreme locus-of-control groups among the el d e r l y (average age = 73.4 years, SD = 6.2 years). H i e r a r c h i c a l multiple regression analyses showed that s t r e s s f u l l i f e events accounted for 9.2% of the variance i n somatic and retarded a c t i v i t i e s scores among extreme internals and f o r 11.1% among extreme externals. Stress accounted for less than 1% of variance i n somatic and retarded a c t i v i t i e s scores among the moderate cont r o l b e l i e f group. Once again, Krause suggested that s e l f -blame may impede constructive coping among extreme i n t e r n a l locus-of-control respondents. Lefcourt, Martin, and Saleh (1984) suggested that a lack of s o c i a l support may jeopardize the coping a b i l i t i e s i n t e r n a l locus-of-control respondents under stress. Using h i e r a r c h i c a l multiple regression anayses, Lefcourt and h i s colleagues studied both locus of control and s o c i a l support as moderators of stress responses among f i r s t - y e a r psychology students (N = 46). Although an i n t e r n a l locus of control was expected to have a moderating e f f e c t on the r e l a t i o n between stress and mood 30 disturbance (Lefcourt, 1982), Lefcourt found that respondents reporting an i n t e r n a l locus of control for a f f i l i a t i o n and low s o c i a l support became more disturbed than did externals as a function of increasing numbers of negative l i f e events. To account for the fact that i n t e r n a l i t y predicts depressive tendencies on the one hand and r e s i l i e n c e on the other, Lefcourt and h i s colleagues (1984) suggested that s o c i a l support may determine the nature of the association between i n t e r n a l i t y and mood disorders. Internals who adopt more stable and global s e l f -blaming tendencies, consistent with the characterological a t t r i b u t i o n s of learned-helplessness depression, may lack the benefit of s o c i a l support. Social support may buffer helplessness among internals either by r e i n f o r c i n g unstable, s p e c i f i c a t t r i b u t i o n s , or by encouraging more external a t t r i b u t i o n s for negative events. Summary. Savicki and Cooley 1s (1983) hypothesis that high control leads to burnout under s t r e s s f u l conditions i s supported by t h e i r own longitudinal research (Wade et a l . , 1986). Learned-helplessness research provides additional support f o r t h i s hypothesis (Krause, 1986; Krause & Stryker, 1984; Lefcourt et a l . , 1984; Pittman & Pittman, 1979). The r e l a t e d hypothesis that high control s h i f t s to low control i n the burnout process i s also supported i n both burnout (Capel, 1987; Cooley & Savicki, 1987; Holt et a l . , 1987; Mclntyre, 1984; McMullen & Krantz, 1988) and locus-of-control research (Krause, 1986; Krause & Stryker, 1984; Lefcourt et a l . , 1984; Pittman & Pittman, 1979). 31 Although both emotional exhaustion and low personal accomplishment have been associated with an external locus of con t r o l (Capel, 1987; Mclntyre, 1984), low personal accomplishment has not been consistently associated with an external locus of control when measures that involve i n t e r n a l a t t r i b u t i o n s have been used (Cooley & Savicki, 1987; McMullen & Krantz, 1988). Implications f o r Emotional Exhaustion Savicki and Cooley (198 3) suggested that depersonalization and low personal accomplishment would be explained by a t t r i b u t i o n s , but they did not expect emotional exhaustion to be explained i n the same way. According to Savicki and Cooley, only workers measuring high on a l l three MBI subscales should be considered burned out. Workers who are exclusively emotionally exhausted may be merely j o b - d i s s a t i s f i e d , i . e . , worn down (emotionally) by an unfavourable job s i t u a t i o n . Consistent with Savicki and Cooley's (1983) assessment, Capel (1987) found that emotional exhaustion was more influenced than were the other two MBI subscales by working conditions, i . e . , by number of years at the present p o s i t i o n , by years of teaching experience, by r o l e c o n f l i c t , and by r o l e ambiguity among secondary-school teachers (N = 78) . Crane and Iwanicki (1986) found, however, that emotional exhaustion could not be c l e a r l y distinguished from depersonalization through reference to job-related stress. Among special-education teachers (N = 443), emotional exhaustion, frequency, and depersonalization, frequency, were both affected by r o l e c o n f l i c t (14% of variance explained). And both were less affected by r o l e ambiguity (1% 32 and 2% of variance, res p e c t i v e l y ) , than was personal accomplishment, frequency, (7% of variance). Role c o n f l i c t and r o l e ambiguity, i n combination, accounted for greater variance i n emotional exhaustion, frequency, and depersonalization, frequency (14% of variance) than they did i n personal accomplishment, frequency (7% of variance) . In some studies, emotional exhaustion and depersonalization have loaded on the same factor (Brookings, Bolton, Brown, & McEvoy, 1985; Dignam, Barrera, & West, 1986). However, co r r e l a t i o n s between emotional exhaustion and depersonalization have generally been moderate, ranging from r = .40 (Maslach & Jackson, 1981) to r = .58 (Lee & Ashforth, 1990). Savicki and Cooley 1s (1983) model notwithstanding, emotional exhaustion has tended to be more strongly ( a l b e i t p o s i t i v e l y ) associated with involvement than has depersonalization (Cooley & Savicki, 1987; McMullen & Krantz, 1988; Maslach & Jackson, 1981; Quigley et a l . , 1987) and more strongly associated with an external locus of control than has personal accomplishment (Capel, 1987; Mclntyre, 1984). I f emotional exhaustion i s afforded i t s central p o s i t i o n i n the burnout process (cf. Lee and Ashforth, i n press), then the burnout, locus-of-contol, and learned-helplessness l i t e r a t u r e s together i d e n t i f y three a t t r i b u t i o n p r o f i l e s l i k e l y be associated with emotional exhaustion under stress. F i r s t , Freudenberger (1977) i d e n t i f i e d the enthusiastic high-achievers or "true" burnouts who increase t h e i r e f f o r t i n the face of stress. If these individuals maintain i n t e r n a l a t t r i b u t i o n s , while adopting progressively more stable ones i n 33 the face of stress, they may come to resemble the personally helpless or learned-helplessness depressed. Hochreich (1974) distinguished between true i n t e r n a l s and those i n t e r n a l s who defend against f a i l u r e by adopting more external a t t r i b u t i o n s i n the face of stress. Under low-stress conditions, both "defensive externals" and t h e i r more congruent counterparts would f i t Freudenberger 1 s desc r i p t i o n of enthusiastic high-achievers. Under high-stress conditions, however, low-trust "externals" would be expected to be more h o s t i l e and blaming than other internals or congruent externals (Hochreich, 1974). F i n a l l y , Pittman and Pittman (1979) i d e n t i f i e d "true" externals who are expected to suffer less-extreme d e f i c i t s under s t r e s s f u l conditions. Although female undergraduate students (N = 87) seem to adopt more internal-stable a t t r i b u t i o n s i n the face of repeated negative events (Peterson et a l . , 1981), some external-locus-of-control respondents seem nevertheless to maintain external a t t r i b u t i o n s under s t r e s s f u l conditions. Under high-stress conditions, external a t t r i b u t o r s are expected to su f f e r the moderate d e f i c i t s associated with universal helplessness (Abramson et a l . , 1978; Pittman & Pittman, 1979). Emotional exhaustion, l i k e depression, has been associated with the self-esteem d e f i c i t s of personal helplessness. S e l f -esteem, as measured by the Rosenberg Self-Esteem Scale (Rosenberg, 1965), correlated s i g n i f i c a n t l y with emotional exhaustion, r = -.32 p_ < .001, depersonalization, r = -.30, p .001, and low personal accomplishment, r = -.21, p_ < .05, among female human-service professionals (N = 135) (Brookings et a l . , 34 1986) . Among female day-care workers, self-esteem, correlated s i g n i f i c a n t l y with emotional exhaustion, r = -.27, g < .05, and depersonalization, r = -.23, p < .05, although i t did not cor r e l a t e s i g n i f i c a n t l y with low personal accomplishment (McMullen & Krantz, 1988). Implications for Job D i s s a t i s f a c t i o n In helping to explain emotional exhaustion, a t t r i b u t i o n s may also help to di s t i n g u i s h between emotional exhaustion and job d i s s a t i s f a c t i o n . Should emotional exhaustion be explained by characterological (internal-stable) and task d i f f i c u l t y (external-stable) at t r i b u t i o n s , i n keeping with the t h e o r e t i c a l overlap between personal and universal helplessness, and should job s a t i s f a c t i o n f a i l to be associated with these a t t r i b u t i o n s , then Savicki and Cooley 1s (1983) suggestion that emotional exhaustion may be l i t t l e more than a state of being worn down (emotionally) by an unfavourable job s i t u a t i o n (Savicki & Cooley, 1983) would not be supported. Job d i s s a t i s f a c t i o n , though seemingly d i s t i n c t from burnout (Maslach & Jackson, 1981), has, l i k e emotional exhaustion, also been associated with self-esteem d e f i c i t s . Among female human-service professionals (N = 135), job ( d i s ) s a t i s f a c t i o n correlated with emotional exhaustion, r = .59, with depersonalization, r = .50, and with low personal accomplishment, r = .32 ( a l l ps < .001) (Brookings et a l . , 1986). The association between s e l f -esteem and job ( d i s ) s a t i s f a c t i o n , r = -.36, was as strong as was that between emotional exhaustion and self-esteem, r = -.32. Thus job d i s s a t i s f a c t i o n does not seem to be uniquely associated with universal helplessness. 35 Additi o n a l Burnout/Learned-Helplessness Links Through reference to a t t r i b u t i o n s , Savicki and Cooley (1983) suggested s i m i l a r i t i e s between burnout and learned-helplessness This association has found additional support i n the burnout, learned-helplessness, and coping l i t e r a t u r e s . Burnout research. Apparent inconsistencies among the personal variables associated with burnout may be explained with reference to the learned-helplessness model. In cross-sectional studies, burned-out teachers have been found to e x h i b i t a va r i e t y of negative c h a r a c t e r i s t i c s , including a Type-A, workaholic personality (Nagy & Davis, 1985) , excessive involvement i n work (Quigley et a l . , 1987), perceived low control over students (Gold, 1985), low l e v e l of s e l f - a c t u a l i z a t i o n (Malanowski & Wood, 1984), and an external locus of control (Holt et a l . , 1987; Mclntyre, 1984). No t h e o r e t i c a l explanation has been given i n the burnout l i t e r a t u r e for finding a population of workaholic high-achievers who exhibit a predominantly external locus of c o n t r o l . A high achievement-orientation and a low o r i e n t a t i o n to co n t r o l would seem to be mutually exclusive a t t r i b u t e s . The learned-helplessness model can account, however, for the f a c t that excessively-involved, Type-A p e r s o n a l i t i e s and those e x h i b i t i n g low perceived control and low s e l f - a c t u a l i z a t i o n are associated with the same syndrome. Under low-stress conditions, i n t e r n a l a t t r i b u t o r s would be expected to be highly achievement-oriented. I f they were to adopt characterological or external a t t r i b u t i o n s i n the face of uncontrollable negative events, however, they would then come to experience low s e l f -36 a c t u a l i z a t i o n and low perceived control as helplessness developed. Lee and Ashforth (1990) d i r e c t l y linked burnout and helplessness. Using a 6-item work-related helplessness scale, they found s i g n i f i c a n t correlations between helplessness and burnout: r = .43, p < .001, emotional exhaustion, r = .26 p < .001, depersonalization, and r = .20 p < .05 low personal accomplishment. Learned-helplessness research. Studying learned-helplessness i n the workplace, Seligman and Schulman (1986) found associations between attributions and both perseverence and performance behaviours among life-insurance sales agents. Consistent with personal and universal-helplessness, they found that agents who made more "pessimistic" a t t r i b u t i o n s , i . e . , more i n t e r n a l , stable, and global a t t r i b u t i o n s for negative events and more external, unstable, and s p e c i f i c a t t r i b u t i o n s f o r p o s i t i v e events, were more prone to low productivity and q u i t t i n g than were those with a more optimistic s t y l e . A p e s s i m i s t i c s t y l e both correlated with and predicted a lack of perseverence on the job. Learned-helplessness i s sometimes assessed behaviourally as a performance d e f i c i t (e.g., Hiroto, 1974; Hiroto & Seligman, 1975) or as low productivity or low perseverence (Seligman & Schulman, 1986). Yet i t i s most often assessed as depression. Depression i s both the c r i t e r i o n most consistent associated with personal helplessness and the emotional symptom most consistently linked with burnout ( K a h i l i , 1988). 37 Coping. Other l i n k s between burnout and learned-helplessness are to be found with reference to coping. The end-states of both syndromes are described i n ways that suggest s p e c i f i c coping behaviours. The f i n a l stage of burnout has been described as apathy (Edelwich & Brodsky, 1982; Freudenberger, 1980) and as depersonalization coping (Lee & Ashforth, i n press). Learned-helplessness was o r i g i n a l l y observed as a f a i l u r e to respond to aversive s t i m u l i (Abramson et a l . , 1978). Later, the cognitive and motivational d e f i c i t s associated with personal and universal helplessness remained l o g i c a l l y consistent with the apathy of burnout (Abramson et a l . , 1978) . Peterson and Seligman (1983) e x p l i c i t l y proposed a place for coping i n reformulated learned-helplessness theory. They suggested that victims who lack control i n the l i t e r a l sense may attempt to gain control i n a secondary or cognitive sense by f i n d i n g meaning or value i n u n c o n t r o l l a b i l i t y . Victims may cope c o g n i t i v e l y by a l i g n i n g themselves with chance fac t o r s , always expecting the worst and thereby creating the i l l u s i o n of c o n t r o l l i n g t h e i r world. Cannon's (1929) " f i g h t or f l i g h t " response set, a conceptual model used to categorize coping strategies (Latack, 1986), has been applied both to learned-helplessness and to burnout. Folkman and Lazarus (1985) included such f l i g h t - o r i e n t e d responses as distancing and escape-avoidance i n t h e i r revised version of the Ways of Coping scale. Consistent with Seligman's hypothesis, Folkman and her colleagues (Folkman, Lazarus, Dunkel-Schetter, Delongis, & Gruen, 1986) found that spouses (N = 170) who assessed negative situations p e s s i m i s t i c a l l y as stable or as 38 "having to be accepted" engaged less i n situation-focused coping s t r a t e g i e s . They engaged more i n escape-avoidance and distancing coping. In a recent study of the ef f e c t s of a t t r i b u t i o n on coping and performance, Mikulincer (1989a) found that undergraduate s o c i a l sciences students (N = 90) who attributed f a i l u r e to internal/general (a composite score of s t a b i l i t y and glob a l i t y ) causes were also less l i k e l y to use problem-focused coping ( i . e . , coping aimed at solving the stress-creating problem) and more l i k e l y to use emotion-focused coping ( i . e . , coping aimed at easing the tension aroused by the threat) than were those who i d e n t i f i e d e x t e r n a l / s p e c i f i c causes. Students who assessed the causes of f a i l u r e p e s s i m i s t i c a l l y were also more l i k e l y to use distancing coping. Mikulincer (1989a) found that both low problem-focused coping and high distancing coping were related to performance d e f i c i t s and that distancing coping acted as an intervening l i n k between a t t r i b u t i o n and performance. As an intervening l i n k between a t t r i b u t i o n and performance, coping may mediate the process by which causal or locus-of-control a t t r i b u t i o n s a f f e c t performance. Consistent with Wortman and Brehm's (1975) integration of reactance and the learned-helplessness theory, Pittman and Pittman (1979) hypothesized that i n t e r n a l locus-of-control individuals t r y harder, or cope more aggressively under the f r u s t r a t i o n of low-helplessness t r a i n i n g , yet under high-helplessness t r a i n i n g , they seem to cope less aggressively and become more resigned and depressed. In causal a t t r i b u t i o n terms, those who make behavioural (internal-unstable) 39 a t t r i b u t i o n s f o r negative events may i n i t i a l l y cope more aggressively or a c t i v e l y , i n accordance with reactance theory, when negative events occur. But as negative events p e r s i s t , they may adopt more characterological (internal-stable) a t t r i b u t i o n s and may abandon t h e i r control-coping strategies more completely than do external a t t r i b u t o r s . Holt and her colleagues (1987) found that female high-stress/ low emotional exhaustion elementary teachers (N = 25) chose more active types of coping strategies (get involved i n a hobby or a c t i v i t y , adopt a humorous attitude, keep persisting) than d i d t h e i r high-stress/high-emotional exhaustion counterparts (N = 40), who chose more passive (e.g., think about i t , get angry, cut out a c t i v i t i e s ) strategies. Among human-service professionals (N = 503), Etzion and Pines (1986) found that, whereas more a c t i v e / d i r e c t behaviours were associated with lower burnout, as measured by Pines' burnout measure (Pines, 1985), more i n a c t i v e or i n d i r e c t coping strategies were associated with higher burnout scores. In developing workshops, Pines and her colleagues (1981) promoted avoidant a c t i v i t i e s as réponses both consistent with and appropriate to high-stress, burnout s i t u a t i o n s . With respect to work-related coping, Latack (1986) found that greater control coping was associated with greater job s a t i s f a c t i o n and less propensity to leave the job among managers and professionals (N = 109). By extension, greater escape coping could be expected to be associated with greater job ( d i s ) s a t i s f a c t i o n and greater job-leaving tendencies. Job ( d i s ) s a t i s f a c t i o n , burnout, and learned-helplessness have a l l been linked to job turn-over ( K a h i l i , 1988; Seligman & Schulman, 1986). However, i n discussing teacher burnout, Dworkin (1987) claimed that entrapment may be more pervasive than job turnover among burned-out teachers. If burned-out teachers were to a t t r i b u t e the cause of t h e i r f a i l u r e s either to t h e i r characters or to t h e i r l e v e l of competence, they would not attempt to escape, any more than learned-helplessness victims attempt to escape from aversive s t i m u l i . Insofar as the burnout victims are personally helpless, t h e i r escape coping w i l l l i k e l y be of the passive, emotion-focused, or cognitive type attributed by Peterson and Seligman (1984) to helplessness victims. Short of changing t h e i r careers, burnout victims may not contemplate leaving t h e i r jobs. In developing a work-related coping measure, Latack (1986) subdivided escape and control coping. She included cognitive reappraisal and active subtypes. According to Dworkin 1s (1987) entrapment theory, burnout should be more strongly associated with escape-cognitive than with escape-active coping. Lee and Ashforth (1990) found nonsignificant c o r r e l a t i o n s between Latack 1s (1986) general escape-coping subscale and the MBI subscales, r = .13, emotional exhaustion, r = .14, depersonalization, and r = -.08 personal accomplishment, among supervisors and managers from a large public welfare agency (N = 219). Yet they found stronger correlations between control coping and burnout, r = -.22, p < .05, emotional exhaustion, r = -.18 depersonalization, and r = .42, p < .001, personal accomplishment. I t seems that burned-out respondents were less 41 consistent i n adopting escape coping than they were i n avoiding control-coping strategies. Perhaps, consistent with Dworkin (1987), they were d i s i n c l i n e d to adopt the more active escape-coping s t r a t e g i e s . L e i t e r (1991) s i m i l a r l y found only weak c o r r e l a t i o n s between Latack's general escape coping and the MBI subscales, r = .16, p < .05, emotional exhaustion, r = .18, p = < .05, depersonalization, and r = .10, ns, personal accomplishment, among workers i n a mental hospital (N = 177). He found stronger c o r r e l a t i o n s between control coping and burnout, r = -.34, p < .01, emotional exhaustion, r = -.26 p < .01, depersonalization, and r = .46, p < .01, personal accomplishment. Again, these findings may indicate that escape-cognitive coping i s the preferred strategy among burnout victims. Implications for Personal Accomplishment Although Savicki and Cooley's (1983) hypothesis that low personal accomplishment i s associated with external a t t r i b u t i o n s has found only l i m i t e d support (Capel, 1987; Cooley & Savicki, 1987; Mclntyre, 1984; McMullen & Krantz, 1988), findings reported by Lee and Ashforth (1990) and by L e i t e r (1991) nevertheless suggest that personal accomplishment may be the burnout subscale most strongly associated with control coping s t r a t e g i e s . Summary and Integration The s i m i l a r i t i e s between burnout and learned-helplessness depression, documented i n the burnout, learned-helplessness, and coping l i t e r a t u r e s , give r i s e to the hypothesis that burnout may be a work-related form of learned helplessness, s i m i l a r to 42 personal and universal helplessness, but d i s t i n c t from job d i s s a t i s f a c t i o n (Maslach & Jackson, 1981). Although some association between constructs measured as c l u s t e r s of emotional symptoms i s to be expected ( K a h i l i , 1988), the association between burnout and depression would be more compelling i f burnout were shown to be associated with those a t t r i b u t i o n s tested by means of the learned-helplessness model. I f respondents most vulnerable to burnout were found to make int e r n a l - s t a b l e a t t r i b u t i o n s , possible remediation strategies for burnout might be reconsidered. Because of the association