Citation
Gender differences in job stress, burnout, and job satisfaction as mediated by coping style of veterinarians in private equine practice

Material Information

Title:
Gender differences in job stress, burnout, and job satisfaction as mediated by coping style of veterinarians in private equine practice
Creator:
Welsch, Barbara Brewer
Publication Date:
Language:
English
Physical Description:
vii, 101 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Burnout ( jstor )
Employment discrimination ( jstor )
Health professions ( jstor )
Horses ( jstor )
Job satisfaction ( jstor )
Job stress ( jstor )
Men ( jstor )
Psychological stress ( jstor )
Veterinary medicine ( jstor )
Women ( jstor )
Dissertations, Academic -- Psychology -- UF ( lcsh )
Psychology thesis, Ph.D ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1998.
Bibliography:
Includes bibliographical references (leaves 89-99).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Barbara Brewer Welsch.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
0029543316 ( ALEPH )
40476900 ( OCLC )

Downloads

This item has the following downloads:


Full Text











GENDER DIFFERENCES IN JOB STRESS, BURNOUT, AND JOB SATISFACTION
AS MEDIATED BY COPING STYLE OF VETERINARIANS IN PRIVATE EQUINE PRACTICE









By

BARBARA BREWER WELSCH








A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL, FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1998


























This work is dedicated to my wonderful mother, Helen M.

Dean, and my understanding husband, Boyd B. Welsch. Their generosity, respect, and love truly made this achievement possible. It is also dedicated to my daughter, Collyn Kimberly Welsch, who made the experience all the more worthwhile.








ACKNOWLEDGEMENTS


1 would like to express my most sincere thanks and

appreciation to all who have helped me complete this dissertation. I am grateful to my advisor and friend, Dr. Carolyn Tucker, for her support, expertise, commitment to excellence, and attention to detail. I am thankful to Dr, Ellen Amatea, who suggested the exploration of coping in the veterinarians i was intending to study. Further, I am indebted to Dr. John Dixon for his invaluable statistical expertise which was provided with patience and incredible promptness.

I will also be forever grateful to my former colleagues, the

members of the Association of Equine Practitioners who responded to my survey, and to their director Gary Carpenter, who provided both the funding and the enthusiasm for the undertaking of this work. Lastly, I owe thanks to my dear friend and veterinary colleague, Dr. Anne Koterba, who suggested that the AAEP might be interested in working with me, to Dr. Ed Blach of the AAEP who advised me on the marketing aspects of survey research, and to Dave Foley of the AAEP who coordinated the printing, labeling, and mailing process.

iii








TABLE OF CONTENTS



ACKNOW LEDGEMENTS .... ........................... iii

A bstra ct . . . . . . . . . . . . . . . . . .. . . . v i

CHAPTER 1 INTRODUCTION ................ ............ 1

CHAPTER 2 REVIEW OF THE RELATED LITERATURE .... -.14

Job S tress .................................. 14

C o p ing . . . . . . . . . . . . . . . . . . 2 2

B u rno ut . . . . . . . . . . . . . . .. . . . 2 7

Job Satisfaction ............... .............. 30

Rationale and Hypotheses of the Present Study ... .34 CHAPTER 3 METHOD ................................ 36

P articipants ................................. 36

M easures ......................... ......... 48

P rocedure .......................... .... ... 44

CHAPTER 4 RESULTS ................................. 46

Descriptive Statistics ......................... 46

Data A nalysis .......................... ..... 47

Hypothesis 1 ............... ........ ..... 47

Hypothesis 2 .......................... .. .48

Hypothesis 3 ................................ 53





iv








Hypothesis 4 ............................. 54

Hypothesis 5 .............. .............. 56

Summary of the Results .................... 61

CHAPTER 5 DISCUSSION ................. ......... 62

Implications of the Research Findings ........ 72 Limitations of This Study .................... 75

Future Research .......................... 78

C onclusions ............................. 80

APPENDIX A SURVEY ............................. 82

APPENDIX B COVER LETTER ....................... 87

APPENDIX C FOLLOW UP POSTCARD ................ 88

REFERENCE LIST .................................. 89

BIOGRAPHICAL SKETCH ............................ 100








Abstract of Dissertation Presented to the Graduate School of the University of
Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

GENDER DIFFERENCES IN JOB STRESS, BURNOUT, AND JOB
SATISFACTION AS MEDIATED BY COPING STYLE OF VETERINARIANS IN PRIVATE EQUINE PRACTICE

By

Barbara B. Welsch

August, 1998

Chairman: Carolyn M. Tucker, Ph.D. Major Department: Psychology

A mail survey of 457 practicing equine veterinarians investigated gender differences in job stress, burnout, job satisfaction, and coping style. Participants included 205 men and 252 women who completed the Health Professions Stress Inventory (HPSI-T), the Burnout Measure (BOS), the Coping Styles Questionnaire (CSQ), a measure of Job Satisfaction (SAT), a single question assessing overall life stress and a demographics questionnaire.

A moderate correlation between job stress and burnout and between job stress and job satisfaction was found. Women reported significantly higher levels of job satisfaction, but were significantly more stressed and more burned out than their male colleagues. These gender differences remained when controlling for the effects of age, years in the profession, years in the current job, income, hours worked, marital status, number of children in the household, and life satisfaction.

Vi








Based on the results of the CSQ, men and women had a tendency to cope differently with job stress. Men used significantly more rational and detached coping, while women used significantly more emotional coping. Rational and detached coping neither amplified nor moderated the effect of job stress on burnout in men orwomen; however, more frequent usage of emotional coping strengthened the positive association between job stress and burnout in men, but not women.

A striking outcome was that women perceived nearly half of the assessed stressors (16 of 35 assessed on the HPSI-T) as significantly more stressful than did the men (who reported no stressor as significantly more stressful than did the womenn. Men and women significantly differed in their perceptions of frequency of occurrence in only 4 of 35 assessed stressors. In each of these four, the women reported the stressor as occurring more frequently.

Women reported considerably more stress concerning their income than did men. As has been documented previously, these women reported average salaries which were considerably lower than their male counterparts.

The results are discussed in the context of previous research completed on other occupations. Implications for the practice of counseling and for veterinary education are also discussed.







Vii















CHAPTER

INTRODUCTION



The veterinary profession, like other health-related professions including medicine and psychology (Pion et al., 1996), is undergoing many changes (Clark, 1996). It is becoming increasingly difficult to establish profitable veterinary practices; furthermore, corporations are beginning to take over established practices in order to create large referral centers with satellite clinics. Employed veterinarians will have far less autonomy than their practice-owning predecessors. Veterinary students are graduating with huge debts (mean = $45,251) but accepting starting salaries (mean = $29,854; Gehrke, 1995) which make loan repayment difficult, and which, in fact, are extremely low in proportion to the number of years of education required.

A previously male-dominated profession, veterinary medicine is rapidly becoming fernininized, with women constituting up to 75% of currently enrolled freshman classes (Gehrke, 1996). Significantly, Bird (1992) showed that female veterinarians had an income disadvantage of $13,000 per year compared to male veterinarians. Forty-two percent of the sex differences in income Bird documented were correlated with gender differences in veterinarians' characteristics (differences


I








2


in education, years of experience, board certification, practice type, investment in private practice, and hours worked). The remaining 58% ($7,400) could not be .explained, indicating that veterinarians who are female, like their counterparts in medicine and dentistry, receive substantially lower incomes than male veterinarians (Bird, 1992).

In summary, veterinary medicine is a profession which requires many years of education typically culminating in high debt. Yet, its practitioners receive comparatively low starting salaries and are losing the ability to own their own businesses. Additionally, an increasing proportion of its practitioners who are women appear to suffer from salary discrimination which, if perceived, is likely stressful. Satisfaction gleaned from the work itself, therefore, is presumably a strong motivator in attracting both males and females to the profession and preventing their attrition.

With the exception of two studies on burnout (Elkins & Elkins, 1987- Elkins & Kearney, 1992) and one (unpublished) study on gender differences in job satisfaction (Welsch, 1991), no research on job satisfaction and stress within the veterinary profession has been conducted. The purposes of this exploratory study are (1) to examine the relationships among job stress, job satisfaction, and burnout in veterinarians in private, predominantly equine, practice and (2) to determine if relationships among these variables are moderated by gender and mediated by











coping styles. Since the stresses and rewards of veterinary work vary among the different areas of practice (government, industry, academia, private practice), this study will focus on the largest segment of the veterinary population--private practice. Equine practice was chosen because of the author's personal familiarity with the area, the perception that this particular population of veterinarians may be the most stressed, and the cooperation of the American Association of Equine Practitioners (AAEP) in the endeavor. This information will be vital to the profession, particularly as it continues to undergo feminization.

During the past 15 years, numerous studies on job satisfaction have been conducted with many focusing on gender differences. Results have been inconclusive, but the bulk of the most recent research suggests that men and women do not differ significantly in overall job satisfaction ( Mobley, Jaret, Marsh, & Lim, 1994; Mottaz, 1986; Russ & McNeilly, 1995; Schneer & Reitman, 1995; Smith & Plant, 1982; Summers & DeCotiis, 1988). Many researchers suggest that when gender differences have been demonstrated, such differences have been an artifact of variables other than gender per se (Neil & Snizek, 1988). However, measurement of and conclusions about overall job satisfaction fail to consider the gender differences in satisfaction (or dissatisfaction) with various job components, and methodological limitations such as lack of control of variables including salary, promotion, age, and occupation status (professional versus nonprofessional) have








4


made it difficult to generalize from one study to the next or to be certain that gender differences are, indeed, nonexistent.

As earlier mentioned, no research on job satisfaction. has been conducted in the veterinary population. Several studies on the job satisfaction and status of physicians and dentists have been published, however (Bickel, 1988; Carr, Friedman, Moskowitz, & Kazis, 1993; Dial, Bickel, & Lewicki, 1989; Lenhart & Evans, 1991; Lewis, Carson, Nace, Howard, & Barnhart, 1994). Richardsen and Burke (1991) found no differences in overall job satisfaction between male and female physicians, but noted differences in sources of satisfaction that predicted overall job satisfaction. For women, satisfaction with time available for family and personal life was an important predictor of overall satisfaction, whereas for men, several work setting factors (e.g. own working conditions, access to support staff, and equipment) were related to job satisfaction.

Kevorkian and Tuel (1994) surveyed 49 medical faculty members who had recently resigned. While they found no significant differences in reasons for leaving, there were gender differences noted when the physicians were asked what it would have taken to induce them to stay in their present position. Men rated more protected time for academics and more support for research more highly than did women, and women ranked effective mentors or role models and more support for patient care more highly than did men.








5


Two major competing theories of job satisfaction in the literature include the "fit hypothesis" (Locke, 1976) and the work attribute hypothesis (Herzberg, Mausner, & Snyderman, 1959). The fit hypothesis addresses the compatibility of external work features and individual worker characteristics. This theory emphasizes the differences in how workers react to their jobs and the preponderant importance of these variations in influencing satisfaction and/or dissatisfaction with work. In contrast, the work attribute hypothesis states that the features of the work itself coupled with advancement, recognition, responsibility, and growth are of preeminent importance for job satisfaction, and individual attributes make little difference. Many authors have defined job satisfaction over the years (Hoppock, 1935; Super, 1942; Hopkins, 1983; Dawis & Lofquist, 1969, 1984). Regardless of whether the concept of needs and values or individual perceptions is employed, the common underlying premise of the most recent perspectives on job satisfaction is that there is something within an individual which corresponds to his or her reaction to (and consequent satisfaction with) the job.

Despite the plethora of job satisfaction research, much of it ignores the concept of job stress. There is an even larger body of literature concerning job stress. Typically, this research addresses the issue of job satisfaction minimally by asking the participants one or two global questions aimed at measuring job satisfaction. There is no end to the number of studies showing that stress in the








6


workplace adversely affects productivity, absenteeism, worker turnover, and employee health and well-being (Perrewe, 1991; Quick, Murphy, & Hurrell, 1992; Spielberger & Reheiser, 1995).

The concept of "stress" can be viewed from several perspectives. The stimulus definition sees stress as a characteristic of the environment that is disturbing or disruptive. The response definition of stress states that stress is a pattern of physiological and/or psychological reactions exhibited by a person who is under some sort of pressure. Neither definition of stress clearly explains individual differences, however (Cox, 1978). A third approach to conceptualizing stress is the transactional model (Lazarus & Folkman, 1984). This view combines the stimulus and the response approaches and defines stress as the consequence of the interaction between environmental stimuli and individual responses. Intervening psychological processes such as perception and cognitive appraisal are emphasized. Stress is considered to occur only when a person perceives a demand as exceeding his or her capacity to deal with it. The major criticism of this theoretical model is that it is incomplete since it fails to consider traits or personality factors (Ben-Porath & Tellegen, 1990; Costa & McCrae, 1990).

When the effects of gender differences on job stress have been studied, the results have been conflicting. While much of the early work found no sex differences, the researchers failed to look at the frequency or the intensity of the








7


stress experienced. Later studies which took the above factors into account were, indeed, able to detect gender differences. There are no studies evaluating job stress in the veterinarian, but several were found which investigated other health related professions (Wolfgang, 1988, 1995; Simpson & Grant, 1991; Richardsen & Burke, 1991). Three of the four studies found no gender-related differences, but Richardsen and Burke (1991) reported significant sex differences in terms of the variables that predicted job stress. Women rated the need to maintain their own level of knowledge and providing counseling for nonmedical problems as more stressful than did men, whereas male physicians reported more stress related to maintaining an adequate income.

Even though consistent differences in the amount or types of stress experienced on the job have not been shown between men and women, it is virtually uncontested that women in Western cultures have higher rates of affective and anxiety disorders than do men, in both clinical and nonclinical populations (Miller & Kirsch, 1987). It is generally agreed that many professional women may experience one or more of the following (Smith, 1993): a hostile work environment in which their values, working style and job contributions are trivialized, unappreciated, or unrewarded; an unsupportive home environment in which it is their primary responsibility to care for the home and the children; a disapproving social environment (i.e., their house is not clean enough and they rarely produce








8


healthy meals; they should not put their children into daycare); and major conflicts with their self-concept (worker vs. mother vs. wife). Two theoretical perspectives have been advanced to address these issues: the role conflict perspective (scarcity hypothesis) and the role expansion perspective. The former suggests that the above conflicts add to the stress of the female worker (Goode, 1960), while the latter model finds that if a female sees her roles to be both meaningful and of high quality, multiple roles lead to positive outcomes and may even buffer the impact of negative life events (Repetti, Mathews, & Waldron, 1989).

Although gender differences in job stress are inconsistently reported, the reported effects of job stress have been consistent. Burnout is a syndrome of emotional exhaustion, depersonalization, and reduced personal accomplishment that can occur among individuals who work with people in some capacity (Maslach, 1986). Burnout appears to be a response to chronic sources of emotional and interpersonal stress on the job. It is a particular problem for men and women in caregiving professions, especially when the provision of care or service occurs in emotionally charged situations (Maslach, 1986). Higher levels of burnout are linked to lower levels of job satisfaction.

Burnout is the one psychological outcome variable that actually has been studied in veterinarians (Elkins & Elkins, 1987; Elkins & Kearney, 1992). (Neither of these two referenced studies was limited to veterinarians in private practice.)








9


Fifty-three percent of males in the 1987 study (which consisted of >95% male respondents) and 67% of female veterinarians surveyed in the all-female 1992 study, achieved scores on a burnout scale (Pines & Aronson, 1988) which were indicative of burnout syndrome. Women scored significantly higher than men, but the study of women was completed nearly 10 years later than the study of men so a direct comparison of these two groups is not valid. Elkins and Kearney (1992), in addition to administering a burnout scale, asked the 613 responding female veterinarians if they were happy in their current position and if their position was stressful. Responses indicated that 83% of the veterinarians in the study reported being happy in their job and 87% reported their work to be stressful.

The confusing and often inconsistent relationship found between job stress, job satisfaction, burnout, and gender may well exist because moderator and/or mediator variables have not been studied. Folkman and Lazarus (1988) distinguish between moderators and mediators in the following way. Moderators are antecedent conditions (such as gender) which affect the perceptions of the situations and how it is appraised. Mediators, however, are generated in the stressful encounter and are hypothesized to change the relationship between the antecedent and the outcome variable. It is likely that the predictor variable of job stress impacts the outcome variables of job satisfaction and burnout only after having been acted upon by other variables Self-efficacy (Bandura, 1977), locus








10


of control (Lefcourt, 1982), and causal attributions (Doherty & Baldwin, 1985) may each play a role in moderating or mediating the effects of job stress and individual differences on job dissatisfaction, burnout, and commitment. Other variables which may be relevant include the need for achievement, level of self-esteem, tolerance of ambiguity, Type A vs. Type B personality (1vancevich & Matteson, 1980), and negative effectivity (Parkes, 1990). A very pragmatic avenue of investigation, however, may be identification of veterinarians' coping styles since interventions aimed at altering a style which is associated with burnout or dissatisfaction could subsequently be offered.

Lazarus' transactional theory of stress (1991), which was cited earlier, reflects the multidimensional, dynamic, and complex nature of stress, including the interactions between the individual and the context in which he or she is working. Two processes, cognitive appraisal and coping, are identified as crucial mediators of stressful person-environment relationships and their immediate and long-term outcomes (Lazarus & Folkman, 1984). Personal meanings of an event itself and the choice and effectiveness of one's coping actions are considered integral to the stress process.

The existing literature presents a confusing view of the role of coping in the workplace. Various researchers label the forms of coping differently, but three general types of coping seem to be cited most frequently: problem-focused,











emotion-focused and symptom relieving (avoidance). Problem focused coping refers to attempts to modify or eliminate the source of stress by taking instrumental actions (Miller & Kirsch, 1987). This could be in the form of direct action or helpseeking behavior (Havlovic & Keenan, 1995). In contrast, emotion-focused coping refers to attempts to control or repress stress-related emotions and maintain affective equilibrium (Billings & Moos, 1984). This form of coping could be in the form of either positive thinking or avoidance/resig nation (Havlovic & Keenan, 1995). Symptom focused coping is defined as the use of exercise, meditation, alcohol, or drugs to alleviate the stress response (Schuler, 1985). Other authors use the term It avoidance coping" to include additional methods of simply ignoring the problem. Roger et. al. (1993), through their work with stress management programs and the development of the Coping Styles Questionnaire (CSQ), identified a fourth mechanism of coping which they term "detachment" coping. This form of coping does not involve denial or attempts to avoid stress, but it can also be distinguished from task-oriented strategies (Roger, Jarvis, & Najarian, 1993). This strategy involves feeling independent of the event and the emotion associated with it.

Much of the occupational-stress literature reports data only from male employees (Parkes, 1990). Gender issues are not addressed and relatively little is known about gender differences in work-related coping. More attention has been given to gender in the general coping literature, but the results are very inconsistent.








12


Several groups have shown no gender differences in coping strategies (Hamilton & Fagot, 1988; Osipow & Davis, 1988). In contrast Pearlin and Schooler (1978) found that women reported less use of effective strategies, a finding corroborated by Billings and Moos (1981). Roger et al. (1993), using the CSQ which they developed, showed a tendency for men to use more detached and rational and less emotional coping styles than women, but the differences were not statistically significant.

In contrast, in two studies conducted in actual work settings, women were shown to cope in more positive ways. Parasuraman and Cleek (1984) found that female managers reported greater use of adaptive coping strategies (planning, information seeking, priority setting) than did males. Parkes (1990) found that men used significantly more suppression strategies such as withdrawal, restraint, compromise, and ignoring the problem. Miller and Kirsch (1987) summarize the literature by concluding that "it appears that the categories of emotion-focused and problem-focused coping may yield significant sex differences in how people cognitively cope with stress .... It is possible that women appraise threatening events as more stressful than men do. However, given the methodological constraints of these studies, further research is needed before firm conclusions can be drawn" (p. 295). The veterinary profession is changing drastically as a result of the changing economy and changing demographics (feminization). Occupations








13


which have become feminized have experienced a decline in salary and prestige leading to decreased job satisfaction. This research was the first to explore job stress, job satisfaction, and coping styles of veterinarians of either gender. Findings from this research will have implications for the future of the profession as it becomes female dominated. More importantly, findings suggest ideas for potential interventions at the level of both the veterinary student and practicing veterinarian. Consequently, the results of this study may enhance job satisfaction and decrease job turnover, burnout, and professional attrition among veterinarians in private practice.















CHAPTER 2

REVIEW OF THE RELATED LITERATURE



The literature review presents research which will aid in the understanding of the background upon which this study was based. There are five sections in this chapter. These include sections on job stress, job satisfaction, burnout, and coping and cover theories of each along with gender and other differences which have been noted. The final section concludes with the five research hypotheses investigated in this study.

Job Stress

The concept of "stress" has been viewed from several, evolving perspectives over the past century. Although many different definitions of stress have been entertained by researchers, most seem to fall into three general categories. In the stimulus definition, stress is seen as an external force that causes a reaction of strain within the individual. Holt (1982) defines 57 different work stressors (which include work overload, role ambiguity, monotony, and lack of control).

The response definition of stress takes the difference in stressors into account. The Canadian endocrinologist, Hans Selye (1950), profoundly shaped the 14









15


modern concept of stress as a response by discovering that tissue damage could result from diverse stimuli and was thus a nonspecific response to all noxious stimuli. It is clear, however, that different stressors culminate in differing response intensities. Although Selye was looking at purely physiological reactions, the response definition of stress has grown to include the psychological reactions exhibited by a person who is under some sort of pressure (Maslach, 1986). Neither definition is capable of explaining individual differences, however (Cox, 1978).

An important advance in the thinking about stress was supplied by French, Caplan, and Van Harrison (1982). Their ideas about person-environment fit combine the separate causal variables of environment and person into an ongoing relationship characterized by fit or misfit. Occupational stress is defined in terms of job characteristics that pose a threat to the individual because of a poor match between the abilities of the employee and the demands of the job. But this approach falls short, because it is basically a static approach. It emphasizes stable relationships between a person and the workplace rather than a process during which stress changes over time and varies within similar contexts. It is clear that even when a person and his or her environment are a good "fit," some stress will occur. In fact, most work researchers agree that some degree of stress is beneficial.

Lazarus' transaction I process theory of stress (1995) distinguishes between stressful antecedent conditions and how these are perceived and cognitively









16


appraised by a particular person. The consequent emotional reactions which follow when a stressor is perceived as threatening and the individual is not able to cope effectively with it form the basis of stress. Thus, Lazarus looks at the stressful stimulus and looks at the response, but, in addition, places great emphasis on the intervening variables of appraisal and coping ability, adding dimensions which others have not considered.

Although Lazarus' theory of stress is the most widely known and accepted theoretical model, it is not without its critics. The major criticism of this theoretical model is that it is incomplete since it fails to consider traits or personality factors (Ben-Porath & Tellegen, 1990; Costa & McCrae, 1990). Further, Brief and George (1994) criticize Lazarus' emphasis on the idiographic nature of occupational stress. They argue that it is especially important to discover those working conditions that are likely to adversely affect groups of employees who are exposed to them. In a similar critique of the model, Harris (1994) has noted that the occupational stressors associated with the climate and the culture of an organization can have profound effects on employees, and that these may differ as a function of gender and individual differences in personality as well as coping skills. Lazarus' transactional model of occupational stress will be discussed in more detail under the section entitled "Coping."

Most of the occupational stress research has centered upon the effects of job stress on the worker's health--both physical and mental. In regard to physical









17


health, coronary heart disease has been clearly linked to job stress along with the Type A behavior pattern (Maslach, 1986). Gastric ulceration has less consistently been linked to job stress as well. Mortality rates have been found to differ by occupation, and although unproved, it is reasonable to implicate stress on the job as one explanatory factor. In a recent letter to the editor of the Journal of the American Veterinary Medical Association (Prichard, 1996), the author observed that in the monthly obituary column there appeared to be a disproportionate number of relatively young veterinarians compared to the obituaries listed in a stockbrokers' journal. In fact, when surveyed, fully 87% of female veterinarians considered their jobs to be very stressful (Elkins & Kearney, 1992).

Stress in the workplace may well be increasing. The proportion of workers who reported "feeling highly stressed" on the job had more than doubled from 1985 to 1990 (Northwestern National Life, 1991). Sixty-nine percent of the 600 workers surveyed in the Northwestern study reported that their productivity was reduced due to high stress levels, and "one in three say job stress is the single greatest stress in their lives" (1991, p. 2).

Job stress has been linked to impairments in psychological well-being and implicated in reports of depression, dissatisfaction, anxiety, tension, lowered selfesteem, and alcohol and drug abuse (Maslach, 1986). Absenteeism, employee turnover, decreased productivity and diminished client-relations are obvious costs to employers (Spielberger & Reheiser, 1995). According to Matteson and









18


Ivancevich (1987) preventable costs relating to diminished productivity, turnover, and absenteeism are in the realm of $2800 per person employed. This figure does not even include such things as workers' compensation, disability payments, and early retirement benefits.

Despite a burgeoning of research in the occupational stress area, the measurement of job stress has been inconsistent and confusing. Most of the literature has relied on answers to only one or a few questions about experienced stress (Barone, Caddey, Katell, Roselione, & Hamilton ,1988). Further, job stress measures tend to confound the frequency of occurrence of a stressful event with its perceived severity (Spielberger & Reheiser, 1995). Some authors have contended that research in work stress should focus on specific aspects of the job which workers find problematic (Jackson & Schuler, 1985), while others have argued for the development of generic measures to facilitate comparing stress levels across occupational groups (Murphy & Hurrell, 1987).

In response to the criticisms of past measures, Spielberger and colleagues developed the Job Stress Survey (JSS), adapting it from earlier surveys which they had designed to measure job stress in policeman and later, teachers (Spielberger & Reheiser, 1995). This 30-item psychometric instrument was designed to assess the perceived intensity and frequency of occurrence of working conditions which could adversely affect the psychological well-being of employees. The survey was intended to form a generic job stress measure and has been used repeatedly in









19


university and corporate work settings (Spielberger & Reheiser, 1995).

It was this author's opinion, in agreement with Jackson and Schuler (1985), that an instrument specific to the health related professions would be more valuable in this particular study. Medical professionals face a different set of stressors than do other professionals, and recently, increasing attention has been focused on the adverse effects of job stress on providers and ultimately on the overall quality of health care (Wolfgang, 1988). Although various instruments and surveys have assessed job stress in physicians, none have looked at veterinarian's work stress. One instrument, the Health Professions Stress Inventory (HPSI) has been used to compare levels of stress within the occupations of mediciine, nursing, and pharmacy, and it was felt that this instrument would provide useful information regarding the veterinary profession while allowing comparisons to other healthrelated professions.

The general job stress literature produces inconsistent findings regarding the effect of gender on job stress. Beehr and Schuler (1980) concluded that women and men show no differences in stress-related symptoms in the workplace, and DiSlavo, Lubbers, Rosi, and Lewis (1994) found that "from a broad perspective, men and women perceive stressors quite similarly" (p. 48). On the basis of a metaanalytic evaluation of 15 studies which examined gender differences in occupational stress, Martocchio and O'Leary (1989) found "no sex differences in experienced and perceived work stress" (p. 495).









20


It has been suggested that failure to identify gender-related differences in occupational stress may be due to sampling problems (men over-rep resented in management positions with women holding clerical and service jobs (Jick & Mitz, 1985). Further, these authors conclude that gender may act "not only as a direct predictor of the source of stress, but also as a moderator affecting how stress is perceived, what coping skills are called upon, and how stress is manifest" (p. 409). In agreement with the above authors, Spielberger and Reheiser (1995), using the JSS, noted striking gender differences in the severity or frequency of items identified as stressful. Men and women rated the same six items as highest in perceived severity and the same 5 items as lowest in severity, however.

Consequently (although somewhat surprisingly), gender related differences were not statistically confirmed in the JSS severity, frequency or overall scores despite the numerous differences reported in the severity and frequency of occurrence of specific JSS stressors. The authors concluded that "gender-related differences in occupational stress appear to be determined by differences in the perceived severity of specific stressors and in the frequency that these stressors are experienced by men and women" (p. 64).

There has been some speculation but little research concerning differences in stress experienced by male and female physicians (Simpson & Grant, 1991). More research has been carried out with medical students than with physicians in practice.









21


Simpson and Grant (1991) reported that, despite speculation to the contrary, physicians in their study evidenced no difference by gender in the types and magnitude of stressors experienced. As in the Richardsen and Burke (1991) study (see below), the authors used their own set of identified stressors, not a standardized instrument. Richardsen and Burke (1991) found that women physicians reported slightly higher overall stress than did men. They also found significant differences in three sources of stress reported: women rated the need to maintain their own level of knowledge and counseling nonmedical problems as more stressful than did men, whereas male physicians reported more stress related to maintaining an adequate income.

Age and years of experience may have an impact on job stress and job satisfaction and other work attitudes such as commitment and turnover intention. In a study of job stress, satisfaction, and life satisfaction in physicians, Linn, Yager, Cope, and Leake (1985) found an inverse relationship between age and stress, and positive relationships among age and job and life satisfaction. Age differences possibly accounted for the difference in overall stress between men and women physicians in the Richardsen and Burke (199 1) study, since men in their study were generally older and had more years of experience than women. No statistics were done to confirm this possibility, however. Similarly, researchers have suggested that experience (or closely related constructs such as career stage) may moderate relationships among sales force members' perceptions, attitudes, and behavior









22


(Russ & McNeilly, 1995).

Coping

As discussed in the previous section, the psychological stress literature is permeated with discussions about the potentially deleterious effects of stress on physical and mental health as well as on individual attitudes and behaviors. Therefore, it follows that much emphasis has also been placed on the investigation of how people manage or cope with stress in all of life's domains including the workplace.

Lazarus' transactional model of stress and coping (Lazarus, 1995) is one of several models which have been used to study the nature of coping. Most researchers who have studied coping cite Lazarus' work and many use it to build slightly different and more complex models. For example, Schuler's (1985) integrative transactional process model of coping (which is specific for organizations) also addresses the impact of the uncertainty and the importance of the situation to the individual and includes a process of feedback and evaluation.



The transactional theory states that a transaction between the person and the environment is stressful only when it is evaluated by an individual as a harm, threat, or challenge to that person's well-being. Lazarus terms this evaluation an appraisal. Within the context of the theory, distinction is made between "harm," "threat," and "challenge." Harm refers to damage which has already occurred (loss









23


of job, poor job evaluation, lack of promotion, disapproval), while threat refers to a harm that has not yet happened, but is anticipated. Challenge refers to a condition of high demand in which the emphasis is on mastering the demands, overcoming obstacles, and growing and expanding as an individual. During threat, the focus is on avoiding or protecting against harm. In challenge, the emphasis is on the positive outcome possibilities. Individuals, particularly those who have chosen medical professions, thrive on challenge while at the same time dislike (and attempt to evade) threat.

Lazarus' early field work demonstrated that the way people evaluate what is happening with respect to their well-being, and the way they cope with it, influences whether or not psychological stress will result, and its intensity (Lazarus, 1966). He terms this "appraisal" and identifies two types: primary appraisal concerns whether or not there is any personal stake in the encounter and secondary appraisal concerns the available coping options for dealing with harm, threat, or challenge. Lazarus defines coping as "the cognitive and behavioral efforts a person makes to manage demands that tax or exceed his or her personal resources" (Lazarus & Launier, 1978). He believes that coping is a process because the act of coping results in changes which provide new information that will alter subsequent appraisals. He has also shown that some coping strategies are stable across encounters while others are unstable and heavily influenced by the environmental conditions of the encounter (Lazarus, 1995).









24


The meaning of work clearly varies widely among workers in our society. For some individuals, work is merely a means of earning money for survival. For others, work is their life--their sole reason for surviving. It is my contention, biased through my own participation in multiple areas within the field of veterinary medicine, and based on the comparatively low income achieved by veterinarians compared to other health professionals, that their work has considerable meaning to most veterinarians.

Locke and Taylor (1990) define meaning of work as "the totality of values, including their importance, that individuals seek and expect to derive from work. Values themselves are what individuals desire or consider to be good or beneficial" (p. 135). These authors propose that individuals may expect and desire to fulfill five different categories of values from their work-- material values, achievement-related values, a sense of purpose, social relationships, and the enhancement or maintenance of self-concept.