between burnout and an external locus of control, Mclntyre (1981) suggested therapeutically conditioning an i n t e r n a l locus of c o n t r o l . However, such an intervention might increase the v u l n e r a b i l i t y to burnout were i t to encourage an exaggerated expenditure of e f f o r t under low-stress conditions or more characterological a t t r i b u t i o n s under high stress conditions. Main expectations. The depersonalization and low personal accomplishment subscales of the MBI were not used i n t h i s analysis because of t h e i r suspected confounding with a t t r i b u t i o n s (Savicki & Cooley, 1983). Attributions s t y l e s and coping stragegies consistent with both personal and universal helplessness were expected, however, to be associated with emotional exhaustion. Teacher stress (Crane & Iwanicki, 1986; Russell, Altmaier, & Van Velzenet, 1987), a t t r i b u t i o n s t y l e (Wade et a l . , 1986), and coping strategies (Etzion & Pines, 1986; Holt et a l . , 1987) have a l l been associated with burnout. S p e c i f i c a l l y , high teacher stress and characterological (internal-stable) a t t r i b u t i o n s for 43 negative events, generally associated with personal helplessness (McMullen & Krantz, 1988), have both been associated with high emotional exhaustion. These associations are consistent with learned-helplessness theory (Sweeney et a l . , 1986) and with l o n g i t u d i n a l burnout research (Wade et a l . , 1986). Figure 2 represents the r e l a t i o n s h i p between locus and s t a b i l i t y a t t r i b u t i o n s , indicating the coping st r a t e g i e s most l i k e l y to be used with each. High escape-cognitive coping was expected to be associated with high emotional exhaustion. This expectation i s consistent with both the hypothesized sense of entrapment experienced by burnout victims (Dworkin, 1987) and with learned-helplessness theory. Personal-helplessness theory holds that active-coping strategies, including active attempts to escape, diminish when attributions are i n t e r n a l and stable for negative events (Abramson et a l . , 1978; Peterson & Seligman, 1984). Escape-cognitive coping, i n p a r t i c u l a r , i s l o g i c a l l y consistent with an internal-stable a t t r i b u t i o n s t y l e f o r negative events and with a b e l i e f that the causes of problems are embedded i n one's character. Because of the t h e o r e t i c a l overlap between personal and universal helplessness (Abramson et a l . , 1978), those a t t r i b u t i o n s and coping strategies consistent with universal helplessness were also expected to be associated with burnout. S p e c i f i c a l l y , high external-stable a t t r i b u t i o n s and escape-active coping were both expected to be associated with high emotional exhaustion. Peterson and his colleagues (Peterson et a l . , 1981) found that external attributions were inconsistent with learned-helplessness depression. Nevertheless, a t t r i b u t i o n s consistent 44 LOCUS Control coping Internal-unstable (behavioural) External-unstable (luck) STABILITY (CONTROL) Escape coping Internal-stable (charactero-logical) External-stable (task d i f f i c u l t y ) Cognitive coping Active coping Figure 2. Model integrating a t t r i b u t i o n s and coping 45 with universal helplessness are t h e o r e t i c a l l y expected to be associated with some of the d e f i c i t s of personal helplessness and depression (Abramson et a l . , 1978). External-stable a t t r i b u t i o n s and escape-active coping were therefore expected to contribute, a l b e i t l e s s strongly than internal-stable a t t r i b u t i o n s and escape-cognitive coping, to emotional exhaustion. Composite a t t r i b u t i o n variables. Generally, a t t r i b u t i o n s are taken to be hypothetical constructs representing psychological processes (Peterson & Seligman, 1984). Internal-stable a t t r i b u t i o n s for negative events are taken, f o r example, to r e f l e c t a psychological r i s k of personal helplessness (Peterson et a l . , 1981). In the learned-helplessness l i t e r a t u r e , a t t r i b u t i o n variables have been constructed mathematically i n various ways to r e f l e c t psychological processes. Peterson and h i s colleagues (1982) suggested creating mathematical composites of a t t r i b u t i o n variables by summing the three dimensions on the A t t r i b u t i o n Style Questionnaire. Because both i n t e r n a l - s t a b l e a t t r i b u t i o n s for negative events and external-unstable a t t r i b u t i o n s for p o s i t i v e events have been found to p r e d i c t depression (Seligman et a l . , 1979; Sweeney et a l . , 1986), Seligman and Schulman used a composite-positive minus composite-negative (CPCN) variable to maximize the e f f e c t of a t t r i b u t i o n s on helplessness. G l o b a l i t y has been the most, and s t a b i l i t y the l e a s t powerful predictor of depression (Peterson et a l . , 1982). Abramson and her colleagues (Abramson, Al l o y , & Metalsky, 1986) therefore suggested using a simple average of s t a b i l i t y and g l o b a l i t y scores to create a mathematical generality v a r i a b l e . 46 However, Mikulincer (1989a) instead used a m u l t i p l i c a t i v e generality factor to r e f l e c t the assumed j o i n t psychological action of s t a b i l i t y and g l o b a l i t y . The two dimensions of in t e r e s t here, locus and s t a b i l i t y , were used m u l t i p l i c a t i v e l y to r e f l e c t the assumed psychological j o i n t action of i n t e r n a l and stable a t t r i b u t i o n s . The e f f e c t of mathematically maximizing extreme scores on the dimensions considered e s s e n t i a l to r i s k was consistent with the expectation that extreme a t t r i b u t i o n s represent the greatest psychological r i s k s (Savicki & Cooley, 1983). A high score on the composite characterological a t t r i b u t i o n s variable (see Table 1) was taken to indicate that predominately int e r n a l and stable responses had been recorded. Locus scores were reversed on a composite task-d i f f i c u l t y a t t r i b u t i o n variable (see Table 1), with a high score being taken to indicate that predominately external and stable responses had been recorded. Product-terms. Whereas composite variables were entered into the regressions as single variables, product-terms were entered a f t e r the manner of interactions. Stress-by-attributions product-terms were used to r e f l e c t the hypothesized moderating e f f e c t of a t t r i b u t i o n s on stress and burnout r e l a t i o n s h i p s . According to the Cherniss 1 (1980) model of burnout, work settings and personal attributes combine to cause burnout outcomes. A personal-helplessness product-term, teacher stress by characterological a t t r i b u t i o n s (see Table 1), was used to r e f l e c t conditions i n l i f e thought to be conducive to personal-helplessness decisions. Only when characterological a t t r i b u t i o n s 47 Composite Variables Behavioural a t t r i b u t i o n s composite: external by stable a t t r i b u t i o n s Characterological a t t r i b u t i o n s composite: i n t e r n a l by stable a t t r i b u t i o n s Luck a t t r i b u t i o n s composite: external by unstable a t t r i b u t i o n s T a s k - d i f f i c u l t y a t t r i b u t i o n s composite: i n t e r n a l by unstable a t t r i b u t i o n s Product-Terms Enthusiasm product-term: teacher stress by behavioural (internal-unstable) composite variable Morale product-term: teacher stress by luck (external unstable) composite variable Personal-helplessness product-term: teacher s t r e s s by characterological (internal-stable) composite variable Universal-helplessness product-term: teacher stress by t a s k - d i f f i c u l t y (external-stable) composite variable Table 1. Glossary of Composite Variables and Product-Terms 48 a t t r i b u t i o n s for negative events and high stress co-exist are depressive interpretations expected (Peterson & Seligman, 1984). Task d i f f i c u t l y (external-stable) a t t r i b u t i o n s f o r negative events were also expected to enhance, a l b e i t l e s s strongly, the e f f e c t of high teacher stress on burnout. The u n i v e r s a l -helplessness product-term, teacher stress by task d i f f i c u l t y a t t r i b u t i o n s (see Table 1), was taken to r e f l e c t conditions i n l i f e thought to be conducive to universal-helplessness decisions. Although external locus-of-control respondents have been reported to be less depressed than internal respondents under high-helplessness conditions (Pittman & Pittman, 1979), learned-helplessness theory holds that external a t t r i b u t o r s w i l l nevertheless experience cognitive and motivational d e f i c i t s under high-stress conditions (Abramson et a l . , 1978). Corollary expectations. Behavioural (internal-unstable) and luck (external-unstable) attributions were expected to be negatively associated with burnout. The task of p r e d i c t i n g the r e l a t i v e strength of these associations was complicated, however, by the mix of a t t r i b u t i o n scales and c r i t e r i o n v a riables used within the learned-helplessness l i t e r a t u r e . Generally, a t t r i b u t i o n s t y l e i s measured as the sum of the locus, s t a b i l i t y , and g l o b a l i t y a t t r i b u t i o n scores. When in t e r n a l - s t a b l e - g l o b a l a t t r i b u t i o n s are associated with learned-helplessness depression, external-unstable-specific a t t r i b u t i o n s are simultaneously associated with low learned-helplessness depression. In a meta-analytic review, Sweeney and h i s colleagues (1986) confirmed that some of the findings reported using dimensionalized a t t r i b u t i o n ratings are consistent with factor 49 studies. A t t r i b u t i o n s made to luck factors (external-unstable a t t r i b u t i o n s ) had a reported e f f e c t s i z e of -.17 on depression, compared to a reported e f f e c t s i z e of .32 for a b i l i t y factors ( i n t e r n a l - s t a b l e a t t r i b u t i o n s ) . The r e l a t e d associations between depression and both behavioural (internal-unstable) and task d i f f i c u l t y (external-stable) a t t r i b u t i o n s are not tested by the learned-helplessness-depression hypothesis. Nevertheless, Peterson and h i s colleagues (1981) observed that behavioural (internal-unstable) and external (task d i f f i c u l t y and luck) explanations were negatively associated with depression, r = -.44 and r = -.47 (ps < .001), repectively, when the at t r i b u t i o n s of female undergraduate students (N= 87) were independently coded. In t h e i r meta-a n a l y t i c review of two-dimensional f a c t o r a l studies, Sweeney and h i s colleagues (1986) reported, however, that e f f o r t ( i n t e r n a l -unstable) factors had an e f f e c t s i z e of .08 on depression, and t a s k - d i f f i c u l t y (external-stable attributions) an e f f e c t s i z e of -.07. Neither group of researchers commented on the expectation that task d i f f i c u l t y (external-stable) a t t r i b u t i o n s would be p o s i t i v e l y associated with universal helplessness (Abramson et a l . , 1978). Over-all, the weight of learned-helplessness research (Sweeney et a l . , 1986) seems to indicate that luck (external-unstable) a t t r i b u t i o n s are l i k e l y to be more strongly and negatively associated with depression than are behavioural (internal-unstable) a t t r i b u t i o n s . Under low-stress conditions, behavioural a t t r i b u t i o n s may actually contribute to helplessness (Pittman & Pittman, 1979; Sweeney et a l . , 1986). In the burnout 50 l i t e r a t u r e , behavioural at t r i b u t i o n s are taken to constitute a burnout r i s k . Freudenberger (1980) considered i n d i v i d u a l s with high achievement needs to be most at r i s k for burnout, and both Edelwich and Brodsky (1980) and Savicki and Cooley (1983) suggested prospective burnout victims may be characterized by t h e i r i d e a l i s t i c enthusiasm. High behavioural (internal-unstable) a t t r i b u t i o n s under low-stress conditions were expected here to contribute to high emotional exhaustion more than were luck (external-stable) a t t r i b u t i o n s . Behavioural a t t r i b u t i o n s under low-stress conditions were assumed to be associated with increases i n e f f o r t and with greater persistence, consistent with Pittman and Pittman 1s (1979) finding that int e r n a l locus-of-control respondents perform better under low-helplessness t r a i n i n g . Although low control coping has been associated with both high emotional exhaustion and low personal accomplishment (Lee & Ashforth, 1990; L e i t e r , 1991), based on Pittman and Pittman's (1979) reported c u r v i l i n e a r r e l a t i o n s h i p between stress and performance among internals, control-active coping was suspected of having, among internals, a c u r v i l i n e a r r e l a t i o n s h i p with stress and burnout. Control-cognitive coping, on the other hand, was expected to have a more li n e a r r e l a t i o n s h i p with d i s t r e s s outcomes among internals and externals a l i k e . Among externals, performance has been observed to decrease i n a l i n e a r fashion with increasing stress (Pittman & Pittman, 1979). Among behavioural (internal-unstable) a t t r i b u t o r s , a high r a t i o of control-cognitive, "I think I can", coping was expected to help explain the i n i t i a l over-exertion i n the face of stress thought 51 to lead to the d e b i l i t a t i o n of burnout (Savicki & Cooley, 1983). Such coping would be expected to diminish i n a l i n e a r fashion with increased stress. Because control-cognitive coping was expected to have a more consistent l i n e a r r e l a t i o n s h i p with emotional exhaustion than was c o n t r o l - a c t i v e coping, high control-cognitive coping was expected to contribute more strongly to high emotional exhaustion. To t e s t the expected association between behavioural a t t r i b u t i o n s and burnout under low-stress conditions, a composite behavioural (in t e r n a l by unstable) variable was mathematically constructed (see Table 1), modelled on the composite characterological variable. In addition, an enthusiasm (stress by behavioural attributions) product-term, modelled on the personal helplessness product-term, was used to r e f l e c t the expected moderating e f f e c t of internal-unstable a t t r i b u t i o n s on stress and s t r a i n r e l a t i o n s h i p s . S i m i l a r l y , a composite luck (external-unstable) variable was constructed (see Table 1), modelled on the composite task d i f f i c u l t y v a r i a b l e , and a morale (stress by luck attributions) product-term (see Table 1), modelled on the universal-helplessness product-term, was used to r e f l e c t the expected moderating e f f e c t of external-unstable a t t r i b u t i o n s . Longitudinal implications. The personal change component of burnout, i d e n t i f i e d i n the descriptive l i t e r a t u r e (Freudenberger, 1974), has been larg e l y ignored i n the context of cross-sectional research designs. Because of limited resources, t h i s study too was cross-sectional i n design. The s t a b i l i t y a t t r i b u t i o n factor was included, however, i n an attempt to overcome t h i s l i m i t a t i o n 52 of design. Peterson and his colleagues (1981) have demonstrated that, among female undergraduate students, those who reported repeated exposure to negative events over the previous year reported more characterological explanations f o r negative events. Characterological a t t r i b u t i o n s , i n turn, have been found to be s i g n i f i c a n t l y more "stable" than behavioural a t t r i b u t i o n s . In t h i s study, those a t t r i b u t i n g negative events to more stable causes were assumed to have suffered more prolonged exposure to negative events. Although stable a t t r i b u t i o n s , as reported here, were made in response to both g e n e r a l - l i f e and work-related hypothetical events, they were nevertheless taken to r e f l e c t , to some extent, the impact of long-term work-related s t r e s s . Sample. Post-secondary instructors were studied. This group, though assumed to be stressed, has been under-represented i n the l i t e r a t u r e (Seiler & Pearson, 1984-1985). At the i n s t i t u t i o n selected, only male instructors were sampled. Males and females d i f f e r with respect burnout, with men generally reporting less burnout (Etzion & Pines, 1986), less emotional exhaustion (Maslach & Jackson, 1981), and more depersonalization than women (Anderson & Iwanicki, 1984; Greenglass & Burke, 1989; Ogus, Greenglass, & Burke, 1990; Schwab & Iwanicki, 1982). However, among teachers, men have been reported to be at greater r i s k of burnout (Schlansker, 1987). Because the population studied was predominantly male, only a sample of males was expected to provide an adequate number of respondents. Hypotheses 53 Main Hypothesis S i g n i f i c a n t l i n e a r relationships were expected between emotional exhaustion and the predictor variables (a) teacher stress (b) inte r n a l - s t a b l e (characterological) and external-stable (task d i f f i c u l t y ) a t t r i b u t i o n s for negative events, (c) the personal-helplessness (teacher stress by in t e r n a l - s t a b l e a t t r i b u t i o n s for negative events) and universal-helplessness (teacher stress by external-stable a t t r i b u t i o n s f o r negative events) product-terms, and (d) the r a t i o s of escape-cognitive coping to t o t a l coping and of escape-active coping to t o t a l coping. S p e c i f i c a l l y , I hypothesized that personal helplessness (teacher stress by internal-stable a t t r i b u t i o n s f o r negative events) and universal helplessness (teacher stres s by external-stable a t t r i b u t i o n s for negative events) would make s i g n i f i c a n t contributions to emotional exhaustion above and beyond those made by teacher stress and by internal-stable (charactrological) and external-stable (task d i f f i c u l t y ) a t t r i b u t i o n s f o r negative events. The r a t i o s of escape-cognitive coping to t o t a l coping and of escape-active coping to t o t a l coping were expected to make ad d i t i o n a l contributions to emotional exhaustion. Although internal-stable and external-stable a t t r i b u t i o n s f o r negative events were both expected to be p o s i t i v e l y associated with emotional exhaustion under high-stress conditions, the association between emotional exhaustion and in t e r n a l - s t a b l e a t t r i b u t i o n s was expected to be stronger. 54 S i m i l a r l y , although both coping r a t i o s were expected to be p o s i t i v e l y associated with emotional exhaustion, the escape-cognitive coping r a t i o was expected to account for more variance i n emotional exhaustion than was the escape-active coping r a t i o . C o r o l lary Hypothesis In addition, s i g n i f i c a n t l i n e a r r e l a t i o n s h i p s were expected between emotional exhaustion and the predictor variables (a) teacher stress (b) external-unstable (luck) and internal-unstable (behavioural) a t t r i b u t i o n s for negative events, (c) the morale (teacher stress by external-unstable a t t r i b u t i o n s f o r negative events) and enthusiasm (teacher stress by internal-unstable a t t r i b u t i o n s for negative events) product-terms, and (d) the r a t i o s of control-cognitive coping to t o t a l coping and of control- a c t i v e coping to t o t a l coping. S p e c i f i c a l l y , I hypothesized that morale (teacher stress by external-unstable a t t r i b u t i o n s for negative events) and enthusiasm (teacher stress by internal-unstable a t t r i b u t i o n s f or negative events) would make s i g n i f i c a n t contributions to emotional exhaustion above and beyond those made by teacher stress and by external-unstable (luck) and internal-unstable (behavioural) a t t r i b u t i o n s for negative events. The r a t i o s of control-cognitive coping to t o t a l coping and of control-active coping to t o t a l coping were expected to make ad d i t i o n a l contributions to emotional exhaustion. Internal-unstable attributions were expected to be p o s i t i v e l y associated with emotional exhaustion under low-stress conditions. This association was expected to be stronger than 55 was the association between external-unstable a t t r i b u t i o n s and emotional exhaustion under low-stress conditions. Although both coping r a t i o s were expected to be negatively associated with emotional exhaustion, the control-cognitive coping r a t i o was expected to account for more variance i n emotional exhaustion than was the control-active coping r a t i o . Exploratory Questions The r e l a t i o n s h i p s described i n the hypotheses above were explored with respect to personal accomplishment. Personal accomplishment and depersonalization were excluded from the main burnout hypotheses because of t h e i r suspected confounding with s t a b i l i t y and locus a t t r i b u t i o n s , respectively. Personal accomplishment was nevertheless explored as a c r i t e r i o n both because of i t s expected association with control coping and because of i t s weaker association with emotional exhaustion. Job ( d i s ) s a t i s f a c t i o n was also explored as a c r i t e r i o n of secondary i n t e r e s t because of i t s expected association with escape-active coping. Job ( d i s ) s a t i s f a c t i o n was expected not to be predicted by a t t r i b u t i o n s , and i n t h i s respect, i t was expected to d i f f e r from emotional exhaustion. 56 Method Respondents The respondents were a volunteer sample of i n s t r u c t i o n a l f a c u l t y at a post-secondary i n s t i t u t i o n i n a large westcoast urban centre i n Canada. During the 1980s, the i n s t i t u t i o n studied was affected by recession and organizational change. The i n s t i t u t i o n a l context was therefore one of li m i t e d and uncertain resources, with r e s u l t i n g increases i n demands on f a c u l t y and decreases i n morale. The climate could be expected to fo s t e r burnout. At the time of the study, the i n s t i t u t i o n was under exceptional s t r a i n , having been s p e c i f i c a l l y targeted f o r cutbacks during the weeks previous. An intense p u b l i c i t y and lobbying campaign had been i n i t i a t e d by the f a c u l t y union. The i n s t r u c t i o n a l faculty i s represented by two unions. Including both groups, the faculty consists of approximately 513 members, 438 men and 75 women. Because men comprise the bulk of the population and because burnout may be re l a t e d to sex (Maslach & Jackson, 1981), only men were sampled, s p e c i f i c a l l y those 30-55 years of age. Maslach and Jackson (1981) reported that burnout tends to decrease with age, either because those who are vulnerable leave or because they adjust. The t o t a l sample consisted of 316 men. A 47% response rate yielded 148 respondents. Of those respondents, 108 returned questionnaires with no more than 20% of responses missing on any scale (38 were eliminated), with an appropriate age-range response (1 was eliminated), and with a v a l i d response set (2 57 were eliminated because of impossible responses on the Expanded Mû) . Procedure Questionnaires were provided to a l l male i n s t r u c t i o n a l f a c u l t y between the ages of 30 and 55 through the c e n t r a l mail service. Faculty members were in v i t e d to return t h e i r questionnaires to the union o f f i c e i n person, through the c e n t r a l mail service, or through t h e i r chief bargaining representative. (For informed consent attachments, reminders, and endorsement, see Appendix A). The study was advertised through the union and the i n s t i t u t e newsletters. (See Appendix B for texts.) Instruments Along with a demographic-data survey, the questionnaire contained f i v e s e l f - r e p o r t instruments (see Appendix C), including seven subscales from the Teacher Stress Questionnaire (TSI, Pettegrew & Wolf, 1982), the Expanded A t t r i b u t i o n Style Questionnaire (Expanded ASQ, Peterson & Villanova, 1988), the f i r s t two scales of Latack's Coping Measure (Latack, 1986), and two factors from the Maslach Burnout Inventory (MBI, Maslach & Jackson, 1981). C r i t e r i o n variables. The main c r i t e r i o n v a r i a b l e was assessed by measuring the frequency of emotional exhaustion (MBI). The frequency of (low) personal accomplishment and the degree of job ( d i s ) s a t i s f a c t i o n (TSI) were also explored as a c r i t e r i o n variables. The Maslach Burnout Inventory (Maslach & Jackson, 1981) measures burnout by assessing three symptom fa c t o r s : (a) 58 emotional e x h a u s t i o n — f e e l i n g emotionally worn down and exhausted by work; (b) personal accomplishment—feelings of competence and achievement i n work; and (c) depersonalization—an unfeeling and impersonal response to c l i e n t s . Burnout i s generally taken to be the configuration of higher emotional exhaustion, lower personal accomplishment, and higher depersonalization, with each item being rated for frequency and in t e n s i t y . Emotional exhaustion i s the longest of the MBI scales (containing 9 of 24 items) and i s the least l i k e l y to be confounded with a t t r i b u t i o n s t y l e (Savicki & Cooley, 1983). I t plays a c r i t i c a l r o l e i n burnout according to a number of t h e o r i s t s (see Lee & Ashforth, i n press), and i s hypothesized to lead d i r e c t l y to both depersonalization, i t s e l f perhaps a defensive coping strategy for dealing with emotional exhaustion ( K a h i l i , 1988; Lee & Ashforth, i n press), and diminished personal accomplishment, a measure of performance competence and a more loos e l y - r e l a t e d construct (Lee & Ashforth). Maslach and Jackson (1981) found that the emotional exhaustion subscale correlated with depersonalization, frequency, r = .44, and with personal accomplishment, frequency, r = .17. Depersonalization was not included i n t h i s study because i t was expected, consistent with Savicki and Cooley's (1983) model, that i t might be confounded with locus a t t r i b u t i o n s . Personal accomplishment was included as an exploratory c r i t e r i o n . The emotional exhaustion subscale consists of nine items (items 1, 2, 3, 5, 7, 10, 11, 12, and 16) and the personal accomplishment subscale of eight items (items 4, 6, 8, 9, 13, 14 15, and 17), each rated for frequency and i n t e n s i t y . For both 59 subscales, the frequency dimension i s labeled at each point, ranging from 0 ("never") to 6 ("every day"). The i n t e n s i t y dimension ranges from 1 ("very mild, barely noticeable") to 7 ("major, very strong"). Maslach and Jackson (1981) report c o r r e l a t i o n s between the in t e n s i t y and frequency dimensions across items ranging from .35 to .73, with a mean of .56. Because both Gaines and Jermier (1983) and Russell and h i s colleagues (1987) have suggested that i n t e r c o r r e l a t i o n s between the i n t e n s i t y and frequency measures are s u f f i c i e n t l y high to warrant using the frequency measure only i n research, only the frequency dimension was used i n t h i s study. Possible scores for emotional exhaustion, frequency ranged from a low of 0 to a high of 54, with a higher score i n d i c a t i n g greater emotional exhaustion. Possible scores f o r personal accomplishment, frequency ranged from a low of 0 to a high of 48, with a higher scores indicating greater personal accomplishment. The personal accomplishment subscale was reversed so that higher scores on both subscales would indicate greater burnout. Maslach and Jackson (1981) have reported good r e l i a b i l i t y f o r both the emotional exhaustion and personal accomplishment subscales. For the frequency subscale, i n t e r n a l consistency (Cronbach's c o e f f i c i e n t alpha) was reported at .89 f o r emotional exhaustion and at .74 for personal accomplishment. Test-retest r e l i a b i l i t y i s moderate to good. Using a two to four week i n t e r v a l , the t e s t was readministered to 53 graduate students i n s o c i a l welfare and administrators i n a health agency, y i e l d i n g a r e l i a b i l i t i y of r = .82 for emotional exhaustion, frequency, and of r = .60 for personal accomplishment, frequency. In t h i s study, the standardized item alpha was calculated at .91 f o r emotional exhaustion, frequency, and at .72 f o r personal accomplishment, frequency. Maslach and Jackson (1981) have reported a series of v a l i d i t y checks which, though encouraging c o l l e c t i v e l y , are unimpressive i n themselves. Through extensive use since 1981 however, the MBI has steadily strengthened i t s v a l i d i t y as a burnout measure. The Teacher Stress Inventory (Pettegrew & Wolf, 1982) was the source of the job s a t i s f a c t i o n subscale, used i n t h i s study as an exploratory variable. Along with the management s t y l e , l i f e s a t i s f a c t i o n , and supervisory support subscales, job s a t i s f a c t i o n was o r i g i n a l l y included i n the inventory to demonstrate i t s construct v a l i d i t y . Job s a t i s f a c t i o n clustered with r o l e - r e l a t e d and task-related stress variables when a Smallest Space Analysis was applied to the inventory (Pettegrew & Wolf, 1982), and i t correlated moderately with other teacher stress subscales (Schutz & Long, 1988). Nevertheless, the subscale was used here as a c r i t e r i o n variable ( K a h i l i , 1988; Wolpin et a l . , 1991). The job s a t i s f a c t i o n scale has f i v e items (items 21-25) and i s responded to on a 5-point Likert-type scale, l a b e l l e d from "strongly disagree" to "strongly agree". Scores were reverse-scored. They ranged from a possible low of 5 to a possible high of 25, with a higher score representing greater job d i s s a t i s f a c t i o n . 61 The i n t e r n a l consistency (Cronbach's c o e f f i c i e n t alpha) had previously been calculated at .86 (Pettegrew & Wolf, 1982). In t h i s study, i t was calculated at .80. Predictor variables. The predictor variables, teacher stress, a t t r i b u t i o n s t y l e for negative events, and coping, were assessed by measuring the degree of teacher stress (TSI), the degree of locus and of s t a b i l i t y a t t r i b u t i o n s f o r negative events (Expanded ASQ) and the r a t i o of escape-cognitive, escape-active, control-cognitive and control-active coping to t o t a l coping (Latack's coping measure). Using'the Teacher Stress Inventory (1982), teacher stress was taken to be the summed score of s i x subscales: r o l e ambiguity (items 1-5), r o l e overload (items 6-10), r o l e c o n f l i c t (items I l -ls) , nonparticipation (items 16-20), task stres s (items 26-32), and supervisory support (items 33-36). These subscales together measure r o l e - r e l a t e d and task-based stress and include the r e l a t e d measure of (lack of) supervisory support. The l i f e s a t i s f a c t i o n and management s t y l e subscales were not used. The l i f e s a t i s f a c t i o n subscale was excluded because i t , l i k e job s a t i s f a c t i o n , was taken to be an outcome measure. Management s t y l e was excluded because i t was not expected to produce s i g n i f i c a n t v a r i a b i l i t y within the organization studied. Both these subscales, l i k e job s a t i s f a c t i o n and supervisory support, were o r i g i n a l l y included i n the TSI to e s t a b l i s h the measure's construct v a l i d i t y . Together, the s i x subscales included i n job stress comprise 31 items, which are responded to on a 5-point Likert-type scale, l a b e l l e d from "strongly disagree" to "strongly agree." 62 P o s i t i v e l y phrased items were reverse-scored, so that a score of 5 represents a high-stressed response for a l l items. Possible t o t a l scores for teacher stress ranged from a low of 31 to high of 155, with a higher score in d i c a t i n g greater teacher stress. Pettegrew and Wolf (1982) have reported i n t e r n a l consistencies (Cronbach's c o e f f i c i e n t alpha) of .79, r o l e ambiguity; .76, r o l e overload; .82, r o l e c o n f l i c t ; .76, nonparticipation; .84, task stress; and .89, supervisory support. In t h i s study, these alphas measured .72, .77, .82, .66, .75, and .77, respectively. The o v e r a l l i n t e r n a l consistency (Cronbach's c o e f f i c i e n t alpha) of the scales used was calculated at .90. The TSI s i g n i f i c a n t l y d i f f e r e n t i a t e s between high and low-stress groups of respondents (Pettegrew & Wolf, 1982). The Expanded ASQ (Peterson & Villanova, 1988) was used to assess a t t r i b u t i o n s t y l e for negative events. A t t r i b u t i o n s t y l e i s conceptualized as the degree of pessimism present i n a respondent's evaluation of causes of negative events. A t t r i b u t i o n s t y l e was measured as the tendency to make in t e r n a l or external and stable or unstable a t t r i b u t i o n s for negative events. Both extreme in t e r n a l and extreme external a t t r i b u t i o n s f o r negative events have been considered dysfunctional (Savicki and Cooley (1983). Peck (1978) c l a s s i f i e d disorders of r e s p o n s i b i l i t y into neuroses and character disorders, with neurotics assuming too much r e s p o n s i b i l i t y , and i n d i v i d u a l s with character disorders not enough. The Expanded ASQ consists of 24 items, 6 negative events taken from the o r i g i n a l A t t r i b u t i o n a l Style Questionnaire (ASQ) (Peterson et a l . , 1982) and 18 more taken from a l i f e events 63 questionnaire designed for college students (Marx, Garrity, & Bowers, 1975). Where some items were inappropriate f o r the s i t u a t i o n or population, minor wording changes were made (see Appendix C). Because only work-related events were of i n t e r e s t i n t h i s study, the Expanded ASQ was factor-analyzed to t e s t i t s domain s p e c i f i c i t y . In t h i s study only the locus and s t a b i l i t y dimensions for negative events were used; the global dimension was excluded. The g l o b a l i t y dimension has been shown to correlate moderately with s t a b i l i t y (r = .55). Both the s t a b i l i t y and g l o b a l i t y dimensions are larg e l y independent of i n t e r n a l i t y (Peterson & Villanova, 1988). Learned-helplessness t h e o r i s t s have not yet concluded whether s t a b i l i t y or g l o b a l i t y i s c r i t i c a l f a c t o r i n d i s t i n g u i s h i n g between attributions that prefigure helplessness and those that contribute to the syndrome (Peterson et a l . , 1981). However, Mikulincer (1988) implicated the s t a b i l i t y dimension i n t h i s process. Although g l o b a l i t y i s considered important i n p r e d i c t i n g depression (Peterson, Villanova, & Raps, 1985), i t was considered to be less important here i n predicting burnout. Burnout i s generally taken to be a more s i t u a t i o n - s p e c i f i c phenomenon than i s depression (Maslach & Jackson, 1977; Pines et a l . , 1981). For each item on the locus and s t a b i l i t y dimensions, the respondent i d e n t i f i e d a major cause of the negative event presented and then, with regard to t h i s cause, responded to two questions on 7-point Likert-type scales. These two questions assessed the i d e n t i f i e d cause on the locus and s t a b i l i t y dimensions. The scales associated with each assessment were 64 scored i n the d i r e c t i o n s of increasing i n t e r n a l i t y and s t a b i l i t y . Possible scores on each scale ranged from a low of 24 to a high of 168, with higher scores i n d i c a t i n g greater i n t e r n a l i t y or s t a b i l i t y of a t t r i b u t i o n s for negative events. Peterson and Villanova (1988) reported i n t e r n a l consistencies (Cronbach's c o e f f i c i e n t alpha) of .66 and .85, r e s p e c t i v e l y , for the locus and s t a b i l i t y factors of the Expanded ASQ. In t h i s study, int e r n a l consistencies were calculated at .84 and .87, respectively. Peterson and Villanova (1988) examined the v a l i d i t y of the Expanded ASQ by c o r r e l a t i n g the dimensions of explanatory s t y l e i n t h i s t e s t with ratings of explanations f o r actual negative events. An i n t e r n a l explanatory s t y l e predicts actual i n t e r n a l explanations to a greater degree than i t predicts stable explanations. A stable explanatory s t y l e predicts stable explanations, but not any better than i t predicts i n t e r n a l explanations. The construct v a l i d i t y for the s t a b l i l i t y dimension therefore presents a problem. Peterson and Villanova (1988) noted that the locus dimension continues to be the least coherent of the three dimensions. This f a c t may be due to the scale's true multidimensionality. Multidimensiality has been claimed for both the locus a t t r i b u t i o n dimension (Cooley & Savicki, 1987; Marsh, 1984) and f o r locus-of-control dimensions (Wong & Sproule, 1981). M u l t i p l i c a t i v e locus by s t a b i l i t y variables were used (cf. Mikulincer, 1989a), with a composite i n t e r n a l - s t a b l e a t t r i b u t i o n variable for negative outcomes ind i c a t i n g characterological s e l f -blame and a composite internal-unstable a t t r i b u t i o n variable 65 i n d i c a t i n g behavioural self-blame. External a t t r i b u t i o n s t y l e s were indicated by composite external-unstable (luck) or external-stable (task d i f f i c u l t y ) composite variables. Latack's coping measure (Latack, 1986) was used to assess coping s t r a t e g i e s . Coping was conceptualized as the e f f o r t to master conditions that tax or exceed adaptive resources (Monat & Lazarus, 1977) . Latack developed her coping categories by integr a t i n g three conceptual frameworks: problem-focused and emotion-focused coping (Folkman, 1982; Folkman & Lazarus, 1980); action, cognitive reappraisal, and symptom management (Latack, 1984; Moos & B i l l i n g s , 1982); and Cannon's (1929) f i g h t or f l i g h t responses. Of the three subscales included i n Latack's coping measure (control coping, escape coping, and symptom management), only the c o n t r o l and escape subscales were used. Conceptually, these subscales f i t most cl o s e l y with kind of behaviours observed i n context of learned-helplessness. Together, the subscales used con s i s t of 28 items. Control coping items include both actions and cognitive reappraisals that are proactive or take charge i n tone; escape coping items include both actions and cognitive reappraisals that suggest an escapist or avoidance mode. L e i t e r (1991) tested, through confirmatory f a c t o r analysis, Latack's (1986) c l u s t e r analysis of the scale, using s t a f f members i n a mental hospital (N = 177). The r e s u l t i n g factor structure was s i m i l a r to one that Latack found to be associated with r o l e ambiguity. Factor loadings were further explored here i n a preliminary factor analysis. 66 According to Latack and L e i t e r , there are 17 control items and 11 escape items. Among the control items, 15 appear active i n character; 4 appear to involve cognitive reappraisals. Among the escape items, 5 appear active i n character; another 6 appear to involve cognitive reappraisals. In t o t a l , four subscales were used: co n t r o l - a c t i v e coping (items 1, 3, 4, 7, 8, 12, 14, 15, 18, 21, 22, 25, and 28), control-cognitive coping (items 5, 9, 13, and 20), escape-active coping (2, 6, 11, 19, and 23), and escape-cognitive coping (items 10, 16, 17, 24, 26, and 27). On a l l subscales, items are assessed using a 5-point response scale (1 = "hardly ever do t h i s " to 5 = "almost always do t h i s " ) . When Latack constructed the tes t , p a r t i c i p a n t s were asked to respond with a s p e c i f i c type of s i t u a t i o n i n mind—role ambiguity, r o l e c o n f l i c t , or ro l e overload. Role ambiguity, for example, was described: "You are uncertain of what you are supposed to do on your job or unsure of how to approach a p a r t i c u l a r assignment" (Latack, 1986, p. 378). Parti c i p a n t s were asked to indicate how frequently they react i n the ways presented by the measure. Items were then grouped into scales or cl u s t e r s based on the consistency of response for each of the stress s i t u a t i o n s . For escape coping under r o l e ambiguity, r o l e c o n f l i c t and r o l e overload conditions, Latack (1986) reported i n t e r n a l consistencies (Cronbach's c o e f f i c i e n t alpha) of .70, .71, and .54 respectively; for control coping under r o l e ambiguity, r o l e c o n f l i c t and role overload, she reported i n t e r n a l consistencies of .79, .85, and .85, respectively. Role overload was the most s i g n i f i c a n t stressor f o r the population studied. Among t h i s sample, the r e l i a b i l i t y of escape 67 coping was .60. The r e l i a b i l i t y of contol coping was measured at .84. No t e s t - r e t e s t r e l i a b i l i t i e s have been established. I n i t i a l c o n s t r u c t - v a l i d i t y evidence has been encouraging for t h i s measure, but the t e s t needs further v a l i d a t i o n with use. Evidence of contruct v a l i d i t y was drawn from the f a c t that Type A p e r s o n a l i t i e s , as measured by a 9-item scale by Caplan, Cobb, Harrison, French, and Pinneau (1975), favour control strategies across s i t u a t i o n s ; those with higher l e v e l s of s o c i a l support, as measured by a 24-item scale from Caplan et a l . (1975), are s i m i l a r l y more l i k e l y to approach s t r e s s f u l s i t u a t i o n s with a control strategy. In the present study, respondents to Latack's Coping Measure were asked to i d e n t i f y t h e i r own s t r e s s f u l s i t u a t i o n and to respond with reference to situations s i m i l a r to the one they had i d e n t i f i e d (see Appendix C). Because the perception of uncontrollable negative outcomes as a source of stress was of p a r t i c u l a r i n t e r e s t , an appropriate s t r e s s f u l experience was defined as one involving a sense that, f o r a job-related task, the respondent f e l t that attaining the desired outcome would be impossible: "Think of a job-related task that has recently been f r u s t r a t i n g to you. By f r u s t r a t i n g , we mean that you've f e l t unable (or l i k e l y to be unable) to achieve an outcome that you want." (See Appendix D for a summary of responses). Latack's coping measure has been used as a measure of d i v e r s i t y of coping and of t o t a l coping. In t h i s study scores were calculated for each of the four subscales, escape-cognitive, escape-active, control-cognitive, and control-active coping. The i n t e r n a l consistencies (Cronbach's c o e f f i c i e n t alpha) of escape-68 cognitive, escape active, control cognitive, and control active coping were calculated at .53, .62, .74, and .80, respectively. The two escape factors were only moderately r e l i a b l e f o r t h i s sample. Possible scores ranged from a low of 6 to a high of 30 for escape-cognitive coping, with a higher score i n d i c a t i n g greater escape-cognitive coping, from a low of 5 to a high of 25 f o r escape-active coping, with a higher score i n d i c a t i n g greater escape-active coping, from a low of 4 to a high of 20 for control-cognitive coping, with a higher score i n d i c a t i n g greater control-cognitive coping, and from a low of 13 to a high of 65 for c o n t r o l - a c t i v e coping, with a higher score i n d i c a t i n g greater co n t r o l - a c t i v e coping. F i n a l l y , for t o t a l coping, possible scores ranged from a low of 28 to a high of 140, with a higher score i n d i c a t i n g greater t o t a l coping. For the MR analyses, coping scales were used i n r a t i o to t o t a l coping (Vitaliano, Maiuro, Russo, & Becker, 1987). Demographic data. Respondents were be asked to complete a b r i e f section of questions concerning demographic data (see Appendix C). This section asked for age, marital status, years of teaching, years of post-secondary education, years of non-teaching employment. Although these data could be personally i d e n t i f y i n g , respondents understood that they had the option of f a i l i n g to complete any section. Data Analysis Most missing data occurred on the Expanded ASQ. However, among the questionnaires included i n t h i s study, only 2.5% of the items on t h i s scale were l e f t unanswered. For a l l missing data, 69 mid-point scores were substituted, i n order to f a c i l i t a t e factor analysis of the scales. Because the Expanded ASQ (Peterson & Vi l l a n o v a , 1988) taps a t t r i b u t i o n s for negative events a r i s i n g from both work-related and general l i f e events, the measure was factor-analyzed to check for domain s p e c i f i c i t y . Latack's coping measure was also factor-analyzed for t h i s sample to inspect t h i s measures' items c l u s t e r s . Even though a l l the reported analyses were conducted using mid-point scores for missing data, regression analyses were also conducted s u b s t i t u t i n g mean scores. Results (not reported) were not appreciably d i f f e r e n t using mean-score substitutions. Descriptive s t a t i s t i c s , including means, standard deviations and Pearson product-moment correlations for pairwise comparisons were calc