Threats to any of the above values (which are important to an individual) cause stress, followed by the activation of coping mechanisms. Material values can be associated with work stress in obvious ways. Workers can base their selfconcept almost entirely on material outcomes. Further, they might exclude other important values (i.e., family relationships) entirely in the pursuit of job success and money. Veterinary practice owners, one of the few veterinary groups able to earn over $100,000, can only do so by working exceedingly long hours, almost always








25


to the detriment of their families.

The key threats to achievement-related values are associated with loss: loss of feelings of competence, loss of interest in (or challenge by) the work itself, or loss of personal control over one's work (Locke & Taylor, 1990). Workers may lose their sense of purpose at work after attaining highly valued, long sought-after goals, or they may gradually begin to question the real value of what they do. Social relationships serve both an expressive and an instrumental function at work (Locke, 1976). Interactions can be pleasurable and/or necessary for completion of the job and difficulties in this area can be particularly disturbing to workers. Some research has shown that this work value may be more important to women than to men. An individual's self-concept is threatened at work when there in an incongruity between the self-concept and what the individual sees as a requirement of the job. The selfconcept is further threatened by perceived failures at work as well.

The stress caused by these threats must be coped with, in one way or another. Coping is typically defined in the literature as the process of adapting to perceived threats, and coping has been repeatedly divided into rational (problemfocused), emotional, and avoidance (symptom-focused) forms with different authors attributing differing meanings to each of the above. Roger, Jarvis, and Najarian (1993) have recently described a scale which measures four types of coping: rational, emotional, avoidance,,and detachment coping. This scale represents an extension of Folkman and Lazarus' (1985) Ways of Coping Questionnaire which









26


was based on their transactional model of stress and coping. Rational coping refers to task oriented coping methods such as working out a plan for dealing with what has happened. Emotional coping refers to attempts at repressing (or expressing) the emotions evoked due to the threat. Avoidance coping involves avoiding action or evoked emotions, and detached coping refers to feeling independent of the event and the emotion associated with it.

Emotional and avoidance coping are considered maladaptive, and rational and detached coping are considered to be adaptive styles of coping. Smith and Sulsky (1995) investigated coping outcomes and found that increased use of emotional and avoidance coping was associated with higher global job stress, more physical symptoms, increased depression, and lower job satisfaction. This finding was repeated over several samples.

It is a strongly held gender-role stereotype that women respond to stress differently from men. That is, they respond passively and/or more emotionally to stressful situations whether at work or elsewhere. Early studies (Billings & Moos, 1981; Folkman & Lazarus, 1980; Pearlin & Schooler, 1978) tended to suggest that men used more rational coping compared with women. However, these studies were flawed because men and women were not matched on occupational level or on organizational power (Long & Kahn, 1993). Roger, Jarvis & Najarian (1993) found a tendency for men to use more detached and rational and less emotional coping styles than women, although the differences were not statistically significant.

I








27


Parasuraman and Cleek (1984) and Tung (1980) found women managers were more likely than men to use adaptive coping behaviors or superior coping. Long and Kahn (1993) attribute these confusing gender differences to three causes. First, women experience more and different work stressors than men (sex discrimination and harassment), Next, gender role stereotypes affect male and female perceptions of appropriate coping (such as disclosing to friends and colleagues). Finally, women are disadvantaged with regard to coping if they lack workplace coping resources such as power, perceived competence, or child care and financial advantages.

In summary, when coping strategies fail, work loses meaning because important values can no longer be attained from it. If this happens, since work usually consumes a considerable part of one's daily living, life, itself, can lose meaning. This is when burnout occurs.

Burnout

According to Schultz and Schultz (11990) "More people are disabled today as a result of stress than at any other time in our history" (p. 539). In 1992, job stress was identified as one of the 10 leading health risks in the United States (Quick, Murphy, & Hurrell, 1992). It has been estimated that fully one third of US workers are currently experiencing burnout or preburnout symptoms (Takooshian, 1994). Burnout is the only psychological outcome variable that has been studied in the veterinary population (Elkins & Elkins, 1987; Elkins & Kearney, 1992).









28


Maslach (1986), who was the first to coin the term "burnout" defines it as a syndrome of emotional exhaustion, depersonalization, and reduced personal accomplishment that can occur among individuals who work with people in some (usually care-giving) capacity, often in emotionally charged situations. Many sources of burnout have been proposed, with no single cause identified. Burnout is viewed as the result of multiple causes, the impact of which differs according to the personality of the worker. According to Maslach (1986) the sources of burnout can be divided into three categories: personal (expectations, motivations, personality), interpersonal (client contact, relationships with coworkers and supervisors, outside relationships), and organizational (workload, bureaucracy, feedback, work pressure). A key factor does seem to be the amount of client contact experienced on the job. As time spent in direct client contact increases, particularly if such contact is upsetting, frustrating, or difficult, burnout is more likely to occur (Pines & Aronson, 1988).

Two studies have been conducted on burnout in veterinarians. Both used Pines and Aronson's self-diagnosis burnout scale (Pines & Aronson, 1988). A mail survey of 1000 veterinarians in 1981 yielded a 33% response rate, of which 91 % of the respondents were male (Elkins & Elkins, 1987). Fifty-three percent of the respondents reported scores over 3 (mean = 3.09), the point at which clear-cut signs of the burnout syndrome are said to occur. The youngest age group (25-29) reported significantly higher burnout scores, as did the veterinarians in practice for









29


10-14 years. A high correlation between reported stress and burnout score was observed, with client relations being the most stressful aspect of practice cited. Solo owners of multiveterinarian practices had significantly higher burnout scores than employees or veterinarians owning solo practices.

The Elkins and Elkins (1987) study was replicated in 1992 in a mail survey of 1000 female veterinarians. (Elkins & Kearney, 1992). Interestingly, this survey yielded a 61 % response rate, using only one mailing as in the previous survey. The fernale respondents reported a burnout score mean of 3.41, with 67% scoring over 3. There was a very strong positive correlation between burnout and hours worked per week and burnout and stress. A strong negative correlation was evident between job satisfaction and burnout. Female practice owners had the lowest burnout scores, although most were owners of solo practices rather than multiveterinarian practices.

The female veterinarians did exhibit higher burnout scores than did their male counterparts. Further, a greater percentage of women than men scored in the clinical range. Conclusions cannot be drawn from such direct comparisons since the studies were widely spaced in time. Lowman (1993) concludes after reviewing the general work literature that no inherent gender differences have been found in research on job-related burnout, but that there may well be differences in the causes of burnout for men and women and in their coping strategies for handling burnout symptoms.









30


In regards to the outcomes of job burnout, job satisfaction has received much attention. As would be predicted, higher levels of burnout are linked to lower levels of satisfaction. Symptomatically, people experiencing burnout express a desire to spend less time working directly with clients and are more likely to be absent from work (Maslach, 1986). Although it would be expected that burnout would readily lead to a deterioration in the quality of client services provided, this area has not been well researched. Jones (1981) did show that burnout was associated with more on-the-job mistakes, more "inhumanistic" counseling practices, more aggressive behavior toward clients, and more disciplinary action by supervisors.

Job Satisfaction

Theories of job satisfaction were mentioned in Chapter 1. The "fit hypothesis" suggests that the real meaning of work to an individual actually lies in the congruence between the worker and his or her job (Locke, 1976). Herzberg, Mausner, and Snyderman (1959) had previously argued that the features of work itself are of preeminent importance, disregarding individual characteristics. Job satisfaction is clearly a difficult concept to define accurately, but in general, some amount of satisfaction exists when the costs of working are exceeded by the benefits in a degree seen as acceptable by the worker (Fraser, 1983). Clearly, an individual's level of job satisfaction lies on a continuum, which is subject to change and is neither constant nor static.

It is noteworthy that much of the job satisfaction literature ignores the









31


concept of job stress while researchers of job stress typically ask their subjects at least one question concerning their level of job satisfaction. Fraser (1983) believes that the basic relationship between job stress and job satisfaction is illustrated by a bell shaped curve, indicating that some degree of stress is optimum for the achievement of maximal job satisfaction.

As discussed in Chapter 1, there has been a plethora of research concerning gender differences in job satisfaction (Andrisani, 1978; Cassidy & Warren, 1991; Martin & Hanson, 1985; Mottaz, 1986; Murray & Atkinson, 1981; Shapiro & Stern, 1975; Weaver, 1974; Zanna, Crosby, & Loewenstein, 1987; Mortimer, Finch, & Maruyama, 1988). Although conflicting findings are evident, in general, men and women do not differ significantly in reported levels of job satisfaction. This is paradoxical, in that gender differences in work values have been more consistently reported. Women, more consistently than men, have placed significant importance on hours worked per week and amount of travel time to and from work (Quinn, Staines, & McCullough, 1974). Presumably, this relates to compatibility with family responsibilities. Miller (1990) showed that men exhibit more concern for leadership, opportunities for advancement, and decision-making ability than do women, whereas women find such occupational concerns as job pressure, work hours and job exhaustion as more salient than do men. Similar findings were replicated among physicians, where no gender differences in overall satisfaction were reported. Yet predictors of job satisfaction varied between the sexes. Satisfaction









32


with time available for family and personal life was an important predictor for women whereas work setting factors were most important to male physicians (Richardsen & Burke, 1991). Similarly, Kevorkian and Tuel (1994) found that male and female academic physicians had different priorities when it came to inducements to remain in academic positions.

Three theories can be utilized to inform the study of gender differences in job satisfaction (Mason, 1995). The socialization view holds that women are socialized into values, attitudes, and behaviors that are less individual in nature and more communal than male values and behaviors. In direct contrast to this view, the structural view suggests that any observed differences in job satisfaction are attributable to factors which co-vary with gender and not to inherent gender differences in socialization. The social role theory suggests that when social role is more salient than gender role, as it is in managerial or other professional positions, male and female attitudes, behaviors and values will be similar. Impoverished opportunity structures (such as clerical or other "blue-collar" positions) will evoke gender role salience, in which gender differences will be notable. Mason's (1995) study of nearly 15,000 workers, provided no evidence for the socialization view, and most support for the structural view (although in some instances the social role view was supported). She compared male and female managers and male and female clerical workers using data collected over a twoyear period by a consulting firm. An important finding of her study, which contrasts









33


with what little has been observed in physicians, is that men and women in management did not appear to differ from one another in their sources of satisfaction at work.

The relationship between education and job satisfaction has been explored. Early research suggested that increasing education might condition workers to expect higher levels of extrinsic rewards (Mortimer, Finch, & Maruyama, 1988) or intrinsic rewards (Seybolt, 1976). On the one hand, employees with higher educational attainment were prepared for more challenging careers which could pay higher salaries, so if this materialized, the workers should be more satisfied. On the other hand, if workers were overeducated and underemployed (or underpaid) they should be less satisfied.

One recent study reported that gender moderated the impact of education on job satisfaction. Educational attainment that had not provided expected levels of extrinsic rewards had a negative impact on job satisfaction of men but not women (Glenn & Weaver, 1982). The authors hypothesized that this might be due to wage earning status--women who are secondary wage earners may place more emphasis on the intrinsic rewards of work. Martin and Shehan (1989) confirmed this hypothesis. They found that the major source of the education-satisfaction relationship was the result of education's connection to higher levels of extrinsic reward only for the primary wage earning women. For secondary wage earning women, intrinsic rewards were the important









34


component of the relationship between schooling and job satisfaction. They also found "no evidence that education that doesn't lead to expected levels of intrinsic and/or extrinsic reward reduces job satisfaction" (p. 192).

Rationale and Hypotheses of the Present Study

The relationships among job stress, burnout and job satisfaction seem relatively straightforward regardless of the types of confounding variables studied. These relationships have not been confirmed in a veterinary population, however. In addition, it is unclear (in any population) how the relationship is influenced by gender and mediated by coping style, and further modified by age and wageearning status.

Based upon the literature review, the following research hypotheses were formulated.

1. Among equine veterinarians in private practice, job stress will be negatively correlated with job satisfaction and positively correlated with burnout.

2. Among equine veterinarians in private practice, there will be no gender differences in the overall level of job stress, job satisfaction, and burnout reported. 3. There will be gender differences in the frequency and severity of specific job stressors reported by equine veterinarians in private practice.

4. There will be no gender differences in coping styles reported by equine veterinarians in private practice.

5. The relationship between job stress and burnout will be mediated by coping style









35


in equine veterinarians private practice*. Specifically, problem-focused and detached coping styles will be associated with a weaker relationship between job stress and burnout while avoidant and emotion-focused coping will be associated with a stronger relationship between job stress and burnout.















CHAPTER 3

METHOD



Participants

Potential participants were 968 (503 men and 465 women) licensed veterinarians in private, predominantly equine practice who were selected from the 1997 membership directory of the American Association of Equine Practitioners (AAEP). The AAEP has approximately 5600 national and international members of which approximately 25% are female. Participants were eligible for selection for the study as long as they were currently practicing, were in private practice (vs. academia, government, or business), lived in the United States, described themselves as working in a practice which is >50% equine, and had a known gender.

Four-hundred ninety-two (492) surveys were returned, resulting in a final response rate of 51 %. Twenty-seven surveys arrived after the cut-off date of June 13, and were not included in the study. Eight surveys were unidentifiable by gender and were also not included. Thus, 457 surveys were usable. Of the usable surveys 45% were from men Q1=205) and 55% were from women (n=252).


36









37


The overall response rate from the males was 43% and the female response rate was 57%. Eighty-nine percent of the actual study participants (EI=407) ranged in age from 30-59 although the age range of all participants was from 25 to over 70 years. The participants were distributed nearly equally in terms of years in the profession of veterinary medicine: 20.8% (n= 95) had been practicing for five years or less, 19.7% Q1=90) between six and 10 years, 23.9% (n= 109 ) between 11 and 15 years, 11.8% (n=54) between 16 and 20 years, and 25% (a_=115) greater than 20 years, Thirty-six percent (n=164) had been working for five years or less at their current job, 22.8% (n =104) had been working from 6 to 10 years, 15.5% (L=7 1), between 11 and 15 years, 6.6% (n=30) between 16 and 20 years, and 19.0% (n=87) longer than 20 years. Sixty percent of the respondents (EI=274) were sole owners of a practice, 13.7% Q1=63) were partners in a practice, and 26.2% (n=120) were employees.

Five percent (n=23) of the respondents were board certified with the colleges of Surgery (n=10 respondents), Theriogenology (n=7 respondents), Veterinary Practice Q1=3 respondents) and Acupuncture (n=3 respondents) being represented. Only 12.3% (fi=56) reported working 40 hours per week or less, with 28% (n=218) of the practitioners averaging greater than 60 hours per week on the job. Fifteen and one-half percent (n=71) annually earned less than $30,000, 28% (n=128) earned between $30,000 and $49,999, 18.6% (n=85) earned between $50,000 and









38


$69,999, 10.9% (n= 50) earned between $70,000 and $89,999, 11.8% (n=54) earned between $90,00 and $109,999 and 15.0% (n=69) earned over $110,000 per year.

Twenty-three percent of respondents (n= 105) had never been married, 54% (n=247) were currently married, 10.5% (n=48) were divorced and unmarried, and 11.6% (n=53) were divorced and remarried once. Sixty-eight percent (f1=31 1) described themselves as the sole wage earner, 13.6% (n=62) described themselves as secondary wager earners, and 18.3% (n = 84) described themselves as equal wage earners in their families. Eighteen percent of respondents (f1=84) had between 1 and 3 children under the age of five in the household, 23.6% (f1=1 08) had between one and three children between the ages of six and 18 living at home, and 5.9% (n=27) of the respondents had one or two adult children (ages 19-25) living at home. Seventy-seven percent (n=350) had no children living at home. Six percent (n=27) of respondents reported having children with significant emotional (11=1 5), disciplinary (n_=1 0), or academic (n=1 5) problems.

Measures

The following four instruments comprised the major part of the mail survey used in this study: (1) the Health Professions Stress Inventory (HPSI), (2) the Coping Styles Questionnaire (CSQ), (3) the Burnout Measure (BM) and (4) Five questions which formed a composite measure of job satisfaction (SAT). (See









39


Appendix A.)

The Health Professions Stress Inventory (HPSI) (Wolfgang, 1988) consists of 30 job situations which health providers might be expected to encounter and find stressful. Respondents indicate how often they have found each situation to be stressful in their work environment, using a scale of 0 to 4 (never to very often). The HPSI has been proven to be both a valid and reliable measure of job stress in the health professions (Wolfgang, 1988). Lyons' (1971) Index of Work Related Tension (IWRT)was used as a measure of concurrent validity for the HPSI. Pearson product-moment correlations between scores on the HPSI and the IWRT were .75 for pharmacists, .76 for physicians, and .78 for nurses (p < .001) (Wolfgang, 1988). Cronbach's alpha coefficients assessing internal consistency of the HPSI were .89, .89, and .88 for pharmacists, nurses, and physicians, respectively.

The HPSI was modified slightly for the veterinary population in this research. The word "patient" was changed to "owner" in three instances (see Appendix A). In accord with the suggestion of Spielberger & Reheiser (1995), the perceived intensity of each stressor was assessed along with its' frequency. An overall job stress index was based on the sum of the cross-products of the Severity and Frequency scores. The overall index (HPSI-T) was compared with the original index (HPSI) (which is hereafter referred to as the frequency index (HPSI-F)) in order to ascertain the necessity of using both intensity and frequency scales. It









40


was concluded that use of both scales enhanced the understanding of gender differences in job stress. Comparisons with the scores reported in the literature for other health professionals (physicians, pharmacists, and nurses) were made based on the frequency index (HPSI--F) alone. Gender differences were assessed on each individual item, as well as on the overall, intensity and frequency scores. In addition, three stressors taken from Simpson and Grant's Measurement of Job Stressors (Simpson & Grant, 1991) were added ("Having owners whose values and beliefs about medical care differ greatly from your own;" "Problems related to business and management aspects of veterinary practice;" and "Having to make difficult ethical and moral

choices.") "Medical" practice was changed to "veterinary" practice in one instance and owner was substituted for "patient" in one instance. One additional item pertinent to the practice of equine medicine ("Worrying about the possibility of zoonotic illness or injury to yourself or your staff') was also added.

Because the HPSI "problematic (HPSI-P)" scale was conceived in addition to the typically used HPSI "frequency (HPSI-F)" scale, Cronbach's alpha was computed to assure internal reliability. Analysis of the responses to the HPSI-F scale indicated adequate internal reliability (Cronbach's alpha = .89) as did analysis of the responses to the HPSI-P scale (Cronbach's alpha = .91).

The Coping Styles Questionnaire (CSQ) was used to assess individual








41


coping style in the participants. The CSQ was developed by Roger, Jarvis, and Najarian in 1993. Responses to 78 items (chosen by the authors from a variety of sources) were subjected to principal factoring and rotated to a four-factor terminal solution following a scree test. The CSQ utilizes the 60 items with the highest loadings. Each item exemplifies ways in which individuals react to stressful situations. Respondents are asked to indicate whether each item illustrates a way they react to stress by circling "always," "often," "sometimes," or "never" for each item. A mean score was calculated for each of the following four coping styles: rational (RAT), detached (DET), emotional (EM), and avoidance (AV). The CSQ has test-retest reliabilities of .80, .79, .76, and .70 for each scale respectively, and it has coefficient alphas of .85, .90, .74 and .69 for each scale respectively. Because of the excessive length of the original questionnaire, only items which had factor loadings of .5 or higher in the original research (Roger, Jarvis, & Najarian, 1993) were used in this study. This procedure cut the CSQ to 30 items and is in accord with accepted means of decreasing lengthy questionnaires while maintaining reliability of measures (Zyzanski, 1993).

To investigate the validity of the Coping Styles Questionnaire as used in this study (thirty items vs. the original sixty), the responses to the thirty items were subjected to principal factoring, and rotated to a four-factor terminal solution following a scree test. With the exception of five items, all loaded on each of four








42


scales which were identical to those identified by the developers of the scale (Roger, Jarvis, & Najarian, 1993). One of the five items which appeared initially on the rational coping scale, loaded slightly more heavily on the detached scale. Four (of a total of 6 ) items intended to measure avoidance coping loaded on other scales. Thus, this scale was not considered to be valid and is not discussed further.

Burnout was measured via the Burnout Measure (Pines & Aronson, 1988; Schaufeli, Maslach, & Marek, 1993). Twenty.-one "feelings' are listed and the participant is asked to relate how often he or she experiences such feelings. A seven point scale is used with choices ranging from 1 (never) to 7 (always). A single score is derived by adding the points attained on the 17 negative feeling items. This is followed by adding the scores for the four positive items and subtracting this from the number 32. This figure is then added to the negative points attained and the sum is then divided by 21. The scale yields scores of 1 to ~7. A score (BOS) of 3 or higher is consistent with some degree of burnout (Pines & Aronson, 1988). The burnout inventory has been shown to be internally consistent (alpha = .92). Test-rettest reliabilities range from .66 to .89 (highest correlations were found when the testing interval was 4 weeks and lower correlations were found when the interval was 4 months). Takooshian (1994) ascribes excellent construct and criterion validity to the instrument as well.








43


Five job satisfaction questions were used following the lead of Martin & Shehan, 1989. Job satisfaction (SAT) was a composite measure of the responses to the five items (1=strongly disagree to 5=strongly agree). The specific items included were as follows: (1) I am satisfied with my current job, (2) 1 cannot think of jobs for which I would exchange, (3) If I had the choice to make again, I would choose the veterinary profession, (4) 1 do not intend to leave this job anytime in the near future, and (5)This job measures up to my initial expectations. The use of five, unweighted items remedies the inherent shortcomings of relying on a single-item measure of the construct (Martin & Shehan, 1989). Values for the scale of satisfaction range from 1 .0 (low satisfaction) to 5.0 (high satisfaction). The job satisfaction scale exhibited adequate reliability (Cronbach's alpha =.85).

In addition to the four above described instruments, the mail survey included a short demographics questionnaire (DQ) and a cover letter. The DQ asked respondents to provide information about their age, gender, years in the profession and in the current job, income, hours worked per week, etc. They were also asked to describe their family wage earning status, marital status and number of children. Finally, they were asked to rate their perceived overall life stress for the past year (see Appendix A). The cover letter (see Appendix B) made an appeal to colleagues for assistance. In an attempt to gain maximal cooperation, the letter was typed on AAEP letterhead stationery and was signed by the marketing director of the AAEP








44


and the principal researcher.

Procedure

The four-page mail survey along with a cover letter was sent by first-class mail to 968 (503 male and 465 female) veterinarians currently in private, predominantly equine practice selected by a stratified random procedure from the AAEP membership list. It was intended that a stratified random sampling procedure would be used to identify the potential participants so that a fairly equal number of potential male participants and female participants would be identified. However, only 465 female members who met the established criteria were identified. Thus, all of these women were included, and a similar number of men meeting all the criteria were randomly chosen.

The survey was mailed in AAEP outer envelopes with a business reply return envelope included. A target date for the receipt of all returned surveys (May 26, 1997) was listed in the survey cover letter. No surveys were returned as undeliverable. Of the 25 surveys which were returned after the deadline and were thus not included in the study, 14 were from women and 11 were from men. Perusal of these surveys revealed no obvious response differences from the surveys which were included in the study.

A reminder, an individually addressed and signed postcard, was mailed firstclass to all participants one week following the initial bailout of surveys. The








45


postcard personally appealed for help for the researcher and the organization in their efforts to get the surveys completed and returned to them. The postcard was signed with a blue pen by the researcher (see Appendix C). The study was terminated within four weeks of the initial mailing, and the data were double keypunched by qualified operators.















CHAPTER 4

RESULTS



Descriptive Statistics

The composite mean Health Professions Stress Inventory score (HPSI-T) Was 7.01 + 2.7, with a range of 1.09 to 16.77. The range of the scale for responding to items on the HPSI-T was 1.0 (not at all stressful) to 25.0 (extremely stressful). The mean score. of the male respondents was 6.2 + 2.4 and the mean score of the women was 7.6 + 2.7.

The composite mean Burnout Score (BOS) was 3.29 + 0.8, with a range of 1. 1 to 6.0. The response scale for the BOS yields scores of 1 to 7. A score of 1 to 2 indicates low burnout; a score of 2 to 3 indicates acceptable levels of stress and manageable level of work dissatisfaction; a score of 3-4 indicates the beginning of the burnout syndrome; a score of 4-5 indicates a state of clinical burnout; and scores over 5 indicate severe problems (Pines & Aronson, 1988). The mean score of the male respondents was 3.07 + 0.8 and the mean score of the female respondents was 3.47 + 0.8.

The composite mean job satisfaction score (SAT) was 3.70 + 1.0, with a range of 1-5. The lowest score (1) indicates extreme dissatisfaction and the highest score (5) indicates extreme satisfaction. The mean job satisfaction score of the men








47


was 3.72 + 1.0 and of the women was 3.68 + 1.0.

Data Analysis

The Statistical Analysis System (SAS) General Linear Model (GLM) procedure was used for all the analyses in this study. The final sample size (E=457) was sufficient to insure a +5% sampling error on a population of approximately 5000 (Salant & Dillman, 1994) when the population is expected to be relatively varied in their responses. Although a power analysis was not performed, according to Cohen (1992) the group sizes (males = 205; females = 252) were sufficiently large to detect medium or large effect sizes using means, correlations, chi-squared analyses, analyses of variance (ANOVA's) and multiple regression analyses at the alpha level of.01. (Minimum numbers needed to achieve the above would have been 95 males and 95 females).

Hypothesis 1

To investigate Hypothesis 1, that among equine veterinarians in private practice, job stress will be negatively correlated with job satisfaction and positively correlated with burnout, Pearson correlation coefficients were calculated between the scores on the Health Professions Stress Inventory (HPSI-T), the scores on the burnout measure (BOS), and the scores on the job satisfaction measure (SAT). Results revealed a moderate positive correlation between job stress and burnout (r = .54; p<.0001) and a moderate negative correlation between job stress and job








48


satisfaction (r = -,39; p< .0001). Therefore, Hypothesis 1 was supported.

Hypothesis 2

Hypothesis 2 stated that among equine veterinarians in private practice, there will be no gender differences in the overall level of job stress, job satisfaction, and burnout experienced. To test this hypothesis, several planned, preliminary analyses were conducted. A preliminary multivariate analysis of variance (MANOVA) and preliminary chi square analyses were conducted to determine if there were gender differences in other independent variables that might be confounds in an examination of gender differences. (These independent variables became dependent variables in the preliminary analyses). The dependent variables in the MANOVA were age, years in the profession, years in the current job, income, hours worked per week, number of veterinarians in the practice, number of children in the home, and perceived life stress. The results of the preliminary MANOVA (Wilks' Lambda) revealed significant main effects for gender in at least one of the dependent variables F(12, 414) = 22.5, p = .0001. The models of six of the seven follow-up univariate analyses of variance (ANOVA's) were significant for gender differences: age F(1, 426) = 101.8, p = .0001; years in the profession F(I, 426) = 135.7, p- = .0001; years in the current job F(1, 426) = 131.2, p = .0001; income F(1, 426) = 121.6, p = .0001; hours worked per week F(1, 426) = 18.4, p = .0001; number of children living at home F(1, 426) = 28, p = .0001; and perceived life








49


stress F(1, 426) = 6.06, p = .01.

Next, as part of the preliminary analyses, Pearson correlation coefficients were computed to determine the correlation between the continuous variables (now

viewed as independent, confounding variables for which gender differences were

found) and job stress (HPSI-T), job satisfaction (SAT) and burnout (BOS). No variable had a correlation which was high enough to potentially account for a significant amount of the variance in the scores. (See Table 4.1).



Table 4.1

Pearson correlation Coefficients of HPSI-T, BOS and SAT scores with the potentially confounding (independent) variables.


HPSI-T SAT BOS
Age -0.29* 0. 16** -0.23*
Years in Profession -0.34* 0.15** -0.22*
Years in job -0.33* 0.16** -0.20*
Income -0.26* 0.21 -0.26*
Hours/Week 0.06 0.01 0.12***
Children <6yrs. 0.01 0.05 -0.07
Children 6-18 -0.09 0.04 -0.03
Children 19-25 -0.06 0.07 -0.07
Total Children Home -0.09 0.08 -0.09
Life Stress 0.25* -0.15** 0.41*
*p < .0001 **p < .001 ***p < .01








50


Although none of the above correlations were high enough to suggest that the investigated independent variables would have an impact on the analysis to test Hypothesis 2, the three highest correlates with the HPSI-T, BOS and/or SAT scores (i.e., perceived life stress, income, and years in the profession ) were entered as covariates in the analysis to be certain that the gender differences found could not be attributed to these covariates. Clearly, if these covariates proved to be nonsignificant as expected, the others with lower correlations with the dependent variables of interest would also be non-significant. Since age and years in the current job were only slightly less correlated with the dependent variables than were years in the profession, it was logical to use only one as a covariate (i.e., years in the profession).

Preliminary chi-squared analyses were performed to determine if there were gender differences in the independent variables that were categorical (i.e. marital status, wage earning status, and board certification status, all of which are serving as dependent variables in the chi-square analyses). Significant gender differences were found for marital status (chi-square = 50.4, df = 5, p = .001) and wage earning status (chi-square = 39,67, df = 2, p = .001) but not for board certification status.

Since gender differences were found in the above specified variables, three MANOVAs were performed: one each with job satisfaction, job stress, and burnout as the dependent variable, gender as the independent variable and marital status,









51


wage earning status. board certification status and their interactions as covariates. No significant main effects were found for job satisfaction IF (32,441) =1.62, P >.02, or burnout F (32,441) 1.46, p > .05. The -main effect found in the model with job stress as the dependent variable IF (32,440), p > .0008 was attributed solely to the independent variable, gender. Thus, it was not necessary to include marital status, wage earning status, and board certification status in the analyses to test Hypothesis 2.

Hypothesis 2 states that there will be no gender differences in the overall levels of job stress, job satisfaction and burnout in veterinarians in private, equine practice. To examine this hypothesis, a MANCOVA was performed with the measures of job stress, job satisfaction and burnout as the dependent variables, gender as the independent variable, and respondents' years in the profession, income, and perceived life stress as the covariates.

The MANCOVA (Wilks' Lambda) revealed significant main effects for years in the profession F(3, 446) = 10.5, p = .0001, perceived life stress F (3, 446) = 29.1, p = .0001, income (F3, 446) = 7.3, p = .0001, and gender F(3,446) = 6.09, p = .0005. All three follow-up univariate ANCOVA's for the dependent variables (using years in the profession, income, and life stress as co-variates) were significant: Job stress F(4, 451) = 4.79, p = .03, Job satisfaction F(4,447) = 10.01, P = .0017, and Burnout F(4, 451) = 5.52, p = .02. Results revealed significant gender differences








52


in job stress (p< .0001), job satisfaction (p < .0001), and burnout (p < .0001) in equine veterinarians in private practice. These differences were shown not to be accounted for by years in the profession, income, or perceived life stress. Table 4.2 shows the least squares means of the scores of male and female veterinarians on the three measures of job stress, job satisfaction and burnout. As shown in this table, the women, as compared to the men in this study, had significantly higher job stress and burnout scores as well as significantly higher job satisfaction scores. Thus, Hypothesis 2 was not supported. (It is noteworthy that if the variance in job satisfaction due to income was not removed from the model, female veterinarians were not significantly more satisfied with their jobs then were male veterinarians.)



Table 4.2

Least Squares Means comparing male and female scores on the HPSI-T, SAT, and BOS.

Gender HPSI-T SAT BOS

Female 7.16 3.87 3.38
Male 6.81 3.52 3.18





Because the covariate income was unexpectedly significant when it

appeared in the model which investigated gender differences in job satisfaction,








5 3


R-square procedures were run to be certain that the variables chosen for use as covariates in the MANCOVA (i.e., life stress, income, years in the profession) were the most appropriate covariates. Thus, three R-square procedures with job stress, job satisfaction or burnout as the dependent variable in each and the 15 possible covariates and their combinations as the predictor variables in each were run to determine the best reduced model for each of the dependent variables. Results revealed that the covariates chosen (years in the profession, income, and life stress) were indeed the most appropriate covariates for the MANCOVA to test Hypothesis 2.

Hypothesis 3

Hypothesis 3 stated that gender differences in the severity and frequency of specific job stressors would be found. Although gender differences in the overall measure of job stress were examined using the covariates of number of years in the profession, income, and perceived life stress, it was concluded that gender differences in each individual stressor would be examined without using covariates. Using seventy separate ANCOVAs would diminish the power of the statistic so much so that potentially significant gender differences would not be identified. Thus, Hypothesis 3 was tested using two-tailed T tests. Since a total of 35 stressors were to be compared on both severity and frequency of occurrence, a









54


particularly stringent test was necessary to avoid finding significant differences which would occur due to chance alone. Adjustments were made, therefore, using permutations of p, a more rigorous criteria than the Bonferroni's adjustment.

Significant differences (p > .01) were found between the means of 20 of 70 items (16 of 35 on the severity scale and 4 of 35 on the frequency scale). In all of these instances, the women found the particular problem to be significantly more stressful than did the men. In fact, although not significant at the p = .01 level, women found all but 6 of the 70 items to be more stressful than did the men. And, the differences between the male and female scores on four of the six items which the men ranked as more stressful were of the magnitude of .03 or less. Notably, both sexes agreed relatively well on the frequency scale, but not well on the severity scale. Thus, Hypothesis 3 was confirmed: there were gender differences in the frequency and severity of specific job stressors reported by equine veterinarians in private practice. See Table 4.3.

Hypothesis 4

To test Hypothesis 4 that there will be no gender differences in coping styles reported by equine veterinarians, a MANCOVA was performed with the three valid coping scales as the dependent variables, gender as the independent variable, and years in the profession, perceived life stress and income as the covariates. These covariates were included given that they were used as covariates in the analyses









55


to test Hypothesis 2 which also was concerned with gender differences.

The MANCOVA (Wilks' Lambda) revealed significant main effects for years in the profession F(4, 440) = 5.4, p = .0003; perceived life stress F(4, 440) = 13.23, F- = .0001; income F(4, 440) = 2.96, p = .020; and gender F(4,440) = 3.98, p = .0035. The models of all three of the follow-up univariate ANCOVA's for the specific types of coping were significant: Rational coping F(3, 447) = 11. 13, p = .0001; Emotional coping F(4, 447) = 22.17, p = .0001; and Detached coping F(4, 447) = 23.89, p =.0001. Table 4.4 lists the least squares means of each gender for each coping style. These means show that men as compared to women used significantly more rational and detached coping styles, while women as compared to men used significantly more emotional coping. Thus, Hypothesis 4 was not supported.



T-able 4.4

Least Squares Means for Coping Styles of Male and Female Veterinarians Gender Rational Coping Emotional Coping Detached Coping

Females 1.96 3.20 2.55
Males 1.86 3.30 2.42

*Lower scores signify higher usage of each particular style of coping.








56


Hypothesis 5

Hypothesis 5 stated that the relationship between job stress and burnout will be mediated by coping style in equine veterinarians in private practice and that rational and detached coping styles will decrease the impact of job stress on burnout while avoidant and emotional coping styles would increase the impact of job stress on burnout. Recall that, due to the results of the factor analysis performed on the Coping Styles Questionnaire (CSQ), avoidant coping was omitted from this comparison.

To test this hypothesis, a multiple regression analysis was used with burnout score (BOS) as the dependent variable and job stress (HPSI-T), frequency of use of each of the three coping styles, and the interactions of job stress with each coping style use frequency as the predictor variables. It was found that 67% of the variance in BOS could be accounted for by job stress and coping styles use frequency and the interactions between job stress and frequency of use of each of the coping styles (F (9, 454) = 100.44, p = .0001).

Typical of multiple regression analysis, the order in which each predictor variable and its interaction is entered into the overall model will affect the outcome. Thus, three separate multiple regression analyses with BOS as the dependent variable in each and with one of the coping style use frequency variables, job stress (HPSI-T) and their interactions as independent variables in each were performed









Table 4.3: 57

Means of Ratings of Health Professions Stress Inventory (HPSI) Items By

Gender


PROBLEMATIC FREQUENT

Female Male Female Male
Having so much work to do that everything cannot be done well 3.50** 2.96 3.53 3.27
Experiencing conflicts with supervisors and/or administrators 2.09** 1.51 1.90* 1.52
Feeling ultimately responsible for patient outcomes 3.46 3.26 3.58 3.55
Not receiving the respect or recognition that you deserve from the public 2.63* 2.24 2.45 2.27 Being uncertain about what to tell an owner about the patient's condition 2.29 2.01 2.23 2.15 Caring for the emotional needs of owners 2.67 2.46 3.07* 2.66
Disagreeing with other professionals concerning the treatment of a patient 2.53* 2.23 2.30 2.22 Not having opportunities to share feelings and experiences with colleagues 2.56* 2.18 2.35 2.15 Experiencing conflicts with coworkers 2.24 1.92 2.12 1.99
Having job duties which conflict with family responsibilities 3.69 3.56 3.64 3.61
Allowing personal feelings/emotions to interfere with the care of patients 2.26** 1.88 2.07 1.93 Keeping up with new developments to maintain professional competence 2.86 2.60 2.63 2.38
Feeling that opportunities for advancement on the job are poor 2.43** 1.94 2.21 1.91
Trying to meet society's expectations for high quality veterinary care 2.29 2.13 2.15 2.12 Supervising the performance of coworkers 2.05 2.14 2.11 2.24
Dealing with "difficult" owners 3.31* 2.97 2.83 2.86
Not being recognized as a true health professional by other health 2.27** 1.74 1.99** 1.63
professionals
Being inadequately prepared to meet the needs of patients 2.45** 1.92 2.04 1.82
Possessing inadequate information regarding a patient's medical condition 2.65** 2.17 2.22 2.09 Not receiving adequate feedback on your job performance 2.48** 1.89 2.22** 1.86
Not having enough staffto adequately provide necessary services 2.95** 2.33 2.85* 2.41
Having nonhealth professionals determine the way you must practice 2.34 2.13 1.95 2.05
Not knowing what type ofjob performance is expected 1.85 1.56 1.65 1.53
**Perm-p > .01 *Perm-p >.05









Table 4.3 (continued) 58


PROBLEMATIC FREQUENT
Female Male Female Male
Being interrupted by phone calls or people while perfonning job duties 3.22**2.75 3.55 3.22 Not being allowed to participate in making decisions about yourjob 1,96** 1.39 1.74* 1.39 Not being challenged by your work 1.95 1.75 1.80 1.87
Feeling that you are inadequately paid as a health professional 3.76** 2.83 3.65**2.92 Caring for terminally ill patients 2.45** 2.00 2.26 2.01
Not being able to use your abilities to the fullest extent on the job 2.71 2.40 2.42 2.26 Fearing that a mistake will be made in the treatment of a patient 2.85** 2.17 2.49**2.13 Having owners whose values and beliefs about care differ greatly from 2.80* 2.43 2.38 2.39

yours
Problems related to business and management of veterinary practice 3.15 2.83 3.06 2.96 Worrying about the possibility ofzoonotic illness or injury to self or staff 2.12** 1.74 1.96 1.78 Having to make difficult ethical or moral choices 2.41** 1.99 2.15 2.01


**Perm-p > .01 *Perm-p >.05








59


to further and more clearly address this research hypothesis.

The multiple regression model with burnout as measured by BOS as the dependent variable and with job stress as measured by HPSI-T, frequency of use of rational coping (RAT), and the interaction of (RAT) with HPSI-T as the predictor variables was significant, F (3, 453) = 108.8, p = .0001, r squared = 42%. In this model, HPSI-T (t = 4.4, p =.001) and rational coping (t = 4.8, p = .001) but not the interaction term were significant predictors of burnout.

The multiple regression model with BOS as the dependent variable and HPSI-T, emotional coping (EM), and the interaction of EM with HPSI-T as the predictor variables was significant, F (3, 454) = 272.5, p = .0001, r squared = 65%. In this model, HPSI-T (t=-2.0, p = .05), EM (t = -11.9, p = .0001), and the interaction between HPSI-T and frequency of use of EM (t= 3.2, p = .0015) were all significant predictors of burnout. Data plots of the interaction between EM and HPSI-T indicate that as frequency of use of EM increased, the position association between stress and BOS strengthened.

The multiple regression model with BOS as the dependent variable and HPSI-T, frequency of use of detached coping (DET) and the interaction between HPSI-T and frequency of use of DET as predictor variables was significant, F(3, 454) = 97.3, p = .0001, r squared = 39%. In this model, HPSI-T (t= 2.6, p = .009) and frequency of use of DIET (t = 3.5, p = .0005) were significant predictors of








60


burnout. The interaction of job stress with frequency of use of detached coping style was not significant.

Coping style was shown to have an impact on burnout in this study. Specifically, higher usage of rational and detached coping styles was associated with decreased amounts of burnout, while higher usage of emotional coping (as well as job stress) was associated with higher amounts of burnout. Hypothesis 5, however, suggested that coping style would mediate (amplify or moderate) the impact of job stress and burnout. This proved to be true only for the frequency of use of the emotional style of coping, which strengthened the positive association of job stress and burnout.

Although Hypothesis 5 did not specify that gender differences would be found, two post-hoc multiple regression analyses identical to the original analysis for this hypothesis were performed on male and female data separately. Results were identical except that in women, the interaction between emotional coping and job stress was not a significant predictor of burnout. Rational and detached coping neither amplified nor moderated the effect of job stress on burnout in males or females; however, more frequent usage of emotional coping strengthened the positive association between job stress and burnout in men but not in women. Thus, Hypothesis 5 was only partially supported.








61


Summary of Results

As is readily predictable, among equine veterinarians in private practice, job stress was found to be negatively correlated with job satisfaction and positively correlated with burnout. Veterinarians who are women reported significantly more job stress and more burnout than did men, while at the same time reporting significantly higher job satisfaction. Although male and female equine practitioners seemed to agree on the frequency with which most job stresses occurred, women found 45.7% of these stressors significantly more stressful than did their male counterparts. Male and female equine practitioners reported using various coping styles with different frequencies: women used significantly more emotional coping than did men, who used significantly more detached and rational coping than did the females. Emotional coping style was the only style shown to interact (in men only) with job stress so as to increase the positive association between job stress and burnout.














CHAPTER 5

DISCUSSION


Four-hundred and ninety-two questionnaires were returned from an initial sample size of 968. This represents a response rate of 51 %, a rather remarkable return rate following receipt of a single four-page questionnaire by a group of healthcare professionals. Single survey bailouts to the general public are typically followed by a 20% response rate (Bailey, 1982).

This response rate compares favorably with previous results obtained from surveys of veterinarians. The American Association of Equine Practitioners typically achieves response rates of between 20-30% from its membership with single bailouts but no reminder postcard. A single bailout (of a considerably shorter survey than the present survey) to 1000 female veterinarians (Elkins & Kearney, 1992) achieved a 61% response rate and a single bailout to 1000 mostly male veterinarians achieved a response rate of 33% (Elkins & Elkins, 1987), Other health care professionals are notoriously difficult to survey. One of the largest studies sent over 3000 questionnaires to nurses, pharmacists and physicians and achieved a 42% response rate with a second, repeat mailing (Wolfgang, 1988).

The response rate of 51 % following a single bailout is particularly high in view of the fact that the survey was mailed in May, the month that 62





63






falls in the middle of the breeding, racing and showing seasons, the busiest times in the lives of equine practitioners, So why might the response of these veterinarians have been so high? It might be because the researcher was a colleague, and a former academic equine practitioner whose name may have been known to some of the participants. Further, the personal appeal made by the researcher on the follow-up postcard was likely influential. This high response rate, however, may have mostly been the result of interest in the topic of the research by the target participants. Indeed, equine veterinarians might be interested in job stress, coping and burnout in their profession--a profession noted for its stagnant, relatively low wages, increasing debt burdens and long working hours (Bovee, 1997).

The response rate of the women (57%) was much higher than that of the men (43%) in this study. This is in accord with the two previous surveys of veterinarians cited (Elkins & Elkins, 1987; Elkins & Kearney, 1992). Why might female veterinarians be so much more willing to respond than men? For one thing, three recent research efforts have substantiated significant gaps in earning between male and female veterinarians which cannot be explained by demographic variables (Bird, 1992; Smith, 1996; Frazier & Howell, 1991). For another, the women in this study were indeed more stressed on the job and more burned out than were the men, and reported significantly higher levels of general life stress. Thus, these





64


topics may be more salient to female veterinarians.

As in other areas of the workforce, women equine practitioners were significantly different from their male counterparts. Not surprisingly, there was only one variable assessed in which women were riot different from male practitioners, and that was the size of the practice in which they worked. This is in contrast to a previous survey (Smith, 1996) in which it was found that female veterinarians (who cared for various species, not solely the horse) tended to work in smaller practices. The significance of this disparity is that it has been shown (Smith, 1996) that practitioners who work in larger hospitals (determined by the number of doctors) tend to earn more than those who work at smaller clinics. The mean number of equine veterinarians per practice in this survey was 1.7, which is typical of most large animal, non-referral practices. In other words, most equine practitioners (regardless of gender) work in small practices.

As in the previous studies cited (Bird, 1992; Smith, 1996), these female practitioners were about 10 years younger and had practiced veterinary medicine for 8 fewer years than the men. As noted in the introduction, this is because of the late entry of women into the veterinary profession, which, until the 1990s was entirely male-dominated. For example, in 1970, 5.3% of the profession was female; this increased to 13.3% in 1980, and 22% in 1990 (Bird, 1992; Elkins & Kearney, 1992).

As in the previous studies, the women in this study worked about 10% fewer hours per week than did the men. Equal percentages of men and women were practice owners, although a greater percentage of women worked as associates in





65


practices. The average female in this study earned about $50,000 whereas the average male reported earnings in the range of $70,000 to $90,000. This finding has also been replicated previously. In the Bird (1992) study, a wage gap of $8000 was found between men and women which could not be explained by number of degrees, board certification, years of experience, hours worked per year, or ownership or partnership status. Smith (1996) found that even when earnings ratios were adjusted for years of experience, number of hours worked per week, number of patients seen per hour, type of practice, hospital size and geographical region, female veterinarians earned only 65% of the salary of their male counterparts.

Hypothesis one stated that among equine veterinarians in private practice, job stress is negatively correlated with job satisfaction and positively correlated with burnout. This hypothesis was supported. The moderate correlations between job stress and burnout (.54) and job stress and job satisfaction (-.39) are in line with previous research completed on female veterinarians (r = .38 between stress and burnout), pharmacists (L = -.46 between stress and satisfaction), and nurses (f = .47 between stress and satisfaction) (Elkins & Kearney, 1992; Wolfgang, 1995).

Hypothesis two stated that among equine veterinarians in private practice, there are no gender differences in the overall amount of job stress reported, and the degree of job satisfaction and burnout experienced. This hypothesis was not supported. Indeed, there were significant gender differences in the amount of job satisfaction reported, and the degree of burnout and job stress experienced. Female equine practitioners experienced significantly more job stress and burnout than men but reported significantly higher job satisfaction when the variance due to years in





66

the profession, life stress and income was taken into account,

The finding of significant gender differences in job satisfaction with women more satisfied than men is not congruent with most of the recent research which found no gender differences in overall job satisfaction (Cassidy & Warren, 1991; Martin & Hanson, 1985; Mortimer, Finch, & Maruyama, 1988). However, it is generally acknowledged that men and women most likely derive different amounts of satisfaction from differing aspects of their work. Unfortunately, due to space limitations the components of job satisfaction were not studied in this sample. These components (which are rarely enumerated and studied) very likely do vary with gender (as compared to a measure of overall job satisfaction). In general, in the few studies which have looked at the various components of satisfaction, women have defined job satisfaction by hours worked per week, travel time to and from work, job pressures, work hours, job exhaustion, satisfaction with time with their families, job complexity, and the equality of social relationships at work. Men, on the other hand, find such factors as leadership opportunities, advancement opportunities, ability to make decisions, work setting factors, autonomy, earning potential, and prestige as the things which make jobs most satisfying to them.

It is intriguing that female equine practitioners (who earn less than their male counterparts and indicated that they feel inadequately paid) report higher levels of job satisfaction (or equal levels when dissatisfaction due to income is not removed from the model) than do male equine practitioners in spite of feeling more burned out and stressed than male equine practitioners, Apparently, some aspects of veterinary medicine are more satisfying to women than to men. Again, this study





67


was unable to determine the specific nature of male and female veterinarian's satisfaction.

Contrary to Hypothesis two, female equine practitioners report being significantly more stressed and burned out on the job than men. The literature on job stress has produced inconsistent findings when assessing gender differences. A large meta-analysis (Martocchio & O'Leary, 1989) found no gender differences in job stress. Jick and Mitz (1985) suggest that failure to identify gender-related differences in the occupational stress literature may be due to sampling problems, with men being over-rep resented in management positions and women in clerical and service jobs. This study, in which the men and women held very similar professional positions, may support the suggestion of Jick & Mitz (1985) that gender related differences in occupational stress could possibly exist and might be found in research that eliminates job position as a confounding variable when examining these gender related differences.

It is actually not surprising that female veterinarians report significantly more burnout than male veterinarians. A number of burnout studies which have looked for gender differences report that women feel more burned out. Henning and Jardim (cited by Pines & Aronson, 1988), based on their interviews of over 100 female executives, believe that women see a career as a source of personal growth and self-fulfillment, and they seek in it the satisfaction of doing what they want to do. These perceptions, according to the authors' theory, may cause women to have higher expectations of their career than men, and if these expectations are not met, women may burn out. This did not prove to be true for the equine veterinarians in





68

the present research. When asked if their current job measured up to their initial expectations both male and female responses to this question were positive and were not significantly different from one another.

Role conflict is a common and logical explanation for female burnout. It has frequently been shown that the more successful a woman manager is in her career, the less successful she feels about her home life. For men, success at work has not been related to success or failure at home (Pines & Aronson, 1988). In this study, men and women reported equal amounts of stress due to "job duties which conflict with family responsibilities." This item was rated the most stressful item for men, and the second most stressful (behind inadequate pay) for women. Yet, women reported significantly more life stress than did men. Nonetheless, even when the variance due to life stress was removed, women still reported significantly more job stress and burnout than did men.

Women probably do pay a higher personal price for their careers in terms of family life, however. This was very much in evidence in this sample. Although the average female in the study was in her thirties (the average male was in his forties), 46% of these women had never married or were divorced and currently single; this compares to only 19.5% of the men. While clearly related to marital status, these female veterinarians had far fewer children than did their male counterparts and yet they report higher overall life stress.

Women's values may make ambition and competition secondary to satisfying work relationships as well as fulfilling relationships at home. Good personal work relationships have been found to be highly negatively correlated with burnout in





69


women but not in men in one study (Pines & Aronson, 1988). Since the average male and female veterinarian in this study came from a very small practice, it can be assumed that both men and women had few supportive work relationships. This relative isolation may well have affected women more than men. In fact, women rated "Not having opportunities to share feelings and experiences with colleagues" as more problematic than did men.

Hypothesis three stated that there are gender differences in the frequency and severity of specific job stressors reported by equine veterinarians in private practice. This hypothesis was supported. Perhaps the most unexpected finding of this study was the fact that female equine practitioners rated 64 of 70 job stressors as more stressful or more frequent in occurrence than did their male colleagues. The 20 items rated as significantly more stressful by women appeared to fall into four categories: 1. conflicts 2. lack of respect, recognition, fair pay, and opportunity

3. inability to deliver optimal patient care, and 4. emotional issues.

Hypothesis four stated that there are no gender differences in coping styles reported by equine veterinarians in private practice. This hypothesis was not supported. Female equine practitioners used emotional coping methods more frequently than did men while men used detached and rational coping methods more frequently than did women. The literature on this matter has been contradictory, although these findings further the observation of Parasuraman and Cleek (1984) who found a tendency (but not a statistically significant difference) for male managers to use more detached and rational and less emotional coping styles than female managers. Long and Kahn (1993) attributed these confusing gender





70

differences to three causes: 1. women experience more and different work stressors than do men. (This was perceived to be true by the female equine practitioners in the present research) 2. women are disadvantaged with regard to coping if they lack workplace coping resources such as power, perceived competence, and financial advantages. (This was also shown to be true for the women in this sample.) 3. gender role stereotypes (Hansen & O'Leary, 1983; Deaux & Ernswiller, 1977) affect male and female perceptions of what is appropriate coping (such as disclosing to friends and colleagues, for example).

Hypothesis five stated that the relationship between job stress and burnout is mediated by coping style in equine veterinarians in private practice. This hypothesis was only partially supported. Emotional coping increased the positive association of job stress with burnout, but only in men. Neither rational nor detached coping mediated the relationship between job stress and burnout. Thus, this hypothesis was only partially supported. In women, no type of coping style mediated the relationship between job stress and burnout.

The coping aspect of this research was based on Lazarus' transactional model of stress and coping which was discussed in Chapter Two. Lazarus' early work (1966) demonstrated that the way people evaluate an event (as it impacts upon their well-being) and how they cope with that event influences the existence and intensity of stress. Thus, if we are interested in studying the impact of stress in the work environment, we need to assess coping style as well. The present study attempted to recognize the impact of stress and coping on the outcome measure, burnout. It was hypothesized that method of coping would mediate the effect of job





71


stress on burnout, and specifically that rational and detached coping would moderate the impact, while emotional coping would amplify the effect. This seems to be a logical extension of Lazarus'theory, and is in agreement with the conception (and findings) of Billings and Moos (1981).

In this study, no interaction between coping style and job stress on the dependent variable, burnout, was observed with the exception of the interaction between job stress and the emotional style of coping in men. Relatedly, Wolfgang (1995) found no interaction between coping style and job stress when the dependent variable was job dissatisfaction. The findings of the current research were in contrast to what might be intuited and to what was hypothesized by Lazarus and his colleagues. Lazarus and Launier (1978), noting the importance of coping by itself, stated that "the ways people cope with stress are even more important. . than the frequency and severity of episodes of stress themselves" (p. 308). In the present study, it was seen that 67% of the variance in burnout was accounted for by job stress, coping and their interactions.

Osipow and Davis (1988) conducted a study to evaluate the relationship of coping resources to occupational stress and strain. By coincidence, the subjects were 213 (of 600) veterinary students who responded to a mail questionnaire. The researchers assessed the impact of four types of personal coping resources (recreation, self-care, social supports, and use of rational coping) on occupational stress and strain. All types of coping resources were effective in reducing job stress, with high levels of social support reducing the impact of all types of stressors. Least in importance was cognitive/rational coping which moderated only





72

the strain of responsibility and role overload.

It is certainly conceivable that the lack of interaction found between job stress and coping in this study (with the exception of emotional coping in men) may relate to the fact that the above types of coping resources were not assessed.

Implications of the Research Findings

This study fails to confirm the findings of the past decade of job satisfaction literature by revealing significant gender differences in job satisfaction among veterinarians in private equine practice. (However, as noted earlier, many of these studies did not control for variables which might have obscured such differences,) Due to financial constraints and survey space limitations, the specific components of job satisfaction in these equine veterinarians were not measured. The current study corrected the usual methodological limitations of failing to consider the impact of potentially confounding demographic variables when assessing gender differences in the work place. Age, years in the profession, years in the current job, marital status, life stress experienced, type of practice, board certification status, number of children (and whether or not they experienced significant disciplinary, academic or emotional problems), salary, family wage earning status, and practice ownership were assessed for gender differences. Although many gender differences in these variables were detected, none were shown to significantly impact the assessment of job stress, job satisfaction or burnout in this population with the exception of income, which impacted job satisfaction.

This study found significant gender differences in job stress and stressors, which is in contrast to three of four studies cited previously which were conducted





73

on physicians, This finding is also in contrast to much of the current literature on job stress in the general population as well. Linn, Yager, Cope and Leak's (1985) observation of the inverse relationship between age and job stress was confirmed. In the current study, years in the profession and years on the job exhibited a slightly stronger inverse correlation with job stress than did age. As in their study, a positive relationship was observed in the present study between age and job satisfaction, but this was a weak correlation only.

The current research confirms Miller and Kirsch' (1987) summary of their review of the gender and coping literature which concluded that the different sexes typically cope in different ways. In this study the styles of emotion-focused, detached, and rational coping all yielded significant gender differences. Further, it seems that these female veterinarians appraised threatening events as more stressful than did male veterinarians.

This study profoundly illustrates the problem noted in Spielberger and Reheiser's (1995) observation that most commonly used job stress measures tend to confound the frequency of occurrence of a stressful event with its perceived severity. The HPSI, as used in most of the health professions' studies of stress, merely asked about the frequency of occurrence of a particularjob stressor. In this study of veterinarians, men and women generally agreed upon the frequency of job stressor occurrence, but evaluated the severity of such occurrences very differently. Had the HPSI not been expanded (by the inclusion of the "problematic" scale in addition to the frequency scale), a great deal of information would have been lost, and gender differences in assessment of the severity of job stressors would have





74

gone undetected.

Another finding of pertinence to survey researchers is the confirmation of the observation of Calvert (1996) who suggested that a larger initial sample size with a single follow-up postcard would result in a comparable response size (albeit a lower response rate) to the mailing of a second questionnaire to non-respondents. Savings in manual labor, postage and printing costs, and response time were considerable in this study, and a more than adequate response rate (51%) was obtained.

A significant implication of this study for the practice of counseling and psychotherapy is that frequent use of emotional coping by male practitioners strengthened the relationship between job stress and burnout. Further, for the men and women in this study, more frequent use of rational and detached coping methods weakened the relationship between job stress and burnout. While many times the constraints inherent in the work environment appear to limit the possibility of constructive action by individuals, how the individual copes will have a huge impact on how he or she feels. This study suggests that for male practitioners when rational coping methods are difficult to enact, detached coping should be attempted, and emotional coping should be avoided. It is clear that the veterinary school curriculum and professional conferences would profit from the addition of seminars which teach students and practitioners about coping styles, enable them to understand how they personally cope with stress, and help them to change their coping styles if necessary.

In terms of veterinary practice, this study makes it clear that female equine





75


veterinarians feel inade quately paid, and earn less than male equine practitioners. These findings confirm the results of several other studies which have concluded that female veterinarians earn considerably less than men even when a number of demographic factors such as number of hours worked, years in the profession etc. are considered (Bird, 1992; Frazier & Howell, 1991; Smith, 1996; Verdon, 1997). Perhaps the women charge less for their services, see fewer patients per hour [although Smith (1996) showed this not to be true of his sample], or work fewer weeks per year. Maybe they give away more of their time to charity cases than do male veterina rians. Possibly they do not negotiate as well as men when buying a practice or buying equipment and drugs. Or, perhaps the general public "discriminates" against female equine practitioners by preferring to work with male veterinarians. It is clear that, for whatever reasons gender-related salary differences persist, female practitioners find these differences to be very stressful, and these salary differences need to be addressed and studied further by the profession.

Limitations of This Study

There are several methodological limitations of this study which relate to the fact that it was a mail survey study. A frequently cited problem of mail survey research is the issue of non-response bias. Is it accurate to assume that the respondents' ideas uniformly represent those of the non-respondents? In this study, a very good response rate (51 %) was attained, comparable to that expected from the general public following four mailings (Dillman, 1983)! If the respondents were the veterinarians most likely to suffer from job stress, then the analysis






76

benefits from a having a rigorous sample, If the respondents were those who found the time to respond because they were experiencing less stress, then significant results would be less likely and again the analysis would be more rigorous.

Barling, Kryl, and Bluen (1990) conducted an organizational survey and hypothesized that the time individuals took to complete and return their questionnaires would be associated with specific work attitudes, experiences and individual differences. Their results revealed that none of the organizational factors nor any demographic variable predicted when respondents would return their questionnaires. In some survey research, it is assumed that late responders fairly represent non-responders. In this study, while a rigorous review of the 25 questionnaires which were received beyond the due date was not conducted, perusal of the responses revealed that these individuals appeared to be no more stressed, dissatisfied or burned out than their more promptly responsive colleagues.

A second methodological limitation of the study is related to history. If the practitioners received the survey at the busiest time of their practice (which is precisely what occurred), they might report less satisfaction and more stress than is typical for them. In general, these veterinarians were fairly unstressed and satisfied, and if, indeed, this was their most stressful time of year, then it provided the best opportunity to study their coping behavior.

Another limitation of mail survey research and thus this study is that individuals may easily misread or misunderstand the questions and/or their intent. The impact of this problem was reduced by first administering the survey to several





77


veterinarians and personally interviewing them after they completed the survey. Areas of consistent confusion were Subsequently clarified and edited in the survey administered in the study.

In regard to coping, all researchers agree that coping is not a static process, although a single questionnaire attempts to portray it as such. Coping styles may change with situations or maturity. It is also likely that individual recollections of how they actually cope might be incorrect or biased.

A significant limitation of the coping portion of this study is that only three coping styles were assessed. Other positive means of dealing with life or work stress ( and therefore means of coping) should be assessed in future research. These could include self-care (eating habits, vitamin and mineral ingestion, meditation), recreation (appropriate amounts of exercising, time spent in enjoyable family activities), and social supports (the degree to which a respondent feels and requests support from co-workers, family, and friends.)

The general izab il ity of this study of equine practitioners to other veterinarians is, of course, questionable. The respondents were all individuals who work predominantly with horses. It is this author's feeling that of all private practice situations, equine practice could be the most stressful. The equine veterinarian spends a large part of the day, usually alone, driving in a truck from place to place. He or she routinely must cover his or her own emergency duty, since emergency clinics such as those commonly established for small animals do not typically exist. These individuals work outside, in extreme temperatures, with animals capable of killing them with a single blow to the head. But these doctors were not





78

overwhelmingly stressed, and again, use of this group may, therefore have made for a more rigorous study, or at least one in which coping was more readily studied.

Conversely, equine veterinarians may have fewer client contact hours than do small animal veterinarians who typically see between three and four patients and their owners per hour, versus equine practitioners who may only average one to two patients and their owners per hour. Since burnout relates highly to hours of direct client contact, small animal practitioners may, therefore suffer from more burnout than equine or other large animal practitioners.

It is definitely inappropriate to compare this population of private practitioners with veterinarians who are not in private practice. This author's earlier work with academic veterinarians (Welsch, 1991) did reveal gender differences in job satisfaction, much like those which have been shown for other university professors (Sandler, 1986; Olsen, 1991; Plascak-Craig & Bean, 1989; Etaugh, 1984).

Future Research

Future research in this area should compare veterinarians in other types of practice with one another. These results could aid veterinary students making decisions regarding further education and training by revealing which types of practice are more stressful or more satisfying than others. The components of job satisfaction should be studied in an effort to better compare male and female practitioners. An important question which needs to be assessed is whether a veterinarian's job is important to him or her most as a source of personal fulfillment or as a source of income. Another important consideration which was neglected in this study is the impact of physical limitations (specifically female size) on job stress





79

in those who must handle large animals.

It truly does appear that a comprehensive understanding of why veterinarians who are female earn significantly less than male veterinarians is needed, since the women in this research study found the salary situation to be more stressful than any other stressor assessed. This should not be a particularly difficult question to answer if enough of the necessary information is obtained (e.g. hours worked per week, weeks worked per year, patients seen per hour, years in the profession, etc.)

For future research on coping styles, the Coping Styles Questionnaire (CSQ) of Roger, Jarvis, and Najarian (1993) was shown to be a reasonable, though imperfect, instrument by the results obtained in this research. It can effectively be shortened from 60 to 30 questions (as was done in this study). The avoidance coping scale, which had the weakest coefficient alpha of the four scales in the authors' study (.69) (Roger, Jarvis, & Najarian, 1993) also had very poor factor loadings in the present study. In this author's opinion, the face validity of the scale is quite weak, and methods which people use to moderate the impact of job stress on the mind or body (such as drug or alcohol usage, TV watching, exercising to an extreme) should be included.

Further, the finding that the interaction between emotional coping and job stress increases burnout in men but not women should be confirmed and then studied further. For example, perhaps higher frequency of use of emotional coping in male veterinarians (since it appears to be more of a female veterinarian coping preference) lowers self-esteem in men more so than in women. Or perhaps the use of emotional coping has a greater negative impact on employees or clients if used





80

by a male than if used by a female, thus leading to increased job stress.