u l a t e d . The correlations describe the r e l a t i o n s among the pr e d i c t o r variables and between each predictor and the c r i t e r i o n . The hypotheses were tested by means of h i e r a r c h i c a l multiple regression (MR) analyses. The predictor variables were entered into the regression equation i n the following order: teacher stress ( l e v e l 1), the composite a t t r i b u t i o n variables (levels 2 and 3), the teacher stress by a t t r i b u t i o n product-terms ( l e v e l 4), and the coping r a t i o s ( l e v e l 5). Product-terms and coping r a t i o s were entered simultaneously within l e v e l s 4 and 5 resp e c t i v e l y . H i e r a r c h i c a l MR models tested the amount of variance i n emotional exhaustion that was explained by coping above and beyond that which was explained by the product-term (teacher 70 stress by attribution) variables, when teacher stress and the a t t r i b u t i o n composites were held constant. Two h i e r a r c h i c a l MR analyses were conducted to t e s t the hypotheses. One MR regressed emotional exhaustion onto (a) teacher stress, (b) internal-stable (characterological) a t t r i b u t i o n s , (c) external-stable (task d i f f i c u l t y ) a t t r i b u t i o n s , (d) the teacher stress by internal-stable a t t r i b u t i o n s product-term, (e) the teacher stress by external-stable a t t r i b u t i o n s product-term, (f) the r a t i o of escape-cognitive coping to t o t a l coping, and (g) the r a t i o of escape-active coping to t o t a l coping. A second MR analysis regressed emotional exhaustion onto (a) teacher stress (b) external-unstable (luck) a t t r i b u t i o n s , (c) internal-unstable (behavioural) a t t r i b u t i o n s , (d) the teacher stress by external-unstable at t r i b u t i o n s product-term, (e) the teacher stress by internal-unstable a t t r i b u t i o n s product-term, (f) the r a t i o of control-cognitive coping to t o t a l coping, and (g) the r a t i o of control-active coping to t o t a l coping. Variance accounted for by teacher stress and f o r a t t r i b u t i o n s was removed f i r s t . A t t ributions are believed to be conditioned by s t r e s s f u l events and are themselves made with reference to antecedent events (Marsh, 1974; Peterson et a l . , 1981). The teacher stress by a t t r i b u t i o n product-terms were entered next. In learned-helplessness theory, i n t e r n a l - s t a b l e and external-stable a t t r i b u t i o n s for negative outcomes are associated with helplessness only i n the face of uncontrollable negative events (Peterson et a l . , 1981). In burnout theory, behavioural 71 a t t r i b u t i o n s are thought to present a v u l n e r a b i l i t y to burnout within s t r e s s f u l work environments (Edelwich & Brodsky, 1980; Freudenberger, 1980; Savicki & Cooley, 1983). Coping was entered l a s t . In the Lazarus and Folkman (1984) t h e o r e t i c a l framework, personality variables and outcome appraisals are antecedents of coping processes. Mikulincer (1989b) found a t t r i b u t i o n s t y l e to be antecedent to coping, which i n turn predicted performance d e f i c i t s . Scales were standardized to remove the confounding e f f e c t of d i f f e r i n g u n i t s . Because a l l scales were standardized, unstandardized regression c o e f f i c i e n t s were reported. For most reported findings, an alpha l e v e l of .05 was adopted. In the case of product-terms, however, the alpha l e v e l f o r the t t e s t l e v e l of the unstandardized regression c o e f f i c i e n t s was set at .10. Both Finney and h i s colleagues (Finney, M i t c h e l l , Cronkite, & Moos, 1984) and Champeau and Peters (1987) have argued that product-terms should be reported at less-stringent l e v e l s to guard against Type II errors. In t h i s way, a larger body of evidence may be gathered to counteract the conservative e f f e c t of predominantly low-power studies. In t h i s study, the r i s k s involved i n reporting relationships as a r e s u l t of Type I errors were considered low. Relationships thus reported are not l i k e l y to be considered of c l i n i c a l importance. Product-terms were graphed following Cohen's method (Cohen & Cohen, 1983). S i g n i f i c a n t product-terms were interpreted by f i r s t c a l c u l a t i n g unstandardized regression c o e f f i c i e n t s from a model including only the product-term (stress by a t t r i b u t i o n ) and i t s composite variables as predictors. 72 Coded values of 1 and -1, representing one standard deviation above and below the grand mean on the stress variable, were taken as the high and low le v e l s when s i g n i f i c a n t contributions were found. Scores on the c r i t e r i o n v a r i a b l e were indicated on the y axis, scores on the composite a t t r i b u t i o n variables on the x axis. High and low teacher-stress means were plott e d at two points, coded at 1 and -1 along the x axis, representing one standard above and below the grand mean on the a t t r i b u t i o n composite. These plotted points anchored two regression l i n e s , one for high-stress and one for low-stress groups, on the c r i t e r i o n variables (see Figures 3 and 4). 73 Results Descriptive S t a t i s t i c s and Preliminary Analyses The demographic make-up of the respondents i s shown i n Table 2. Within the sampled age-range of 30 to 55 years, the mean age was 44.1 years (SD = 6.5). Most were married (88.9%). The mean number of years of post-secondary education was 5.8 years (SD = 3.3). This t a l l y seems to be of doubtful usefulness, however; the range of 0 - 19 years indicates that t h i s question must have been misinterpreted by some. The mean number of years of employment i n the present job was 10.2 years (SD = 6.3); the mean number of years i n non-Table 2 Demographic Charact e r i s t i c s of Sample (N=108) C h a r a c t e r i s t i c % M SD Range Age 44.1 6.5 30-55 Ma r i t a l Status: Single 11.1 Married 88.9 Years of post- a secondary education 5.8 3.3 0-19 Years non-teaching employment 12.1 6.7 0-33 Years teaching at ... 10.2 6.3 0-25 a Some respondents seem to have misunderstood t h i s question. 74 teaching employment was 12.1 years (SD = 6.7). Because post-secondary instructors are under-represented i n the l i t e r a t u r e , there are few points of comparison for these data. Because of the work-related focus of t h i s study, the Expanded ASQ (Peterson & Villanova, 1988) was factor-analyzed i n a preliminary analysis to tes t the measure's domain-specificity. Neither exploratory nor confirmatory analyses yielded an exc l u s i v e l y work-related factor on either the locus or s t a b i l i t y dimensions. Nor did exploratory or confirmatory analyses y i e l d any other factor recognizable by the event's domain. Untapped considerations, such as the perceived importance of an event, may instead have influenced factor loadings. Contrary to Bagby and his colleagues' findings (Bagby, Atkingson, Dickens, & Gavin, 1990), locus and control items did load onto two separate factors i n the context of a rotated oblimin (four-factor) analysis. Therefore, Peterson & Villanova's (1988) dimensions were used i n t h i s study. Latack's coping measure (1986) was also factor-analyzed to explore the loadings of i t s items (cf. L e i t e r , 1991). Although confirmatory analyses supported to some extent Latack's c l u s t e r s , they indicated many discrepancies. Using .35 as a minumum loading c r i t e r i o n , two, three, and four-factor rotated oblimin solutions yielded factors for the present sample that d i f f e r e d from Latack's c l u s t e r s (see Appendix D). Associations between the derived coping factors and burnout tended to be stronger for t h i s sample than were associations r e s u l t i n g from Latack's c l u s t e r s (Appendix D, Table 8). Nevertheless, Latack's c l u s t e r s were used i n t h i s study. 75 Table 3 shows the means and standard deviations f o r the pr e d i c t o r and c r i t e r i o n variables (N = 108). Neither skewness nor kurtosis was a problem with any of the scales used. Emotional exhaustion (M = 22.7, SD = 11.4) and personal accomplishment (M = 32.2, SD = 4.9) scores were s i m i l a r to those reported f o r teachers elsewhere i n the burnout l i t e r a t u r e . Among junior high and elementary school teachers, Nagy and Davis (1985) reported emotional exhaustion means of 22.3 (SD = 11.4) and personal accomplishment means of 37.5 (SD = 7.6). Item emotional exhaustion means (M = 2.5, SD = 1.3) and personal accomplishment means (M = 4.0, SD = 6.1) indicated that t h i s sample had a s i m i l a r l e v e l of emotional exhaustion but seemingly lower personal accoomplishment scores compared with other human services professionals. Although Maslach and Jackson (1981) reported s i m i l a r mean scores for both emotional exhaustion (M = 2.7, SD = 1.3) and personal accomplishment (M = 4.2, SD = 1.0) among human service occupations (N = 420), Lee and Ashforth (1990) found seemingly higher personal accomplishment scores. Among supervisors and managers i n human services (N = 171), they reported an emotional exhaustion mean of 22.3 (SD = 12.9) and a personal accomplishment mean of 46.8 (SD = 12.7). Among service professionals i n a large public welface agency, they reported an emotional exhaustion mean of 23.2 (SD = 13.5) and a personal accomplishment mean of 46.6 (SD = 12.2). Among workers i n a mental hospital (N=177), L e i t e r (1991) reported personal accomplishment scores s i m i l a r to those i n the present study. For Le i t e r * s population the mean personal Table 3 Means and Standard Deviations for Predictor and C r i t e r i o n 76 Variables (N = 108) Measure M SD a 1 Emotional exhaustion 22.7 11.4 a 2 Personal accomplishment 32.2 4.9 3 Job s a t i s f a c t i o n 16.9 3.9 a 4 Teacher stress 90.3 17.5 b 5 Locus a t t r i b u t i o n 94.8 19.8 b 6 S t a b i l i t y a t t r i b u t i o n 113.0 18.5 7 Escape coping 32.8 5.6 8 Control coping 60.0 9.3 9 Escape-cognitive coping 18.7 3.7 10 Escape-active coping 14.1 3.7 11 Control-cognitive coping 13.8 3.2 12 Control-active coping 46.3 7.2 a Items are scored with higher scores i n d i c a t i n g more emotional exhaustion and stress but less personal accomplishment. b Items are scored with higher scores i n d i c a t i n g more i n t e r n a l i t y and s t a b i l i t y . 77 accomplishment score was 33.6 (SD = 5.8); the mean emotional exhaustion score was 20.9 (SD = 9.1). Although personal accomplishment scores f o r both t h i s population and L e i t e r * s (1991) population of mental health workers seem low when compared to scores reported f o r other populations (Maslach & Jackson, 1981; Lee & Ashforth, 1990), the grounds for comparison are not s u f f i c i e n t to determine whether differences are r e a l . The teacher stress mean (M = 90.3, SD = 17.5) f o r t h i s sample was converted to item means for each of the teacher stress factors i n order to f a c i l i t a t e comparison with other samples. Item means f o r r o l e ambiguity (2.5, SD = .8), r o l e overload (2.8, SD = .8), r o l e c o n f l i c t (3.2, SD = .9), non-participation (3.1, SD = .7), task stress (2.8, SD = .8), and supervisory support (3.1, SD = .7) tended to be higher than those reported f o r any of the groups studied by Long and her colleagues (Long, Schutz, Kendall, & Hunt, 1986). These l a t t e r groups included elementary teachers, secondary teachers, administrators, coordinator consultants, supervisory aides, c l e r i c a l and support s t a f f , and cus t o d i a l s t a f f . Secondary school teachers, for example, reported means of 2.0, ro l e ambiguity; 2.5, r o l e overload; 2.4, r o l e c o n f l i c t ; 2.5, non-participation; 3.0, task s t r e s s ; and 2.1, supervisory support. (Standard deviations were not reported). Whereas respondents i s t h i s sample reported job s a t i s f a c t i o n item means of 3.4 (SD = .8), the secondary teachers studied by Long and her colleagues (1986) reported job s a t i s f a c t i o n means of only 2.4. 78 For t h i s sample, item mean responses on locus (M = 4.0, SD = 0.8) and s t a b i l i t y (M = 4.7, SD = 0.8) a t t r i b u t i o n dimensions for negative events indicated that, r e l a t i v e to introductory psychology students, the present population tended to give more external and stable causes for negative events. Among introductory male and female psychology students (N = 140), Peterson and Villanova (1988) reported mean locus a t t r i b u t i o n responses of 4.6 (SD = 0.6) and mean s t a b i l i t y a t t r i b u t i o n responses of 4.4 (SD = 0.7). The escape coping mean (32.8, SD = 5.6) and the control coping mean (60.0, SD = 9.2) for t h i s sample were s i m i l a r to those f o r other populations. Among managerial and professional s t a f f of a manufacturing firm and an osteopathic h o s p i t a l (N = 109), Latack (1986) reported escape coping means of 34.5 (SD = 6.5), 23.4 (SD = 5.0), and 15.4 (SD = 3.3) with respect to r o l e ambiguity, r o l e c o n f l i c t , and role overload stressors, r e s p e c t i v e l y . Again with respect to r o l e ambiguity, r o l e c o n f l i c t , and r o l e overload stressors, she reported control coping means of 63.6 (SD = 7.2), 74.0 (SD = 10.1) and 87.6 (SD = 11.0), respectively. Even though the most s i g n i f i c a n t stressors for t h i s sample were ro l e overload and task stress, the coping means fo r t h i s sample were most consistent with those f o r r o l e ambiguity stressors among managerial and professional s t a f f . Among supervisors and managers i n human services (N = 171), Lee and Ashforth (1990) also reported comparable coping means: escape-coping (M = 27.4, SD = 5.2) and control-coping (M = 66.9, SD = 9.3). Among mental hospital workers (N = 171), L e i t e r 79 (1991) reported comparable coping means as well: escape-coping (M = 31.1, SD = 6.4) and control-coping (M = 62.4, SD = 8.8). Pearson Product Moment Correlations Between Variables The only s i g n i f i c a n t c o r r e l a t i o n between the i n t e r v a l demographic data and any of the predictor variables was a modest p o s i t i v e c o r r e l a t i o n between age and job ( d i s ) s a t i s f a c t i o n , r = .18, p_ < .05 (see Table 4). Analysis of variance (ANOVA) showed that married subjects did not d i f f e r s i g n i f i c a n t l y from single subjects with respect to any of the predictor variables, emotional exhaustion, personal accomplishment and job d i s s a t i s f a c t i o n ( a l l Fs <1). Teacher stress was the predictor variable most strongly associated with emotional exhaustion, r = .62, p < .001 (see Table 5). A l l the TSI factors correlated s i g n i f i c a n t l y with both emotional exhaustion and with job ( d i s ) s a t i s f a c t i o n , although only the task-stress, role-overload, and supervisory-support correlated s i g n i f i c a n t l y with personal accomplishment. Task stress and r o l e overload showed moderate associations with emotional exhaustion (both rs = .59). Role ambiguity (r = .46), r o l e c o n f l i c t (r = .43), and non-participation (r = .34) also showed moderate associations with emotional exhaustion, whereas supervisory support (r = .26) showed only a weak association with emotional exhaustion (see Appendix D, Table 1). The locus a t t r i b u t i o n dimension correlated weakly with teacher stress and with emotional exhaustion, r s = -.17, p < .05 (see Appendix D, Table 2), even though no association between the locus dimension and either stress or burnout had been 80 81 82 anticipated. The s t a b i l i t y dimension f a i l e d to c o r r e l a t e s i g n i f i c a n t l y with either teacher stress or burnout. This lack of association raised concerns about the v a l i d i t y of the s t a b i l i t y dimension. As a v a l i d i t y check on the a t t r i b u t i o n dimensions, respondents had been asked to i d e n t i f y a f r u s t r a t i n g event i n the work s i t u a t i o n and to evaluate the locus and s t a b i l i t y of i t s cause. The r e s u l t i n g single-item scales correlated with the Expanded ASQ scales, r = .37, p <.001, locus, and r = .30, p <.001, s t a b i l i t y (see Appendix D, Table 2). For t h i s s i n g l e item, the locus dimension was unassociated with the burnout c r i t e r i a , as expected. As expected too, the s t a b i l i t y dimension for t h i s s i n g l e item was p o s i t i v e l y associated with both teacher stress and emotional exhaustion, r = .20, p < .05, and r = .18, p < .05, respectively. Of the composite att r i b u t i o n s , only external-stable (task-d i f f i c u l t y ) and internal-unstable (behavioural) a t t r i b u t i o n s correlated s i g n i f i c a n t l y with emotional exhaustion, r = .27 and r = -.19, respectively. Although these a t t r i b u t i o n s also correlated s i g n i f i c a n t l y with job ( d i s ) s a t i s f a c t i o n , no s i g n i f i c a n t c o rrelations were found between any composite a t t r i b u t i o n variables and low personal accomplishment (see Table 5) . The i n t e r c o r r e l a t i o n between external-stable (task-d i f f i c u l t y ) and internal-unstable (behavioural) a t t r i b u t i o n s was very high, r = -.93, as was that between in t e r n a l - s t a b l e (characterological) and external-unstable (luck) a t t r i b u t i o n s , r 83 = -.94. For p r a c t i c a l purposes, these a t t r i b u t i o n composites were taken to be opposite poles of the same scale (see Table 5). Inter c o r r e l a t i o n s between the corresponding product-terms, constructed as weighted variables for the purpose of t h i s matrix, were also very high: r = -.90 between both the u n i v e r s a l -helplessness (stress by external-stable) and enthusiasm (stress by internal-unstable) product-terms and between personal-helplessness (stress by internal-stable) and morale (stress by external-unstable) product-terms. The product-terms were, for p r a c t i c a l purposes, also taken as opposite poles of si n g l e scales (see Appendix D, Table 3). A l l weighted product-terms correlated as expected with the c r i t e r i o n variables (see Appendix D, Table 3) . Both the escape-active and the control-cognitive coping r a t i o s correlated s i g n i f i c a n t l y , as expected, with emotional exhaustion, r = .20 and r = -.17, respectively. The expected associations between emotional exhaustion and both the escape-cognitive and the control-active coping r a t i o s , on the other hand, f a i l e d to reach significance (see Table 5). A l l coping r a t i o s correlated s i g n i f i c a n t l y as expected with low personal accomplishment: escape-cognitive r a t i o , r = .24; escape-active r a t i o , r = .21; control-cognitive r a t i o , r = -.28; and co n t r o l - a c t i v e r a t i o , r = -.21 (see Table 5). None correlated s i g n i f i c a n t l y with job ( d i s ) s a t i s f a c t i o n . Of the coping r a t i o s , only the cognitive r a t i o s correlated s i g n i f i c a n t l y with any of the composite a t t r i b u t i o n scales. Control-cognitive coping correlated p o s i t i v e l y with i n t e r n a l -unstable (behavioural) att r i b u t i o n s , r = .17. Escape-cognitive 84 coping correlated correlated p o s i t i v e l y with i n t e r n a l - s t a b l e (characterological) a t t r i b u t i o n s , r = .16, and negatively with both internal-unstable (behavioural) and external-unstable (luck) a t t r i b u t i o n s , both rs = -.17. The expected negative c o r r e l a t i o n between external-stable (task d i f f i c u l t y ) a t t r i b u t i o n s and control-cognitive coping, r = -.15, and the expected p o s i t i v e c o r r e l a t i o n external-stable (task d i f f i c u l t y ) a t t r i b u t i o n s and escape-cognitive coping, r = .14, f a i l e d to reach significance, as did correlations between the active coping r a t i o s and the composite a t t r i b u t i o n variables (see Table 5). Regression Analyses Two h i e r a r c h i c a l multiple regression analyses were conducted using emotional exhaustion as the c r i t e r i o n . The main regression was s i g n i f i c a n t , F(7,100) = 11.04 p < .001. The predictor variables i n the equation accounted f o r 44% (adjusted R square = .40) of the variance. Table 6 summarizes the findings. In an h i e r a r c h i c a l manner, the variables were entered as follows: l e v e l 1, teacher stress; l e v e l 2, in t e r n a l - s t a b l e a t t r i b u t i o n s ; l e v e l 3, external-stable a t t r i b u t i o n s ; l e v e l 4, the personal-helplessness product-term (stress by in t e r n a l - s t a b l e a t t r i b u t i o n s ) and the universal-helplessness product-term (stress by external-stable a t t r i b u t i o n s ) ; l e v e l 5, the escape-cognitive coping r a t i o and the escape-active coping r a t i o . The product-terms and coping r a t i o s were entered simultaneously i n l e v e l s 4 and 5, respectively. Teacher stress was entered f i r s t . I t accounted f o r 38% of the variance and was p o s i t i v e l y associated, as expected, with 85 86 emotional exhaustion. External-stable a t t r i b u t i o n s accounted for an a d d i t i o n a l 1% of variance i n emotional exhaustion; t h i s contribution was not s i g n i f i c a n t . Internal-stable a t t r i b u t i o n s made no add i t i o n a l contribution to emotional exhaustion. The personal-helplessness and universal-helplessness product-terms, entered simultaneously, accounted f o r an add i t i o n a l 2% of variance, with the contribution of universal helplessness (stress by external-stable attributions) approaching s i g n i f i c a n c e at the p < .103 l e v e l . When coping was added, the unique contribution of universal helplessness reached the p < .07 l e v e l of si g n i f i c a n c e . Escape-cognitive and escape-active coping, entered simultaneously, accounted for an additional 3% of variance i n emotional exhaustion, with escape-active coping making a s i g n i f i c a n t contribution. Escape-cognitive coping was p o s i t i v e l y associated, as expected, with emotional exhaustion. The contribution of the the universal-helplessness (stress by external stable attributions) product-term, considered s i g n i f i c a n t at p < .103, was subsequently graphed f o r int e r p r e t a t i o n . Figure 3 shows that, under high-stress conditions, external-stable at t r i b u t i o n s were p o s i t i v e l y associated with emotional exhaustion; under low stress conditions, there was l i t t l e or no re l a t i o n s h i p between external-stable a t t r i b u t i o n s and emotional exhaustion. The c o r o l l a r y regression was also s i g n i f i c a n t F(7,100) = 11.15 p < .001. The predictor variables i n the equation accounted for 44% (adjusted R square = .40) of the variance. 