Conclusions

It is clear from this study that equine practice is a satisfying profession for both male and female practitioners and that these veteri narians do not exhibit high job stress or burnout.. Further, despite having higher job stress and burnout than male practitioners (and lower income), the female practitioners Studied were as satisfied as their male counterparts. It appears that coping style had a significant impact upon the difference in burnout noted between men and women. The quality of the male practitioners' work lives could most likely be significantly improved if they could learn to use fewer emotional coping strategies and more rational and detached coping strategies. In addition, it is likely that a more comprehensive study of coping as described earlier could readily identify other ways in which male and female veterinarians could improve the quality of their work lives.

Since salary typically impacts job satisfaction, research into the organizational and management factors, work pattern factors and other factors that may be contributing to the lower salaries earned by these female practitioners could be invaluable in understanding (and then correcting if possible) the income disparity noted between male and female equine practitioners. Female equine practitioners' job satisfaction would then most likely increase even further.

In conclusion, further comprehension of the complex processes of job satisfaction, stress, burnout and coping will have implications for the profession as it continues to become feminized. A full understanding of these complicated issues can prove important to professors and counselors of veterinarians as well as to the





81

practitioners themselves, Such research and associated interventions may ultimately insure long-term tenures and healthy quality of lives among equine practitioners, especially women, who may indeed be satisfied with their career choice bUt are at higher risk of stress and burnout.
























APPENDIX A

SURVEY









Instructions: Please circle one number in -hch of the columns that best describes how problematic each situation is and how frequently each condition occurs. Use the keys below.

How problematic is each condition for you (v.h i. happens, even if rely) in your current work situation?
1-not at all problematic; 2=not very problematic; 3=somewhat problematic; 4- problematic; 5--very
problematic.
How frequently did you experience stress flora each of these conditions in the past year?
l=never; occasionally; 3='about once each month; 4-=about once each week; 5=nearly every day HOW PROBLEMATIC? FREQUENT?
Having so much work to do that everything cannot be done well 1 2 3 4 5 1 2 3 4 5
Experiencing conflicts with supervisors and/or administrators 1 2 3 4 5 I 2 3 4 5
Feeling ultimately responsible fbr patient outcomes 1 2 3 4 5 1 2 3 4 5
Not receiving the respect or recognition that yu deserve from the public 1 2 3 4 5 1 2 3 4 5 Being uncertain about what to tell an owner about the patient's condition 1 2 3 4 5 1 2 3 4 5 Caringfortheemotionalneedsofowners 1 2 3 4 5 1 2 3 4 5
Disagreeing with other professionals concerning the treatment of a patient 1 2 3 4 5 1 2 3 4 5 Not having opportunities to share feelings and experiences with colleagues 1 2 3 4 5 1 2 3 4 5 Experiencing conflicts with coworkers 1 2 3 4 5 1 2 3 4 5
Having job duties which conflict with family responsibilities 1 2 3 4 5 1 2 3 4 5
Allowing personal feelings/emotions to interfere witfh the care of patients 1 2 3 4 5 1 2 3 4 5 Keeping up with new developments to maintaii professional competence 1 2 3 4 5 1 2 3 4 5 Feeling that opportunities for advancernrt on the job are poor 1 2 3 4 5 1 2 3 4 5
Trying to meet society's expectations for high quality veterinary care 1 2 3 4 5 1 2 3 4 5 Supervising the performance of coworkers 1 2 3 4 5 1 2 3 4 5
Dealing with "difficult" owners 1 2 3 4 5 1 2 3 4 5
Not being recognized as a true health professional by other health professional 1 2 3 4 5 1 2 3 4 5 Being inadequately prepared to meet the needs ofpatients 1 2 3 4 5 1 2 3 4 5
Possessing inadequate information regarding a patient's medical condition 1 2 3 4 5 1 2 3 4 5 Not receiving adequate feedback on your job performance 1 2 3 4 5 1 2 3 4 5
Not having enough staffto adequately providenecessary services 1 2 3 4 5 1 2 3 4 5
Having nonhealth professionals determine the way you must practice 1 2 3 4 5 1 2 3 45 Not knowing what type ofjob performance is expected 1 2 3 45 1 2 3 4 5
Being interrupted by phone calls orpeoplewhileperforingjob duties 1 2 3 4 5 1 2 3 4 5 Not being allowed to participate in making decisions about yourjob 1 2 3 4 5 1 2 3 4 5











83










84




HOW PROBLEMATIC? FREQUENT?
Not. being cballnged by your work 1 2 3 4 5 1 2 3 4 5
Feeling that you are inadequately paid as a health professional 1 2 3 4 5 1 2 3 4 5
Caring for terminally ill patients 1 2 3 4 5 1 2 3 4 5
Not being able to use your abilities to the fidlest extent on thejob 1 2 3 4 5 1 2 3 45
Fearing that a mistake will be made in the treatment of a patient 1 2 3 4 5 1 2 3 45

Having owners whose values and belief about care differ greatly from yours 1 2 3 4 5 1 2 3 4 5 Problems related to business and management ofveterinary practice 1 2 3 4 5 1 2 3 4 5
Worrying about the possibility ofzoonotic illness or injury to selfor staff 1 2 3 4 5 1 2 3 4 5 Having to make difficult ethical or moral choices 1 2 3 4 5 1 2 3 4 5

Instructions: Circle the appropriate number beside each of the next five statements.
Strongly disagree Strongly Agree
I am sadsfiod with my current job. 1 2 3 4 5
1 cannot think of any jobs for which I would exchange. 1 2 3 4 5
IfI had the choice to make again, I would choose the veterinary profession. 1 2 3 4 5 I do NOT'intend to leave this job anytime in the near future. 1 2 3 4 5
This job measure-i up to my iitial expectations 1 2 3 4 5

Instructions: Although people may react in different ways to different situations, we all tend to have a characteristic way of dealing with things which upset u.s. How would you describe the way you typically react to the stress you encounter during the practice of veterinary medicine?
A=always O=often S=sometimes N-never
See the situation for what it actually is and nothing more A 0 S N
See the problem as something separate from myself so I can deal with it A 0 S N Become miserable or depressed A O S N
Feel that no-one understands A O S N
Become lonely or isolated A 0 S N
Take action to change things A O S N
Avoid family or friends in general A O S N
Feel helpless-there's nothing you can do about it A 0 S N
Try to find out more information to help make a decision about things A 0 S N Keep things tc myselfand not let others know how bad things are for me A 0 S N Feel independent of the circumstances A 0 S N
Cry, or feel like crying A 0 S N
Pretend there's nothing the matter, even if people ask what's bothering me A 0 S N










85






A=always O=often S=sometimes N=never
Get things into proportion-nothing is really that important A 0 S N
Feel that time will sort things out A O S N
Feel completely clear-headed about the whole thing A 0 S N
Believe that I can cope with most things with the minimum of fuss A 0 S N Try not to let my heart rule my head A 0 S N
Try to find a logical, way of explaining the problem A 0 S N
Decide it's useless to get upset and just get on wit; things A 0 S N Feel worthless and unimportant A 0 S N
Trust in fke-things have a way of working out for the best A 0 S N
Use my past experience to try to deal with the situation. A 0 S N
Just take nothing personally A O S N
Criticize or blame myself A O S N
Talk about it as little as possible A 0 S N
Feel completely calm in the face of any adversity A 0 S N
See the thing as a challenge that must be met A 0 S N
Be realistic in my approach to the situation A 0 S N
Try to think about or do something else A 0 S N

Instructions: How often do you have any of the following experiences?

I--Never 2---Once in a great while 3=Rarely 4=Sometimes 5=Often 6-=Usually 7=Always
Feeling depressed 1 2 3 4 5 6 7
Having a good day 1 2 3 4 5 6 7
Being physicallyexhausted 1 2 3 4 5 6 7
Beingemotionallyexhausted 1 2 3 4 5 6 7
Being tired 1 2 3 4 5 6 7
Being happy 1 2 3 4 5 6 7
Being "wiped out" 1 2 3 4 5 6 7
Feeling"bumedout" 1 2 3 4 5 6 7
Being unhappy 1 2 3 4 5 6 7
Feelingrundown 1 2 3 4 5 6 7










86




l=Never 2=Once in a great while 3=Raely 4=Sometimes 5=Often 6=-Usually 7=Always
Feeling trapped 1 2 3 4 5 6 7
Feeling worthless 1 2 3 4 5 6 7
Being weary 1 2 3 4 5 6 7
Being troubled 1 2 3 4 5 6 7
Feeling disillusioned and resentful about people 1 2 3 4 5 6 7 Feeling energetic 1 2 3 4 5 6 7
Feeling way 1 2 3 4 5 6 7
Feeling hopeless 1 2 3 4 5 6 7
Feeling rejected 1 2 3 4 5 6 7
Feeling optimistic 1 2 3 4 5 6 7
Feeling anxious 1 2 3 4 5 6 7
Instructions: Please circle the appropriate answer below. This is a crucial part of the study. Please answer all the items!
Age 25-29 30-39 40-49 50-59 60-69 70+
Gender male female
Years in the profession of veterinary medicine 0-5 6-10 11-15 16-20 >20
Years in the currentjob 0-5 6-10 11-15 16-20 >20
Circle one: Practice owner(sole owner) Practice partner Employee
Yearlyincome <$30,000 30,000-49,999 50,000-69,999 70,000-89,999 90,000-109,999 >110,000 Hours worked per week <20 20-40 41-50 51-60 >60
Practice type small animal mixed animal large animal other (state type)Number of veterinarians in practice 1 2-4 5-7 >7
Board certification? no yes (state type)Marital status never married married divorced & single divorced & re-married remarried>1 widowed Family wage earning status: primary (or sole) wage earner secondary wage earner equal wage earner Number of children living at your home under age 5 between 6-18 between 19-25
Have any of your children had (circle if applicable) emotional disciplinary significant academic problems? On a scale of 1-10, how stressful would you rate your life (excluding your worklife) this past year? Not at all stressful Average for an American Very stressful
1 2 3 4 5 6 7 8 9 10
END OF SURVEY.
PLEASE ENCLOSE IN ENVELOPE PROVIDED.








APPENDIX B
COVER LETTER
May 9, 1997

Dear Colleague:

I am a former large animal internist (Ohio '80), currently in part-time private practice while completing my Ph.D. in Counseling Psychology at the University of Florida. Throughout my 17 years in veterinary medicine, I have observed that some veterinarians are quite happy with our profession and others are dissatisfied. With the help of the AAEP, I am hoping to begin to understand what veterinarians find stressful about their jobs, how satisfied they are with the profession as compared to other health care professionals, and what might distinguish those who are satisfied from those who are dissatisfied with veterinary medicine.

My funding is very limited, so I urge you to please take the time to fill out the next four pages. In pilot studies with small animal practitioners, the survey took an ave,,age of 12 minutes to complete. Your response is very important to the success of this study, as a high return rate is crucial for the results to be meaningful.

This is a completely ANONYMOUS questionnaire. In no way can you be
identified. Your completion of the questionnaire constitutes your agreement to allow the University of Florida to use the results for teaching and research purposes. You need not answer any question you do not wish to answer. I would gladly answer any questions, as would the Project Supervisor, Dr. Carolyn Tucker. You can reach me at 352-376-1790 (or e-mail bbwelsch@aol.com) and Dr. Tucker at 352-392-1532. I would welcome any written comments, either via e-mail or returned with the survey. The results will be presented in December, 1997 at the Phoenix meeting of the AAEP.

Thank you in advance for helping a colleague in need, and for contributing to an understanding of our changing profession!

Sincerely,


Barbara Brewer Welsch, DVM, EdS Ed Blach, DVM, MS, MBA
ACVIM, ACVECC AAEP Marketview
PLEASE REPLY BY MAY 26, 1997

87








APPENDIX C
FOLLOW UP POSTCARD






May 22, 1997

Last week an AAEP/University of Florida survey
addressing issues of equine practice was mailed to you. Your name was randomly selected to represent the membership.

If you have already completed and returned the
survey to us, please accept our sincere thanks. If not, please do so today. Because it has been sent to a small but representative sample of AAEP members, it is extremely important that yours also be included in the study if the results are to accurately represent the responses of equine practitioners.

Thank you so very much!


Barbara Brewer Welsch, DVM Project Director







88














REFERENCES


Andrisani, P. (1978). Job satisfaction among working women. Signs, 3, 558-607.

Bailey, K.D. (1982). Methods of social research. New York: The Free Press, Macmillan.

Bandura, A. (1977). Self-efficacy : Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215.

Barling, J., Kryl, I., & Bluen, S. (1990). "Finally, all the questionnaires are back": Bias in the time taken to return survey questionnaires. Personality and individual differences, 11,177-180.

Barone, D., Caddy, G., Katell, A., Roselione, F., & Hamilton, R. (1988).
The Work Stress Inventory: Organizational stress and job risk. Educational anad Psychological Measurement, 48, 141-154.

Beehr, T., & Schuler, R. (1980). Stress in organizations. In K. Rowland & G. Ferris (Eds.). Personnel management (pp. 390-419). Boston: Allyn and Bacon.

Ben-Porath, Y., & Tellegen, A. (1990). A place for traits in stress research. Psychological Inquiry, 1, 14-17.

Bickel, J. (1988). Women in medical education. A status report. New England Journal of Medicine, 319, 1579-1584.

Billings, A., & Moos, R. (1981). The role of coping responses and social
resources in attenuating the stress of life events. Journal of Behavioral Medicine,
4 139-157.

Bird, C. E. (1992). Comparing women's representation in three health professions: Dentistry, medicine, and veterinary medicine. (Doctoral dissertation, University of Illinois at Urbana-Champaign, 1992). Dissertation

89








90


Abstracts International, 53 (10A), 3678.

Bobula, J. (1980). Work patterns, practice characteristics, and
incomes of male and female physicians. Journal of Medical Education, 55, 826833.

Bovee, K.C. (1997). Stagnant wages, increasing debt burden: 'Profession needs a plan'. DVM: The News Magazine of Veterinary Medicine, 29(3),1.

Brief, A. P., & George, J.M. (1994). Psychological stress and the
workplace: A brief comment on Lazarus' outlook. In P.L. Perrewe & R. Crandall (Eds), Occupational stress: A handbook (pp.15-19). Washington, DC: Taylor & Francis. (Original work published 1991).

Calvert, P. (1996). Gender differences in clinician predictions of working alliance with borderline personality disordered and posttraumatic stress disordered clients (Doctoral Dissertation, University of Florida, 1996). Dissertation Abstracts International, 57(10B), 6561.

Carr, P., Friedman, R., Moskowitz, M., & Kazis, L. (1993). Comparing the status of women and men in academic medicine. Annals of Internal Medicine, 119, 908-913.

Cassidy, M., & Warren, B. (1991). Status consistency and work
satisfaction among professional and managerial women and men. Gender & Society, 5(2), 193-206.

Clark, A. (1996). The real job market. Journal of the American Veterinary Medical Association, 208, 860-862.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155159.
Costa, P. & McCrae, R. (1990). Personality: Another "hidden factor" in stress research. Psychological Inquiry. 22-24.

Cox, T. (1978). Stress. Baltimore, MD: University Park Press.

Dawis, R. V., & Lofquist, L. H. (1969). Adjustment to work. New York: Appleton-Centu ry-Crofts.








91


Dawis, R. V., & Lofquist, L. (1984). A psychological theory of work adjustment. Minneapolis: University of Minnesota Press.

Deaux, K., & Emswiller, M. (1974). Explanations for successful
performance on sex-linked tasks: what is skill for the male is luck for the female. Journal of Personality and Social Psychology, 29., 80-85.

Dial, T., Bickel, J., & Lewicki, A. (1989). Sex differences in rank attainment among radiology and internal medicine Faculty. Academic Medicine, 64, 198-202.

Dillman, D. A. (1983). Mail and other self-administered questionnaires. In P. Rossi, J. Wright, & A. Anderson (Eds.), Handbook of survey research (pp. 359-377). San Diego: Academic Press, Inc.

DiSalvo, B., Lubbers, C., Rossi, A., & Lewis, J. (1994). The impact of gender on work-related stress. In P. L. Perrewe & R. Crandall (Eds.), Occupational stress: A handbook (pp. 39-50). Washington, DC: Taylor & Francis.

Doherty, W., & Baldwin, C. (1985). Shifts and stability in locus of control during the 1970's: Divergence of the sexes. Journal of Personality and Social Psychology, 48, 1048-1053.

Elkins, A., & Elkins, J. (1987). Professional burnout among U.S.
veterinarians: How serious a problem? Veterinary Medicine, 82, 1245-1250.

Elkins, A., & Kearney, M. (1992). Professional burnout among female veterinarians in the United States. Journal of the American Veterinary Medical Association, 200, 604-608.

Etaugh, C. (1984, Fall). Women faculty and administrators in higher
education: Changes in their status since 1972. Journal of National Association for Women Deans, Administrators, and Counselors, pp. 21-25.

Folkman, S., & Lazarus, R. (1980). An analysis of coping in a middleaged community sample. Journal of Helath and Social Behavior, 21, 219-239.

Folkman, S., & Lazarus, R. (1985). If it changes it must be a process: Study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology, 48, 150-170.








92


Folkman, S., & Lazarus, R. (1988). Coping as a mediator of emotion. Journal of Personality and Social Psychology. 54. 466-475.

Fraser, T. (1983). Human stress, work and -job satisfaction. A critical approach. Geneva, Switzerland: International Labour Office.

French, J.R., Caplan, R.D., & Van Harrison, R. (1982). The mechanisms of -job stress and strain. Chichester, England: Wiley.

Glenn, N., & Weaver, C. (1982). Further evidence on education and job satisfaction. Social Forces. 61.46-55.
Gehrke, B. (1995). Employment, starting salaries, and educational
indebtedness of 1995 graduates of US veterinary medical colleges. Journal of the American Veterinary Medical Association, 207, 1290-1291.

Gehrke, B. (1996). Gender redistribution in the veterinary medical
profession. Journal of the American Veterinary Medical Association, 208. 12541255.

Goode, W. (1960). A theory of strain. American Sociological Review, 25. 483-496.

Hamilton, S., & Fagot, B. (1988). Chronic stress and coping styles: A comparison of male and female undergraduates. Journal of Personality and Social Psychology. 55. 819-823.

Hansen, R., & O'Leary, V. (1983). Actresses and actors: The effects of sex on causal attributions. Basic and Applied Social Psychology A4, 209-230.

Harris, J. R. (1994). An examination of the transaction approach in
occupational stress research. In P.L. Perrewe & R. Crandall (Eds.), Occupational stress: A Handbook (pp. 21-28). Washington, DC: Taylor and Francis. (Original work published 1991).

Havlovic, S. & Keenan, J. (1995). Coping with work stress: The influence of individual differences. In R. Crandall & P. Perrewe (Eds.), Occupational stress: A handbook (pp. 179-192). Washington, DC: Taylor & Francis, Ltd.

Herzberg, F., Mausner, B., & Snyderman, B. (1959). The motivation to work. -New York: Wiley.








93


Holt, R. R. (1982). Occupational stress. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress (pp. 419-444). New York: Free Press.

Hopkins, A. (1983). Work and Job satisfaction in the public sector. Totowa, NJ: Rowman & Allan held.

Hoppock, R. (1935). Job satisfaction. New York: National Occupational Conference, Harper.

Ivancevich, J. & Matteson, M. (1980). Stress and work: A managerial perspective. Glenview, IL: Scott, Foresman.

Jackson, S., & Schuler, R. (1985). A meta-analysis and conceptual critique of research on role ambiguity and role conflict in work settings. Organizational Behavior and Human Decision Process. 36. 16-78.

Jick, T., & Mitz, L. (1985). Sex differences in work stress. Academy of Management Review. 10. 408-420.

Jones, J.W. (1981). Diagnosing and treating staff burnout among health professionals. In J.W. Jones (Ed.), The burnout syndrome-(pp. 107-126). Park Ridge, IL: London House Press.

Kevorkian, C., & Tuel, S. (1994). Faculty departure from academic
programs of physical medicine and rehabilitation. American Journal of Physical Medicine & Rehabilitation. 73, 378-386.

Lazarus, R. (1966). Psychological stress and the coping process.. New York: Oxford University Press.

Lazarus, R. (1991). Psychological stress in the workplace. Journal of Social Behavior and Personality 6. 1-13.

Lazarus, R. S. (1995). Psychological stress in the workplace. In R.
Crandall & P. Perrewe (Eds.). Occupational stress., (pp. 3-19) Washington, DC: Taylor and Francis.

Lazarus, R., & Folkman, S. (1984). Stress, appraisal. and coping. New York: Springer.

Lazarus, R. S., & Launier, R. (1978). Stress-related transactions between person and environment. In L. A. Pervin & M. Lewis (Eds.), Perspectives in




Full Text
ACKNOWLEDGEMENTS
I would like to express my most sincere thanks and
appreciation to all who have helped me complete this dissertation. I am
grateful to my advisor and friend, Dr. Carolyn Tucker, for her support,
expertise, commitment to excellence, and attention to detail. I am thankful to
Dr. Ellen Amatea, who suggested the exploration of coping in the
veterinarians I was intending to study. Further, I am indebted to Dr. John
Dixon for his invaluable statistical expertise which was provided with patience
and incredible promptness.
I wifi also be forever grateful to my former colleagues, the
members of the Association of Equine Practitioners who responded to my
survey, and to their director Gary Carpenter, who provided both the funding
and the enthusiasm for the undertaking of this work. Lastly, I owe thanks to
my dear friend and veterinary colleague, Dr. Anne Koterba, who suggested
that the AAEP might be interested in working with me, to Dr. Ed Blach of the
/AEP who advised me on the marketing aspects of survey research, and to
Dave Foley of the AAEP who coordinated the printing, labeling, and mailing
process.


5
Two major competing theories of job satisfaction in the literature include the
fit hypothesis (Locke, 1976) and the work attribute hypothesis (Herzberg,
Mausner, & Snyderman, 1959). The fit hypothesis addresses the compatibility of
externa! work features and individual worker characteristics. This theory
emphasizes the differences in how workers react to their jobs and the preponderant
importance of these variations in influencing satisfaction and/or dissatisfaction with
work. In contrast, the work attribute hypothesis states that the features of the work
itself coupled with advancement, recognition, responsibility, and growth are of
preeminent importance for job satisfaction, and individual attributes make little
difference. Many authors have defined job satisfaction over the years (Hoppock,
1935; Super, 1942; Hopkins, 1983; Dawis & Lofquist, 1969, 1984). Regardless of
whether the concept of needs and values or individual perceptions is employed, the
common underlying premise of the most recent perspectives on job satisfaction is
that there is something within an individual which corresponds to his or her reaction
to (and consequent satisfaction with) the job.
Despite the plethora of job satisfaction research, much of it ignores the
concept of job stress. There is an even larger body of literature concerning job
stress. Typically, this research addresses the issue of job satisfaction minimally by
asking the participants one or two global questions aimed at measuring job
satisfaction. There is no end to the number of studies showing that stress in the


APPENDIX A
SURVEY


18
Ivancevich (1987) preventable costs relating to diminished productivity, turnover,
and absenteeism are in the realm of $2800 per person employed. This figure does
not even include such things as workers compensation, disability payments, and
early retirement benefits.
Despite a burgeoning of research in the occupational stress area, the
measurement of job stress has been inconsistent and confusing. Most of the
literature has relied on answers to only one or a few questions about experienced
stress (Barone, Caddey, Katell, Roselione, & Hamilton 1988). Further, job stress
measures tend to confound the frequency of occurrence of a stressful event with its
perceived severity (Spielberger & Reheiser, 1995). Some authors have contended
that research in work stress should focus on specific aspects of the job which
workers find problematic (Jackson & Schuler, 1985), while others have argued for
the development of generic measures to facilitate comparing stress levels across
occupational groups (Murphy & Hurrell, 1987).
In response to the criticisms of past measures, Spielberger and colleagues
developed the Job Stress Survey (JSS), adapting it from earlier surveys which they
had designed to measure job stress in policeman and later, teachers (Spielberger
& Reheiser, 1995). This 30-item psychometric instrument was designed to assess
the perceived intensity and frequency of occurrence of working conditions which
could adversely affect the psychological well-being of employees. The survey was
intended to form a generic job stress measure and has been used repeatedly in


77
veterinarians and personally interviewing them after they completed the survey.
Areas of consistent confusion were subsequently clarified and edited in the survey
administered in the study.
In regard to coping, all researchers agree that coping is not a static process,
although a single questionnaire attempts to portray it as such. Coping styles may
change with situations or maturity. St is also likely that individual recollections of
how they actually cope might be incorrect or biased.
A significant limitation of the coping portion of this study is that only three
coping styles were assessed. Other positive means of dealing with life or work
stress ( arid therefore means of coping) should be assessed in future research.
These could include self-care (eating habits, vitamin and mineral ingestion,
meditation), recreation (appropriate amounts of exercising, time spent in enjoyable
family activities), and social supports (the degree to which a respondent feels and
requests support from co-workers, family, and friends.)
The generalizability of this study of equine practitioners to other veterinarians
is, of course, questionable. The respondents were all individuals who work
predominantly with horses. It is this authors feeling that of all private practice
situations, equine practice could be the most stressful. The equine veterinarian
spends a large part of the day, usually alone, driving in a truck from place to place.
He or she routinely must cover his or her own emergency duty, since emergency
clinics such as those commonly established for small animals do not typically exist.
These individuals work outside, in extreme temperatures, with animals capable of
killing them with a single blow to the head. But these doctors were not


at the University of Florida in August, 1998, following completion of her
predoctoral internship at the University of Florida Counseling Center. She
continues to work part-time in small animal veterinary practice in Gainesville,
Florida.
101


47
was 3.72 1.0 and of the women was 3.68 1.0.
Data Analysis
The Statistical Analysis System (SAS) General Linear Model (GLM)
procedure was used for all the analyses in this study. The final sample size (n=457)
was sufficient to insure a 5% sampling error on a population of approximately 5000
(Saiant & Dillman, 1994) when the population is expected to be relatively varied in
their responses. Although a power analysis was not performed, according to Cohen
(1992) the group sizes (males = 205; females = 252) were sufficiently large to detect
medium or large effect sizes using means, correlations, chi-squared analyses,
analyses of variance (ANOVA's) and multiple regression analyses at the alpha level
of .01. (Minimum numbers needed to achieve the above would have been 95 males
and 95 females).
Hypothesis 1
To investigate Hypothesis 1, that among equine veterinarians in private
practice, job stress will be negatively correlated with job satisfaction and positively
correlated with burnout, Pearson correlation coefficients were calculated between
the scores on the Health Professions Stress Inventory (HPSI-T), the scores on the
burnout measure (BOS), and the scores on the job satisfaction measure (SAT).
Results revealed a moderate positive correlation between job stress and burnout
(r = .54; p<.0001) and a moderate negative correlation between job stress and job


SI
practitioners themselves. Such research and associated interventions may
ultimately insure long-term tenures and healthy quality of lives among equine
practitioners, especially women, who may indeed be satisfied with their career
choice but are at higher risk of stress and burnout.


66
the profession, life stress and income was taken into account.
The finding of significant gender differences in job satisfaction with women
more satisfied than men is not congruent with most of the recent research which
found no gender differences in overall job satisfaction (Cassidy & Warren, 1991;
Martin & Hanson, 1985; Mortimer, Finch, & Maruyama, 1988). However, it is
generally acknowledged that men and women most likely derive different amounts
of satisfaction from differing aspects of their work. Unfortunately, due to space
limitations the components of job satisfaction were not studied in this sample.
These components (which are rarely enumerated and studied) very likely do vary
with gender (as compared to a measure of overall job satisfaction). In general, in
the few studies which have looked at the various components of satisfaction,
women have defined job satisfaction by hours worked per week, travel time to and
from work, job pressures, work hours, job exhaustion, satisfaction with time with
their families, job complexity, and the equality of social relationships at work. Men,
on the other hand, find such factors as leadership opportunities, advancement
opportunities, ability to make decisions, work setting factors, autonomy, earning
potential, and prestige as the things which make jobs most satisfying to them.
It is intriguing that female equine practitioners (who earn less than their male
counterparts and indicated that they feel inadequately paid) report higher levels of
job satisfaction (or equal levels when dissatisfaction due to income is not removed
from the model) than do male equine practitioners in spite of feeling more burned
out and stressed than male equine practitioners. Apparently, some aspects of
veterinary medicine are more satisfying to women than to men. Again, this study


48
satisfaction (r = -.39; p< .0001). Therefore, Hypothesis 1 was supported.
Hypothesis 2
Hypothesis 2 stated that among equine veterinarians in private practice,
there will be no gender differences in the overall level of job stress, job satisfaction,
and burnout experienced. To test this hypothesis, several planned, preliminary
analyses were conducted. A preliminary multivariate analysis of variance
(MANOVA) and preliminary chi square analyses were conducted to determine if
there were gender differences in other independent variables that might be
confounds in an examination of gender differences. (These independent variables
became dependent variables in the preliminary analyses). The dependent variables
in the MANOVA were age, years in the profession, years in the current job, income,
hours worked per week, number of veterinarians in the practice, number of children
in the home, and perceived life stress. The results of the preliminary MANOVA
(Wilks Lambda) revealed significant main effects for gender in at least one of the
dependent variables F(12, 414) = 22.5, £ = .0001. The models of six of the seven
follow-up univariate analyses of variance (ANOVAs) were significant for gender
differences: age F(1, 426) = 101.8, £ = .0001; years in the profession F(1, 426) =
135.7, ^ = .0001; years in the current job F(1, 426) = 131.2, £ = .0001; income F(1,
426) = 121.6, g = .0001; hours worked per week F(1, 426) = 18.4, £ = .0001;
number of children living at home F(1, 426) = 28, £ = .0001; and perceived life


90
Abstracts International. 53 (10A), 3678.
Bobula, J. (1980). Work patterns, practice characteristics, and
incomes of male and female physicians. Journal of Medical Education. 55. 826-
833.
Bovee, K.C. (1997). Stagnant wages, increasing debt burden: Profession
needs a pian. DVM: The News Magazine of Veterinary Medicine. 29(31.1.
Brief, A. P., & George, J.M. (1994). Psychological stress and the
workplace: A brief comment on Lazarus outlook. In P.L. Perrewe & R. Crandall
(Eds), Occupational stress: A handbook (pp, 15-191. Washington. DC: Taylor &
Francis. (Original work published 1991).
Calvert, P. (1996). Gender differences in clinician predictions of working
alliance with borderline personality disordered and posttraumatic stress
disordered clients (Doctoral Dissertation, University of Florida, 1996).
Dissertation Abstracts international, 57(1 OB), 6561.
Carr, P., Friedman, R., Moskowitz, M., & Kazis, L. (1993). Comparing the
status of women and men in academic medicine. Annals of Internal Medicine.
119. 908-913.
Cassidy, M., & Warren, B. (1991). Status consistency and work
satisfaction among professional and managerial women and men. Gender &
Sodetx^5(2), 193-206.
Clark, A. (1996). The real job market. Journal of the American Veterinary
Medical Association. 208, 860-862.
Cohen, J. (1992). A power primer. Psychological Bulletin. 112(1), 155-
159.
Costa, P. & McCrae, R. (1990). Personality: Another hidden factor in
stress research. Psychological Inquiry,1,22-24.
Cox, T. (1978). Stress. Baltimore, MD: University Park Press.
Dawis, R. V., & Lofquist, L. H. (1969). Adjustment to work. New York:
Appleton-Century-Crofts.


Abstract of Dissertation Presented to the Graduate School of the University of
Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of
Philosophy
GENDER DIFFERENCES IN JOB STRESS, BURNOUT, AND JOB
SATISFACTION AS MEDIATED BY COPING STYLE OF VETERINARIANS
IN PRIVATE EQUINE PRACTICE
By
Barbara B. Welsch
August, 1998
Chairman: Carolyn M. Tucker, Ph.D.
Major Department: Psychology
A mail survey of 457 practicing equine veterinarians investigated gender
differences in job stress, burnout, job satisfaction, and coping style. Participants
included 205 men and 252 women who completed the Health Professions Stress
Inventory (HPSI-T), the Burnout Measure (BOS), the Coping Styles Questionnaire
(CSQ), a measure of Job Satisfaction (SAT), a single question assessing overall life
stress and a demographics questionnaire.
A moderate correlation between job stress and burnout and between job
stress and job satisfaction was found. Women reported significantly higher levels of job
satisfaction, but were significantly more stressed and more burned out than their male
colleagues. These gender differences remained when controlling for the effects of age,
years in the profession, years in the current job, income, hours worked, marital status,
number of children in the household, and life satisfaction.
VI


Instructions: Please circle one number in each of the columns that best describes how
problematic each situation is and how frequently each condition occurs. Use the keys below.
How problematic is each condition for you (when it happens, even if rarely) in your current work situation?
l=not at ali problematic; 2=not very problematic; 3=somewhat problematic; 4= problematic; 5=very
problematic.
.How frequently did you experience stress ora each of these conditions in the past year?
l=never; 2=occasionally; 3=about once each month; 4=about once each week; 5=nearly every day
HOW PROBLEMATIC? FREQUENT?
Having so much work to do that everything cannot be done well 12345 12345
Experiencing conflicts with supervisors and/or administrators 12345 S 2345
Feeling ultimately responsible for patient outcomes 12345 12345
Not receiving the respect or recognition that you deserve from the public 12345 12345
Being uncertain about what to tell an owner about fee patients condition 12345 12 3 45
Caring for the emotional needs of owners 12345 12345
Disagreeing with other professionals concerning the treatment of a patient 1 2 345 1 2 345
Not having opportunities to share feelings and experiences with colleagues 12345 12345
Experiencing conflicts with coworkers 1 2345 12345
Having job duties which conflict with family responsibilities 1 2. 345 12345
Allowing personal feelings/emotions to interfere vrith the care of patients 12345 12345
Keeping up with new developments to maintain professional competence 1 2 3 4 5 12345
Feeling that opportunities for advancement or) the job are poor 12345 12345
Trying to meet societys expectations for high quality veterinary care 1234 5 12345
Supervising the performance of coworkers 1 2 345 12345
Dealing with difficult owners 12345 12345
Not being recognized as a true health professional by other health professional 12345 1 2345
Being inadequately prepared to meet the needs of patients 12345 1234 5
Possessing inadequate information regarding a patient's medical, condition 12345 12345
Not receiving adequate feedback on your job performance 12345 12345
Not having enough staff to adequately provide necessary services 12345 12345
Having nonhealth professionals determine the way you must practice 1 2345 12345
Not knowing what type of job performance is expected 12345 123 4 5
Being interrupted by phone calls or people while performing job duties 12345 12345
Not being allowed to participate in making decisions about your job 12345 12345
83


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
Doctor of Philoso
( 1
DhyT-
A/1
v nfnWtrL v
i 1 /
Professor of Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Ellen S. Amatea
Professor of Counselor Education
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree
Professor Emeritus of Psychology
f Doctor of Philosophy.
t 77'
-Ay
Harry A. Grter
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Dqctor of Philosophy.
/"PC
' Shae Graham Kosch
Professor of Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in^
quality, as a dissertation for the degree Doctor oj^Philosophy.
RobeffC. Ziller
Professor of Ps'


95
Martocchio, J., & OLeary, A. (1989). Sex differences in occupational
stress: A meta-analytic review. Journal of Applied Psychology, 74. 495-501.
Maslach, C. (1986). Stress, burnout, and workaholism. In R. Kilburg, P.
Nathan, & R. Thoresan (Eds.), Professionals in distress. Issues, syndromes and
solutions in psychology, (pp. 51-75) Washington, DC: American Psychological
Association.
Mason, E. S, (1995). Gender differences in job satisfaction. Journal of
Social Psychology, 135, 143-151.
Matteson, M., & Ivancevich, J. (Eds.). (1987). Controlling work stress:
Effective human resource and management strategies. San Francisco: Jossey-
Bass.
Miller, J. (1990). Individual and occupational determinants of job
satisfaction: A focus on gender differences. Sociology of Work and
Occupations.7, 337-366.
Miller, S., & Kirsch, N. (1987). Sex differences in cognitive coping with
stress. In R. Barnett, L, Biener, & G. Baruch (Eds.), Gender and stress (pp. 278-
307). New York: Free Press, Macmillan, Inc.
Mobley, G., Jaret, C., Marsh, K., & Lim, Y. (1994). Mentoring, job
satisfaction, gender, and the legal profession. Sex Roles. 31, 79-98.
Mortimer, J., Finch, M., & Maruyama, G. (1988). Work experience and job
satisfaction: Variation by age and gender. In J. Mortimer & K. Borman (Eds.),
Work experience and psychological development through the life span (pp. 109-
155). Boulder: Westview Press.
Mottaz C. (1986). Gender differences in work satisfaction, work-related
rewards and values, and the determinants of work satisfaction. Fluman
Relations. 39. 359-370.
Murphy, L., & Hurrell, J. (1987). Stress measurement and management in
organizations: Development and current status. In A.Riley & S. Zaccaro (Eds.),
Occupational stress and organizational effectiveness (pp.29-51). New York:
Praeger.
Murray, M. & Atkinson, T. (1981). Gender differences in correlates of job
satisfaction. Canadian Journal of Behavioral Science. 13. 44-52.


42
scales which were identical to those identified by the developers of the scale
(Roger, Jarvis, & Najaran, 1993). One of the five items which appeared initially on
the rational coping scale, loaded slightly more heavily on the detached scale. Four
(of a total of 6 ) items intended to measure avoidance coping loaded on other
scales. Thus, this scale was not considered to be valid and is not discussed further.
Burnout was measured via the Burnout Measure (Pines & Aronson, 1988;
Schaufeli, Maslach, & Marek, 1993). Twenty-one feelings are listed and the
participant is asked to relate how often he or she experiences such feelings. A
seven point scale is used with choices ranging from 1 (never) to 7 (always). A
single score is derived by adding the points attained on the 17 negative feeling
items. This is followed by adding the scores for the four positive items and
subtracting this from the number 32. This figure is then added to the negative
points attained and the sum is then divided by 21. The scale yields scores of 1 to
7. A score (BOS) of 3 or higher is consistent with some degree of burnout (Pines
& Aronson, 1988). The burnout inventory has been shown to be internally
consistent (alpha = .92). Test-rettest reliabilities range from .66 to .89 (highest
correlations were found when the testing interval was 4 weeks and lower
correlations were found when the interval was 4 months). Takooshian (1994)
ascribes excellent construct and criterion validity to the instrument as well.


59
to further and more dearly address this research hypothesis.
The multiple regression model with burnout as measured by BOS as the
dependent variable and with job stress as measured by HPSI-T, frequency of use
of rational coping (RAT), and the interaction of (RAT) with HPSI-T as the predictor
variables was significant, F (3, 453) = 108.8, £ = .0001, r squared = 42%. In this
model, HPSI-T (t = 4.4, £ = .001) and rational coping (t = 4.8, £ = .001) but not the
interaction term were significant predictors of burnout.
The multiple regression model with BOS as the dependent variable and
HPSI-T, emotional coping (EM), and the interaction of EM with HPSI-T as the
predictor variables was significant, F (3, 454) = 272.5, £ = .0001, r squared = 65%.
In this model, HPSI-T (t=-2.0, £ = .05), EM (t = -11.9, £ = .0001), and the
interaction between HPSI-T and frequency of use of EM (t= 3.2, £ = .0015) were all
significant predictors of burnout. Data plots of the interaction between EM and
HPSI-T indicate that as frequency of use of EM increased, the position association
between stress and BOS strengthened.
The multiple regression model with BOS as the dependent variable and
HPSI-T, frequency of use of detached coping (DET) and the interaction between
HPSI-T arid frequency of use of DET as predictor variables was significant, F(3,
454) = 97.3, £ = .0001, r squared = 39%. In this model, HPSI-T (t= 2.6, £ = .009)
and frequency of use of DET (t = 3.5, £ = .0005) were significant predictors of


21
Simpson and Grant (1991) reported that, despite speculation to the contrary,
physicians in their study evidenced no difference by gender in the types and
magnitude of stressors experienced. As in the Richardsen and Burke (1991) study
(see below), the authors used their own set of identified stressors, not a
standardized instrument. Richardsen and Burke (1991) found that women
physicians reported slightly higher overall stress than did men. They also found
significant differences in three sources of stress reported: women rated the need to
maintain their own level of knowledge and counseling nonmedical problems as
more stressful than did men, whereas male physicians reported more stress related
to maintaining an adequate income.
Age and years of experience may have an impact on job stress and job
satisfaction and other work attitudes such as commitment and turnover intention.
In a study of job stress, satisfaction, and life satisfaction in physicians, Linn, Yager,
Cope, and Leake (1985) found an inverse relationship between age and stress, and
positive relationships among age and job and life satisfaction. Age differences
possibly accounted for the difference in overall stress between men and women
physicians in the Richardsen and Burke (1991) study, since men in their study were
generally older and had more years of experience than women. No statistics were
done to confirm this possibility, however. Similarly, researchers have suggested
that experience (or closely related constructs such as career stage) may moderate
relationships among sales force members perceptions, attitudes, and behavior


TABLE OF CONTENTS
ACKNOWLEDGEMENTS iii
Abstract vi
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 REVIEW OF THE RELATED LITERATURE. 14
Job Stress 14
Coping 22
Burnout 27
Job Satisfaction 30
Rationale and Hypotheses of the Present Study 34
CHAPTER 3 METHOD 36
Participants 36
Measures 48
Procedure 44
CHAPTER 4 RESULTS 46
Descriptive Statistics 46
Data Analysis 47
Hypothesis 1 47
Hypothesis 2 48
Hypothesis 3 53
IV


97
Price, J., & Mueller, C. (1981). A causal model of turnover for nurses.
Academy of Management Journal. 24. 543-565.
Prichard, J. T. (1996). Dying too young. Journal of the American
Veterinary Medical Association. 208. 1788.
Quick, J.C., Murphy, L.R., & Hurrell, J.J. (Eds.). (1992) Stress & well
being at work: Assessments and interventions for occupational mental health.
Washington, DC: American Psychological Association.
Quinn, R., Staines, G., & McCullough, M. (1974). Job satisfaction: Is there
a trend? The 1977 quality of employment survey. Ann Arbor: Survey Research
Center, University of Michigan.
Repetti, R., Matthews, K., & Waldron, I. (1989). Employment and womens
health. American Psychologist, 44, 1394-1401.
Richardsen, A., & Burke, R. (1991). Occupational stress and job
satisfaction among physicians: Sex differences. Soc. Sci. Med., 33. 1179-1187.
Roger, D., Jarvis, G., & Najaran, B. (1993). Detachment and coping: The
construction and validation of a new scale for measuring coping strategies.
Personality and Individual Differences. 15. 619-626.
Russ, F., & McNeiliy, K. (1995). Links among satisfaction, commitment,
and turnover intentions: The moderating effect of experience, gender, and
performance. Journal of Business Research. 34. 57-65.
Salant, P., & Dillman, D. (1994). How to conduct vour own survey. New
York: John Wiley & Sons.
Sandler, B. (1986). The campus climate revisited: Chilly for women
faculty, administrators, and graduate students (Project on the Status and
Education of Women). Washington, DC: Association of American Colleges.
Schaufeli, W.B., Maslach, C., & Marek, T. (1993). The burnout measure.
Profession Burnout: Recent developments in theory and research, (pp. 205-207).
Washington, DC: Taylor and Francis.
Schneer, J., & Reitman, F. (1995). The impact of gender as managerial
careers unfold. Journal of Vocational Behavior, 47, 290-315.


69
women but not in men in one study (Pines & Aronson, 1988). Since the average
male and female veterinarian in this study came from a very small practice, it can
be assumed that both men and women had few supportive work relationships. This
relative isolation may well have affected women more than men. In fact, women
rated Not having opportunities to share feelings and experiences with colleagues
as more problematic than did men.
Hypothesis three stated that there are gender differences in the frequency
and severity of specific job stressors reported by equine veterinarians in private
practice. This hypothesis was supported. Perhaps the most unexpected finding of
this study was the fact that female equine practitioners rated 64 of 70 job stressors
as more stressful or more frequent in occurrence than did their male colleagues.
The 20 items rated as significantly more stressful by women appeared to fall into
four categories: 1. conflicts 2. lack of respect, recognition, fair pay, and opportunity
3. inability to deliver optimal patient care, and 4. emotional issues.
Hypothesis four stated that there are no gender differences in coping styles
reported by equine veterinarians in private practice. This hypothesis was not
supported. Female equine practitioners used emotional coping methods more
frequently than did men while men used detached and rational coping methods
more frequently than did women. The literature on this matter has been
contradictory, although these findings further the observation of Parasuraman and
Cleek (1984) who found a tendency (but not a statistically significant difference) for
male managers to use more detached and rational and less emotional coping styles
than female managers. Long and Kahn (1993) attributed these confusing gender


26
was based on their transactional model of stress and coping. Rational coping refers
to task oriented coping methods such as working out a plan for dealing with what
has happened. Emotional coping refers to attempts at repressing (or expressing)
the emotions evoked due to the threat. Avoidance coping involves avoiding action
or evoked emotions, and detached coping refers to feeling independent of the event
and the emotion associated with it.
Emotional and avoidance coping are considered maladaptive, and rational
and detached coping are considered to be adaptive styles of coping. Smith and
Sulsky (1995) investigated coping outcomes and found that increased use of
emotional and avoidance coping was associated with higher global job stress, more
physical symptoms, increased depression, and lower job satisfaction. This finding
was repeated over several samples.
It is a strongly held gender-role stereotype that women respond to stress
differently from men. That is, they respond passively and/or more emotionally to
stressful situations whether at work or elsewhere. Early studies (Billings & Moos,
1981; Folkman & Lazarus, 1980; Pearlin & Schooler, 1978) tended to suggest that
men used more rational coping compared with women. However, these studies
were flawed because men and women were not matched on occupational level or
on organizational power (Long & Kahn, 1993). Roger, Jarvis & Najaran (1993)
found a tendency for men to use more detached and rational and less emotional
coping styles than women, although the differences were not statistically significant.


REFERENCES
Andrisani, P. (1978). Job satisfaction among working women. Signs, 3,
558-607.
Bailey, K.D. (1982). Methods of social research. New York: The Free
Press, Macmillan.
Bandura, A. (1977). Self-efficacy : Toward a unifying theory of behavioral
change. Psychological Review. 84, 191-215.
Barling, J., Kryl, I., & Bluen, S. (1990). Finally, all the questionnaires are
back: Bias in the time taken to return survey questionnaires. Personality and
individual differences. 11,177-180.
Barone, D., Caddy, G., Katell, A., Roselione, F., & Hamilton, R. (1988).
The Work Stress Inventory: Organizational stress and job risk. Educational anad
Psychological Measurement. 48, 141-154.
Beehr, T., & Schuler, R. (1980). Stress in organizations. In K. Rowland &
G. Ferris (Eds.). Personnel management (pp. 390-419). Boston: Allyn and
Bacon.
Ben-Porath, Y., & Tellegen, A. (1990). A place for traits in stress research.
Psychological Inquiry,1, 14-17.
Bickel, J. (1988). Women in medical education. A status report. New
England Journal of Medicine. 319. 1579-1584.
Billings, A., & Moos, R. (1981). The role of coping responses and social
resources in attenuating the stress of life events. Journal of Behavioral Medicine.
4YI39--157.
Bird, C. E. (1992). Comparing women's representation in three health
professions: Dentistry, medicine, and veterinary medicine. (Doctoral
dissertation, University of Illinois at Urbana-Champaign, 1992). Dissertation
89


CHAPTER 4
RESULTS
Descriptive Statistics
The composite mean Health Professions Stress Inventory score (HPSI-T)
was 7.01 + 2.7, with a range of 1.09 to 16.77. The range of the scale for
responding to items on the HPSI-T was 1.0 (not at all stressful) to 25.0 (extremely
stressful). The mean score of the male respondents was 6.2 + 2.4 and the mean
score of the women was 7.6+ 2.7.
The composite mean Burnout Score (BOS) was 3.29 0.8, with a range of
1,1 to 6.0. The response scale for the BOS yields scores of 1 to 7. A score of 1 to
2 indicates low burnout; a score of 2 to 3 indicates acceptable levels of stress arid
manageable level of work dissatisfaction; a score of 3-4 indicates the beginning of
the burnout syndrome; a score of 4-5 indicates a state of clinical burnout; and
scores over 5 indicate severe problems (Pines & Aronson, 1988). The mean score
of the male respondents was 3.07 + 0.8 and the mean score of the female
respondents was 3.47 + 0.8.
The composite mean job satisfaction score (SAT) was 3.70 + 1.0, with a
range of 1-5. The lowest score (1) indicates extreme dissatisfaction and the highest
score (5) indicates extreme satisfaction. The mean job satisfaction score of the men


71
stress on burnout, and specifically that rational and detached coping would
moderate the impact, while emotional coping would amplify the effect. This seems
to be a logical extension of Lazarus' theory, and is in agreement with the conception
(and findings) of Billings and Moos (1981).
In this study, no interaction between coping style and job stress on the
dependent variable, burnout, was observed with the exception of the interaction
between job stress and the emotional style of coping in men. Relatedly, Wolfgang
(1995) found no interaction between coping style and job stress when the
dependent variable was job dissatisfaction. The findings of the current research
were in contrast to what might be intuited and to what was hypothesized by Lazarus
and his colleagues. Lazarus and Launier (1978), noting the importance of coping
by itself, stated that the ways people cope with stress are even more important.
. than the frequency and severity of episodes of stress themselves (p. 308). In the
present study, it was seen that 67% of the variance in burnout was accounted for
by job stress, coping and their interactions.
Osipow and Davis (1988) conducted a study to evaluate the relationship of
coping resources to occupational stress and strain. By coincidence, the subjects
were 213 (of 600) veterinary students who responded to a mail questionnaire. The
researchers assessed the impact of four types of persona! coping resources
(recreation, self-care, social supports, and use of rational coping) on occupational
stress and strain. All types of coping resources were effective in reducing job
stress, with high levels of social support reducing the impact of all types of
stressors. Least in importance was cognitive/rational coping which moderated only


35
in equine veterinarians private practice. Specifically, problem-focused and
detached coping styles will be associated with a weaker relationship between job
stress and burnout while avoidant and emotion-focused coping will be associated
with a stronger relationship between job stress and burnout.


34
component of the relationship between schooling and job satisfaction. They also
found no evidence that education that doesnt lead to expected levels of
intrinsic and/or extrinsic reward reduces job satisfaction (p. 192).
Rationale and Hypotheses of the Present Study
The relationships among job stress, burnout and job satisfaction seem
relatively straightforward regardless of the types of confounding variables studied.
These relationships have not been confirmed in a veterinary population, however.
In addition, it is unclear (in any population) how the relationship is influenced by
gender and mediated by coping style, and further modified by age and wage
earning status.
Based upon the literature review, the following research hypotheses were
formulated.
1. Among equine veterinarians in private practice, job stress will be negatively
correlated with job satisfaction and positively correlated with burnout.
2. Among equine veterinarians in private practice, there will be no gender
differences in the overall level of job stress, job satisfaction, and burnout reported.
3. There will be gender differences in the frequency and severity of specific job
stressors reported by equine veterinarians in private practice.
4. There will be no gender differences in coping styles reported by equine
veterinarians in private practice.
5. The relationship between job stress and burnout will be mediated by coping style


25
to the detriment of their families
The key threats to achievement-related values are associated with loss: loss
of feelings of competence, loss of interest in (or challenge by) the work itself, or loss
of personal control over ones work (Locke & Taylor, 1990). Workers may lose their
sense of purpose at work after attaining highly valued, long sought-after goals, or
they may gradually begin to question the real value of what they do. Social
relationships serve both an expressive and an instrumental function at work (Locke,
1976). Interactions can be pleasurable and/or necessary for completion of the job
and difficulties in this area can be particularly disturbing to workers. Some research
has shown that this work value may be more important to women than to men. An
individuals self-concept is threatened at work when there in an incongruity between
the self-concept and what the individual sees as a requirement of the job. The self-
concept is further threatened by perceived failures at work as well.
The stress caused by these threats must be coped with, in one way or
another. Coping is typically defined in the literature as the process of adapting to
perceived threats, and coping has been repeatedly divided into rational (problem-
focused), emotional, and avoidance (symptom-focused) forms with different authors
attributing differing meanings to each of the above. Roger, Jarvis, and Najaran
(1993) have recently described a scale which measures four types of coping:
rational, emotional, avoidance, and detachment coping. This scale represents an
extension of Folkman and Lazarus (1985) Ways of Coping Questionnaire which


56
Hypothesis 5
Hypothesis 5 stated that the relationship between job stress and burnout will
be mediated by coping style in equine veterinarians in private practice and that
rational and detached coping styles will decrease the impact of job stress on
burnout while avoidant and emotional coping styles would increase the impact of job
stress on burnout. Recall that, due to the results of the factor analysis performed
on the Coping Styles Questionnaire (CSQ), avoidant coping was omitted from this
comparison.
To test this hypothesis, a multiple regression analysis was used with burnout
score (BOS) as the dependent variable and job stress (HPSI-T), frequency of use
of each of the three coping styles, and the interactions of job stress with each
coping style use frequency as the predictor variables. It was found that 67% of the
variance in BOS could be accounted for by job stress and coping styles use
frequency and the interactions between job stress and frequency of use of each of
the coping styles (F (9, 454) = 100.44, £ = .0001).
Typical of multiple regression analysis, the order in which each predictor
variable and its interaction is entered into the overall model will affect the outcome.
Thus, three separate multiple regression analyses with BOS as the dependent
variable in each and with one of the coping style use frequency variables, job stress
(HPSI-T) and their interactions as independent variables in each were performed


CHAPTER 1
INTRODUCTION
The veterinary profession, like other health-related professions including
medicine and psychology (Pin et ai., 1996), is undergoing many changes (Clark,
1996). It is becoming increasingly difficult to establish profitable veterinary practices;
furthermore, corporations are beginning to take over established practices in order
to create large referral centers with satellite clinics. Employed veterinarians will have
far less autonomy than their practice-owning predecessors. Veterinary students
are graduating with huge debts (mean = $45,251) but accepting starting salaries
(mean = $29,854; Gehrke, 1995) which make loan repayment difficult, and which,
in fact, are extremely low in proportion to the number of years of education required.
A previously male-dominated profession, veterinary medicine is rapidly
becoming femininized, with women constituting up to 75% of currently enrolled
freshman classes (Gehrke, 1996). Significantly, Bird (1992) showed that female
veterinarians had an income disadvantage of $13,000 per year compared to male
veterinarians. Forty-two percent of the sex differences in income Bird documented
were correlated with gender differences in veterinarians characteristics (differences
1


GENDER DIFFERENCES IN JOB STRESS, BURNOUT, AND JOB SATISFACTION
AS MEDIATED BY COPING STYLE OF VETERINARIANS IN
PRIVATE EQUINE PRACTICE
Bv
BARBARA BREWER WELSCH
/
/
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1998


12
Several groups have shown no gender differences in coping strategies (Hamilton
& Fagot, 1988; Osipow & Davis, 1988). In contrast Pearlin and Schooler (1978)
found that women reported less use of effective strategies, a finding corroborated
by Billings and Moos (1981). Roger et al. (1993), using the CSQ which they
developed, showed a tendency for men to use more detached and rational and less
emotional coping styles than women, but the differences were not statistically
significant.
In contrast, in two studies conducted in actual work settings, women were
shown to cope in more positive ways. Parasuraman and Cleek (1984) found that
female managers reported greater use of adaptive coping strategies (planning,
information seeking, priority setting) than did males. Parkes (1990) found that men
used significantly more suppression strategies such as withdrawal, restraint,
compromise, and ignoring the problem. Miller and Kirsch (1987) summarize the
literature by concluding that it appears that the categories of emotion-focused and
problem-focused coping may yield significant sex differences in how people
cognitively cope with stress .... It is possible that women appraise threatening
events as more stressful than men do. However, given the methodological
constraints of these studies, further research is needed before firm conclusions can
be drawn (p. 295). The veterinary profession is changing drastically as a result of
the changing economy and changing demographics (feminization). Occupations


19
university and corporate work settings (Spielberger & Reheiser, 1995).
It was this authors opinion, in agreement with Jackson and Schuler (1985),
that an instrument specific to the health related professions would be more valuable
in this particular study. Medical professionals face a different set of stressors than
do other professionals, and recently, increasing attention has been focused on the
adverse effects of job stress on providers and ultimately on the overall quality of
health care (Wolfgang, 1988). Although various instruments and surveys have
assessed job stress in physicians, none have looked at veterinarians work stress.
One instrument, the Health Professions Stress Inventory (HPSI) has been used to
compare levels of stress within the occupations of mediciine, nursing, and
pharmacy, and it was felt that this instrument would provide useful information
regarding the veterinary profession while allowing comparisons to other health-
related professions.
The general job stress literature produces inconsistent findings regarding the
effect of gender on job stress. Beehr and Schuler (1980) concluded that women
and men show no differences in stress-related symptoms in the workplace, and
DiSlavo, Lubbers, Rosi, and Lewis (1994) found that from a broad perspective,
men and women perceive stressors quite similarly (p. 48). On the basis of a meta-
analytic evaluation of 15 studies which examined gender differences in occupational
stress, Martocchio and OLeary (1989) found no sex differences in experienced and
perceived work stress (p. 495).


39
Appendix A.)
The Health Professions Stress inventory (HPSI) (Wolfgang, 1988) consists
of 30 job situations which health providers might be expected to encounter and find
stressful. Respondents indicate how often they have found each situation to be
stressful in their work environment, using a scale of 0 to 4 (never to very often). The
HPSI has been proven to be both a valid and reliable measure of job stress in the
health professions (Wolfgang, 1988). Lyons (1971) Index of Work Related Tension
(IWRT)was used as a measure of concurrent validity for the HPSI. Pearson
product-moment correlations between scores on the HPSI and the IWRT were .75
for pharmacists, .76 for physicians, and .78 for nurses (p < .001) (Wolfgang, 1988).
Cronbachs alpha coefficients assessing internal consistency of the HPSI were .89,
.89, and .88 for pharmacists, nurses, and physicians, respectively.
The HPSI was modified slightly for the veterinary population in this research.
The word patient was changed to owner in three instances (see Appendix A).
In accord with the suggestion of Spielberger & Reheiser (1995), the perceived
intensity of each stressor was assessed along with its frequency. An overall job
stress index was based on the sum of the cross-products of the Severity and
Frequency scores. The overall index (HPSI-T) was compared with the original
index (HPSI) (which is hereafter referred to as the frequency index (HPSI-F)) in
order to ascertain the necessity of using both intensity and frequency scales. It


55
to test Hypothesis 2 which also was concerned with gender differences.
The MANCOVA (Wilks Lambda) revealed significant main effects for years
in the profession F(4, 440) = 5.4, £ = .0003; perceived life stress F(4, 440) = 13.23,
£ .0001; income F(4, 440) = 2.96, £ = .020; and gender F(4,440) = 3.98, £ =
.0035. The models of all three of the follow-up univariate ANCOVAs for the specific
types of coping were significant: Rational coping F(3, 447) = 11.13, £ = .0001;
Emotional coping F(4, 447) = 22.17, £ = .0001; and Detached coping F(4, 447) =
23.89, £ ~ .0001. Table 4.4 lists the least squares means of each gender for each
coping style. These means show that men as compared to women used
significantly more rational and detached coping styles, while women as compared
to men used significantly more emotional coping. Thus, Hypothesis 4 was not
supported.
Table 4.4
Least Squares Means for Coping Styles of Male and Female Veterinarians*
Gender Rational Coping Emotional Coping Detached Coping
Females 1.96 3.20 2.55
Males 1.86 3.30 2.42
*L.ower scores signify higher usage of each particular styie of coping.


37
The overall response raie from the males was 43% and the female response
rate was 57%. Eighty-nine percent of the actual study participants (n=407) ranged
in age from 30-59 although the age range of all participants was from 25 to over
70 years. The participants were distributed nearly equally in terms of years in the
profession of veterinary medicine: 20.8% (n= 95) had been practicing for five years
or less, 19.7% (n=90) between six and 10 years, 23.9% (n=109 ) between 11 and
15 years, 11.8% (n=54) between 16 and 20 years, and 25% (n_=115) greater than
20 years. Thirty-six percent (n=164) had been working for five years or less at their
current job, 22.8% (n_=104) had been working from 6 to 10 years, 15.5% (m=71),
between 11 and 15 years, 6.6% (n=30) between 16 and 20 years, and 19.0%
(n=87) longer than 20 years. Sixty percent of the respondents (n=274) were sole
owners of a practice, 13.7% (n=63) were partners in a practice, and 26.2% (n=120)
were employees.
Five percent (n=23) of the respondents were board certified with the colleges
of Surgery (n=10 respondents), Theriogenology (n=7 respondents), Veterinary
Practice (n=3 respondents) and Acupuncture (n=3 respondents) being represented.
Only 12.3% (n=56) reported working 40 hours per week or less, with 28% (n=218)
of the practitioners averaging greater than 60 hours per week on the job. Fifteen
and one-half percent (n=71) annually earned less than $30,000, 28% (n=128)
earned between $30,000 and $49,999, 18.6% (n=85) earned between $50,000 and


APPENDIX B
COVER LETTER
May 9, 1997
Dear Colleague:
! am a former large animal internist (Ohio 80), currently in part-time private
practice while completing my Ph.D. in Counseling Psychology at the University of
Florida. Throughout my 17 years in veterinary medicines I have observed that some
veterinarians are quite happy with our profession and others are dissatisfied. With the
help of the AAEP, S am hoping to begin to understand what veterinarians find stressful
about their jobs, how satisfied they are with the profession as compared to other health
care professionals, and what might distinguish those who are satisfied from those who
are dissatisfied with veterinary medicine.
My funding is very limited, so I urge you to please take the time to fill out the
next four pages. In pilot studies with small animal' practitioners, the survey took an
average of 12 minutes to complete. Your response is very important to the success
of this study, as a high return rate is crucial for the results to be meaningful.
This is a completely ANONYMOUS questionnaire. In no way can you be
identified. Your completion of the questionnaire constitutes your agreement to allow the
University of Florida to use the results for teaching and research purposes. You need
not answer any question you do not wish to answer. I would gladly answer any
questions, as would the Project Supervisor, Dr. Carolyn Tucker. You can reach me at
352-376-1790 (or e-mail bbwelsch@aol.com) and Dr. Tucker at 352-392-1532. I
would welcome any written comments, either via e-mail or returned with the survey.
The results will be presented in December, 1997 at the Phoenix meeting of the AAEP.
Thank you in advance for helping a colleague in need, and for contributing to an
understanding of our changing profession!
Sincerely,
Barbara Brewer Welsch, DVM, EdS Ed Blach, DVM, MS, MBA
ACVIM, ACVECC AAEP Markefview
PLEASE REPLY BY MAY 26, 1997
87


33
with what little has been observed in physicians, is that men and women in
management did not appear to differ from one another in their sources of
satisfaction at work.
The relationship between education and job satisfaction has been explored.
Early research suggested that increasing education might condition workers to
expect higher levels of extrinsic rewards (Mortimer, Finch, & Maruyama, 1988) or
intrinsic rewards (Seybolt, 1976). On the one hand, employees with higher
educational attainment were prepared for more challenging careers which could pay
higher salaries, so if this materialized, the workers should be more satisfied. On the
other hand, if workers were overeducated and underemployed (or underpaid) they
should be less satisfied.
One recent study reported that gender moderated the impact of education
on job satisfaction. Educational attainment that had not provided expected
levels of extrinsic rewards had a negative impact on job satisfaction of men but
not women (Glenn & Weaver, 1982). The authors hypothesized that this might
be due to wage earning status-women who are secondary wage earners may
place more emphasis on the intrinsic rewards of work. Martin and Shehan
(1989) confirmed this hypothesis. They found that the major source of the
education-satisfaction relationship was the result of educations connection to
higher levels of extrinsic reward only for the primary wage earning women. For
secondary wage earning women, intrinsic rewards were the important