87 *=high stress o=low stress E m 0 t 1 0 n a 1 E X h a u s t i o n 34 32 30 28 26 24 22 20 18 16 14 12 10 // 0 31.52 26.58 16.03 -o 15.09 low (-1) high (+1) External-Stable A t t r i b u t i o n s Figure 3. Regression of emotional exhaustion on external-stable a t t r i b u t i o n s within high-stress and low-stress subgroups. (High stress: Y = 2.47x + 29.05; low stress: Y = -.47X + 15.56). Note. Emotional exhaustion M = 22.7, SD = 11.4. 88 Table 7 summarizes the findings. In an h i e r a r c h i c a l manner, the var i a b l e s were entered as follows: l e v e l 1, teacher s t r e s s ; l e v e l 2, external-unstable a t t r i b u t i o n s ; l e v e l 3, i n t e r n a l -unstable a t t r i b u t i o n s ; l e v e l 4, the morale product-term (stress by external-unstable attributions) and the enthusiasm product-term (stress by internal-unstable a t t r i b u t i o n s ) ; l e v e l 5, the control-cognitive coping r a t i o and the control-active coping r a t i o . The product-terms and coping r a t i o s were entered simultaneously within l e v e l s 4 and 5, respectively. Teacher stress was entered f i r s t . Again i t accounted for 38% of the variance and was p o s i t i v e l y associated, as expected, with emotional exhaustion. Composite external-stable and i n t e r n a l - s t a b l e a t t r i b u t i o n s accounted for no a d d i t i o n a l variance i n emotional exhaustion. The morale and enthusiasm product-terms, entered simultaneously, accounted for an additional 3% of variance, with enthusiasm making a contribution s i g n i f i c a n t at the p < .03 l e v e l . When coping was added the unique contribution of the enthusiasm product-term reached the p < .02 l e v e l of s i g n i f i c a n c e . Control-cognitive and control-active coping, entered simultaneously, added an additional 3% of variance to emotional exhaustion, with the contribution of control-cognitive coping approaching significance at the p < .053 l e v e l . Control-cognitive coping was negatively associated, as expected, with emotional exhaustion. The s i g n i f i c a n t contribution of the enthusiasm (stress by internal-unstable attributions) product-term was graphed for 89 90 in t e r p r e t a t i o n . Figure 4 shows that internal-unstable a t t r i b u t i o n s were p o s i t i v e l y associated, as expected, withemotional exhaustion under low-stress conditions; they werenegatively associated with emotional exhaustion under high-stress conditions. Exploratory Analyses Both hypotheses were explored using personal accomplishment and job ( d i s ) s a t i s f a c t i o n ) as c r i t e r i a . For personal accomplishment, both regressions were s i g n i f i c a n t (see Appendix D). The anticipated associations with stress and a t t r i b u t i o n s tended to be weaker for personal accomplishment than they were for emotional exhaustion; the associations with coping tended to be stronger (see Appendix D, Tables 4 & 5). A l l coping str a t e g i e s contributed as expected to personal accomplishment. As expected, greater escape-cognitive and escape-active coping were associated with lower personal accomplishment, and greater control-cognitive and control-active coping were associated with greater personal accomplishment. Cognitive coping strategies made larger contributions than did active s t r a t e g i e s (see Appendix D). For job ( d i s ) s a t i s f a c t i o n , both regressions were again s i g n i f i c a n t (see Appendix D, Tables 6 & 7). Only teacher stress contributed s i g n i f i c a n t l y to job d i s s a t i s f a c t i o n . Post-Hoc Analysis To explore the possible e f f e c t of a concurrent i n s t i t u t i o n a l c r i s i s on the present findings, correlations among predictor and c r i t e r i o n variables were calculated for those questionnaires (n = 16) d i s t r i b u t e d a f t e r budgetary issues had been resolved. For 91 *=high stress o=low stress Figure 4. Regression of emotional exhaustion on i n t e r n a l -unstable at t r i b u t i o n s within high-stress and low-stress subgroups. (High stress: Y = -3.06x + 29.01; low stress: Y = 1.46x + 15.53). Note. Emotional exhaustion M = 22.7, SD = 11.4. 92 t h i s greatly reduced sample, internal-stable (characterological) a t t r i b u t i o n s were p o s i t i v e l y associated with both emotional exhaustion (r = .57, p < .05) and low personal accomplishment (r = .43, p < .05), as had o r i g i n a l l y been expected. In addition, luck a t t r i b u t i o n s were negatively associated with emotional exhaustion (r = -.55, p < '° 5) a n ^ low personal accomplishment (r = -.40), although the l a t t e r association f a i l e d to reach s i g n i f i c a n c e (see Appendix D, Table 9). No such s i g n i f i c a n t associations were found for the t o t a l sample between indi c a t o r s of personal helplessness and the burnout c r i t e r i a . On the other hand, associations between indicators of universal helplessness and emotional exhaustion, found for the t o t a l sample, were not found f o r t h i s reduced sample. 93 Discussion The hypothesis that burnout i s a form of psychological helplessness was tested by adapting a learned-helplessness model to burnout. As expected, at t r i b u t i o n s and coping st r a t e g i e s consistent with universal helplessness contributed to emotional exhaustion. However, those at t r i b u t i o n s and coping strategies consistent with personal helplessness f a i l e d to make s i g n i f i c a n t contributions. These r e s u l t s indicate that burnout may resemble universal more than personal helplessness. Universal Helplessness As expected, the tendency under high-stress conditions to make task d i f f i c u l t y (external-stable) a t t r i b u t i o n s f o r negative events was associated with greater emotional exhaustion. This fin d i n g , along with the stronger contibution of escape-active coping r e l a t i v e to escape-cognitive coping, supported the hypothesis that emotional exhaustion may be a j o b - s p e c i f i c form of universal helplessness. Associations were modest, however. Task d i f f i c u l t y (external-stable) a t t r i b u t i o n s correlated only weakly with emotional exhaustion, r = .27, p < .05, accounting f o r only 7% of the variance i n emotional exhaustion. When teacher stres s was f i r s t removed, characterological (internal-stable) and task d i f f i c u l t y (external-stable) a t t r i b u t i o n s together contributed only 3% (unadjusted R square) to the explained variance i n emotional exhaustion, a small contribution (Cohen, 1988). Correlations between escape-active coping and emotional exhaustion were also weak, r = .20, p < .05, with escape-active 94 coping accounting for only 4% of the variance i n emotional exhaustion. When teacher stress and both characterological (internal-stable) and task d i f f i c u l t y (external-stable) a t t r i b u t i o n s were f i r s t removed, escape-active coping and escape-cognitive coping together contributed only 3% (unadjusted R square) to explained variance i n emotional exhaustion. By supporting a universal-helplessness explanation of burnout, these findings nevertheless support Savicki and Cooley's (1983) burnout hypotheses. Savicki and Cooley suggested that burnout, measured as depersonalization and low personal accomplishment, would d i f f e r from job d i s s a t i s f a c t i o n i n being more strongly associated with a defensive loss of involvement and low perceived control under high-stress conditions. The present findings indicate that burnout, measured as emotional exhaustion, does d i f f e r from job d i s s a t i s f a c t i o n i n being better explained by a t t r i b u t i o n s . Under high-stress conditions, burnout was more strongly associated with task d i f f i c u l t y (external-stable) a t t r i b u t i o n s than was job d i s s a t i s f a c t i o n . Had burnout been measured as depersonalization rather than as emotional exhaustion, the contribution of task d i f f i c u l t y (external-stable) a t t r i b u t i o n s to burnout might have been stronger (Cooley & Savi c k i , 1987; McMullen & Krantz, 1988; Maslach & Jackson, 1981; Quigley et a l . , 1987). The tendency to make behavioural (internal-unstable) a t t r i b u t i o n s f o r negative events under low-stress conditions was also associated with greater emotional exhaustion. So too was lower control-cognitive coping. Again, however, associations were modest. When teacher stress was f i r s t removed, luck 95 (external-unstable) and behavioural (internal-unstable) a t t r i b u t i o n s together contributed only 3% (unadjusted) to emotional exhaustion variance, a small contribution (Cohen, 1988). The negative c o r r e l a t i o n between control-cognitive coping and emotional exhaustion was weak, r = -.17, p < .05, with control-cognitive coping accounted for only 2.9% of the variance i n emotional exhaustion. When teacher stress and both luck (external-unstable) and behavioural (internal-unstable) a t t r i b u t i o n s were f i r s t removed, control-cognitive coping and escape-active coping together contributed only 3% (unadjusted R square) to explained variance i n emotional exhaustion. Whereas behavioural attributions under low-stress conditions had been expected to represent a v u l n e r a b i l i t y to personal helplessness and to burnout, the present findings suggest that behavioural a t t r i b u t i o n s may instead represent a v u l n e r a b i l i t y to universal helplessness and to burnout. Again, t h i s f i n d i n g i s consistent with Savicki and Cooley 1s (1983) model of burnout (cf. Edelwich & Brodsky, 1980; Freudenberger, 1980). Testing causal implications of the present findings i s beyond scope of t h i s study. Nevertheless, the association between behavioural attributions and control-cognitive coping, r = .17, p < .05, lends support to a suggestion that over-enthusiasm leads to burnout by t r i g g e r i n g exaggerated e f f o r t (Edelwich & Brodsky, 1980; Freudenberger, 1980; Savicki & Cooley, 1983). Both Brickman and his colleagues (Brickman et a l . , 1982) and Savicki and Cooley (1983) have suggested a complementary explanation for the association between behavioural a t t r i b u t i o n s and burnout under low-stress conditions. By decreasing the 96 competence of helping professionals, exaggerated behavioural a t t r i b u t i o n s may increase the frequency with which negative outcomes occur. Personal Helplessness More s t r i k i n g than the contributions to emotional exhaustion made by the universal-helplessness and enthusiasm product-terms i s the f a i l u r e of the personal-helplessness and high-morale product-terms to make sim i l a r contributions. Predictions associated with the learned-helplessness-depression hypothesis were made with greater confidence than were those associated with universal helplessness. Nevertheless, support for the more established hypotheses was extremely tenuous. Despite the suspected confounding of attr i b u t i o n s with personal accomplishment, the personal-helplessness product-term f a i l e d to make any unique contribution to personal accomplishment. The greater contribution to personal accomplishment made by escape-cognitive r e l a t i v e to escape-active coping had been anticipated on the assumption that such coping would be associated with personal helplessness. Such, however, seemed not to be the case. Several explanations may account for the f a i l u r e of the more established learned-helplessness-depression hypothesis. F i r s t , a s i g n i f i c a n t percentage of the male respondents sampled may have been "defensive externals" (Hochreich, 1974), i . e . , i n t e r n a l s who reported external a t t r i b u t i o n s f a l s e l y to save face when threatened with f a i l u r e . In such a case, t a s k - d i f f i c u l t y (external-stable) a t t r i b u t i o n s would have masked underlying characterological (internal-stable) a t t r i b u t i o n s . Defensive externals would be expected to exhibit exaggerated blame-97 pro j e c t i o n and h o s t i l i t y and might be expected to cope by means of depersonalization (cf. Lee & Ashforth, i n press). Other studies (Anderson & Iwanicki, 1984; Etzion & Pines, 1986; Greenglass & Burke, 1988; Greenglass et a l . , 1990; Maslach & Jackson, 1981; Schwab & Iwanicki; 1982) have indicated that c h a r a c t e r i s t i c s of burnout may be gender-specific. Alternately, the anger prevalent i n the i n s t i t u t i o n when the data were c o l l e c t e d may have temporarily reduced characterological a t t r i b u t i o n s for negative events. Questionnaires (n = 16) d i s t r i b u t e d a f t e r the term and budget c r i s i s ended showed that, for t h i s much reduced sample, characterological (internal-stable) and luck (external-unstable) a t t r i b u t i o n s correlated as expected with burnout scores (see Appendix D, Table 7). The c r i s i s and non-crisis groups may a c t u a l l y have been d i f f e r e n t i n t h i s respect. A t t r i b u t i n g negative events to one's character or to luck may have affected the perception of stress and burnout i n other unexpected ways. Not only did characterological ( i n t e r n a l -stable) and luck (external-unstable) a t t r i b u t i o n s f a i l to c o r r e l a t e s i g n i f i c a n t l y with burnout, they also f a i l e d to c o r r e l a t e s i g n i f i c a n t l y with teacher stress and job d i s s a t i s f a c t i o n as well. Those who made characterological a t t r i b u t i o n s may have been less i n c l i n e d to i d e n t i f y external sources of d i s t r e s s . For the instructors sampled (N = 108), only "disengagement" coping was associated as expected with characterological and luck a t t r i b u t i o n s (see Appendix D, Table 6) . Disengagement or withdrawal i s generally associated depression (Shreeve, 1984). 98 I t i s also possible, however, that burnout and depression are d i s t i n c t constructs. Those who report greater emotional exhaustion and lower personal accomplishment may be more i n c l i n e d than those who are depressed to believe that t h e i r peers would encounter f a i l u r e s s i m i l a r to t h e i r own. Such a b e l i e f i n un i v e r s a l f a i l u r e i s the distinguishing c h a r a c t e r i s t i c of universal helplessness (Abramson et a l . , 1978). Should the tendency to a t t r i b u t e blame externally d i s t i n g u i s h burnout victims from those who are depressed, burnout would be expected to involve l e s s damage to self-esteem than does depression (Abramson et a l . , 1978). Burnout victims might then be expected to f i t Freudenberger's description of i n d i v i d u a l s who " f i n d f a u l t with everything and everyone around them" (Freudenberger, 1977, p. 26) better than Maslach 1s description of victims "prone to lay blame on some flaw within themselves" (Maslach, 1982a, p. 239). Maslach he r s e l f noted that workers she observed "experienced a sense of f a i l u r e and a loss of self-esteem, and a state of depression would often set i n " (p. 239). Limitations The present findings were lim i t e d by the v a l i d i t y of the study's measuring instruments and by the nature of the study's design and methodology. They generalize properly only to volunteers from the population of male ins t r u c t o r s (N = 316) studied at one urban post-secondary se t t i n g . Volunteers (47%) i n t h i s study may have been either less burned out or more i n c l i n e d to blame others than those who declined to p a r t i c i p a t e . A l l the measures used were s e l f - r e p o r t questionnaires. In a s e l f - r e p o r t study, those who are burned out may perceive and 99 report stress, a t t r i b u t i o n s , and coping d i f f e r e n t l y from those who are not. Teacher stress and a t t r i b u t i o n s t y l e may be a d d i t i o n a l l y confounded. Respondents who blame external causes may i d e n t i f y more stress i n t h e i r environments than those who blame i n t e r n a l causes. Among t h i s sample (N = 108), greater teacher stress scores were associated with greater task d i f f i c u t l y (external-stable) a t t r i b u t i o n s for negative events, r = .26, p <.05, and lower luck (internal-unstable) a t t r i b u t i o n s f o r negative events, r = -.20, p < .05. Although i n t e r c o r r e l a t i o n s of t h i s magnitude were not of concern here, they do indicate that confounding between personal and situations v a r i a b l e s may occur i n s e l f - r e p o r t studies. The findings of such studies should not, therefore, be u n c r i t i c a l l y taken to support s i t u a t i o n a l (or personal) explanations of burnout. Peterson and Seligman (1984) have advised p a r t i c u l a r caution when using s e l f - r e p o r t assessments of a t t r i b u t i o n s . The cognitive causes of depression may be automatic rather than conscious or voluntary responses, and as such they may be more accessible i n interview or c l i n i c a l settings than i n questionnaire surveys. Internal a t t r i b u t i o n s f o r negative events may be e s p e c i a l l y vulnerable to unpredictable reporting among men. I f defensive reporting, such as Hochriech (1974) attributed to low-trust male internals, were to occur i n burnout conditions, such defensiveness would undermine the v a l i d i t y of both the characterological (internal-stable) and task d i f f i c u l t y (external-unstable) a t t r i b u t i o n composites. A t t r i b u t i o n measures assessing a general s t y l e are subject to the futher c r i t i c i s m that they may y i e l d r e s u l t s that are not 100 relevant to p a r t i c u l a r situations, e s p e c i a l l y i f the context of those s i t u a t i o n s f a i r l y c l e a r l y defines c a u s a l i t y . In t h i s study, respondents were asked, i n the introduction to the Latack Coping Measure (Latack, 1986), to i d e n t i f y a s p e c i f i c f a i l u r e on the job and to indicate a locus and s t a b i l i t y a t t r i b u t i o n with respect to that event. The r e s u l t i n g single-item scales correlated with the Expanded ASQ scales, r = .37, p < .001, locus and r = .30, p < .001, s t a b i l i t y , i n d i c a t i n g some degree of consistency between a general s t y l e and p a r t i c u l a r s i t u a t i o n s . However, the s i g n i f i c a n t correlations between the single-item s t a b i l i t y scale and both teacher stress and emotional exhaustion, r = 20, p < .05, and r = .18, p < .05, respectively, i n the absence of s i g n i f i c a n t correlations for f u l l - s c a l e s t a b l i l i t y a t t r i b u t i o n s , suggest that a more s p e c i f i c and work-related scale might have greater v a l i d i t y than the Expanded ASQ. The Expanded ASQ may be c r i t i c i z e d too for p l a c i n g i n t e r n a l i t y and exte r n a l i t y on the same dimension, as does Rotter's (1966) locus of control scale. This p r a c t i c e runs contrary to evidence that i n t e r n a l i t y and e x t e r n a l i t y represent two separate dimensions for both a t t r i b u t i o n s and locus of cont r o l (Cooley & Savicki, 1987; Wong & Sproule, 1984). By fo r c i n g i n t e r n a l i t y and externality onto the same scale, the Expanded ASQ a r t i f i c i a l l y l i n k s characterological ( i n t e r n a l -stable) to luck (external-unstable) a t t r i b u t i o n s and task d i f f i c u l t y (external-stable) to behavioural (internal-unstable) a t t r i b u t i o n s . Studies using separate factors ratings have obtained r e s u l t s somewhat inconsistent with those based on the 101 a t t r i b u t i o n dimensions of the Expanded ASQ (Peterson et a l . , 1981; Sweeney et a l . , 1986). Respondents who may have had a demonstrable a t t r i b u t i o n s t y l e were not distinguished here from those who made diverse a t t r i b u t i o n s . Eliminating respondents with large standard deviations on a t t r i b u t i o n s has been found to improve correlations with depression (Peterson & Seligman, 1984). Neither was the l e v e l of teacher stress that represents an experience of non-contingency determined. Repeated f a i l u r e alone i s not expected to a f f e c t helplessness outcomes as do experiences of non-contingency between responses and outcomes (Kofta & Sedek, 1989). Although burnout was expected to vary unequally across subgroups for stress (Kofta & Sedek, 1989; Pittman & Pittman, 1979) and a t t r i b u t i o n s (Krause, 1986; Savicki & Cooley, 1983), a basic product-term analysis was used. Such an approach assumes monotonie and uniform changes across a l l l e v e l s of the moderator variable (Finney et a l . , 1984; House, 1981). The use of more sen s i t i v e approaches was precluded by the study's sample s i z e . Product-terms were interpreted i n t h i s study despite i t s non-experimental design. Cohen and Cohen (1983) have claimed that product-terms carry interaction information. Like weighted-variables, however, product-terms are combined variables when used a f t e r the f a c t . When not controlled by treatments, the constituent variables i n product-terms may be i n t e r c o r r e l a t e d or even causally linked (Pedhazur, 1982). In addition, they may be correlated with other variables not included i n the design, or they may serve as proxy variables, as do a t t r i b u t i o n s (Peterson & Seligman, 1984). If mathematically induced to function i n 102 concert, such variables may together produce a s i g n i f i c a n t e f f e c t on a predictor, such as was produced i n t h i s study by the universal-helplessness and enthusiasm product-terms. But when no psychological combination has been represented, each v a r i a b l e i n a product-term t e c h n i c a l l y a f f e c t s the predictor independently. Because neither the assumption of independence nor the assumption of i n t e r a c t i o n i s adequately met i n non-experimental studies using product-terms, findings are extremely d i f f i c u l t to int e r p r e t , simple though they may be to represent (Pedhazur, 1982) . According to Pedhazur, the best antidote against erroneous interpretations may be reference to a sound t h e o r e t i c a l model. Though based on an explanatory model, t h i s study d i d not e s t a b l i s h c a u s a l i t y between any of i t s variables, nor any progressive patterns of stress, a t t r i b u t i o n s f o r negative events, or coping associated with the burnout syndrome. Nevertheless, the study did provide i n t e r e s t i n g differences and re l a t i o n s h i p s that i n v i t e further explanation. Suggestions for Future Studies To t e s t the causal implications of the present findings l o n g i t u d i n a l studies would be required. Preferably, such studies would sample both male and female respondents to control f o r possible defensiveness among males when reporting a t t r i b u t i o n s f o r negative