28
Maslach (1986), who was the first to coin the term burnout defines it as a
syndrome of emotional exhaustion, depersonalization, and reduced personal
accomplishment that can occur among individuals who work with people in some
(usually care-giving) capacity, often in emotionally charged situations. Many
sources of burnout have been proposed, with no single cause identified. Burnout
is viewed as the result of multiple causes, the impact of which differs according to
the personality of the worker. According to Maslach (1986) the sources of burnout
can be divided into three categories: personal (expectations, motivations,
personality), interpersonal (client contact, relationships with coworkers and
supervisors, outside relationships), and organizational (workload, bureaucracy,
feedback, work pressure). A key factor does seem to be the amount of client
contact experienced on the job. As time spent in direct client contact increases,
particularly if such contact is upsetting, frustrating, or difficult, burnout is more likely
to occur (Pines & Aronson, 1988).
Two studies have been conducted on burnout in veterinarians. Both used
Pines and Aronsons self-diagnosis burnout scale (Pines & Aronson, 1988). A mail
survey of 1000 veterinarians in 1981 yielded a 33% response rate, of which 91%
of the respondents were male (Elkins & Elkins, 1987). Fifty-three percent of the
respondents reported scores over 3 (mean = 3.09), the point at which clear-cut
signs of the burnout syndrome are said to occur. The youngest age group (25-29)
reported significantly higher burnout scores, as did the veterinarians in practice for


51
wage earning status, board certification status and their interactions as covariates.
No significant main effects were found for job satisfaction F (32,441) 1.62, p >.02,
or burnout F (32,441) = 1.46, p > .05. The main effect found in the model with job
stress as the dependent variable F (32,440), p > .0008 was attributed solely to the
independent variable, gender. Thus, it was not necessary to include marital status,
wage earning status, and board certification status in the analyses to test
Flypothesis 2.
Flypothesis 2 states that, there will be no gender differences in the overall
levels of job stress, job satisfaction and burnout in veterinarians in private, equine
practice. To examine this hypothesis, a MANCOVA was performed with the
measures of job stress, job satisfaction and burnout as the dependent variables,
gender as the independent variable, and respondents years in the profession,
income, and perceived life stress as the covariates.
The MANCOVA (Wilks Lambda) revealed significant main effects for years
in the profession F(3, 446) = 10.5, p = .0001, perceived life stress F (3, 446) = 29.1,
P = .0001, income (F3, 446) = 7.3, p = .0001, and gender F(3,446) = 6.09, p =
.0005. All three follow-up univariate ANCOVAs for the dependent variables (using
years in the profession, income, and life stress as co-variates) were significant: Job
stress F(4, 451) = 4.79, p = .03, Job satisfaction F(4,447) = 10.01, p = .0017, and
Burnout F(4, 451) = 5.52, p = .02. Results revealed significant gender differences


84
HOW PROBLEMATIC? FREQUENT?
Not. being challenged by your work
1 2 3 4 5
1 2 3 4 5
Feeling that you are inadequately paid as a health professional
1 2 3 4 5
1 2 3 4 5
Caring for terminally ill patients
1 2 3 4 5
1 2 3 4 5
Not being able to use your abilities to tire fullest extort on the job
1 2 3 4 5
1 2 3 45
Fearing that a mistake will be made in tre treatment of a patient
1 2 3 4 5
1 2 3 45
Having owners whose values and beliefs about care differ greatly from yours
1 2 3 4 5
1 2 3 4 5
Problems related to business and management of veterinary practice
1 2 3 4 5
1 2 3 4 5
Worrying about the possibility of zoonotic, illness or injury to self or staff
1 2 3 4 5
1 2 3 4 5
Having to make difficult ethical or moral choices
1 2 3 4 5
1 2 3 4 5
instructions: Circle the appropriate number beside each of the next five statements.
Strongly disagree
Strongly Agree
I am satisfied with my current job.
1 2
3
4
5
I cannot think of any jobs for which would exchange.
1 2
3
4
5
If I liad the choice to make a,gain, I would choose the veterinary profession.
1 2
3
4
5
I do NOT intend to lea ve this job anytime in the near future.
1 2
3
4
5
This job measures up to my initial expectations
1 2
3
4
5
Instructions: Although people may react in different ways to different situations, we all tend to
have a characteristic way of dealing with things which upset us. How would you describe the
way you typically react to the stress you encounter during the practice of veterinary medicine?
A=always 0=often S=somerimes N=never
See the situation for what it actually is and nothing more A O S N
See the problem as something separate from myself so I can deal with it A O S N
Become miserable or depressed A O S N
Feel that no-one understands A O S N
Become lonely or isolated A O S N
Take action to change things A O S N
Avoid family or friends in general A O S N
Feel helpless-heres nothing you can do about it A O S N
Try to find out more information, to help make a decision about things A O S N
Keep tilings to myself and not let others know how bad things are for me A O S N
Feel independent of the circumstances A O S N
Cry, or feel like crying A O S N
Pretend theres nothing the matter, even if people ask whats bothering me A O S N


73
on physicians. This finding is a!so in contrast to much, of the current literature on job
stress in the general population as well, Linn, Yager, Cope and Leaks (1985)
observation of the inverse relationship between age and job stress was confirmed.
In the current study, years in the profession and years on the job exhibited a slightly
stronger inverse correlation with job stress than did age. As in their study, a positive
relationship was observed in the present study between age and job satisfaction,
but this was a weak correlation only.
The current research confirms Miller and Kirsch (1987) summary of their
review of the gender and coping literature which concluded that the different sexes
typically cope in different ways, in this study the styles of emotion-focused,
detached, and rational coping all yielded significant gender differences. Further, it
seems that these female veterinarians appraised threatening events as more
stressful than did male veterinarians.
This study profoundly illustrates the problem noted in Spielberger and
Reheisers (1995) observation that most commonly used job stress measures tend
to confound the frequency of occurrence of a stressful event with its perceived
severity. The HP-SI, as used in most of the health professions studies of stress,
merely asked about the frequency of occurrence of a particular job stressor. In this
study of veterinar ians, men and women generally agreed upon the frequency of job
stressor occurrence, but evaluated the severity of such occurrences very differently.
Had the HPSI not been expanded (by the inclusion of the problematic scale in
addition to the frequency scale), a great deal of information would have been lost,
and gender differences in assessment of the severity of job stressors would have


7
stress experienced. Later studies which took the above factors into account were,
indeed, able to detect gender differences. There are no studies evaluating job
stress in the veterinarian, but several were found which investigated other health
related professions (Wolfgang, 1988, 1995; Simpson & Grant, 1991; Richardsen &
Burke, 1991). Three of the four studies found no gender-related differences, but
Richardsen and Burke (1991) reported significant sex differences in terms of the
variables that predicted job stress. Women rated the need to maintain their own
level of knowledge and providing counseling for nonmedical problems as more
stressful than did men, whereas male physicians reported more stress related to
maintaining an adequate income.
Even though consistent differences in the amount or types of stress
experienced on the job have not been shown between men and women, it is
virtually uncontested that women in Western cultures have higher rates of affective
and anxiety disorders than do men, in both clinical and nonclinical populations
(Miller & Kirsch, 1987). It is generally agreed that many professional women may
experience one or more of the following (Smith, 1993): a hostile work environment
in which their values, working style and job contributions are trivialized,
unappreciated, or unrewarded; an unsupportive home environment in which it is
their primary responsibility to care for the home and the children; a disapproving
social environment (i.e., their house is not clean enough and they rarely produce


96
Neil, C., & Snizek, W. (1988). Gender as a moderater of job satisfaction.
Work and Occupations. 15. 201-219.
Northwestern National Life. (1991). Employee burnout: Americas newest
epidemic. Minneapolis. MN: Author.
Numerof, R.E., & Abrams, M.N. (1984). Sources of stress among nurses:
An empirical investigation. Journal of Human Stress. 10. 88-100.
Olsen, D. (1991). Women and minority faculty job satisfaction: A structural
model examining the effect of professional role interests. Boston, MA:
Association for the study of Higher Education. (ERIC document Reproduction
Service No. ED 339 323)
Osipow, S., & Davis, A. (1988). The relationship of coping resources to
occupational stress and strain. Journal of Vocational Behavior. 32, 1-15.
Parasuraman, S., & Cleek, M. (1984). Coping behaviors and managers
affective reactions to role stressors. Journal of Vocational Behavior. 24. 179-193.
Parkes, K. (1990). Coping, negative affectivity, and the work environment:
Additive and interactive predictors of mental health. Journal of Applied
Psychology. 75.399-409.
Pearlin, L., & Schooler, C. (1978). The structure of coping. Journal of
Health and Social Behavior. 19, 2-21.
Perrewe, P.L. (Ed.). (1991). Handbook on job stress (Special issues).
Journal of Social Behavior and Personality,6(7).
Pines, A., & Aronson, E. (1988). Career burnout: Causes and cures, (pp.
113-130; p. 219). New York: The Free Press.
Pin, G., Mednick, M., Astin, H., Hall, C., Kenkel, M., Keita, G., Kohout, L.,
& Kelleher, J. (1996). The shifting gender composition of psychology. Trends and
implications for the discipline. American Psychologist, 51, 509-528.
Plascak-Craig, F., & Bean, J. (1989). Education faculty job satisfaction in
major research universities. Atlanta: Association for the Study of Higher
Education. (ERIC Document Reproduction Service No. ED 313 978)


22
(Russ & McNeilly, 1995).
Coping
As discussed in the previous section, the psychological stress literature is
permeated with discussions about the potentially deleterious effects of stress on
physical and mental health as well as on individual attitudes and behaviors.
Therefore, it follows that much emphasis has also been placed on the investigation
of how people manage or cope with stress in all of lifes domains including the
workplace.
Lazarus transactional model of stress and coping (Lazarus, 1995) is one of
several models which have been used to study the nature of coping. Most
researchers who have studied coping cite Lazarus work and many use it to build
slightly different and more complex models. For example, Schulers (1985)
integrative transactional process model of coping (which is specific for
organizations) also addresses the impact of the uncertainty and the importance of
the situation to the individual and includes a process of feedback and evaluation.
The transactional theory states that a transaction between the person and
the environment is stressful only when it is evaluated by an individual as a harm,
threat, or challenge to that persons well-being. Lazarus terms this evaluation an
appraisal. Within the context of the theory, distinction is made between harm,
threat, and challenge. Harm refers to damage which has already occurred (loss


27
Parasuraman and Cleek (1S84) and Tung (1980) found women managers were
more likely than men to use adaptive coping behaviors or superior coping. Long
and Kahn (1993) attribute these confusing gender differences to three causes.
First, women experience more and different work stressors than men (sex
discrimination and harassment). Next, gender role stereotypes affect male and
female perceptions of appropriate coping (such as disclosing to friends and
colleagues). Finally, women are disadvantaged with regard to coping if they lack
workplace coping resources such as power, perceived competence, or child care
and financial advantages.
In summary, when coping strategies fail, work loses meaning because
important values can no longer be attained from it. If this happens, since work
usually consumes a considerable part of ones daily living, life, itself, can lose
meaning. This is when burnout occurs.
Burnout
According to Schultz and Schultz (1990) More people are disabled today as
a result of stress than at any other time in our history (p. 539). In 1992, job stress
was identified as one of the 10 leading health risks in the United States (Quick,
Murphy, & Hurrell, 1992). It has been estimated that fully one third of US workers
are currently experiencing burnout or preburnout symptoms (Takooshian, 1994).
Burnout is the only psychological outcome variable that has been studied in the
veterinary population (Elkins & Elkins, 1987; Elkins & Kearney, 1992).


8
healthy meals; they should not put their children into daycare); and major conflicts
with their self-concept (worker vs. mother vs. wife). Two theoretical perspectives
have been advanced to address these issues: the role conflict perspective (scarcity
hypothesis) and the role expansion perspective. The former suggests that the
above conflicts add to the stress of the female worker (Goode, 1960), while the
latter model finds that if a female sees her roles to be both meaningful and of high
quality, multiple roles lead to positive outcomes and may even buffer the impact of
negative life events (Repetti, Mathews, & Waldron, 1989).
Although gender differences in job stress are inconsistently reported, the
reported effects of job stress have been consistent. Burnout is a syndrome of
emotional exhaustion, depersonalization, and reduced personal accomplishment
that can occur among individuals who work with people in some capacity (Maslach,
1986). Burnout appears to be a response to chronic sources of emotional and
interpersonal stress on the job. It is a particular problem for men and women in
caregiving professions, especially when the provision of care or service occurs in
emotionally charged situations (Maslach, 1986). Higher levels of burnout are linked
to lower levels of job satisfaction.
Burnout is the one psychological outcome variable that actually has been
studied in veterinarians (Elkins & Elkins, 1987; Elkins & Kearney, 1992). (Neither
of these two referenced studies was limited to veterinarians in private practice.)


This work is dedicated to my wonderful mother, Helen M.
Dean, and my understanding husband, Boyd B. Welsch. Their generosity,
respect, and love truly made this achievement possible. It is also dedicated
to my daughter, Collyn Kimberly Welsch, who made the experience all the
more worthwhile.


78
overwhelmingly stressed, and again, use of this group may, therefore have made
for a more rigorous study, or at least one in which coping was more readily studied.
Conversely, equine veterinarians may have fewer client contact hours than
do small animal veterinarians who typically see between three and four patients and
their owners per hour, versus equine practitioners who may only average one to two
patients and their owners per hour. Since burnout relates highly to hours of direct
client contact, small animal practitioners may, therefore suffer from more burnout
than equine or other large animal practitioners.
it is definitely inappropriate to compare this population of private practitioners
with veterinarians who are not in private practice. This authors earlier work with
academic veterinarians (Welsch, 1991) did reveal gender differences in job
satisfaction, much like those which have been shown for other university professors
(Sandler, 1986; Olsen, 1991; Plascak-Craig & Bean, 1989; Etaugh, 1984).
Future Research
Future research in this area should compare veterinarians in other types of
practice with one another. These results could aid veterinary students making
decisions regarding further education and training by revealing which types of
practice are more stressful or more satisfying than others. The components of job
satisfaction should be studied in an effort to better compare male and female
practitioners. An important question which needs to be assessed is whether a
veterinarians job is important to him or her most as a source of personal fulfillment
or as a source of income. Another important consideration which was neglected in
this study is the impact of physical limitations (specifically female size) on job stress


40
was concluded that use of both scales enhanced the understanding of gender
differences in job stress. Comparisons with the scores reported in the literature for
other health professionals (physicians, pharmacists, and nurses) were made based
on the frequency index (HPSI-F) alone. Gender differences were assessed on
each individual item, as well as on the overall, intensity and frequency scores. In
addition, three stressors taken from Simpson and Grants Measurement of Job
Stressors (Simpson & Grant, 1991) were added (Having owners whose values and
beliefs about medical care differ greatly from your own; Problems related to
business and management aspects of veterinary practice; and Having to make
difficult ethical and moral
choices.) Medical practice was changed to veterinary practice in one instance
and owner was substituted for patient in one instance. One additional item
pertinent to the practice of equine medicine (Worrying about the possibility of
zoonotic illness or injury to yourself or your staff) was also added.
Because the HPSI problematic (HPSI-P) scale was conceived in addition
to the typically used HPSI frequency (HPSI-F) scale, Cronbachs alpha was
computed to assure internal reliability. Analysis of the responses to the HPSI-F
scale indicated adequate internal reliability (Cronbachs alpha = .89) as did analysis
of the responses to the HPSI-P scale (Cronbachs alpha = .91).
The Coping Styles Questionnaire (CSQ) was used to assess individual


BIOGRAPHICAL SKETCH
Barbara (Dean) Brewer Welsch was born in Youngstown, Ohio, and grew
up in Pittsburgh, Pennsylvania. She obtained a B.A. in psychology with a
teaching certificate in French, general Science and mathematics from Marietta
College, Marietta, Ohio, in 1972. After teaching in the Columbus Public School
System for three years, she obtained an M.A. degree in science education from
the Ohio State University, followed by a Doctor of Veterinary Medicine degree,
also from the Ohio State University. Barbara interned in large animal medicine,
surgery and theriogenology at the University of Georgia and was a resident in
large animal medicine and neurology at the University of Florida where she later
obtained a faculty position. She is a Diplmate of both the American College of
Veterinary Internal Medicine and the American College of Veterinary Emergency
and Critical Care. She left the University of Florida as a tenured, Associate
Professor of Medicine after nearly 10 years and obtained an M.Ed. and Ed.S.
in student personnel in higher education from the Department of Counselor
Education at the University of Florida. Dr. Welsch will receive her Ph.D.
degree in counseling psychology from the Department of Psychology
100


24
The meaning of work clearly varies widely among workers in our society. For
some individuals, work is merely a means of earning money for survival. For others,
work is their life--their sole reason for surviving. It is my contention, biased through
my own participation in multiple areas within the field of veterinary medicine, and
based on the comparatively low income achieved by veterinarians compared to
other health professionals, that their work has considerable meaning to most
veterinarians.
Locke and Taylor (1990) define meaning of work as the totality of values,
including their importance, that individuals seek and expect to derive from work.
Values themselves are what individuals desire or consider to be good or beneficial
(p. 135). These authors propose that individuals may expect and desire to fulfill five
different categories of values from their work-- material values, achievement-related
values, a sense of purpose, social relationships, and the enhancement or
maintenance of self-concept.
Threats to any of the above values (which are important to an individual)
cause stress, followed by the activation of coping mechanisms. Material values can
be associated with work stress in obvious ways. Workers can base their self-
concept almost entirely on material outcomes. Further, they might exclude other
important values (i.e., family relationships) entirely in the pursuit of job success and
money. Veterinary practice owners, one of the few veterinary groups able to earn
over $100,000, can only do so by working exceedingly long hours, almost always


98
Schuler, R. (1985). Integrative transactional process model of coping with
stress in organizations. In T. Beehr, & R. Bhagat (Eds.), Human stress and
cognition in organizations (pp. 347-374). New York: Wiley.
Schultz, D. P. & Schultz, S. E. (1990). Psychology and industry today (5th
ed.). New York: Macmillan.
Selye, H. (1950). Stress. Montreal: Acta.
Seybolt, J. (1977). Work satisfaction as a function of the person-
environment interaction. Organizational Behavior and Human Performance. 17.
66-75.
Shapiro, J., & Stern, L. (1975). Job satisfaction: Male and female,
professional and nonprofessional workers. Personnel Journal. 54. 388-389.
Simpson L, & Grant, L. (1991). Sources and magnitude of job stress
among physicians. Journal of Behavioral Medicine. 14, 27-41.
Smith, C. (1993). Evaluations of whats at stake and what I can do. In B.
Long & S. Kahn (Eds.), Women, work and coping (pp.238-265) Buffalo: McGill-
Queens University Press.
Smith, C. S., & Sulsky, L. (1995). An investigation of job-related coping
strategies across multiple stressors and samples. In L. Murphy, J. Hurrell, S.
Sauter, & G. Keita (Eds.), Job stress interventions. Washington, DC: American
Psychological Association.
Smith, D.M. (1996, June). The gender gap: Why do female doctors earn
less? Veterinary Economics, pp. 44-51.
Smith, D., & Plant, W. (1982). Sex differences in job satisfaction of
university professors. Journal of Applied Psychology, 67. 249-251.
Spielberger, C., & Reheiser, E. C. (1995). Measuring occupational stress:
The job stress survey. In R. Crandall & P. Perrewe (Eds.), Occupational stress
(pp. 51-69) Washington, DC: Taylor and Francis.
Summers, T., & DeCotiis, T. (1988). An investigation of sex differences in
job satisfaction. Sex Roles. 18. 679-689.


32
with time available for family and personal life was an important predictor for women
whereas work setting factors were most important to male physicians (Richardsen
& Burke, 1991). Similarly, Kevorkian and Tuel (1994) found that male and female
academic physicians had different priorities when it came to inducements to remain
in academic positions.
Three theories can be utilized to inform the study of gender differences in job
satisfaction (Mason, 1995). The socialization view holds that women are socialized
into values, attitudes, and behaviors that are less individual in nature and more
communal than male values and behaviors, in direct contrast to this view, the
structural view suggests that any observed differences in job satisfaction are
attributable to factors which co-vary with gender and not to inherent gender
differences in socialization. The social role theory suggests that when social role
is more salient than gender role, as it is in managerial or other professional
positions, male and female attitudes, behaviors and values will be similar.
Impoverished opportunity structures (such as clerical or other blue-collar positions)
will evoke gender role salience, in which gender differences will be notable.
Masons (1995) study of nearly 15,000 workers, provided no evidence for the
socialization view, and most support for the structural view (although in some
instances the social role view was supported). She compared male and female
managers and male and female clerical workers using data collected over a two-
year period by a consulting firm. An important finding of her study, which contrasts


6
workplace adversely affects productivity, absenteeism, worker turnover, and
employee health and well-being (Perrewe, 1991; Quick, Murphy, & Hurrell, 1992;
Spielberger & Reheiser, 1995).
The concept of stress can be viewed from several perspectives. The
stimulus definition sees stress as a characteristic of the environment that is
disturbing or disruptive. The response definition of stress states that stress is a
pattern of physiological and/or psychological reactions exhibited by a person who
is under some sort of pressure. Neither definition of stress clearly explains
individual differences, however (Cox, 1978). A third approach to conceptualizing
stress is the transactional model (Lazarus & Folkman, 1984). This view combines
the stimulus and the response approaches and defines stress as the consequence
of the interaction between environmental stimuli and individual responses.
Intervening psychological processes such as perception and cognitive appraisal are
emphasized. Stress is considered to occur only when a person perceives a
demand as exceeding his or her capacity to deal with it. The major criticism of this
theoretical model is that it is incomplete since it fails to consider traits or personality
factors (Ben-Porath & Tellegen, 1990; Costa & McCrae, 1990).
When the effects of gender differences on job stress have been studied, the
results have been conflicting. While much of the early work found no sex
differences, the researchers failed to look at the frequency or the intensity of the


11
emotion-focused and symptom relieving (avoidance). Problem focused coping
refers to attempts to modify or eliminate the source of stress by taking instrumental
actions (Miller & Kirsch, 1987). This could be in the form of direct action or help
seeking behavior (Havlovic & Keenan, 1995). In contrast, emotion-focused coping
refers to attempts to control or repress stress-related emotions and maintain
affective equilibrium (Billings & Moos, 1984). This form of coping could be in the
form of either positive thinking or avoidance/resignation (Havlovic & Keenan, 1995).
Symptom focused coping is defined as the use of exercise, meditation, alcohol, or
drugs to alleviate the stress response (Schuler, 1985). Other authors use the term
avoidance coping to include additional methods of simply ignoring the problem.
Roger et. al. (1993), through their work with stress management programs and the
development of the Coping Styles Questionnaire (CSQ), identified a fourth
mechanism of coping which they term detachment" coping. This form of coping
does not involve denial or attempts to avoid stress, but it can also be distinguished
from task-oriented strategies (Roger, Jarvis, & Najaran, 1993). This strategy
involves feeling independent of the event and the emotion associated with it.
Much of the occupational-stress literature reports data only from male
employees (Parkes, 1990). Gender issues are not addressed and relatively little
is known about gender differences in work-related coping. More attention has been
given to gender in the general coping literature, but the results are very inconsistent.


23
of job, poor job evaluation, lack of promotion, disapproval), while threat refers to a
harm that has not yet happened, but is anticipated. Challenge refers to a condition
of high demand in which the emphasis is on mastering the demands, overcoming
obstacles, and growing and expanding as an individual. During threat, the focus is
on avoiding or protecting against harm. In challenge, the emphasis is on the
positive outcome possibilities. Individuals, particularly those who have chosen
medical professions, thrive on challenge while at the same time dislike (and attempt
to evade) threat.
Lazarus early field work demonstrated that the way people evaluate what is
happening with respect to their well-being, and the way they cope with it, influences
whether or not psychological stress will result, and its intensity (Lazarus, 1966). He
terms this appraisal and identifies two types: primary appraisal concerns whether
or not there is any personal stake in the encounter and secondary appraisal
concerns the available coping options for dealing with harm, threat, or challenge.
Lazarus defines coping as the cognitive and behavioral efforts a person makes to
manage demands that tax or exceed his or her personal resources (Lazarus &
Launier, 1978). He believes that coping is a process because the act of coping
results in changes which provide new information that will alter subsequent
appraisals. He has also shown that some coping strategies are stable across
encounters while others are unstable and heavily influenced by the environmental
conditions of the encounter (Lazarus, 1995).


9
Fifty-three percent of males in the 1987 study (which consisted of >95% male
respondents) and 67% of female veterinarians surveyed in the all-female 1992
study, achieved scores on a burnout scale (Pines & Aronson, 1988) which were
indicative of burnout syndrome. Women scored significantly higher than men, but
the study of women was completed nearly 10 years later than the study of men so
a direct comparison of these two groups is not valid. Elkins and Kearney (1992),
in addition to administering a burnout scale, asked the 613 responding female
veterinarians if they were happy in their current position and if their position was
stressful. Responses indicated that 83% of the veterinarians in the study reported
being happy in their job and 87% reported their work to be stressful.
The confusing and often inconsistent relationship found between job stress,
job satisfaction, burnout, and gender may well exist because moderator and/or
mediator variables have not been studied. Folkman and Lazarus (1988) distinguish
between moderators and mediators in the following way. Moderators are
antecedent conditions (such as gender) which affect the perceptions of the
situations and how it is appraised. Mediators, however, are generated in the
stressful encounter and are hypothesized to change the relationship between the
antecedent and the outcome variable. It is likely that the predictor variable of job
stress impacts the outcome variables of job satisfaction and burnout only after
having been acted upon by other variables Self-efficacy (Bandura, 1977), locus


30
In regards to the outcomes of job burnout, job satisfaction has received much
attention. As would be predicted, higher levels of burnout are linked to lower levels
of satisfaction. Symptomatically, people experiencing burnout express a desire to
spend less time working directly with clients and are more likely to be absent from
work (Maslach, 1986). Although it would be expected that burnout would readily
lead to a deterioration in the quality of client services provided, this area has not
been well researched. Jones (1981) did show that burnout was associated with
more on-the-job mistakes, more inhumanistic counseling practices, more
aggressive behavior toward clients, and more disciplinary action, by supervisors.
Job Satisfaction
Theories of job satisfaction were mentioned in Chapter 1. The fit
hypothesis suggests that the real meaning of work to an individual actually lies in
the congruence between the worker and his or her job (Locke, 1976). Herzberg,
Mausner, and Snyderman (1959) had previously argued that the features of work
itself are of preeminent importance, disregarding individual characteristics. Job
satisfaction is clearly a difficult concept to define accurately, but in general, some
amount of satisfaction exists when the costs of working are exceeded by the
benefits in a degree seen as acceptable by the worker (Fraser, 1983). Clearly, an
individuals level of job satisfaction lies on a continuum, which is subject to change
and is neither constant nor static.
It is noteworthy that much of the job satisfaction literature ignores the


72
the strain of responsibility and role overload.
it is certainly conceivable that the lack of interaction found between job stress
and coping in this study (with the exception of emotional coping in men) may relate
to the fact that the above types of coping resources were not assessed.
Implications of the Research Findings
This study fails to confirm the findings of the past decade of job satisfaction
literature by revealing significant gender differences in job satisfaction among
veterinarians in private equine practice. (However, as noted earlier, many of these
studies did not control for variables which might have obscured such differences.)
Due to financial constraints and survey space limitations, the specific components
of job satisfaction in these equine veterinarians were not measured. The current
study corrected the usual methodological limitations of failing to consider the impact
of potentially confounding demographic variables when assessing gender
differences in the work place. Age, years in the profession, years in the current job,
marital status, life stress experienced, type of practice, board certification status,
number of children (and whether or not they experienced significant disciplinary,
academic or emotional problems), salary, family wage earning status, and practice
ownership were assessed for gender differences. Although many gender
differences in these variables were detected, none were shown to significantly
Impact the assessment of job stress, job satisfaction or burnout in this population
with the exception of income, which impacted job satisfaction.
This study found significant gender differences in job stress and stressors,
which is in contrast to three of four studies cited previously which were conducted


74
gone undetected.
Another finding of pertinence to survey researchers is the confirmation of the
observation of Calvert (1996) who suggested that a larger initial sample size with
a single follow-up postcard would result in a comparable response size (albeit a
lower response rate) to the mailing of a second questionnaire to non-respondents
Savings in manual labor, postage and printing costs, and response time were
considerable in this study, and a more than adequate response rate (51%) was
obtained.
A significant implication of this study for the practice of counseling and
psychotherapy is that frequent use of emotional coping by male practitioners
strengthened the relationship between job stress and burnout. Further, for the men
and women in this study, more frequent use of rational and detached coping
methods weakened the relationship between job stress and burnout. While many
times the constraints inherent in the work environment appear to limit the possibility
of constructive action by individuals, how the individual copes will have a huge
impact on how he or she feels. This study suggests that for male practitioners when
rational coping methods are difficult to enact, detached coping should be attempted,
and emotional coping should be avoided. It is clear that the veterinary school
curriculum and professional conferences would profit from the addition of seminars
which teach students and practitioners about coping styles, enable them to
understand how they personally cope with stress, and help them to change their
coping styles if necessary.
In terms of veterinary practice, this study makes it clear that female equine


J
coping styles. Since the stresses and rewards of veterinary work vary among the
different areas of practice (government, industry, academia, private practice), this
study will focus on the largest segment of the veterinary population-private practice.
Equine practice was chosen because of the authors personal familiarity with the
area, the perception that this particular population of veterinarians may be the most
stressed, and the cooperation of the American Association of Equine Practitioners
(AAEP) in the endeavor. This information will be vital to the profession, particularly
as it continues to undergo feminization.
During the past 15 years, numerous studies on job satisfaction have been
conducted with many focusing on gender differences. Results have been
inconclusive, but the bulk of the most recent research suggests that men and
women do not differ significantly in overall job satisfaction ( Mobley, Jaret, Marsh,
& Lim, 1994; Mottaz, 1986; Russ & McNeilly, 1995; Schneer & Reitman, 1995;
Smith & Plant, 1982; Summers & DeCotiis, 1988). Many researchers suggest that
when gender differences have been demonstrated, such differences have been an
artifact of variables other than gender per se (Neil & Snizek, 1988). However,
measurement of and conclusions about overall job satisfaction fail to consider the
gender differences in satisfaction (or dissatisfaction) with various job components,
and methodological limitations such as lack of control of variables including salary,
promotion, age, and occupation status (professional versus nonprofessional) have


31
concept of job stress while researchers of job stress typically ask their subjects at
least one question concerning their level of job satisfaction. Fraser (1983) believes
that the basic relationship between job stress and job satisfaction is illustrated by
a bell shaped curve, indicating that some degree of stress is optimum for the
achievement of maximal job satisfaction.
As discussed in Chapter 1, there has been a plethora of research concerning
gender differences in job satisfaction (Andrisani, 1978; Cassidy & Warren, 1991;
Martin & Hanson, 1985; Mottaz, 1986; Murray & Atkinson, 1981; Shapiro & Stern,
1975; Weaver, 1974; Zanna, Crosby, & Loewenstein, 1987; Mortimer, Finch, &
Maruyama, 1988). Although conflicting findings are evident, in general, men and
women do not differ significantly in reported levels of job satisfaction. This is
paradoxical, in that gender differences in work values have been more consistently
reported. Women, more consistently than men, have placed significant importance
on hours worked per week and amount of travel time to and from work (Quinn,
Staines, & McCullough, 1974). Presumably, this relates to compatibility with family
responsibilities. Miller (1990) showed that men exhibit more concern for leadership,
opportunities for advancement, and decision-making ability than do women,
whereas women find such occupational concerns as job pressure, work hours and
job exhaustion as more salient than do men. Similar findings were replicated
among physicians, where no gender differences in overall satisfaction were
reported. Yet predictors of job satisfaction varied between the sexes. Satisfaction


Based on the results of the CSQ, men and women had a tendency to
cope differently with job stress. Men used significantly more rational and detached
coping, while women used significantly more emotional coping. Rational and detached
coping neither amplified nor moderated the effect of job stress on burnout in men or
women; however, more frequent usage of emotional coping strengthened the positive
association between job stress and burnout in men, but not women.
A striking outcome was that women perceived nearly half of the
assessed stressors (16 of 35 assessed on the HPSI-T) as significantly more stressful
than did the men (who reported no stressor as significantly more stressful than did the
women). Men and women significantly differed in their perceptions of frequency of
occurrence in only 4 of 35 assessed stressors. In each of these four, the women
reported the stressor as occurring more frequently.
Women reported considerably more stress concerning their income than
did men. As has been documented previously, these women reported average salaries
which were considerably lower than their male counterparts.
The results are discussed in the context of previous research completed
on other occupations. Implications for the practice of counseling and for veterinary
V
education are also discussed.


49
stress F(1,426) = 6.06, £ = .01.
Next, as part of the preliminary analyses, Pearson correlation coefficients
were computed to determine the correlation between the continuous variables (now
viewed as independent, confounding variables for which gender differences were
found) and job stress (HPSI-T), job satisfaction (SAT) and burnout (BOS). No
variable had a correlation which was high enough to potentially account for a
significant amount of the variance in the scores. (See Table 4.1).
Table 4.1
Pearson correlation Coefficients of HPSI-T, BOS and SAT scores with the
potentially confounding (independent) variables.
HPSI-T
SAT
BOS
Age
-0.29*
0.16**
-0.23*
Years in Profession
-0.34*
0.15**
-0.22*
Years in job
-0.33*
0.16**
-0.20*
Income
-0.26*
0.21*
-0.26*
Hours/Week
0.06
0.01
0.12***
Children <6yrs.
0.01
0.05
-0.07
Children 6-18
-0.09
0.04
-0.03
Children 19-25
-0.06
0.07
-0.07
Total Children Home
-0.09
0.08
-0.09
Life Stress
0.25*
-0.15**
0.41*
*£<.0001 **£<.001
***£< 01


53
R-square procedures were run to be certain that the variables chosen for use as
covariates in the MANCOVA (i.e., life stress, income, years in the profession)
were the most appropriate covariates. Thus, three R-square procedures with
job stress, job satisfaction or burnout as the dependent variable in each and the
15 possible covariates and their combinations as the predictor variables in each
were run to determine the best reduced model for each of the dependent
variables. Results revealed that the covariates chosen (years in the profession,
income, and life stress) were indeed the most appropriate covariates for the
MANCOVA to test Hypothesis 2.
Hypothesis 3
Hypothesis 3 stated that gender differences in the severity and frequency of
specific job stressors would be found. Although gender differences in the overall
measure of job stress were examined using the covariates of number of years in the
profession, income, and perceived life stress, it was concluded that gender
differences in each individual stressor would be examined without using covariates.
Using seventy separate ANCOVAs would diminish the power of the statistic so
much so that potentially significant gender differences would not be identified.
Thus, Hypothesis 3 was tested using two-tailed T tests. Since a total of 35
stressors were to be compared on both severity and frequency of occurrence, a


91
Dawis, R. V., & Lofquis, L. (1984). A psychological theory of work
adjustment. Minneapolis: University of Minnesota Press.
Deaux, K., & Emswiller, M. (1974). Explanations for successful
performance on sex-linked tasks: what is skill for the male is luck for the female.
Journal of Personality and Social Psychology. 29. 80-85.
Dial, T., Bickel, J., & Lewicki, A. (1989). Sex differences in rank attainment
among radiology and internal medicine Faculty. Academic Medicine. 64.
198-202.
Dillman, D. A. (1983). Mail and other self-administered questionnaires. In
P. Rossi, J. Wright, & A. Anderson (Eds.), Handbook of survey research (pp.
359-377). San Diego: Academic Press, Inc.
DiSalvo, B., Lubbers, C., Rossi, A., & Lewis, J. (1994). The impact of
gender on work-related stress. In P. L. Perrewe & R. Crandall (Eds.),
Occupational stress: A handbook (pp. 39-50). Washington, DC: Taylor & Francis.
Doherty, W., & Baldwin, C. (1985). Shifts and stability in locus of control
during the 1970s: Divergence of the sexes. Journal of Personality and Social
Psychology. 48, 1048-1053.
Elkins, A., & Elkins, J. (1987). Professional burnout among U.S.
veterinarians: How serious a problem? Veterinary Medicine. 82, 1245-1250.
Elkins, A., & Kearney, M. (1992). Professional burnout among female
veterinarians in the United States. Journal of the American Veterinary Medical
Association. 200. 604-608.
Etaugh, C. (1984, Fall). Women faculty and administrators in higher
education: Changes in their status since 1972. Journal of National Association
for Women Deans. Administrators, and Counselors, pp. 21-25.
Folkman, S., & Lazarus, R. (1980). An analysis of coping in a middle-
aged community sample. Journal of Helath and Social Behavior. 21, 219-239.
Folkman, S., & Lazarus, R. (1985). If it changes it must be a process:
Study of emotion and coping during three stages of a college examination.
Journal of Personality and Social Psychology, 48, 150-170.


GENDER DIFFERENCES IN JOB STRESS, BURNOUT, AND JOB SATISFACTION
AS MEDIATED BY COPING STYLE OF VETERINARIANS IN
PRIVATE EQUINE PRACTICE
Bv
BARBARA BREWER WELSCH
/
/
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1998

This work is dedicated to my wonderful mother, Helen M.
Dean, and my understanding husband, Boyd B. Welsch. Their generosity,
respect, and love truly made this achievement possible. It is also dedicated
to my daughter, Collyn Kimberly Welsch, who made the experience all the
more worthwhile.

ACKNOWLEDGEMENTS
I would like to express my most sincere thanks and
appreciation to all who have helped me complete this dissertation. I am
grateful to my advisor and friend, Dr. Carolyn Tucker, for her support,
expertise, commitment to excellence, and attention to detail. I am thankful to
Dr. Ellen Amatea, who suggested the exploration of coping in the
veterinarians I was intending to study. Further, I am indebted to Dr. John
Dixon for his invaluable statistical expertise which was provided with patience
and incredible promptness.
I wifi also be forever grateful to my former colleagues, the
members of the Association of Equine Practitioners who responded to my
survey, and to their director Gary Carpenter, who provided both the funding
and the enthusiasm for the undertaking of this work. Lastly, I owe thanks to
my dear friend and veterinary colleague, Dr. Anne Koterba, who suggested
that the AAEP might be interested in working with me, to Dr. Ed Blach of the
/AEP who advised me on the marketing aspects of survey research, and to
Dave Foley of the AAEP who coordinated the printing, labeling, and mailing
process.

TABLE OF CONTENTS
ACKNOWLEDGEMENTS iii
Abstract vi
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 REVIEW OF THE RELATED LITERATURE. 14
Job Stress 14
Coping 22
Burnout 27
Job Satisfaction 30
Rationale and Hypotheses of the Present Study 34
CHAPTER 3 METHOD 36
Participants 36
Measures 48
Procedure 44
CHAPTER 4 RESULTS 46
Descriptive Statistics 46
Data Analysis 47
Hypothesis 1 47
Hypothesis 2 48
Hypothesis 3 53
IV

Hypothesis 4 54
Hypothesis 5 56
Summary of the Resuits 61
CHAPTER 5 DISCUSSION 62
Implications of the Research Findings 72
Limitations of This Study 75
Future Research 78
Conclusions 80
APPENDIX A SURVEY 82
APPENDIX B COVER LETTER 87
APPENDIX C FOLLOW UP POSTCARD 88
REFERENCE LIST 89
BIOGRAPHICAL SKETCH 100
V

Abstract of Dissertation Presented to the Graduate School of the University of
Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of
Philosophy
GENDER DIFFERENCES IN JOB STRESS, BURNOUT, AND JOB
SATISFACTION AS MEDIATED BY COPING STYLE OF VETERINARIANS
IN PRIVATE EQUINE PRACTICE
By
Barbara B. Welsch
August, 1998
Chairman: Carolyn M. Tucker, Ph.D.
Major Department: Psychology
A mail survey of 457 practicing equine veterinarians investigated gender
differences in job stress, burnout, job satisfaction, and coping style. Participants
included 205 men and 252 women who completed the Health Professions Stress
Inventory (HPSI-T), the Burnout Measure (BOS), the Coping Styles Questionnaire
(CSQ), a measure of Job Satisfaction (SAT), a single question assessing overall life
stress and a demographics questionnaire.
A moderate correlation between job stress and burnout and between job
stress and job satisfaction was found. Women reported significantly higher levels of job
satisfaction, but were significantly more stressed and more burned out than their male
colleagues. These gender differences remained when controlling for the effects of age,
years in the profession, years in the current job, income, hours worked, marital status,
number of children in the household, and life satisfaction.
VI

Based on the results of the CSQ, men and women had a tendency to
cope differently with job stress. Men used significantly more rational and detached
coping, while women used significantly more emotional coping. Rational and detached
coping neither amplified nor moderated the effect of job stress on burnout in men or
women; however, more frequent usage of emotional coping strengthened the positive
association between job stress and burnout in men, but not women.
A striking outcome was that women perceived nearly half of the
assessed stressors (16 of 35 assessed on the HPSI-T) as significantly more stressful
than did the men (who reported no stressor as significantly more stressful than did the
women). Men and women significantly differed in their perceptions of frequency of
occurrence in only 4 of 35 assessed stressors. In each of these four, the women
reported the stressor as occurring more frequently.
Women reported considerably more stress concerning their income than
did men. As has been documented previously, these women reported average salaries
which were considerably lower than their male counterparts.
The results are discussed in the context of previous research completed
on other occupations. Implications for the practice of counseling and for veterinary
V
education are also discussed.

CHAPTER 1
INTRODUCTION
The veterinary profession, like other health-related professions including
medicine and psychology (Pin et ai., 1996), is undergoing many changes (Clark,
1996). It is becoming increasingly difficult to establish profitable veterinary practices;
furthermore, corporations are beginning to take over established practices in order
to create large referral centers with satellite clinics. Employed veterinarians will have
far less autonomy than their practice-owning predecessors. Veterinary students
are graduating with huge debts (mean = $45,251) but accepting starting salaries
(mean = $29,854; Gehrke, 1995) which make loan repayment difficult, and which,
in fact, are extremely low in proportion to the number of years of education required.
A previously male-dominated profession, veterinary medicine is rapidly
becoming femininized, with women constituting up to 75% of currently enrolled
freshman classes (Gehrke, 1996). Significantly, Bird (1992) showed that female
veterinarians had an income disadvantage of $13,000 per year compared to male
veterinarians. Forty-two percent of the sex differences in income Bird documented
were correlated with gender differences in veterinarians characteristics (differences
1

?
in education, years of experience, board certification, practice type, investment in
private practice, and hours worked). The remaining 58% ($7,400) could not be
explained, indicating that veterinarians who are female, like their counterparts in
medicine and dentistry, receive substantially lower incomes than male veterinarians
(Bird, 1992).
In summary, veterinary medicine is a profession which requires many years
of education typically culminating in high debt. Yet, its practitioners receive
comparatively low starting salaries and are losing the ability to own their own
businesses. Additionally, an increasing proportion of its practitioners who are
women appear to suffer from salary discrimination which, if perceived, is likely
stressful. Satisfaction gleaned from the work itself, therefore, is presumably a
strong motivator in attracting both males and females to the profession and
preventing their attrition.
With the exception of two studies on burnout (Elkins & Elkins, 1987; Elkins
& Kearney, 1992) and one (unpublished) study on gender differences in job
satisfaction (Welsch, 1991), no research on job satisfaction and stress within the
veterinary profession has been conducted. The purposes of this exploratory study
are (1) to examine the relationships among job stress, job satisfaction, and burnout
in veterinarians in private, predominantly equine, practice and (2) to determine if
relationships among these variables are moderated by gender and mediated by

J
coping styles. Since the stresses and rewards of veterinary work vary among the
different areas of practice (government, industry, academia, private practice), this
study will focus on the largest segment of the veterinary population-private practice.
Equine practice was chosen because of the authors personal familiarity with the
area, the perception that this particular population of veterinarians may be the most
stressed, and the cooperation of the American Association of Equine Practitioners
(AAEP) in the endeavor. This information will be vital to the profession, particularly
as it continues to undergo feminization.
During the past 15 years, numerous studies on job satisfaction have been
conducted with many focusing on gender differences. Results have been
inconclusive, but the bulk of the most recent research suggests that men and
women do not differ significantly in overall job satisfaction ( Mobley, Jaret, Marsh,
& Lim, 1994; Mottaz, 1986; Russ & McNeilly, 1995; Schneer & Reitman, 1995;
Smith & Plant, 1982; Summers & DeCotiis, 1988). Many researchers suggest that
when gender differences have been demonstrated, such differences have been an
artifact of variables other than gender per se (Neil & Snizek, 1988). However,
measurement of and conclusions about overall job satisfaction fail to consider the
gender differences in satisfaction (or dissatisfaction) with various job components,
and methodological limitations such as lack of control of variables including salary,
promotion, age, and occupation status (professional versus nonprofessional) have

4
made it difficult to generalize from one study to the next or to be certain that gender
differences are, indeed, nonexistent.
As earlier mentioned, no research on job satisfaction has been conducted in
the veterinary population. Several studies on the job satisfaction and status of
physicians and dentists have been published, however (Bickel, 1988; Carr,
Friedman, Moskowitz, & Kazis, 1993; Dial, Bickel, & Lewicki, 1989; Lenhart &
Evans, 1991; Lewis, Carson, Nace, Howard, & Barnhart, 1994). Richardsen and
Burke (1991) found no differences in overall job satisfaction between male and
female physicians, but noted differences in sources of satisfaction that predicted
overall job satisfaction. For women, satisfaction with time available for family and
personal life was an important predictor of overall satisfaction, whereas for men,
several work setting factors (e.g. own working conditions, access to support staff,
and equipment) were related to job satisfaction.
Kevorkian and Tuel (1994) surveyed 49 medical faculty members who had
recently resigned. While they found no significant differences in reasons for leaving,
there were gender differences noted when the physicians were asked what it would
have taken to induce them to stay in their present position. Men rated more
protected time for academics and more support for research more highly than did
women, and women ranked effective mentors or role models and more support for
patient care more highly than did men.

5
Two major competing theories of job satisfaction in the literature include the
fit hypothesis (Locke, 1976) and the work attribute hypothesis (Herzberg,
Mausner, & Snyderman, 1959). The fit hypothesis addresses the compatibility of
externa! work features and individual worker characteristics. This theory
emphasizes the differences in how workers react to their jobs and the preponderant
importance of these variations in influencing satisfaction and/or dissatisfaction with
work. In contrast, the work attribute hypothesis states that the features of the work
itself coupled with advancement, recognition, responsibility, and growth are of
preeminent importance for job satisfaction, and individual attributes make little
difference. Many authors have defined job satisfaction over the years (Hoppock,
1935; Super, 1942; Hopkins, 1983; Dawis & Lofquist, 1969, 1984). Regardless of
whether the concept of needs and values or individual perceptions is employed, the
common underlying premise of the most recent perspectives on job satisfaction is
that there is something within an individual which corresponds to his or her reaction
to (and consequent satisfaction with) the job.
Despite the plethora of job satisfaction research, much of it ignores the
concept of job stress. There is an even larger body of literature concerning job
stress. Typically, this research addresses the issue of job satisfaction minimally by
asking the participants one or two global questions aimed at measuring job
satisfaction. There is no end to the number of studies showing that stress in the

6
workplace adversely affects productivity, absenteeism, worker turnover, and
employee health and well-being (Perrewe, 1991; Quick, Murphy, & Hurrell, 1992;
Spielberger & Reheiser, 1995).
The concept of stress can be viewed from several perspectives. The
stimulus definition sees stress as a characteristic of the environment that is
disturbing or disruptive. The response definition of stress states that stress is a
pattern of physiological and/or psychological reactions exhibited by a person who
is under some sort of pressure. Neither definition of stress clearly explains
individual differences, however (Cox, 1978). A third approach to conceptualizing
stress is the transactional model (Lazarus & Folkman, 1984). This view combines
the stimulus and the response approaches and defines stress as the consequence
of the interaction between environmental stimuli and individual responses.
Intervening psychological processes such as perception and cognitive appraisal are
emphasized. Stress is considered to occur only when a person perceives a
demand as exceeding his or her capacity to deal with it. The major criticism of this
theoretical model is that it is incomplete since it fails to consider traits or personality
factors (Ben-Porath & Tellegen, 1990; Costa & McCrae, 1990).
When the effects of gender differences on job stress have been studied, the
results have been conflicting. While much of the early work found no sex
differences, the researchers failed to look at the frequency or the intensity of the

7
stress experienced. Later studies which took the above factors into account were,
indeed, able to detect gender differences. There are no studies evaluating job
stress in the veterinarian, but several were found which investigated other health
related professions (Wolfgang, 1988, 1995; Simpson & Grant, 1991; Richardsen &
Burke, 1991). Three of the four studies found no gender-related differences, but
Richardsen and Burke (1991) reported significant sex differences in terms of the
variables that predicted job stress. Women rated the need to maintain their own
level of knowledge and providing counseling for nonmedical problems as more
stressful than did men, whereas male physicians reported more stress related to
maintaining an adequate income.
Even though consistent differences in the amount or types of stress
experienced on the job have not been shown between men and women, it is
virtually uncontested that women in Western cultures have higher rates of affective
and anxiety disorders than do men, in both clinical and nonclinical populations
(Miller & Kirsch, 1987). It is generally agreed that many professional women may
experience one or more of the following (Smith, 1993): a hostile work environment
in which their values, working style and job contributions are trivialized,
unappreciated, or unrewarded; an unsupportive home environment in which it is
their primary responsibility to care for the home and the children; a disapproving
social environment (i.e., their house is not clean enough and they rarely produce

8
healthy meals; they should not put their children into daycare); and major conflicts
with their self-concept (worker vs. mother vs. wife). Two theoretical perspectives
have been advanced to address these issues: the role conflict perspective (scarcity
hypothesis) and the role expansion perspective. The former suggests that the
above conflicts add to the stress of the female worker (Goode, 1960), while the
latter model finds that if a female sees her roles to be both meaningful and of high
quality, multiple roles lead to positive outcomes and may even buffer the impact of
negative life events (Repetti, Mathews, & Waldron, 1989).
Although gender differences in job stress are inconsistently reported, the
reported effects of job stress have been consistent. Burnout is a syndrome of
emotional exhaustion, depersonalization, and reduced personal accomplishment
that can occur among individuals who work with people in some capacity (Maslach,
1986). Burnout appears to be a response to chronic sources of emotional and
interpersonal stress on the job. It is a particular problem for men and women in
caregiving professions, especially when the provision of care or service occurs in
emotionally charged situations (Maslach, 1986). Higher levels of burnout are linked
to lower levels of job satisfaction.
Burnout is the one psychological outcome variable that actually has been
studied in veterinarians (Elkins & Elkins, 1987; Elkins & Kearney, 1992). (Neither
of these two referenced studies was limited to veterinarians in private practice.)

9
Fifty-three percent of males in the 1987 study (which consisted of >95% male
respondents) and 67% of female veterinarians surveyed in the all-female 1992
study, achieved scores on a burnout scale (Pines & Aronson, 1988) which were
indicative of burnout syndrome. Women scored significantly higher than men, but
the study of women was completed nearly 10 years later than the study of men so
a direct comparison of these two groups is not valid. Elkins and Kearney (1992),
in addition to administering a burnout scale, asked the 613 responding female
veterinarians if they were happy in their current position and if their position was
stressful. Responses indicated that 83% of the veterinarians in the study reported
being happy in their job and 87% reported their work to be stressful.
The confusing and often inconsistent relationship found between job stress,
job satisfaction, burnout, and gender may well exist because moderator and/or
mediator variables have not been studied. Folkman and Lazarus (1988) distinguish
between moderators and mediators in the following way. Moderators are
antecedent conditions (such as gender) which affect the perceptions of the
situations and how it is appraised. Mediators, however, are generated in the
stressful encounter and are hypothesized to change the relationship between the
antecedent and the outcome variable. It is likely that the predictor variable of job
stress impacts the outcome variables of job satisfaction and burnout only after
having been acted upon by other variables Self-efficacy (Bandura, 1977), locus

10
of control (Lefcourt, 1982), and causal attributions (Doherty & Baldwin, 1985) may
each play a role in moderating or mediating the effects of job stress and individual
differences on job dissatisfaction, burnout, and commitment. Other variables which
may be relevant include the need for achievement, level of self-esteem, tolerance
of ambiguity, Type A vs. Type B personality (Ivancevich & Matteson, 1980), and
negative affectivity (Parkes, 1990). A very pragmatic avenue of investigation,
however, may be identification of veterinarians coping styles since interventions
aimed at altering a style which is associated with burnout or dissatisfaction could
subsequently be offered.
Lazarus transactional theory of stress (1991), which was cited earlier,
reflects the multidimensional, dynamic, and complex nature of stress, including the
interactions between the individual and the context in which he or she is working.
Two processes, cognitive appraisal and coping, are identified as crucial mediators
of stressful person-environment relationships and their immediate and long-term
outcomes (Lazarus & Folkman, 1984). Personal meanings of an event itself and the
choice and effectiveness of ones coping actions are considered integral to the
stress process.
The existing literature presents a confusing view of the role of coping in the
workplace. Various researchers label the forms of coping differently, but three
general types of coping seem to be cited most frequently: problem-focused,

11
emotion-focused and symptom relieving (avoidance). Problem focused coping
refers to attempts to modify or eliminate the source of stress by taking instrumental
actions (Miller & Kirsch, 1987). This could be in the form of direct action or help
seeking behavior (Havlovic & Keenan, 1995). In contrast, emotion-focused coping
refers to attempts to control or repress stress-related emotions and maintain
affective equilibrium (Billings & Moos, 1984). This form of coping could be in the
form of either positive thinking or avoidance/resignation (Havlovic & Keenan, 1995).
Symptom focused coping is defined as the use of exercise, meditation, alcohol, or
drugs to alleviate the stress response (Schuler, 1985). Other authors use the term
avoidance coping to include additional methods of simply ignoring the problem.
Roger et. al. (1993), through their work with stress management programs and the
development of the Coping Styles Questionnaire (CSQ), identified a fourth
mechanism of coping which they term detachment" coping. This form of coping
does not involve denial or attempts to avoid stress, but it can also be distinguished
from task-oriented strategies (Roger, Jarvis, & Najaran, 1993). This strategy
involves feeling independent of the event and the emotion associated with it.
Much of the occupational-stress literature reports data only from male
employees (Parkes, 1990). Gender issues are not addressed and relatively little
is known about gender differences in work-related coping. More attention has been
given to gender in the general coping literature, but the results are very inconsistent.

12
Several groups have shown no gender differences in coping strategies (Hamilton
& Fagot, 1988; Osipow & Davis, 1988). In contrast Pearlin and Schooler (1978)
found that women reported less use of effective strategies, a finding corroborated
by Billings and Moos (1981). Roger et al. (1993), using the CSQ which they
developed, showed a tendency for men to use more detached and rational and less
emotional coping styles than women, but the differences were not statistically
significant.
In contrast, in two studies conducted in actual work settings, women were
shown to cope in more positive ways. Parasuraman and Cleek (1984) found that
female managers reported greater use of adaptive coping strategies (planning,
information seeking, priority setting) than did males. Parkes (1990) found that men
used significantly more suppression strategies such as withdrawal, restraint,
compromise, and ignoring the problem. Miller and Kirsch (1987) summarize the
literature by concluding that it appears that the categories of emotion-focused and
problem-focused coping may yield significant sex differences in how people
cognitively cope with stress .... It is possible that women appraise threatening
events as more stressful than men do. However, given the methodological
constraints of these studies, further research is needed before firm conclusions can
be drawn (p. 295). The veterinary profession is changing drastically as a result of
the changing economy and changing demographics (feminization). Occupations

13
which have become feminized have experienced a decline in salary and prestige
leading to decreased job satisfaction. This research was the first to explore job
stress, job satisfaction, and coping styles of veterinarians of either gender. Findings
from this research will have implications for the future of the profession as it
becomes female dominated. More importantly, findings suggest ideas for potential
interventions at the level of both the veterinary student and practicing veterinarian.
Consequently, the results of this study may enhance job satisfaction and decrease
job turnover, burnout, and professional attrition among veterinarians in private
practice.

CHAPTER 2
REVIEW OF THE RELATED LITERATURE
The literature review presents research which will aid in the understanding
of the background upon which this study was based. There are five sections in this
chapter. These include sections on job stress, job satisfaction, burnout, and coping
and cover theories of each along with gender and other differences which have
been noted. The final section concludes with the five research hypotheses
investigated in this study.
Job Stress
The concept of stress has been viewed from several, evolving perspectives
over the past century. Although many different definitions of stress have been
entertained by researchers, most seem to fall into three general categories. In the
stimulus definition, stress is seen as an external force that causes a reaction of
strain within the individual. Holt (1982) defines 57 different work stressors (which
include work overload, role ambiguity, monotony, and lack of control).
The response definition of stress takes the difference in stressors into
account. The Canadian endocrinologist, Hans Selye (1950), profoundly shaped the
14

15
modern concept of stress as a response by discovering that tissue damage could
result from diverse stimuli and was thus a nonspecific response to all noxious
stimuli. It is clear, however, that different stressors culminate in differing response
intensities. Although Selye was looking at purely physiological reactions, the
response definition of stress has grown to include the psychological reactions
exhibited by a person who is under some sort of pressure (Maslach, 1986). Neither
definition is capable of explaining individual differences, however (Cox, 1978).
An important advance in the thinking about stress was supplied by French,
Caplan, and Van Harrison (1982). Their ideas about person-environment fit
combine the separate causal variables of environment and person into an ongoing
relationship characterized by fit or misfit. Occupational stress is defined in terms of
job characteristics that pose a threat to the individual because of a poor match
between the abilities of the employee and the demands of the job. But this
approach falls short, because it is basically a static approach. It emphasizes stable
relationships between a person and the workplace rather than a process during
which stress changes over time and varies within similar contexts. It is clear that
even when a person and his or her environment are a good fit, some stress will
occur. In fact, most work researchers agree that some degree of stress is
beneficial.
Lazarus transactional process theory of stress (1995) distinguishes between
stressful antecedent conditions and how these are perceived and cognitively

16
appraised by a particular person. The consequent emotional reactions which follow
when a stressor is perceived as threatening and the individual is not able to cope
effectively with it form the basis of stress. Thus, Lazarus looks at the stressful
stimulus and looks at the response, but, in addition, places great emphasis on the
intervening variables of appraisal and coping ability, adding dimensions which
others have not considered.
Although Lazarus theory of stress is the most widely known and accepted
theoretical model, it is not without its critics. The major criticism of this theoretical
model is that it is incomplete since it fails to consider traits or personality factors
(Ben-Porath &Tellegen, 1990; Costa & McCrae, 1990). Further, Brief and George
(1994) criticize Lazarus emphasis on the idiographic nature of occupational stress.
They argue that it is especially important to discover those working conditions that
are likely to adversely affect groups of employees who are exposed to them. In a
similar critique of the model, Harris (1994) has noted that the occupational stressors
associated with the climate and the culture of an organization can have profound
effects on employees, and that these may differ as a function of gender and
individual differences in personality as well as coping skills. Lazarus transactional
model of occupational stress will be discussed in more detail under the section
entitled Coping.
Most of the occupational stress research has centered upon the effects of job
stress on the workers health--both physical and mental. In regard to physical

17
health, coronary heart disease has been clearly linked to job stress along with the
Type A behavior pattern (Maslach, 1986). Gastric ulceration has less consistently
been linked to job stress as well. Mortality rates have been found to differ by
occupation, and although unproved, it is reasonable to implicate stress on the job
as one explanatory factor. In a recent letter to the editor of the Journal of the
American Veterinary Medical Association (Prichard, 1996), the author observed that
in the monthly obituary column there appeared to be a disproportionate number of
relatively young veterinarians compared to the obituaries listed in a stockbrokers
journal, in fact, when surveyed, fully 87% of female veterinarians considered their
jobs to be veiy' stressful (Elkins & Kearney, 1992).
Stress in the workplace may well be increasing. The proportion of workers
who reported feeling highly stressed on the job had more than doubled from 1985
to 1990 (Northwestern National Life, 1991). Sixty-nine percent of the 600 workers
surveyed in the Northwestern study reported that their productivity was reduced due
to high stress levels, and one in three say job stress is the single greatest stress
in their lives (1991, p. 2).
Job stress has been linked to impairments in psychological well-being and
implicated in reports of depression, dissatisfaction, anxiety, tension, lowered self
esteem, and alcohol and drug abuse (Maslach, 1986). Absenteeism, employee
turnover, decreased productivity and diminished client-relations are obvious costs
to employers (Spielberger & Reheiser, 1995). According to Matteson and

18
Ivancevich (1987) preventable costs relating to diminished productivity, turnover,
and absenteeism are in the realm of $2800 per person employed. This figure does
not even include such things as workers compensation, disability payments, and
early retirement benefits.
Despite a burgeoning of research in the occupational stress area, the
measurement of job stress has been inconsistent and confusing. Most of the
literature has relied on answers to only one or a few questions about experienced
stress (Barone, Caddey, Katell, Roselione, & Hamilton 1988). Further, job stress
measures tend to confound the frequency of occurrence of a stressful event with its
perceived severity (Spielberger & Reheiser, 1995). Some authors have contended
that research in work stress should focus on specific aspects of the job which
workers find problematic (Jackson & Schuler, 1985), while others have argued for
the development of generic measures to facilitate comparing stress levels across
occupational groups (Murphy & Hurrell, 1987).
In response to the criticisms of past measures, Spielberger and colleagues
developed the Job Stress Survey (JSS), adapting it from earlier surveys which they
had designed to measure job stress in policeman and later, teachers (Spielberger
& Reheiser, 1995). This 30-item psychometric instrument was designed to assess
the perceived intensity and frequency of occurrence of working conditions which
could adversely affect the psychological well-being of employees. The survey was
intended to form a generic job stress measure and has been used repeatedly in

19
university and corporate work settings (Spielberger & Reheiser, 1995).
It was this authors opinion, in agreement with Jackson and Schuler (1985),
that an instrument specific to the health related professions would be more valuable
in this particular study. Medical professionals face a different set of stressors than
do other professionals, and recently, increasing attention has been focused on the
adverse effects of job stress on providers and ultimately on the overall quality of
health care (Wolfgang, 1988). Although various instruments and surveys have
assessed job stress in physicians, none have looked at veterinarians work stress.
One instrument, the Health Professions Stress Inventory (HPSI) has been used to
compare levels of stress within the occupations of mediciine, nursing, and
pharmacy, and it was felt that this instrument would provide useful information
regarding the veterinary profession while allowing comparisons to other health-
related professions.
The general job stress literature produces inconsistent findings regarding the
effect of gender on job stress. Beehr and Schuler (1980) concluded that women
and men show no differences in stress-related symptoms in the workplace, and
DiSlavo, Lubbers, Rosi, and Lewis (1994) found that from a broad perspective,
men and women perceive stressors quite similarly (p. 48). On the basis of a meta-
analytic evaluation of 15 studies which examined gender differences in occupational
stress, Martocchio and OLeary (1989) found no sex differences in experienced and
perceived work stress (p. 495).

20
It has been suggested that failure to identify gender-related differences in
occupational stress may be due to sampling problems (men over-represented in
management positions with women holding clerical and service jobs (Jick & Mitz,
1985). Further, these authors conclude that gender may act not only as a direct
predictor of the source of stress, but also as a moderator affecting how stress is
perceived, what coping skills are called upon, and how stress is manifest (p. 409).
In agreement with the above authors, Spielberger and Reheiser (1995), using the
JSS, noted striking gender differences in the severity or frequency of items identified
as stressful. Men and women rated the same six items as highest in perceived
severity and the same 5 items as lowest in severity, however.
Consequently (although somewhat surprisingly), gender related differences
were not statistically confirmed in the JSS severity, frequency or overall scores
despite the numerous differences reported in the severity and frequency of
occurrence of specific JSS stressors. The authors concluded that gender-related
differences in occupational stress appear to be determined by differences in the
perceived severity of specific stressors and in the frequency that these stressors are
experienced by men and women (p. 64).
There has been some speculation but little research concerning differences
in stress experienced by male and female physicians (Simpson & Grant, 1991).
More research has been carried out with medical students than with physicians in
practice.