events (Hochreich, 1974). C o n t r o l l i n g f o r defensiveness may be p a r t i c u l a r l y important when using a measure as seemingly transparent as the Expanded ASQ (Bagby et a l . , 1990). Future studies should also include depersonalization i n 103 t h e i r assessment of burnout (Greenglass & Burke, 1988; Ogus et a l . , 1990). Future research involving stress and a t t r i b u t i o n s t y l e should attempt to avoid the conceptual confounding between constructs that may have occurred here. Such confounding would occur when stress was assessed by a s e l f - r e p o r t questionnaire, i f reporting high stress implies an external a t t r i b u t i o n . Should the Expanded ASQ be used i n future work-related studies, averaging the s t a b i l i t y / g l o b a l i t y scores should be considered instead of using only the locus and s t a b i l i t y score i n composite a t t r i b u t i o n variables. The g l o b a l i t y dimension was omitted here because of the length of the Expanded ASQ. I t was assumed that generalizing causes would be less important i n the context of a work-specific construct than i n the context of depression. Results l i n k i n g burnout to universal helplessness tend to confirm that burnout generalizes less than does depression. Nevertheless, the v a l i d i t y of the s t a b i l i t y dimension continues to be a concern. Peterson and V i l l a n o v a (1988) reported stronger v a l i d i t y f o r the g l o b a l i t y dimension than f o r the s t a b i l i t y dimension. The i n c l u s i o n of the g l o b a l i t y dimension might strengthen the v a l i d i t y of future findings. In future work-related studies using the Expanded ASQ, u t i l i z i n g a more domain-specific version of the scale should be considered. A post-hoc factor analysis of the locus and s t a b i l i t y dimensions revealed no domain-specificity f o r the Expanded ASQ. However, informal feedback indicated that the Expanded ASQ was perceived by some respondents to be excessively personal i n a work-related context. The a t t r i b u t i o n s t y l e 104 questionnaire contained more missing data than d i d any of the other scales administered. A scale having greater face v a l i d i t y would l i k e l y have produced subs t a n t i a l l y improved response rates. In future studies, consideration should also be given to using an a t t r i b u t i o n measure that separates the intern a l / e x t e r n a l , stable/unstable a t t r i b u t i o n dimensions onto separate scales, as do factor measures. In addition, future studies using Latack's (1986) coping meansure should explore further the c l u s t e r s reported by Latack (1986) confirmed by L e i t e r (1991). Latack 1s loadings were not c o n s i s t e n t l y confirmed i n t h i s study. P r a c t i c a l Implications Based on comparisons with other populations (Long et a l , 1986), the post-secondary male instructors studied here might be considered a stressed group. However, t h e i r burnout scores were s i m i l a r to those of other teachers (Nagy & Davis, 1985) and other human service populations (Maslach & Jackson, 1981). Because r o l e overload and task stress were the organizational factors that contributed most to emotional exhaustion and low personal accomplishment for t h i s population, workload reductions could be considered among possible burnout remedies for such a group. Universal helplessness i s expected to be l e s s damaging to self-esteem than i s personal helplessness. Nevertheless, as a contributor to burnout, universal helplessness would be expected to a f f e c t workers and t h e i r environments negatively. Preliminary studies l i n k i n g learned-helplessness and work environments have suggested that organizationally-induced helplessness may be p a r t i c u l a r l y damaging i n contexts where personal control i s 105 assumed. Within a context of control over choice of problem and procedural control ( i . e . , control over agendas and procedures), the lack of outcome control ( i . e . , control over the situation) may have ex p e c i a l l y d e b i l i t a t i n g e f f e c t s (Lacey, 1979; Peterson & Seligman, 1983). Apparent control over choice of problem and procedural control are l i k e l y to characterize the work of post-secondary in s t r u c t o r s and perhaps of other helping professionals as w e l l . Should future studies show that behavioural ( i n t e r n a l -unstable) a t t r i b u t i o n s play a role i n the etiology of burnout, such a fi n d i n g would c a l l into question Mclntyre's (1984) suggested remedy for burnout. Mclntyre suggested therapeutically encouraging i n t e r n a l attributions among burned-out teachers. Wong and Sproule (1984) have suggested, on the other hand, that we abandon our control-orientation and learn greater balance between i n t e r n a l and external a t t r i b u t i o n s . Like Wong and Sproule (1984), presenters of burnout workshops have advocated balanced a t t r i b u t i o n s (Pines et a l . , 1981). Maslach (1982b) attributed the remedial power of burnout workshops to t h e i r a b i l i t y to foster a sense of di f f u s e d or shared r e s p o n s i b i l i t y , by s h i f t i n g the focus of the problem from highly personal to more s i t u a t i o n a l causes. Within the population studied, an organizational c r i s i s may have encouraged such shared r e s p o n s i b i l i t y among in t e r n a l a t t r i b u t o r s . Instead of r e l y i n g on either workshops or c r i s e s to modify a t t r i b u t i o n s , Savicki and Cooley (1982) suggested that professional t r a i n i n g should foster a t t r i b u t i o n a l balance i n the assessment of successes and f a i l u r e s . Organizations could 106 s i m i l a r l y encourage a t t r i b u t i o n a l balance by paying attention to expectations of accountability i n performance appraisal systems (Savicki & Cooley, 1983). Individuals with r i s k - r e l a t e d a t t r i b u t i o n a l s t y l e s might be encouraged to locate the cause of f r u s t r a t i o n s more fu n c t i o n a l l y as they process evaluative feedback. I f future studies warrant, cognitive r e t r a i n i n g might also be contemplated for those deemed at greatest r i s k . Evidence suggests that a t t r i b u t i o n s t y l e responds to cognitive intervention i n the treatment of depression. One study (Rush, Beck, Kovacs, & Hollon, 1977) found that modifying a depressive i n d i v i d u a l ' s cognitive s t y l e was more e f f e c t i v e i n a l l e v i a t i n g depressive symptoms than was antidepressive medication. Another study (Shaw, 1977) found that cognitive modification was more e f f e c t i v e i n reducing depressive symptomatology than was ei t h e r behaviour therapy, no treatment, or an attention-placebo therapy. The b e l i e f that a t t r i b u t i o n s r e f l e c t i n g shared r e s p o n s i b i l i t y foster greater emotional health and co-operation (Wong & Sproule, 1984) may be consistent with Freudenberger's (1980) view that intimacy and self-awareness remediate burnout. Nevertheless, u n t i l c l i n i c a l l y s i g n i f i c a n t findings are reported i n the context of longitudinal studies, personal interventions for burnout should not be i n i t i a t e d . Freudenberger (1980) believed that a l l our values and s o c i a l systems would ultimately be challenged i n our attempt to understand, prevent, and remediate burnout. For the moment, however, the l e s s - i n t r u s i v e organizational interventions remain those most c l e a r l y indicated by the l i t e r a t u r e . 107 References Abramson, L. Y., A l l o y . L. B., & Metalsky, G. I. (1986). The cognitive diathesis-stress theories of depression: Toward an adquate evaluation of the theories' v a l i d i t i e s . In L. B. A l l o y (Ed.), Cognitive processes i n depression (pp. 3-30). New York: Gu i l f o r d Press. Abramson, L. Y., Seligman, M. E. P., & Teasdale, J . D. (1978). Learned helplessness i n humans: Crit i q u e and reformulation. Journal of Abnormal Psychology. 87, 49-74. Anderson, M. B. G., & Iwanicki, E. F. (1984). Teacher motivation and i t s r e l a t i o n s h i p to burnout. Educational Administration Quarterly. 2JD(2) , 109-132. Ashforth, B. E. (1989). The experience of powerlessness i n organizations. Organizational Behavior and Human Decision Processes, 43, 207-242. Bagby, R. M., Atkingson, L., Dickens, S., & Gavin, D. (1990). Dimensional analysis of the a t t r i b u t i o n a l s t y l e questionnaire: Attributions or outcomes and events. Canadian Journal of Behavioural Science. 22.(2) , 140-150. Barad, C. B. (1979). Study of burnout syndrome among SSA f i e l d p u b l i c contact employees. Washington, DC: S o c i a l Security Administration, O f f i c e of Management, Budget and Personnel, O f f i c e of Human Resources. Berkeley Planning Associates. (1977). Evaluation of c h i l d abuse and neglect demonstration projects 1974-1977, Volume IX. Project management and worker burnout. Washington, DC: U.W. Department of Commerce. (NTIS not. PB-278 446). Brickman, P., Rabinowitz, V. C , Karuza, J. J r . , Coates, D., Cohn, E., & Kidder, L. (1982). Models of helping and coping. American Psychologist. 37. 368-384. Brookings, J., Bolton, B., Brown, C., & McEvoy, A. (1986). Self-reported job burnout among female human service professionals. Journal of Occupational Behaviour. 6(2), 143-150. Cannon, W. B. (1929). Organization for p h y s i o l o g i c a l homeostasis. Psychological Review. 9, 339-430. Capel, S. (1987). The incidence of and influences on stress and burnout i n secondary school teachers. B r i t i s h Journal of Educational Psychology. 57. 279-288. 108 Caplan, R. D., Cobb, S., French, J. R. P., J r . , Harisson, R. V., & Pineau, S. R., J r . (1975). Job demands and worker health: Main e f f e c t s and occupational diff e r e n c e s . U.S. Dept. of H.E.W., Washington, DC. C a r r o l l , J . , & White, W. (1982). Theory b u i l d i n g : Intregrating i n d i v i d u a l and environmental factors with an e c o l o g i c a l framework. In W. S. Paine (Ed.), Job stress and burnout: Research, theory and intervention perspectives (pp. 41-60). Beverly H i l l s , CA: Sage. Champoux, J . E., & Peters, W. (1987). Form, e f f e c t s i z e and power i n moderated regression analysis. Journal of Occupational Psychology. 60, 243-255. Cherniss, C. (1980). Staff burnout: Job stress i n the human services. Beverly H i l l s , CA: Sage. Chronbach, L. J . , & Snow, R., E., (1977). Aptitudes and i n s t r u c t i o n a l methods: A handbook for research on in t e r a c t i o n s . New York: Irvington. Cohen, J . , & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences. H i l l s d a l e , NJ: Erlbaum. Cohen, J . (1988). S t a t i s t i c a l power analysis f o r the behavioral sciences, 2nd Ed. New York: Academic Press. Constable, J . R., & Russell, D. (1986). The e f f e c t of s o c i a l support and the work environment upon burnout among nurses. Journal of Human Stress. 12., 20-26 Cooley, E. J . , & Savicki, V. (1987). Therapist's a t t r i b u t i o n s about treatment outcome: The development of a scale. Unpublished manuscript. Western Oregon State College, Monmouth, Ore. Coyne, J . , Metalsky, G., & Lavelle, T. L., (1980). Learned helplessness as experimenter-induced f a i l u r e and i t s a l l e v i a t i o n with attentional redeployment. Journal of Abnormal Psychology. 89, 350-357. Crane, S. (1981). An analysis of the r e l a t i o n s h i p among personal and professional variables, r o l e ambiguity, r o l e c o n f l i c t , and percieved burnout of s p e c i a l education teachers. Unpublished doctoral d i s s e r t a t i o n , University of Connecticut. Crane, S., & Iwanicki, E. F. (1986). Perceived r o l e c o n f l i c t , r o l e ambiguity and burnout among sp e c i a l education teachers. RASE: Remedial and Special Education. 7(2), 24-31. 109 Cunningham, W. G. (1983). Teacher burnout - solutions f o r the 1980's : A review of the l i t e r a t u r e . Urban Review. 15, 37-51. Dignam, J . T., Barrera, M., J r . , & West, S. G., (1986). Occupational stress, s o c i a l support, and burnout among cor r e c t i o n a l o f f i c e r s . American Journal of Community Psychology. 14, 177-193. Dweck, C. S., & Licht, B. (1980). Learned helplessness and i n t e l l e c t u a l achievement. In J. Garber & M. E. P. Seligman (Eds.), Human helplessness (pp. 197-221). New York: Academic Press. Dworkin, A. G. (1987). Teacher burnout i n the p u b l i c schools: Structural causes and consequences for c h i l d r e n . New York: State University of New York Press. Edelwich, J . , & Brodsky, A. (1980). Burn-out: Stages of disillusionment i n the helping professions. New York & London: Human Sciences Press. E i n s i e d e l , A. A., J r . , & T u l l y , H. A. (1981). Methodological considerations i n studying the burnout phenomenon. In J. W. Jones (Ed.), The burnout syndome: Current research, theory, interventions (pp. 89-106). Park Ridge, 111: London House Press. Etzion, D., & Pines, A. (1986). Sex and culture i n burnout and coping among human service professionals. Journal of Cross-Cultural Psychology. 17(2), 191-209. Finney, J . W., M i t c h e l l , R., E., Cronkite, R. C., & Moos, R. H. (1984). Methodological issues i n estimating main and i n t e r a c t i v e e f f e c t s : Examples from coping/ s o c i a l support and stress f i e l d . Journal of Health and S o c i a l Behavior. 25. 85-98. Folkman, S. (1982). An approach to the measurement of coping. Journal of Occupational Behavior. 3., 95-108. Folkman, S., & Lazarus, R. S. (1980). An analysis of coping i n a middle-aged community sample. Journal of Health and S o c i a l Behavior. 21. 219-239. Folkman. S., & Lazarus, R. S. (1985). I f i s changes i t must be a process: Study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology. 50. 571-579. Folkman, S., Lazarus, R. S., Dunkel-Schetter, C., Delongis, A., & Gruen, R. J. (1986). Dynamics of a s t r e s s f u l encounter: Cognitive appraisal, coping, and encounter outcomes. Journal of Personality and Social Psychology. 50. 992-1003. 110 Freudenberger, H. J . (1974). Staff burn-out. Journal of Social Issues. 30(1), 159-165. Freudenberger, H, J . (1977). Burn-out: The organizational menace. Training and Development Journal. 31. 26-27. Freudenberger, H. J. (1980). Burn-out: The high cost of high achievement. New York: Anchor Press. Freudenberger, H. J. (1983). Burnout: Contemporary issues, trends, and concerns. In B. A. Farber (Ed.), Stress and burnout i n the human services professions (pp. 23-28). New York: Permagon Press. Gaines, J., & Jermier, J. M. (1983). Emotional exhaustion i n a high stress organization. Academy of Management Journal. 26, 567-586. Gold, Y. (1985). The relationship of s i x personal and l i f e h i s t o r y variables to standing on three dimensions of the Maslach Burnout Inventory i n a sample of elementary and junior high school teachers. Educational and Psychological Measurement. 45. 377-387. Golembiewski, R.T., Munzenrider, R. F., & Stevenson, J . G. (1986). Stress i n organizations: Toward a phase model of burnout. New York: Praeger. Greenglass, E. R., & Burke, R. J. (1988). Work and family precursors of burnout i n teachers: Sex differences. Sex Roles. 18. 215-229. Hackman, J . R., & Oldham, G. R. (1974). The job diagnostic survey: An instrument for the diagnosis of jobs and the evaluation of job redesign projects. JSAS Catalog of Selected Documents i n Psychology. 4., 148. (Ms. No. 810). Hammen, C., & deMayo, R. (1982). Cognitive co r r e l a t e s of teacher stress and depressive symptoms: Implications for a t t r i b u t i o n a l models of depression. Journal of Abnormal Psychology. 91, 96-101. Herr, E. L., & Cramer, S. H. (1984). Career guidance and counselling throughout the l i f e span: Systematic approaches. 2nd ed. Boston: L i t t l e , Bronw & Company. Hiroto, D. S. (1974). Locus of control and learned helplessness. Journal of Experimental Psychology. 102. 187-193. Hiroto, D. S., & Seligman, M. E. P. (1975). Generality of learned helplessness i n man. Journal of Personal and Social Psychology. 31, 311-327. I l l Hochreich, D. J. (1974). Defensive e x t e r n a l i t y and a t t r i b u t i o n of r e s p o n s i b i l i t y . Journal of Personality. 42. 543-557. Holt, P., Fine, M. J., & Tollefson, N. (1987). Mediating str e s s : Survival of the hardy. Psychology i n the Schools f 24/ 51-58. House, J . S. (1981). Work stress and s o c i a l support. Reading, MA: Addison-Wesley. Iwanicki, E. R., & Schwab, R. L. (1981). A cr o s s - v a l i d a t i o n study of the Maslach Burnout Inventory. Educational and Psychological Measurement. 41, 1167-1174. K a h i l i , S. (1988). Symptoms of professional burnout: A review of the empirical evidence. Canadian Psychology. 2j9(3) , 284-297. Kanner, A. D., Coyne, J. C., Schaefer, C , & Lazarus, R. S. (1981). Comparison of two modes of stress measurement: Daily hassles and u p l i f t s versus major l i f e events. Journal of Behavioral Medicine. 4(1), 1-37. Kle i n , D. C., & Seligman, M. E. P. (1976). Reversal of performance d e f i c i t s i n learned helplessness and depression. Journal of Abnormal Psychology. 85, 11-26. Kofta, M., & Sedek, G. (1989). Repeated f a i l u r e : A source of helplessness or a factor i r r e l e v a n t to i t s emergence? Journal of Experimental Psychology: General. 118, 3-12. Krause, N. (1986). Stress and coping: Reconceptualizing the r o l e of locus of control b e l i e f s . Journal of Gerontology. 41(5), 617-622. Krause, N., & Stryker, S. (1984). Stress and well-being: The buffering r o l e of locus of control b e l i e f s . S o c i a l Science Medicine. 18. 783-790. Kremer-Hayon, L., & Kurtz, H. (1985). The r e l a t i o n of personal and environmental variables to teacher burnout. Teaching and Teacher Education: An International Journal of Research and Studies. 1, 243-249. Kyriacou, C. (1987). Teacher stress and burnout: An int e r n a t i o n a l review. Educational Research. 29(2), 146-152. Lacey, H. M. (1979). Control, perceived control and the methodological r o l e of cognitive constructs. In L. C. Perlmuter & R. A. Monty (Eds.), Choice and perceived control (pp. 5-16). H i l l s d a l e , NJ: Erlbaum. 112 Latack, J . C. (1984). Career t r a n s i t i o n within organizations: An exploratory study of work, nonwork and coping strat e g i e s . Organizational Behavior and Human Performance. 34, 296-322. Latack, J . C. (1986). Coping with job stress: Measures and future d i r e c t i o n s for scale development. Journal of Applied Psychology. 71, 377-385. Lazarus, R. S., & Folkman, S., (1984). Stress, appraisal, and coping. New York: Behavioral Science Books. Lee, R. T., & Ashforth, B. E. (1990). On the meaning of Maslach's three dimensions of burnout. Journal of Applied Psychology. 75. 743-747. Lee, R. T., & Ashforth, B. E. (In press). A l o n g i t u d i n a l study of burnout among supervisors and managers: Comparisons between the L e i t e r and Maslach (1988) and Golembiewski et a l . , (1986) models. Organizational Behavior and Human Decision Processes. Lefcourt, H. M. (1982). Locus of control: Current trends i n theory and research (2nd ed.). H i l l s d a l e , NJ: Erlbaum. Lefcourt, H. M., Martin, R. A., & Saleh, W. E. (1984). Locus of co n t r o l and s o c i a l support: Interactive moderators of s t r e s s . Journal of Personality and Social Psychology. 47, 378-398. Lefcourt, H. M., Martin, R. A., & Ware, E. E. (1984). Locus of co n t r o l , causal a t t r i b u t i o n s , and a f f e c t s i n achievement-related contexts. Canadian Journal of Behavioural Sciences. 16(1), 57-64. L e i t e r , M. P. (1991). Coping patterns as predictors of burnout: The function of control and escapist coping patterns. Journal of Organizational Behaviour. 12, 123-144. L e i t e r , M. P., & Maslach, C. (1988). The impact of interpersonal environment on burnout and organizational commitment. Journal of Organizational Behavior. 9, 297-308. Long, B. C., Schutz, R. W., Kendall, P. R., & Hunt C. G. (1986). Employee health service for teaching and nonteaching s t a f f : A needs assessment. Evaluation and Program Planning. 9, 355-366. Malanowski, G. R., & Wood, P. H. (1984). Burnout and s e l f - a c t u a l i z a t i o n i n public school teachers. Journal of Psychology. 117. 23-26. Marsh, H. W. (1984). Relations among dimensions of s e l f -a t t r i b u t i o n , dimensions of self-concept, and academic achievements. Journal of Educational Psychology. 76, 1291-1308. 113 Marx, M. B., Garrity, T. F., & Bowers, F. R. (1975). The influence of recent l i f e experiences on the health of co l l e g e freshmen. Journal of Psychosomatic Research. 19, 87-98. Maslach, C. (1977 August). Burn-out: A s o c i a l psychological a n a l y s i s . Paper presented at annual convention of American Psychological Association, San Francisco. Maslach, C. (1982a). Burnout i n health professionals: A s o c i a l psychological analysis. In G. S. Sanders & J . Seels (Eds.), S o c i a l Psychology of Health and I l l n e s s (pp. 227-251). London: Lawrence Erlbaum Associates. Maslach, C. (1982b). Burnout: The cost of caring. Englewood C l i f f s , NJ: Prentice-Hall. Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of Occupational Behavior. 2, 99-113. Maslach, C., & Pines, A. (1977). A burn-out syndrome i n the day care s e t t i n g . Child Care Quarterly. 6, 100-113. Mclntyre, T. (1984). The r e l a t i o n s h i p between locus of control and teacher burnout. B r i t i s h Journal of Educational Psychology. 5(2) 235-238. McMullen, M. B., & Krantz, M. (1988). Burnout i n day care workers: The e f f e c t s of learned helplessness and s e l f -esteem. Child & Youth Care Quarterly. 17(4), 275-280. Metalsky, G. I., Halberstadt, L. J., & Abramson, L Y. (1987). V u l n e r a b i l i t y to depressive mood reactions: Toward a more powerful t e s t of the diathesis-stress and causal mediation components of the reformulated theory of depression. Journal of Personality and Social Psychology. 52, 386-393. Mikulincer, M. (1988). The r e l a t i o n between stable/unstable a t t r i b u t i o n and learned helplessness. B r i t i s h Journal of S o c i a l Psychology. 27. 221-230. Mikulincer, M. (1989a). Causal a t t r i b u t i o n , coping strategies, and learned helplessness. Cognitive Therapy and Research. 13, 565-582. Mikulincer, M. (1989b). Coping and learned helplessness: E f f e c t s of coping strategies on performance following unsolvable problems. European Journal of Personality. 3., 181-194. Moos, R. H., & B i l l i n g s , A. G. (1982) . Conceptualizing and measuring coping resources and process. In L. Goldberg & S. Breznitz (Eds.) Handbook of stress: Theoretical and c l i n i c a l aspects (pp. 212-230). New York: Free Press. 114 Monat, A., & Lazarus, R. S. (Eds.). (1977). Stress and coping; An anthology. New York: Columbia University Press. Nagy, S., & Davis, L. G. (1985). Burnout: A comparative analysis of personality and environmental v a r i a b l e s . Psychological Report. 57, 1319-1326. Nowicki, S., & Strickland, B. R. (1973). A locus of control scale for adults. Journal of Consulting and C l i n i c a l Psychology. 40. 148-154. Ogus, E. D., Greenglass, E. R., & Burke, R. J. (1990). Gender-r o l e differences, work stress and depersonalization. Journal of Social Behavior and Personality. 5, 387-398. Overmier, J . B., & Seligman, M. E. P. (1967). E f f e c t s of inescapable shock upon subsequent escape and avoidance learning. Journal of Comparative and Ph y s i o l o g i c a l Psychology, 63. 28-33. Peck, M. S. (1978). The road less traveled: A new psychology of love, t r a d i t i o n a l values and s p i r i t u a l growth. New York: Simon and Schuster. Pedhazur, E. J. (1982). Multiple regression i n behavioral research: Explanation and prediction (2nd ed). NY: Holt, Rinehart & Winston. Perlman, B., & Hartman, E. A. (1982). Burnout: Summary and future research. Human Relations. 35(4), 283-305. Peterson, C , Schwartz, S. A., & Seligman, M. E. P. (1981). Self-blame and depressive symptoms. Journal of Personality and S o c i a l Psychology. 41, 253-259. Peterson, C., & Seligman, M. E. P. (1983). Learned helplessness and v i c t i m i z a t i o n . Journal of S o c i a l Issues. 2, 103-116. Peterson, C , & Seligman, M. E. P. (1984). Causal explanations as a r i s k factor for depression: Theory and evidence, Psychological Review. 91, 347-374. Peterson, C., Semmel, A., von Baeyer, E., Abramson, L. Y., Metalsky, G. I., & Seligman, M. E. P. (1982). The A t t r i b u t i o n a l Style Questionnaire. Cognitive Therapy and Research. 6, 287-299. Peterson, C , & Villanova, P. (1988). An expanded a t t r i b u t i o n a l s t y l e questionnaire. Journal of Abnormal Psychology. 97. 87-89. 115 Peterson, C., Villanova, P., & Raps, C. S. (1985). Depression and a t t r i b u t i o n s : Factors responsible for inconsistent r e s u l t s i n the published l i t e r a t u r e . Journal of Abnormal Psychology. 94, 165-168. Pettegrew, L. S., & Wolf, G. E. (1982). V a l i d a t i n g measures of teacher s t r e s s . American Educational Research Journal. 19, 373-396. Pines, A. (1985). The burnout measure. In J . W. Jones (Ed.), P o l i c e burnout: Theory, research, and a p p l i c a t i o n . Park Ridge, IL: London Management Press. Pines, A. M., & Kafry, D. (1978). Occupational tedium i n the s o c i a l services. Social Work. 23. 499-507. Pines, A. M., Aranson, E., & Kafry, D. (1981). Burnout: From tedium to personal growth. New York: The Free Press. Pines, A. M., & Maslach, C. (1978). C h a r a c t e r i s t i c s of s t a f f burn-out i n mental health settings. Hospital & Community Psychiatry. 29, 233-237. Pittman, N. L., & Pittman, T. S. (1979). E f f e c t s of amount of helplessness t r a i n i n g and internal-external locus of control on mood and performance. Journal of Personality and Social Psychology. 37, 39-47. Quigley, T. A., Slack, T., & Smith, G. J . (1987). Burnout i n secondary school teacher coaches. The Alberta Journal of Educational Research. 33.(4), 260-274. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Rotter, J . B. (1966). Generalized expectancies f o r i n t e r n a l versus external control of reinforcement. Psychological Monographs: General and Applied. 80(1), 1-28. Rotter, J . B. (1975). Some problems and misconceptions related to the construct of internal versus external control of reinforcement. Journal of Consulting and C l i n i c a l Psychology. 43., 56-67. Rothbaum, F., Wesz, J. R., & Snyder, S. S. (1982). Changing the world and changing the s e l f : A two-process theory of perceived control. Journal of Personality and S o c i a l Psychology. 2, 5-37. Rush, A. J., Beck, A. T., Kovacs, M., & Hollon, S. (1977). Comparative e f f i c a c y of cognitive therapy and pharmacotherapy i n the treatment of depressed out-patients. Cognitive Therapy and Research. 1, 17-37. 116 Rus s e l l , C. W., Altmaier, E., & Van Velzen, D. (1987). Job-related stress, s o c i a l support, and burnout among classroom teachers. Journal of Applied Psychology. 72. 269-274. Russell, D. (1982). The causal dimension scale: A measure of how in d i v i d u a l s perceive causes. Journal of Personality and Soc i a l Psychology. 42. 1137-1145. Sacks, C. H., & Bugental, B. (1987). A t t r i b u t i o n s as moderators of a f f e c t i v e and behavioral responses to s o c i a l f a i l u r e . Journal of Personality and Social Psychology. 53., 939-947. Sandler, I., Reese, F., Spencer, L. , & Harpin, P. (1981). Person by environment int e r a c t i o n and locus of co n t r o l : Laboratory, therapy, and classroom studies. In H. M. Lefcourt (Ed.), Research with the locus of control construct (Vol. 3) (pp. 309-360). New York: Academic Press. Savicki, V., & Cooley, E. J . (1982). Implications of burnout research and theory for counselor educators. The Personnel and Guidance Journal. 60(7), 415-419. Sav i c k i , V., & Cooley, E. J . (1983). Theoretical and research considerations of burnout. Children and Youth Services Review, 5, 227-238. Schlansker, B. (1987). A p r i n c i a l ' s guide to teacher s t r e s s . P r i n c i p a l . 66(5), 32-34. Schutz, R. W., & Long, B. C. (1988). Confirmatory fa c t o r analysis, v a l i d a t i o n and r e v i s i o n of a teacher stress inventory, Educational and Psychological Measurement. 48, 154-165. Schwab, R. L. (1983). Teacher burnout: Moving beyond "psychobabble". Theory into Practice. 22.(1), 21-26. Schwab, R. L., & Iwanicki, E. F. (1982). Who are the burned out teachers? Educational Research Quarterly. 7, 5-16. S e i l e r , R. , & Pearson, D. (1984-1985). Dysfunctional stress among un i v e r s i t y faculty. Educational Research Quarterly. 9(2), 15-26. Seligman, M. E. P. (1975). Helplessness: On depression, development, and death. San Francisco, CA: Freeman. Seligman, M. E. P., Abramson, L. Y., Semmel, A., & von Baeyer, C. (1979). Depressive a t t r i b u t i o n a l s t y l e . Journal of Abnormal Psychology. 88. 242-247. 117 Seligman, M. E. P., & Maier, S. F. (1967). F a i l u r e to escape traumatic shock. Journal of Experimental Psychology, 74, 1-9. Seligman, M. E. P., & Schulman, P. (1986). Explanatory s t y l e as a predictor of productivity and q u i t t i n g among l i f e insurance sales agents. Journal of Personality and Social Psychology. 50, 832-838. Shaw, B. F. (1977). Comparison of cognitive therapy and behavior therapy i n the treatment of depression. Journal of Consulting and C l i n i c a l Psychology. 45. 543-551. Shirom, A. (1989). Burnout i n work organizations. In C. L. Cooper & I. Robertson (Eds.), International review of i n d u s t r i a l and organizational psychology (pp. 25-48). Chichester: Wiley. Shreeve, C. (1984). Depression: Its causes and how to overcome i t . Wellingborough, Northamptonshire: Turnstone Press Limited. Sweeney, P. D., Anderson, K., & Bailey, S. (1986). A t t r i b u i t o n a l s t y l e i n depression: A meta-analytical review. Journal of Personality and Soc i a l Psychology. 50, 974-991. Teasdale, J . D. (1978). Effects of r e a l and r e c a l l e d success on learned helplessness and depression. Journal of Abnormal Psychology. 87, 155-164. V i t a l i a n o , P. P., Maiuro, R. D., Russo, J., & Becker, J . (1987). Raw versus r e l a t i v e scores i n the assessment of coping st r a t e g i e s . Journal of Behavioral Medicine. 10, 1-18. Wade, D. C., Cooley, E., & Savicki, V. (1986). A long i t u d i n a l study of burnout. Children and Youth Services Review. 8(2), 161-173. Weiner, B. (1972). Theories of motivation: From mechanism to cognition. Chicago: Rand McNally. Weiner, B. (1974). Achievement motivation and a t t r i b u t i o n theory• Morristown, NJ: General Learning Press. Weiner, B., Frieze, I., Kukla, A., Reek, L., Rest, S., & Rosenbaum, R. M. (1971). Perceiving the causes of success and f a i l u r e . In E. E. Jones, D. E. Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, & B. Weiner (Eds.), A t t r i b u t i o n : Perceiving the causes of behavior (pp. 95-120). Morristown, NJ: General Learning Press. Wolpin, J . , Burke, R. J., & Greenglass, E. R. (1991). Is job s a t i s f a c t i o n an antecedent or a consequence of psychological burnout? Human Relations. 44(10), 193-209. 118 Wong, P. T. P., & Sproule, C. F. (1984). An a t t r i b u t i o n analysis of the locus of control construct and the Trent A t t r i b u t i o n P r o f i l e . In H. M. Lefcourt (Ed.), Research with the locus of control construct (Vol. 3) (pp. 309-360). New York: Academic Press. Wong, P. T. P., & Weiner, B. (1981). When people ask "why" questions, and the h e u r i s t i c of a t t r i b u t i o n a l research. Journal of Personality and Social Psychology. 40, 650-663. Wortman, C. B., & Brehm, J. W. (1975). Responses to uncontrollable outcomes: An integration of reactance theory and the learned helplessness model. In L. Berdowitz (Ed.), Advances i n experimental s o c i a l psychology (Vol. 8) (pp. 277-336). New York: Academic Press. 119 Appendices Appendix A: Informed consent attachments, reminders and i n s t i t u t e endorsement. Cover Letter - f i r s t questionnaire d i s t r i b u t i o n JOB STRESS AND COPING AT ... May, 1988 Purpose of t h i s study: This study i s part of a Masters' thesis i n v e s t i g a t i n g job stress and coping at.... I t s purpose i s to explore ways of reducing i n s t r u c t o r stress i n the ... environment. The questions attached assess both the ways in d i v i d u a l s deal with s t r e s s f u l events and the way the environment a f f e c t s l e v e l s of stress and i n d i v i d u a l reactions to i t . Procedure: As a p a r t i c i p a n t , you are asked to complete t h i s paper and pencil questionnaire. I t w i l l take you approximately 3 0 minutes. Conf ident i a 1 i t y : Please submit your questionnaire anonymously. A l l raw data w i l l be kept c o n f i d e n t i a l and w i l l be destroyed when i t has been analyzed. Results w i l l be reported i n group averages. Consent: By returning your completed questionnaire you w i l l have consented to p a r t i c i p a t e i n the investigation. You may refuse to p a r t i c i p a t e or to answer any s p e c i f i c items or questions i n the questionnaire. Results: Analyzed c o l l e c t i v e l y , the r e s u l t s w i l l be av a i l a b l e to in d i v i d u a l s through t h e i r union and management representatives. If you wish to receive a summary of the r e s u l t s , please return the attached memo separately. Returns : Please return your completed questionnaire by May 16 to ... eit h e r i n person or through the inter n a l mail. Questions: If you have any questions, please contact the investigators: 120 E i l e e n Stephens Dr. Bonita Long Counselling Psychology, UBC (228-4756) F i r s t Reminder Questionnaire Reminder Have you completed your "Stress and Coping" Questionnaire yet? Your response i s important. I f you've already returned your questionnaire, thank you; please disregard t h i s notice. I f not, remember, your response i s important. To report accurately on stress and i t s impact on i n s t r u c t o r morale at we need a representative response. Do you need a fresh copy? I f you need a fresh copy, please c a l l and I ' l l be happy to send one to you. Or you can pick one up yourself i n .... Your decision to p a r t i c i p a t e . Your p a r t i c i p a t i o n i n t h i s survey i s voluntary. I f you do p a r t i c i p a t e , you may leave blank any s p e c i f i c questions you do not want to answer. Conf i d e n t i a 1 i t y • Your questionnaire i s anonymous. I t w i l l be handled c o n f i d e n t i a l l y u n t i l the data i s analyzed; then i t w i l l be destroyed. Only group averages w i l l appear i n the f i n a l report. Returning the questionnaire. Please return your completed questionnaire to ... e i t h e r i n person or through the i n t e r n a l mail. Questions? I f you need anything c l a r i f i e d , please contact E i l e e n Stephens Dr. Bonita Long Counselling Psychology, UBC (228-4756) 121 Second Reminder "Stress and Coping" Questionnaire Reminder #2 Did your questionnaire get l o s t i n the press of recent events? Your response i s needed. To answer the research questions underlying the study, I s t i l l need 15 more su b s t a n t i a l l y completed returns. If you're among those who have responded already, thank you. I hope to have some int e r e s t i n g findings to report to you. I f not, please take some time now to complete the questionnaire. Do you need a fresh copy? I f you need a fresh copy, please leave a message with ..., and I ' l l be happy to send one to you. Or, i f you prefer, you can pick one up yourself i n .... C o n f i d e n t i a l i t y . Your questionnaire i s anonymous. I t w i l l be handled c o n f i d e n t i a l l y u n t i l the data i s analyzed; then i t w i l l be destroyed. Only group averages w i l l appear i n the f i n a l report. P a r t i c i p a t i o n . Your p a r t i c i p a t i o n i s voluntary. As a p a r t i c i p a n t , you may refuse to answer any s p e c i f i c questions i n the questionnaire. Returning the questionnaire. Please return your completed questionnaire by to ..., e i t h e r i n person or through the i n t e r n a l mail. Questions? If you need anything c l a r i f i e d , please c a l l me at .... 122 Endorsement - included with second questionnaire d i s t r i b u t i o n May, 1988 Stress Survey Endorsement E a r l i e r t h i s month you were selected as a member of the sample group for a study on stress and coping among in s t r u c t o r s at .... Having reviewed the study's questionnaire and the research proposal guiding i t s use, we want to encourage you to take the time, i f you haven't already done so, to p a r t i c i p a t e i n t h i s study. The events of the past month have understandably diminished response rates. However, we want to remind you that the r e s u l t s , i f representative, w i l l be of value to the ... community. The study uses questions developed by North American researchers. The questions included are being used to t e s t some s p e c i f i c hypotheses concerning the ways you and your environment inter a c t to a f f e c t your wellbeing at .... I f you have already responded, thank you f o r your cooperation. There i s no need to send i n a second response. I f not, a questionnaire i s enclosed here for your convenience. 123 Cover Letter - second questionnaire d i s t r i b u t i o n JOB STRESS AND COPING AT ... May, 1988 Purpose of t h i s study: This study i s part of a Masters' thesis i n v e s t i g a t i n g job stress and coping at .... Its purpose i s to explore ways of reducing i n s t r u c t o r stress i n the ... environment. The questions attached assess both the ways in d i v i d u a l s deal with s t r e s s f u l events and the way the environment a f f e c t s l e v e l s of stress and i n d i v i d u a l reactions to i t . Procedure : As a p a r t i c i p a n t , you are asked to complete t h i s paper and pe n c i l questionnaire. I t w i l l take you approximately 30 minutes. C o n f i d e n t i a l i t y : Please submit your questionnaire anonymously. A l l raw data w i l l be kept c o n f i d e n t i a l and w i l l be destroyed when i t has been analyzed. Results w i l l be reported i n group averages. P a r t i c i p a t i o n : Your p a r t i c i p a t i o n i s voluntary. As a p a r t i c i p a n t , you may refuse to answer any s p e c i f i c items or questions i n the questionnaire. Results: Analyzed c o l l e c t i v e l y , the res u l t s w i l l be a v a i l a b l e to in d i v i d u a l s through t h e i r union and management representatives. If you wish to receive a summary of the r e s u l t s , please return the attached memo separately. Returns : Please return your completed questionnarie by to ..., either i n person or through the inter n a l mail. Questions: If you need anything c l a r i f i e d , please c a l l me at.... 124 Appendix B: Advertising. A r t i c l e - May 2, 1988 INSTRUCTOR LOOKS AT ... STRESS LEVELS With stress at i t s present l e v e l , a timely study i s being conducted on how ... faculty copes with stress at the .... The study forms part of a M.A. thesis being conducted by ... i n s t r u c t o r E i l e e n Stephens who says that while numerous studies have examined stress i n an educational s e t t i n g , few have dealt with i t i n the context of highter education. Dr. Bonita Long of the UBC Counselling Psychology Department i s Stephens' co-researcher on the project. "We w i l l investigate not only the ways i n d i v i d u a l s i n the ... environment deal with s t r e s s f u l events, but also how the environment a f f e c t s l e v e l s of stress and i n d i v i d u a l reactions to i t , " says Stephens. She i s currently d i s t r i b u t i n g a questionnaire to a sample of 250 ... f a c u l t y members. Summaries of the findings w i l l be mailed to questionnaire respondents and w i l l also be available to f a c u l t y through t h e i r management or union representatives. If you receive one on Eileen's questionnaires, please complete i t (without g i v i n g your name) and return i t to .... The questionnaire takes approximately 45 minutes to complete. For further information c a l l Eileen Stephens ... or Dr. Bonnie Long (228-4657). A r i t i c l e - May 5, 1988 HOW ARE YOU COPING? Management and both unions ... have given t h e i r blessing to a study on coping with work-related stress at .... The study w i l l draw on a body or research which attempts to i s o l a t e the most useful ways of dealing i t h work pressure and i t s impact on the i n d i v i d u a l . Most research on teacher stress has been done i n the school system where burnout i s a growing concern. L i t t l e attention has been paid to i n s t i t u t i o n s of higher education. A ... -based study i s timely and should be of inte r e s t to us a l l . I f you receive a questionnaire, please take some time to complete i t . ... w i l l be c o l l e c t i n g c o n f i d e n t i a l returns. A r t i c l e - May 24, 1988 STRESS ... i n s t r u c t o r Eileen Stephens s t i l l needs 15 more s u b s t a n t i a l l y completed questionnaires for her survery on stress and coping. I f you receive a questionnaire and have not yet f i l l e d i t out, please take a few minutes to do so. Your help i s much appreciated by Eileen, who w i l l be using the information for her M.A. t h e s i s on stress among ... i n s t r u c t o r s . 125 Appendix C: Instruments. A t t r i b u t i o n s of Treatment Outcomes Factor 1: External Source of Change Some r e c i p i e n t s do not want to change and there i s nothing I can do to modify t h i s . Some c l i e n t s cannot be reached no matter what you do. The source of improvement i s a person under my care comes completely from within them, when they are ready to change, they w i l l . I f e e l l i k e helping some c l i e n t s i s mostly a matter of luck getting them just at the r i g h t time. I f e e l I have l i t t l e power to change most of my c l i e n t s . If I am not getting r e s u l t s with a r e c i p i e n t I usually know i t s due to factors i n the reci p i e n t themselves or i n t h e i r l i v e s over which I have l i t t l e control. Some of my c l i e n t s seem to get better because of placebo e f f e c t s rather than any s p e c i f i c thing I do. Some r e c i p i e n t s know where they want to go and I don't have much impact upon these decisions. External factors (Family, Friend, Job) usually explain most of the improvements I see i n my c l i e n t s . Factor 2 : Personal Impact When people under my care show improvement I f e e l a sense of personal accomplishment. I have a d i r e c t and powerful impact upon the people under my care. Frequently, when a c l i e n t gets better, i t ' s because of what I have done with them. Almost anything I do with c l i e n t s can have an impact upon t h e i r mental health. If I am not getting r e s u l t s with a r e c i p i e n t I t r y to work harder with t h i s person. 126 Factor 3: Personal Responsibility I f e e l i t i s my personal r e s p o n s i b i l i t y to see that r e c i p i e n t s show some improvements. I can't help but f e e l that I have f a i l e d when one of my r e c i p i e n t s gets worse. If one of the people I work with has f a i l e d to change, i t ' s problably because I have not worked hard enough or c a r e f u l l y enough with them. At times I f e e l helpless to do anything to help the people under my care. (Reversed scoring) Factor 4: Do No Harm If I behave c a r e l e s s l y or say the wrong thing with c l i e n t s i t can cause them to get worse. Very l i t t l e of what I say or do with r e c i p i e n t s could cause them to get worse. (Reversed scoring) I have to be very c a r e f u l about what I say and do with r e c i p i e n t s because I can cause more pain and su f f e r i n g . Questionnaire Administered STRESS AND COPING QUESTIONNAIRE This questionnaire consists of five sections: job stress, interpretation of l i f e events, coping, emotional response, and demographic data. Job Stress Please indicate how much you agree or disagree with the following statements by c i r c l i n g the appropriate number. I can predict what w i l l be expected of me in my work tomorrow. I am unclear on what the scope and r e s p o n s i b i l i t i e s of my job are. I am uncertain what the c r i t e r i a for evaluating my performance actually are. I receive enough information to carry out my job effectively. When asked, I am able to t e l l someone exactly what the demands of my job are. I feel that my job interferes with my family l i f e . I feel constant pressure from others to improve the quality of my work. I find that I have extra work beyond what should normally be expected of me. 01 >1 u r H <2> r H >1 > IT) r H c •p 18 0 c >i e - H <u •P r H 0 01 3 in •0 •o ifl o< 0 U r H 0 <u B 10 0) o u r-l X w o < 1 2 3 4 5 9. The c r i t e r i a of performance for my job are too high. 10. I am given too much re-sponsibility without adequate authority to carry i t out. 11. I receive c o n f l i c t i n g demands from two or more people or groups in the i n s t i t u t e setting. 12. I have to buck a rule or policy in order to carry out an assignment. 13. I have a hard time satisfying the c o n f l i c t i n g demands of students, administrators and instructors. 14. I am given work-related duties without adequate resources and materials to carry them out. 15. There i s a difference between the way my administrative head thinks things should be done and the way I think they should be done. 16. My fellow faculty members and I regularly have time during work hours to discuss job-related issues. 17. I have influence over what goes on in the i n s t i t u t e . 18. I'm informed of important things that are happening in the i n s t i t u t e . 129 u > CD U 10 S >i rH i - l >1 « rH C •P 0 C e • H a 0 Cfl 3 rs (0 D 1 rH 0 a» a) 0 u w o ta 10 •P ta o B 19. My administrative head asks my opinion on decisions that d i r e c t l y affect me. 20. I fee l that i t i s useless to make suggestions about my work because decisions are made regardless of my attempts to influence them. 1 2 3 4 5 21. A l l i n a l l , I would say that I am extremely s a t i s f i e d with my job. 22. My job i s extremely important in comparison with other interests in my l i f e . 23. Knowing what I know now, i f I had to decide a l l over again whether to take this job, I would d e f i n i t e l y do so. 24. In general, my job measures up extremely well with the sort of job I wanted before I took i t . 25. If a good friend told me that (s)he was interested in taking a job here, I would have serious reservations about recommending i t . 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 26. Trying to complete marks and paper work on time causes me a lot of stress. 27. I find that dealing with student di s c i p l i n e problems puts a lot of stress on me. 1 2 3 4 5 1 2 3 4 5 130 01 >. >i U rH m rH 3 > «J rH r H m on nt (0 e • H ai •p r H 0 01 3 01 •a Tt (0 O1 0 u r H 0 ai e <a ai 0 vi rH x w o b 28. Trying to provide a good education in an atmosphere of decreasing financial support i s very s t r e s s f u l . 29. There i s a l o t of stress just keeping up with changing professional standards. 30. Trying to keep my work from being too routine and boring puts a lot of stress on me. 31. Having to participate in inst i t u t e a c t i v i t i e s outside of the normal working hours i s very stressful to me. 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 32. I find that trying to be attentive to the problems and needs of fellow faculty i s very str e s s f u l . 1 2 3 4 5 33. When I r e a l l y need to talk to my administrative head, (s)he wi l l i n g to l i s t e n . 1 2 3 4 5 34. My administrative head pays attention to what I am saying. 35. My administrative head stands up to outsiders for the people (s)he supervises. 36. When I have co n f l i c t s with students my administrative head gives me the kind of support I need. 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 131 Interpretation of L i f e Events Please try to imagine yourself in the situations that follow. If such a situation happened to you, what would you feel would have caused i t ? While both work-related and general l i f e events may have many causes, we want you to pick only one—THE MAJOR CAUSE IF THIS EVENT HAPPENED TO YOU. Please write the cause in the blank provided after each event. Next we want you to answer two questions about the cause you provided. F i r s t , i s the cause of this event something about you or something about other people or circumstances? Second, i s the cause of this event something that w i l l persist across time or something that w i l l never again be present? To summarize, we want you to: 1. Read each situation and vi v i d l y imagine i t happening to you. 2. Decide what you feel would be the one major cause of the situation i f i t happened to you. 3. Write the cause in the blank provided. 4. Answer two questions about the cause. 1. You have been looking for new employment unsuccessfully for some time. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 132 A colleague comes to you with a problem, and you don't help. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present You give a lecture, and the audience reacts negatively. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) to t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 133 4. You meet a colleague who acts hostilely to you. A. Write down the one major cause: B. Is th i s cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 5. You can't get a l l the work done that others expect of you. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 134 You go out s o c i a l l y and the encounter goes badly. A. Write down the one major cause: B. Is th i s cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present A relationship that i s important to you ends. A. Write down the one major cause: B. Is th i s cause due to something about you or something about other people or circumstances? (ci r c l e one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 135 You experience a major personal injury. A. Write down the one major cause: B. Is the cause of this due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present You are found to have violated a guideline or policy. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (ci r c l e one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 136 10. You and your family have a serious argument. A. Write down the one major cause: B. Is th i s cause due to something about you or something about other people or circumstances: ( c i r c l e one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 11. You lose your job. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (ci r c l e one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 137 12. After your f i r s t term at work, you are on extended probation. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances: (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 13. A friend t e l l s you that you are not to be trusted. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 138 You have a lot of trouble understanding what your employer requires of you. A. Write down the one major cause: B. Is th i s cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never - always present 1 2 3 4 5 6 7 present You cannot sleep soundly. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances: (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 139 16. You experience sexual d i f f i c u l t i e s . A. Write down the one major cause: B. Is th i s cause due to something about you or something about other people or circumstances? t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. ^  In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 17. You confront a serious confl i c t in your values. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 140 Your office-mate t e l l s you (s)he i s switiching to a room down the h a l l . A. Write down one major cause: B. Is this cause due to something about you or something about other people or circumstances? (ci r c l e one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present: ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present There are few recreational a c t i v i t i e s in which you are interested. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 141 20. Your Christmas vacation plans are cancelled. A. Write down one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 21. You have trouble with your administrative head. A. Write down the one major cause: B. Is th i s cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 142 22. You experience financial d i f f i c u l t i e s . A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 23. Your attempt to be friendly with a person of the opposite sex i s a failure. A. Write down the one major cause: B. Is this cause due to something about you or something about other people or circumstances? (ci r c l e one number) to t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present 143 24. You fee l sick and t i r e d a l l of the time. A. Write down one major cause. B. Is this cause due to something about you or something about other people of circumstances? (circle one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me C. In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present Coping People in the human services professions t y p i c a l l y report experiences of frustration in connection with their work. Think of a job-related task that has recently been frustrating to you. By frustrating, we mean that you've f e l t unable (or l i k e l y to be unable) to achieve an outcome that you want. A. Identify the task: (describe briefly) B. Identify the frustrating outcome you've encountered or expect: C. Write down the one major cause: D. Is this cause due to something about you or something about other people or circumstances? ( c i r c l e one number) t o t a l l y due t o t a l l y due to others 1 2 3 4 5 6 7 to me 144 In the future, w i l l this cause again be present? ( c i r c l e one number) never always present 1 2 3 4 5 6 7 present Keeping in mind situations l i k e the one you've just described, please respond to the following 28 items. In situations of t h i s kind, how frequently do you react in the following ways? (circle one number) > i <u i -( o> en u C O i 0 (0 u in P - H to "O 01 <u u to 01 •r-l a (0 u •u 3 <H •z. 0) (0 u < C CI 0 01 u u •P 01 W 10 1. Get together with my supervisor to discuss th i s . 2. Do my best to get out of the situation gracefully. 3. Request help from people who have the power to do something for me. 4. Throw myself into my work and work harder, longer hours. 5. Try to see th i s situation as an opportunity to learn and develop new s k i l l s . 6. Try to keep away from this type of situation. 7. Talk with people (other than my supervisor) who are involved. 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 145 >i <u 0) r - l q) 0) c n J-i C t n D> u 0 (0 (0 4-1 U M Id 3 4-> - H - H a) W 73 a 2 0) r< 0> C A) 0 <1) U U 4-> C> CO 10 8. Put extra attention on planning and scheduling. 9. Try to think of myself as a winner — as someone who always comes through. 10. Try not to get concerned about i t . 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 11. Separate myself as much as possible from the people who created this situation. 1 2 3 4 5 12. Seek advice from people outside the situation who may not have power but who can help me think of ways to do what i s expected of me. 13. T e l l myself that I can probably work things out to my advantage. 14. Devote more time and energy to doing my job. 15. Try to be very organized so that I can keep on top of things. 16. Accept this situation because there i s nothing I can do to change i t . 17. Set my own p r i o r i t i e s based on what I l i k e to do. 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 18. Try to get additional people involved in the situation. 1 2 3 4 5 19. Avoid being i n this situation i f I can. 20. Think about the challenges I can find in this situation. 21. Decide what I think should be done and explain this to the people who are affected. 22. Give i t my best ef f o r t to do what I think i s expected of me. 23. Delegate work to others. 24. T e l l myself that time takes care of situations l i k e t h i s . 25. Try to work faster and more e f f i c i e n t l y . 26. Anticipate the negative consequences so that I'm prepared for the worst. 27. Remind myself that work isn't everything. 28. Work on changing po l i c i e s which caused this situation. 146 01 rH 01 01 rH rH c n u u (0 c n C c n c n U 0) G 0) 0 10 10 P 01 0 0) U W V) 3 u Vl Vl • P - H - H 0) c n •p c n W T J Q z < (A 10 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 147 Emotional Response* Please read each statement carefully and decide i f you ever fe e l this way about your iob. If you have never had this feeling, write a "0" (zero) before the statement. If you have had this feeling, indicate how often you feel i t by writing the number (from 1 to 6) that best describes how frequently you feel that way. An example i s shown below. Example: HOW OFTEN: 0 1 2 3 4 5 6 Never A few Once a A few Once A few Every times month times a times day a year or a week a or less less month week HOW OFTEN 0-6 Statement I f e e l depressed at work. If you never fe e l depressed at work, you would write the number "0" under the heading "HOW OFTEN." If you rarely feel depressed at work (a few times a year or less), you would write the number "1." If your feelings of depression are f a i r l y frequent (a few times a week, but not daily) you would write a "5." HOW OFTEN 0-6 Statement 1. I fe e l emotionally drained from my work. 2. I feel used up at the end of the workday. 3. I f e e l fatigued when I get up in the morning and have to face another day on the job. 4. I can easily understand how my students feel about things. 5. Working with people a l l day is really a strain for me. 6. I deal very effectively with the problems of my students. 148 HOW OFTEN: 0 1 2 3 4 5 6 Never A few Once a A few Once A few Every times month times a times day a year or a week a or less less month week HOW OFTEN 0-6 Statement 7. I fee l burned out from my work. 8. I fee l I'm positively influencing other people's li v e s through my work. 9. I fee l very energetic. 10. I fee l frustrated by my job. 11. I fee l I'm working too hard on my job. 12. Working with people directly puts too much stress on me. 13. I can easily create a relaxed atmosphere with my students. 14. I fee l exhilarated after working closely with my students. 15. I have accomplished many worthwhile things i n this job. 16. I fee l l i k e I'm at the end of my rope. 17. In my work, I deal with emotional problems very calmly. •Reproduced by special permission of the publisher. Maslach and Jackson, 1981. Further reproduction i s prohibited without the publisher's consent. Demographic Data The following information w i l l be used to decribe the group of subjects Your sex: male female Your age: years Your marital status: single married Years of post-secondary education: years Years of non-teaching employment: years Years of teaching at ... : years 150 Wording Changes on Extended A t t r i b u t i o n Style Questionnaire Changes are indicated i n parentheses: 2. A f r i e n d (colleague) comes to you with a problem, and you don't t r y to help. 4. You meet a f r i e n d (colleague) who acts h o s t i l e l y to you. 6. You go out on a date, and i t goes badly. (You go out s o c i a l l y and the encounter goes badly. 7. Your steady romantic rela t i o n s h i p ends. (A r e l a t i o n s h i p that i s important to you ends. 8. You are found g u i l t y of a minor v i o l a t i o n of the law. (You are found to have vi o l a t e d a guideline or p o l i c y . 11. You are f i r e d from your job. (You lose your job.) 12. A f t e r your f i r s t term at school (work), you are on academic (extended) probation. 13. Your best f r i e n d (A friend) t e l l s you that you are not to be trusted. 14. You have a l o t of trouble understanding what your new employer (your employer) requires of you. 18. Your roommate (office-mate) t e l l s you he/she i s switching to a room down the h a l l . 21. You have trouble with one of your in s t r u c t o r s (with your administrative head). 151 Appendix D: A n c i l l i a r y r e s u l t s . Responses to the introduction to Latack's coping measure. Respondents i d e n t i f i e d such f r u s t r a t i o n s as heavy marking loads, lack of curriculum development, outdated equipment and f a c i l i t i e s , unmotivated students, and poor marketing of t h e i r programs. A lack of funding, a lack of time, and i n s u f f i c i e n t management support were frequently-cited causes. 152 153 154 155 Exploratory analyses for personal accomplishment The main regression was s i g n i f i c a n t F(7,100) = 3.22 p < .004. The predictor variables accounted for 18% (adjusted R square = .13) of the variance i n low personal accomplishment. Table 4 summarizes the findings. The variables were entered i n an h i e r a r c h i c a l manner as follows: l e v e l 1, teacher s t r e s s ; l e v e l 2, inte r n a l - s t a b l e a t t r i b u t i o n s ; l e v e l 3, external-stable a t t r i b u t i o n s ; l e v e l 4, the personal-helplessness product-term (stress by internal-stable attribution) and the u n i v e r s a l -helplessness product-term (stress by external-stable a t t r i b u t i o n s ) ; l e v e l 5, the escape-cognitive and the escape-active coping r a t i o s . The product-terms and coping r a t i o s were entered simultaneously within lev e l s 4 and 5, res p e c t i v e l y . Teacher stress was entered f i r s t . I t accounted f o r 5% of the variance and was p o s i t i v e l y associated, as expected, with low personal accomplishment. Internal-stable a t t r i b u t i o n s accounted fo r an add i t i o n a l 1% of the variance i n low personal accomplishment; t h i s contribution was not s i g n i f i c a n t . External-unstable a t t r i b u t i o n s , entered next, made no ad d i t i o n a l contribution. The personal-helplessness and the universal-helplessness product-terms, entered simultaneously, added an a d d i t i o n a l 1% of variance to low personal accomplishment. Although these product-terms f a i l e d to make s i g n i f i c a n t contributions, the m u l t i p l i c a t i v e model was nevertheless retained (Finney et a l . , 1984) . 156 157 Escape-cognitive and escape-active coping, entered simultaneously, accounted for an additional 11% of variance, with both coping r a t i o s being p o s i t i v e l y associated, as expected, with low personal accomplishment. These contributions were s i g n i f i c a n t . As expected, escape-cognitive coping (unstandardized regression c o e f f i c i e n t = 1.54) contributed more than d i d escape-active coping (unstandardized regression c o e f f i c i e n t = 1.33). The c o r o l l a r y regression was also s i g n i f i c a n t , F(7,100) = 3.32 p < .003. The predictor variables accounted f o r 19% (adjusted R square = .13) of the variance i n low personal accomplishment. Table 5 summarizes the findings. The variables were entered i n an h i e r a r c h i c a l manner as follows: l e v e l 1, teacher stre s s ; l e v e l 2, external-unstable a t t r i b u t i o n s ; l e v e l 3, internal-unstable a t t r i b u t i o n s ; l e v e l 4, the morale product-term (stress by external-unstable attributions) and the enthusiasm product-term (stress by internal-unstable a t t r i b u t i o n s ) ; l e v e l 5, the control-cognitive coping r a t i o and the c o n t r o l - a c t i v e coping r a t i o . The product-terms and coping r a t i o s were entered simultaneously within lev e l s 4 and 5, res p e c t i v e l y . Teacher stress was entered f i r s t . Again i t accounted for 5% of the variance and was p o s i t i v e l y associated, as expected, with low personal accomplishment. External-unstable a t t r i b u t i o n s accounted for 1% of additional variance; t h i s contribution was not s i g n i f i c a n t . Internal-unstable a t t r i b u t i o n s , entered next, made no unique contribution. The morale and enthusiasm product-terms, entered simultaneously, added an additional 2% of variance; neither 158 159 contribution was s i g n i f i c a n t . Nevertheless, the m u l t i p l i c a t i v e model was retained (Finney et a l . , 1984). When coping was added, the contribution of the enthusiasm product-term reached the p < .067 l e v e l of s i g n i f i c a n c e . Control-cognitive and control-active coping, entered simultaneously, adding an additional 11% of variance to low personal accomplishment. Only the contribution of c o n t r o l -cognitive coping was s i g n i f i c a n t . Control-cognitive coping was negatively associated, as expected, with low personal accomplishment. 160 161 162 Factor analysis of Latack's coping measure (1986). Using .35 as a minumum loading c r i t e r i o n , a two-factor rotated oblimin solution yielded r e s u l t s for the present sample d i f f e r i n g from Latack's c l u s t e r s . On the f i r s t f actor, recognizable as "other-focused coping", p o s i t i v e items r e f l e c t e d "engagement" strategies and negative items "avoidance" s t r a t e g i e s . On the second factor, recognizable as "self-focused coping", p o s i t i v e items r e f l e c t e d p o s i t i v e work-habits. A three-factor rotated oblimin solution loaded the p o s i t i v e and negative items from "other-focused coping" onto separate "engagement" and "avoidance" factors; a four-factor rotated oblimin sol u t i o n loaded items from "self-focused coping" onto separate factors, recognizable as "positive work-habits" and "cognitive reframing." Loadings on "engagement coping" factor 1. Get together with my supervisor to discuss t h i s (.81). 2. Request help from people who have the power to do something f o r me (.69). 3. Talk with people (other than my supervisor) who are involved (.49) . 4. Seek advice from people outside the s i t u a t i o n who may not have power but who can help me think of ways to do what i s expected of me (.35). 5. Try to get additional people involved i n the s i t u a t i o n (.73) . 6. Decide what I think should be done and explain t h i s to the people who are affected (.54). 7. Work on changing p o l i c i e s which caused t h i s s i t u a t i o n (.51). 163 Loadings on "p o s i t i v e work habits" coping factor 1. Throw myself into my work and work harder, longer hours (.74). 2. Try to see t h i s s i t u a t i o n as an opportunity to learn and develop new s k i l l s (.42). 3. Put extra attention on planning and scheduling (.43). 4. T e l l myself that I can probably work things out to my advantage (.43). 5. Devote more time and energy to doing my job (.75). 6. Try to be very organized so that I can keep on top of things (.43) . 7. Give my best e f f o r t to do what i s expected of me (.50). 8. Try to work fas t e r and more e f f i c i e n t l y (.70). Loadings on "avoidance coping" factor 1. Try to keep away from t h i s type of s i t u a t i o n (.80). 2. Separate myself as much as possible from the people who caused t h i s s i t u a t i o n (.45). 3. Avoid being i n t h i s s i t u a t i o n i f I can (.81). Loadings on "reframing" coping factor 1. Try to think of myself as a winner, as someone who always comes through (.59). 2. Think about the challenges I can f i n d i n t h i s s i t u a t i o n (.44) . 3. T e l l myself that time takes care of s i t u a t i o n s l i k e t h i s ( .35). 3. Remind myself that work i s n ' t everything (.46) 164 Correlations between coping factors and c r i t e r i o n v a r i a b l e s . The f i r s t factor, 11 (dis) engagement coping" correlated s i g n i f i c a n t l y with emotional exhaustion, r = .24, and with low personal accomplishment, r = .31. I t also correlated s i g n i f i c a n t l y with characterological a t t r i b u t i o n s , r = .21, and with luck a t t r i b u t i o n s , r = -.23. "Cognitive reframing" correlated s i g n i f i c a n t l y with emotional exhaustion, r = -.19, and with low personal accomplishment, r = -.35, and "positive work habits" correlated s i g n i f i c a n t l y with low personal accomplishment, r = -.24. Associations between these derived coping factors and the burnout subscales were i n the directions expected. They tended to be stronger f o r t h i s sample than were those r e s u l t i n g from Latack's 37 c l u s t e r s . 165 166 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0054151/manifest

Comment

Related Items