21
Simpson and Grant (1991) reported that, despite speculation to the contrary,
physicians in their study evidenced no difference by gender in the types and
magnitude of stressors experienced. As in the Richardsen and Burke (1991) study
(see below), the authors used their own set of identified stressors, not a
standardized instrument. Richardsen and Burke (1991) found that women
physicians reported slightly higher overall stress than did men. They also found
significant differences in three sources of stress reported: women rated the need to
maintain their own level of knowledge and counseling nonmedical problems as
more stressful than did men, whereas male physicians reported more stress related
to maintaining an adequate income.
Age and years of experience may have an impact on job stress and job
satisfaction and other work attitudes such as commitment and turnover intention.
In a study of job stress, satisfaction, and life satisfaction in physicians, Linn, Yager,
Cope, and Leake (1985) found an inverse relationship between age and stress, and
positive relationships among age and job and life satisfaction. Age differences
possibly accounted for the difference in overall stress between men and women
physicians in the Richardsen and Burke (1991) study, since men in their study were
generally older and had more years of experience than women. No statistics were
done to confirm this possibility, however. Similarly, researchers have suggested
that experience (or closely related constructs such as career stage) may moderate
relationships among sales force members perceptions, attitudes, and behavior

22
(Russ & McNeilly, 1995).
Coping
As discussed in the previous section, the psychological stress literature is
permeated with discussions about the potentially deleterious effects of stress on
physical and mental health as well as on individual attitudes and behaviors.
Therefore, it follows that much emphasis has also been placed on the investigation
of how people manage or cope with stress in all of lifes domains including the
workplace.
Lazarus transactional model of stress and coping (Lazarus, 1995) is one of
several models which have been used to study the nature of coping. Most
researchers who have studied coping cite Lazarus work and many use it to build
slightly different and more complex models. For example, Schulers (1985)
integrative transactional process model of coping (which is specific for
organizations) also addresses the impact of the uncertainty and the importance of
the situation to the individual and includes a process of feedback and evaluation.
The transactional theory states that a transaction between the person and
the environment is stressful only when it is evaluated by an individual as a harm,
threat, or challenge to that persons well-being. Lazarus terms this evaluation an
appraisal. Within the context of the theory, distinction is made between harm,
threat, and challenge. Harm refers to damage which has already occurred (loss

23
of job, poor job evaluation, lack of promotion, disapproval), while threat refers to a
harm that has not yet happened, but is anticipated. Challenge refers to a condition
of high demand in which the emphasis is on mastering the demands, overcoming
obstacles, and growing and expanding as an individual. During threat, the focus is
on avoiding or protecting against harm. In challenge, the emphasis is on the
positive outcome possibilities. Individuals, particularly those who have chosen
medical professions, thrive on challenge while at the same time dislike (and attempt
to evade) threat.
Lazarus early field work demonstrated that the way people evaluate what is
happening with respect to their well-being, and the way they cope with it, influences
whether or not psychological stress will result, and its intensity (Lazarus, 1966). He
terms this appraisal and identifies two types: primary appraisal concerns whether
or not there is any personal stake in the encounter and secondary appraisal
concerns the available coping options for dealing with harm, threat, or challenge.
Lazarus defines coping as the cognitive and behavioral efforts a person makes to
manage demands that tax or exceed his or her personal resources (Lazarus &
Launier, 1978). He believes that coping is a process because the act of coping
results in changes which provide new information that will alter subsequent
appraisals. He has also shown that some coping strategies are stable across
encounters while others are unstable and heavily influenced by the environmental
conditions of the encounter (Lazarus, 1995).

24
The meaning of work clearly varies widely among workers in our society. For
some individuals, work is merely a means of earning money for survival. For others,
work is their life--their sole reason for surviving. It is my contention, biased through
my own participation in multiple areas within the field of veterinary medicine, and
based on the comparatively low income achieved by veterinarians compared to
other health professionals, that their work has considerable meaning to most
veterinarians.
Locke and Taylor (1990) define meaning of work as the totality of values,
including their importance, that individuals seek and expect to derive from work.
Values themselves are what individuals desire or consider to be good or beneficial
(p. 135). These authors propose that individuals may expect and desire to fulfill five
different categories of values from their work-- material values, achievement-related
values, a sense of purpose, social relationships, and the enhancement or
maintenance of self-concept.
Threats to any of the above values (which are important to an individual)
cause stress, followed by the activation of coping mechanisms. Material values can
be associated with work stress in obvious ways. Workers can base their self-
concept almost entirely on material outcomes. Further, they might exclude other
important values (i.e., family relationships) entirely in the pursuit of job success and
money. Veterinary practice owners, one of the few veterinary groups able to earn
over $100,000, can only do so by working exceedingly long hours, almost always

25
to the detriment of their families
The key threats to achievement-related values are associated with loss: loss
of feelings of competence, loss of interest in (or challenge by) the work itself, or loss
of personal control over ones work (Locke & Taylor, 1990). Workers may lose their
sense of purpose at work after attaining highly valued, long sought-after goals, or
they may gradually begin to question the real value of what they do. Social
relationships serve both an expressive and an instrumental function at work (Locke,
1976). Interactions can be pleasurable and/or necessary for completion of the job
and difficulties in this area can be particularly disturbing to workers. Some research
has shown that this work value may be more important to women than to men. An
individuals self-concept is threatened at work when there in an incongruity between
the self-concept and what the individual sees as a requirement of the job. The self-
concept is further threatened by perceived failures at work as well.
The stress caused by these threats must be coped with, in one way or
another. Coping is typically defined in the literature as the process of adapting to
perceived threats, and coping has been repeatedly divided into rational (problem-
focused), emotional, and avoidance (symptom-focused) forms with different authors
attributing differing meanings to each of the above. Roger, Jarvis, and Najaran
(1993) have recently described a scale which measures four types of coping:
rational, emotional, avoidance, and detachment coping. This scale represents an
extension of Folkman and Lazarus (1985) Ways of Coping Questionnaire which

26
was based on their transactional model of stress and coping. Rational coping refers
to task oriented coping methods such as working out a plan for dealing with what
has happened. Emotional coping refers to attempts at repressing (or expressing)
the emotions evoked due to the threat. Avoidance coping involves avoiding action
or evoked emotions, and detached coping refers to feeling independent of the event
and the emotion associated with it.
Emotional and avoidance coping are considered maladaptive, and rational
and detached coping are considered to be adaptive styles of coping. Smith and
Sulsky (1995) investigated coping outcomes and found that increased use of
emotional and avoidance coping was associated with higher global job stress, more
physical symptoms, increased depression, and lower job satisfaction. This finding
was repeated over several samples.
It is a strongly held gender-role stereotype that women respond to stress
differently from men. That is, they respond passively and/or more emotionally to
stressful situations whether at work or elsewhere. Early studies (Billings & Moos,
1981; Folkman & Lazarus, 1980; Pearlin & Schooler, 1978) tended to suggest that
men used more rational coping compared with women. However, these studies
were flawed because men and women were not matched on occupational level or
on organizational power (Long & Kahn, 1993). Roger, Jarvis & Najaran (1993)
found a tendency for men to use more detached and rational and less emotional
coping styles than women, although the differences were not statistically significant.

27
Parasuraman and Cleek (1S84) and Tung (1980) found women managers were
more likely than men to use adaptive coping behaviors or superior coping. Long
and Kahn (1993) attribute these confusing gender differences to three causes.
First, women experience more and different work stressors than men (sex
discrimination and harassment). Next, gender role stereotypes affect male and
female perceptions of appropriate coping (such as disclosing to friends and
colleagues). Finally, women are disadvantaged with regard to coping if they lack
workplace coping resources such as power, perceived competence, or child care
and financial advantages.
In summary, when coping strategies fail, work loses meaning because
important values can no longer be attained from it. If this happens, since work
usually consumes a considerable part of ones daily living, life, itself, can lose
meaning. This is when burnout occurs.
Burnout
According to Schultz and Schultz (1990) More people are disabled today as
a result of stress than at any other time in our history (p. 539). In 1992, job stress
was identified as one of the 10 leading health risks in the United States (Quick,
Murphy, & Hurrell, 1992). It has been estimated that fully one third of US workers
are currently experiencing burnout or preburnout symptoms (Takooshian, 1994).
Burnout is the only psychological outcome variable that has been studied in the
veterinary population (Elkins & Elkins, 1987; Elkins & Kearney, 1992).

28
Maslach (1986), who was the first to coin the term burnout defines it as a
syndrome of emotional exhaustion, depersonalization, and reduced personal
accomplishment that can occur among individuals who work with people in some
(usually care-giving) capacity, often in emotionally charged situations. Many
sources of burnout have been proposed, with no single cause identified. Burnout
is viewed as the result of multiple causes, the impact of which differs according to
the personality of the worker. According to Maslach (1986) the sources of burnout
can be divided into three categories: personal (expectations, motivations,
personality), interpersonal (client contact, relationships with coworkers and
supervisors, outside relationships), and organizational (workload, bureaucracy,
feedback, work pressure). A key factor does seem to be the amount of client
contact experienced on the job. As time spent in direct client contact increases,
particularly if such contact is upsetting, frustrating, or difficult, burnout is more likely
to occur (Pines & Aronson, 1988).
Two studies have been conducted on burnout in veterinarians. Both used
Pines and Aronsons self-diagnosis burnout scale (Pines & Aronson, 1988). A mail
survey of 1000 veterinarians in 1981 yielded a 33% response rate, of which 91%
of the respondents were male (Elkins & Elkins, 1987). Fifty-three percent of the
respondents reported scores over 3 (mean = 3.09), the point at which clear-cut
signs of the burnout syndrome are said to occur. The youngest age group (25-29)
reported significantly higher burnout scores, as did the veterinarians in practice for

29
10-14 years. A high correlation between reported stress and burnout score was
observed, with client relations being the most stressful aspect of practice cited.
Solo owners of multiveterinarian practices had significantly higher burnout scores
than employees or veterinarians owning solo practices.
The Elkins and Elkins (1987) study was replicated in 1992 in a mail survey
of 1000 female veterinarians. (Elkins & Kearney, 1992). Interestingly, this survey
yielded a 61% response rate, using only one mailing as in the previous survey. The
female respondents reported a burnout score mean of 3.41, with 67% scoring over
3. There was a very strong positive correlation between burnout and hours worked
per week and burnout and stress. A strong negative correlation was evident
between job satisfaction and burnout. Female practice owners had the lowest
burnout scores, although most were owners of solo practices rather than multi
veterinarian practices.
The female veterinarians did exhibit higher burnout scores than did their male
counterparts. Further, a greater percentage of women than men scored in the
clinical range. Conclusions cannot be drawn from such direct comparisons since the
studies were widely spaced in time. Lowman (1993) concludes after reviewing the
general work literature that no inherent gender differences have been found in
research on job-related burnout, but that there may well be differences in the
causes of burnout for men and women and in their coping strategies for handling
burnout symptoms.

30
In regards to the outcomes of job burnout, job satisfaction has received much
attention. As would be predicted, higher levels of burnout are linked to lower levels
of satisfaction. Symptomatically, people experiencing burnout express a desire to
spend less time working directly with clients and are more likely to be absent from
work (Maslach, 1986). Although it would be expected that burnout would readily
lead to a deterioration in the quality of client services provided, this area has not
been well researched. Jones (1981) did show that burnout was associated with
more on-the-job mistakes, more inhumanistic counseling practices, more
aggressive behavior toward clients, and more disciplinary action, by supervisors.
Job Satisfaction
Theories of job satisfaction were mentioned in Chapter 1. The fit
hypothesis suggests that the real meaning of work to an individual actually lies in
the congruence between the worker and his or her job (Locke, 1976). Herzberg,
Mausner, and Snyderman (1959) had previously argued that the features of work
itself are of preeminent importance, disregarding individual characteristics. Job
satisfaction is clearly a difficult concept to define accurately, but in general, some
amount of satisfaction exists when the costs of working are exceeded by the
benefits in a degree seen as acceptable by the worker (Fraser, 1983). Clearly, an
individuals level of job satisfaction lies on a continuum, which is subject to change
and is neither constant nor static.
It is noteworthy that much of the job satisfaction literature ignores the

31
concept of job stress while researchers of job stress typically ask their subjects at
least one question concerning their level of job satisfaction. Fraser (1983) believes
that the basic relationship between job stress and job satisfaction is illustrated by
a bell shaped curve, indicating that some degree of stress is optimum for the
achievement of maximal job satisfaction.
As discussed in Chapter 1, there has been a plethora of research concerning
gender differences in job satisfaction (Andrisani, 1978; Cassidy & Warren, 1991;
Martin & Hanson, 1985; Mottaz, 1986; Murray & Atkinson, 1981; Shapiro & Stern,
1975; Weaver, 1974; Zanna, Crosby, & Loewenstein, 1987; Mortimer, Finch, &
Maruyama, 1988). Although conflicting findings are evident, in general, men and
women do not differ significantly in reported levels of job satisfaction. This is
paradoxical, in that gender differences in work values have been more consistently
reported. Women, more consistently than men, have placed significant importance
on hours worked per week and amount of travel time to and from work (Quinn,
Staines, & McCullough, 1974). Presumably, this relates to compatibility with family
responsibilities. Miller (1990) showed that men exhibit more concern for leadership,
opportunities for advancement, and decision-making ability than do women,
whereas women find such occupational concerns as job pressure, work hours and
job exhaustion as more salient than do men. Similar findings were replicated
among physicians, where no gender differences in overall satisfaction were
reported. Yet predictors of job satisfaction varied between the sexes. Satisfaction

32
with time available for family and personal life was an important predictor for women
whereas work setting factors were most important to male physicians (Richardsen
& Burke, 1991). Similarly, Kevorkian and Tuel (1994) found that male and female
academic physicians had different priorities when it came to inducements to remain
in academic positions.
Three theories can be utilized to inform the study of gender differences in job
satisfaction (Mason, 1995). The socialization view holds that women are socialized
into values, attitudes, and behaviors that are less individual in nature and more
communal than male values and behaviors, in direct contrast to this view, the
structural view suggests that any observed differences in job satisfaction are
attributable to factors which co-vary with gender and not to inherent gender
differences in socialization. The social role theory suggests that when social role
is more salient than gender role, as it is in managerial or other professional
positions, male and female attitudes, behaviors and values will be similar.
Impoverished opportunity structures (such as clerical or other blue-collar positions)
will evoke gender role salience, in which gender differences will be notable.
Masons (1995) study of nearly 15,000 workers, provided no evidence for the
socialization view, and most support for the structural view (although in some
instances the social role view was supported). She compared male and female
managers and male and female clerical workers using data collected over a two-
year period by a consulting firm. An important finding of her study, which contrasts

33
with what little has been observed in physicians, is that men and women in
management did not appear to differ from one another in their sources of
satisfaction at work.
The relationship between education and job satisfaction has been explored.
Early research suggested that increasing education might condition workers to
expect higher levels of extrinsic rewards (Mortimer, Finch, & Maruyama, 1988) or
intrinsic rewards (Seybolt, 1976). On the one hand, employees with higher
educational attainment were prepared for more challenging careers which could pay
higher salaries, so if this materialized, the workers should be more satisfied. On the
other hand, if workers were overeducated and underemployed (or underpaid) they
should be less satisfied.
One recent study reported that gender moderated the impact of education
on job satisfaction. Educational attainment that had not provided expected
levels of extrinsic rewards had a negative impact on job satisfaction of men but
not women (Glenn & Weaver, 1982). The authors hypothesized that this might
be due to wage earning status-women who are secondary wage earners may
place more emphasis on the intrinsic rewards of work. Martin and Shehan
(1989) confirmed this hypothesis. They found that the major source of the
education-satisfaction relationship was the result of educations connection to
higher levels of extrinsic reward only for the primary wage earning women. For
secondary wage earning women, intrinsic rewards were the important

34
component of the relationship between schooling and job satisfaction. They also
found no evidence that education that doesnt lead to expected levels of
intrinsic and/or extrinsic reward reduces job satisfaction (p. 192).
Rationale and Hypotheses of the Present Study
The relationships among job stress, burnout and job satisfaction seem
relatively straightforward regardless of the types of confounding variables studied.
These relationships have not been confirmed in a veterinary population, however.
In addition, it is unclear (in any population) how the relationship is influenced by
gender and mediated by coping style, and further modified by age and wage
earning status.
Based upon the literature review, the following research hypotheses were
formulated.
1. Among equine veterinarians in private practice, job stress will be negatively
correlated with job satisfaction and positively correlated with burnout.
2. Among equine veterinarians in private practice, there will be no gender
differences in the overall level of job stress, job satisfaction, and burnout reported.
3. There will be gender differences in the frequency and severity of specific job
stressors reported by equine veterinarians in private practice.
4. There will be no gender differences in coping styles reported by equine
veterinarians in private practice.
5. The relationship between job stress and burnout will be mediated by coping style

35
in equine veterinarians private practice. Specifically, problem-focused and
detached coping styles will be associated with a weaker relationship between job
stress and burnout while avoidant and emotion-focused coping will be associated
with a stronger relationship between job stress and burnout.

CHAPTER 3
METHOD
Participants
Potential participants were 968 (503 men and 465 women) licensed
veterinarians in private, predominantly equine practice who were selected from the
1997 membership directory of the American Association of Equine Practitioners
(AAEP). The AAEP has approximately 5600 national and international members
of which approximately 25% are female. Participants were eligible for selection for
the study as long as they were currently practicing, were in private practice (vs.
academia, government, or business), lived in the United States, described
themselves as working in a practice which is >50% equine, and had a known
gender.
Four-hundred ninety-two (492) surveys were returned, resulting in a final
response rate of 51%. Twenty-seven surveys arrived after the cut-off date of June
13, and were not included in the study. Eight surveys were unidentifiable by gender
and were also not included. Thus, 457 surveys were usable. Of the usable surveys
45% were from men (n=205) and 55% were from women (n=252).
36

37
The overall response raie from the males was 43% and the female response
rate was 57%. Eighty-nine percent of the actual study participants (n=407) ranged
in age from 30-59 although the age range of all participants was from 25 to over
70 years. The participants were distributed nearly equally in terms of years in the
profession of veterinary medicine: 20.8% (n= 95) had been practicing for five years
or less, 19.7% (n=90) between six and 10 years, 23.9% (n=109 ) between 11 and
15 years, 11.8% (n=54) between 16 and 20 years, and 25% (n_=115) greater than
20 years. Thirty-six percent (n=164) had been working for five years or less at their
current job, 22.8% (n_=104) had been working from 6 to 10 years, 15.5% (m=71),
between 11 and 15 years, 6.6% (n=30) between 16 and 20 years, and 19.0%
(n=87) longer than 20 years. Sixty percent of the respondents (n=274) were sole
owners of a practice, 13.7% (n=63) were partners in a practice, and 26.2% (n=120)
were employees.
Five percent (n=23) of the respondents were board certified with the colleges
of Surgery (n=10 respondents), Theriogenology (n=7 respondents), Veterinary
Practice (n=3 respondents) and Acupuncture (n=3 respondents) being represented.
Only 12.3% (n=56) reported working 40 hours per week or less, with 28% (n=218)
of the practitioners averaging greater than 60 hours per week on the job. Fifteen
and one-half percent (n=71) annually earned less than $30,000, 28% (n=128)
earned between $30,000 and $49,999, 18.6% (n=85) earned between $50,000 and

38
$69:999, 10.9% (n= 50) earned between $70,000 and $89,999, 11.8% (n=54)
earned between $90,00 and $109,999 and 15.0% (n=69) earned over $110,000
per year.
Twenty-three percent of respondents (n=105) had never been married, 54%
(n-247) were currently married, 10.5% (n=48) were divorced and unmarried, and
11.6% (n~53) were divorced and remarried once. Sixty-eight percent (n=311)
described themselves as the sole wage earner, 13.6% (n=62) described themselves
as secondary wager earners, and 18.3% (n_= 84) described themselves as equal
wage earners in their families. Eighteen percent of respondents (n=84) had
between 1 and 3 children under the age of five in the household, 23.6% (n=108)
had between one and three children between the ages of six and 18 living at home,
and 5.9% (n=27) of the respondents had one or two adult children (ages 19-25)
living at home. Seventy-seven percent (n=350) had no children living at home. Six
percent (n-27) of respondents reported having children with significant emotional
(n=15), disciplinary (n_=10), or academic (n=15) problems.
Measures
The following four instruments comprised the major part of the mail survey
used in this study: (1) the Health Professions Stress Inventory (HPSI), (2) the
Coping Styles Questionnaire (CSQ), (3) the Burnout Measure (BM) and (4) Five
questions which formed a composite measure of job satisfaction (SAT). (See

39
Appendix A.)
The Health Professions Stress inventory (HPSI) (Wolfgang, 1988) consists
of 30 job situations which health providers might be expected to encounter and find
stressful. Respondents indicate how often they have found each situation to be
stressful in their work environment, using a scale of 0 to 4 (never to very often). The
HPSI has been proven to be both a valid and reliable measure of job stress in the
health professions (Wolfgang, 1988). Lyons (1971) Index of Work Related Tension
(IWRT)was used as a measure of concurrent validity for the HPSI. Pearson
product-moment correlations between scores on the HPSI and the IWRT were .75
for pharmacists, .76 for physicians, and .78 for nurses (p < .001) (Wolfgang, 1988).
Cronbachs alpha coefficients assessing internal consistency of the HPSI were .89,
.89, and .88 for pharmacists, nurses, and physicians, respectively.
The HPSI was modified slightly for the veterinary population in this research.
The word patient was changed to owner in three instances (see Appendix A).
In accord with the suggestion of Spielberger & Reheiser (1995), the perceived
intensity of each stressor was assessed along with its frequency. An overall job
stress index was based on the sum of the cross-products of the Severity and
Frequency scores. The overall index (HPSI-T) was compared with the original
index (HPSI) (which is hereafter referred to as the frequency index (HPSI-F)) in
order to ascertain the necessity of using both intensity and frequency scales. It

40
was concluded that use of both scales enhanced the understanding of gender
differences in job stress. Comparisons with the scores reported in the literature for
other health professionals (physicians, pharmacists, and nurses) were made based
on the frequency index (HPSI-F) alone. Gender differences were assessed on
each individual item, as well as on the overall, intensity and frequency scores. In
addition, three stressors taken from Simpson and Grants Measurement of Job
Stressors (Simpson & Grant, 1991) were added (Having owners whose values and
beliefs about medical care differ greatly from your own; Problems related to
business and management aspects of veterinary practice; and Having to make
difficult ethical and moral
choices.) Medical practice was changed to veterinary practice in one instance
and owner was substituted for patient in one instance. One additional item
pertinent to the practice of equine medicine (Worrying about the possibility of
zoonotic illness or injury to yourself or your staff) was also added.
Because the HPSI problematic (HPSI-P) scale was conceived in addition
to the typically used HPSI frequency (HPSI-F) scale, Cronbachs alpha was
computed to assure internal reliability. Analysis of the responses to the HPSI-F
scale indicated adequate internal reliability (Cronbachs alpha = .89) as did analysis
of the responses to the HPSI-P scale (Cronbachs alpha = .91).
The Coping Styles Questionnaire (CSQ) was used to assess individual

41
coping style in the participants. The CSQ was developed by Roger, Jarvis, and
Najaran in 1993. Responses to 78 items (chosen by the authors from a variety of
sources) were subjected to principal factoring and rotated to a four-factor terminal
solution following a scree test. The CSQ utilizes the 60 items with the highest
loadings. Each item exemplifies ways in which individuals react to stressful
situations. Respondents are asked to indicate whether each item illustrates a way
they react to stress by circling always, often, sometimes, or never for each
item. A mean score was calculated for each of the following four coping styles:
rational (RAT), detached (DET), emotional (EM), and avoidance (AV). The CSQ
has test-retest reliabilities of .80, .79, .76, and .70 for each scale respectively, and
it has coefficient alphas of .85, .90, .74 and .69 for each scale respectively.
Because of the excessive length of the original questionnaire, only items which had
factor loadings of .5 or higher in the original research (Roger, Jarvis, & Najaran,
1993) were used in this study. This procedure cut the CSQ to 30 items and is in
accord with accepted means of decreasing lengthy questionnaires while maintaining
reliability of measures (Zyzanski, 1993).
To investigate the validity of the Coping Styles Questionnaire as used in this
study (thirty items vs. the original sixty), the responses to the thirty items were
subjected to principal factoring, and rotated to a four-factor terminal solution
following a scree test. With the exception of five items, all loaded on each of four

42
scales which were identical to those identified by the developers of the scale
(Roger, Jarvis, & Najaran, 1993). One of the five items which appeared initially on
the rational coping scale, loaded slightly more heavily on the detached scale. Four
(of a total of 6 ) items intended to measure avoidance coping loaded on other
scales. Thus, this scale was not considered to be valid and is not discussed further.
Burnout was measured via the Burnout Measure (Pines & Aronson, 1988;
Schaufeli, Maslach, & Marek, 1993). Twenty-one feelings are listed and the
participant is asked to relate how often he or she experiences such feelings. A
seven point scale is used with choices ranging from 1 (never) to 7 (always). A
single score is derived by adding the points attained on the 17 negative feeling
items. This is followed by adding the scores for the four positive items and
subtracting this from the number 32. This figure is then added to the negative
points attained and the sum is then divided by 21. The scale yields scores of 1 to
7. A score (BOS) of 3 or higher is consistent with some degree of burnout (Pines
& Aronson, 1988). The burnout inventory has been shown to be internally
consistent (alpha = .92). Test-rettest reliabilities range from .66 to .89 (highest
correlations were found when the testing interval was 4 weeks and lower
correlations were found when the interval was 4 months). Takooshian (1994)
ascribes excellent construct and criterion validity to the instrument as well.

43
Five job satisfaction questions were used following the lead of Martin &
Shehan, 1989. Job satisfaction (SAT) was a composite measure of the responses
to the five items (1=strongly disagree to 5=strongly agree). The specific items
included were as follows: (1) am satisfied with my current job, (2) I cannot think
of jobs for which I would exchange, (3) If had the choice to make again, I would
choose the veterinary profession, (4) I do not intend to leave this job anytime in the
near future, and (5)This job measures up to my initial expectations. The use of five,
unweighted items remedies the inherent shortcomings of relying on a single-item
measure of the construct (Martin & Shehan, 1989). Values for the scale of
satisfaction range from 1.0 (low satisfaction) to 5.0 (high satisfaction). The job
satisfaction scale exhibited adequate reliability (Cronbachs alpha = .85).
In addition to the four above described instruments, the mail survey included
a short demographics questionnaire (DQ) and a cover letter. The DQ asked
respondents to provide information about their age, gender, years in the profession
and in the current job, income, hours worked per week, etc. They were also asked
to describe their family wage earning status, marital status and number of children.
Finally, they were asked to rate their perceived overall life stress for the past year
(see Appendix A). The cover letter (see Appendix B) made an appeal to colleagues
for assistance. In an attempt to gain maximal cooperation, the letter was typed on
AAEP letterhead stationery and was signed by the marketing director of the AAEP

44
and the principal researcher.
Procedure
The four-page mail survey along with a cover letter was sent by first-class
mail to 968 (503 male and 465 female) veterinarians currently in private,
predominantly equine practice selected by a stratified random procedure from the
AAEP membership list. It was intended that a stratified random sampling procedure
would be used to identify the potential participants so that a fairly equal number of
potential male participants and female participants would be identified. However,
only 465 female members who met the established criteria were identified. Thus,
ail of these women were included, and a similar number of men meeting all the
criteria were randomly chosen.
The survey was mailed in AAEP outer envelopes with a business reply
return envelope included. A target date for the receipt of all returned surveys (May
26, 1997) was listed in the survey cover letter. No surveys were returned as
undeliverable. Of the 25 surveys which were returned after the deadline and were
thus not included in the study, 14 were from women and 11 were from men.
Perusal of these surveys revealed no obvious response differences from the
surveys which were included in the study.
A reminder, an individually addressed and signed postcard, was mailed first-
class to all participants one week following the initial mailout of surveys. The

45
postcard personally appealed for help for the researcher and the organization In
their efforts to get the surveys completed and returned to them. The postcard was
signed with a blue pen by the researcher (see Appendix C). The study was
terminated within four weeks of the initial mailing, and the data were double key
punched by qualified operators.

CHAPTER 4
RESULTS
Descriptive Statistics
The composite mean Health Professions Stress Inventory score (HPSI-T)
was 7.01 + 2.7, with a range of 1.09 to 16.77. The range of the scale for
responding to items on the HPSI-T was 1.0 (not at all stressful) to 25.0 (extremely
stressful). The mean score of the male respondents was 6.2 + 2.4 and the mean
score of the women was 7.6+ 2.7.
The composite mean Burnout Score (BOS) was 3.29 0.8, with a range of
1,1 to 6.0. The response scale for the BOS yields scores of 1 to 7. A score of 1 to
2 indicates low burnout; a score of 2 to 3 indicates acceptable levels of stress arid
manageable level of work dissatisfaction; a score of 3-4 indicates the beginning of
the burnout syndrome; a score of 4-5 indicates a state of clinical burnout; and
scores over 5 indicate severe problems (Pines & Aronson, 1988). The mean score
of the male respondents was 3.07 + 0.8 and the mean score of the female
respondents was 3.47 + 0.8.
The composite mean job satisfaction score (SAT) was 3.70 + 1.0, with a
range of 1-5. The lowest score (1) indicates extreme dissatisfaction and the highest
score (5) indicates extreme satisfaction. The mean job satisfaction score of the men

47
was 3.72 1.0 and of the women was 3.68 1.0.
Data Analysis
The Statistical Analysis System (SAS) General Linear Model (GLM)
procedure was used for all the analyses in this study. The final sample size (n=457)
was sufficient to insure a 5% sampling error on a population of approximately 5000
(Saiant & Dillman, 1994) when the population is expected to be relatively varied in
their responses. Although a power analysis was not performed, according to Cohen
(1992) the group sizes (males = 205; females = 252) were sufficiently large to detect
medium or large effect sizes using means, correlations, chi-squared analyses,
analyses of variance (ANOVA's) and multiple regression analyses at the alpha level
of .01. (Minimum numbers needed to achieve the above would have been 95 males
and 95 females).
Hypothesis 1
To investigate Hypothesis 1, that among equine veterinarians in private
practice, job stress will be negatively correlated with job satisfaction and positively
correlated with burnout, Pearson correlation coefficients were calculated between
the scores on the Health Professions Stress Inventory (HPSI-T), the scores on the
burnout measure (BOS), and the scores on the job satisfaction measure (SAT).
Results revealed a moderate positive correlation between job stress and burnout
(r = .54; p<.0001) and a moderate negative correlation between job stress and job

48
satisfaction (r = -.39; p< .0001). Therefore, Hypothesis 1 was supported.
Hypothesis 2
Hypothesis 2 stated that among equine veterinarians in private practice,
there will be no gender differences in the overall level of job stress, job satisfaction,
and burnout experienced. To test this hypothesis, several planned, preliminary
analyses were conducted. A preliminary multivariate analysis of variance
(MANOVA) and preliminary chi square analyses were conducted to determine if
there were gender differences in other independent variables that might be
confounds in an examination of gender differences. (These independent variables
became dependent variables in the preliminary analyses). The dependent variables
in the MANOVA were age, years in the profession, years in the current job, income,
hours worked per week, number of veterinarians in the practice, number of children
in the home, and perceived life stress. The results of the preliminary MANOVA
(Wilks Lambda) revealed significant main effects for gender in at least one of the
dependent variables F(12, 414) = 22.5, £ = .0001. The models of six of the seven
follow-up univariate analyses of variance (ANOVAs) were significant for gender
differences: age F(1, 426) = 101.8, £ = .0001; years in the profession F(1, 426) =
135.7, ^ = .0001; years in the current job F(1, 426) = 131.2, £ = .0001; income F(1,
426) = 121.6, g = .0001; hours worked per week F(1, 426) = 18.4, £ = .0001;
number of children living at home F(1, 426) = 28, £ = .0001; and perceived life

49
stress F(1,426) = 6.06, £ = .01.
Next, as part of the preliminary analyses, Pearson correlation coefficients
were computed to determine the correlation between the continuous variables (now
viewed as independent, confounding variables for which gender differences were
found) and job stress (HPSI-T), job satisfaction (SAT) and burnout (BOS). No
variable had a correlation which was high enough to potentially account for a
significant amount of the variance in the scores. (See Table 4.1).
Table 4.1
Pearson correlation Coefficients of HPSI-T, BOS and SAT scores with the
potentially confounding (independent) variables.
HPSI-T
SAT
BOS
Age
-0.29*
0.16**
-0.23*
Years in Profession
-0.34*
0.15**
-0.22*
Years in job
-0.33*
0.16**
-0.20*
Income
-0.26*
0.21*
-0.26*
Hours/Week
0.06
0.01
0.12***
Children <6yrs.
0.01
0.05
-0.07
Children 6-18
-0.09
0.04
-0.03
Children 19-25
-0.06
0.07
-0.07
Total Children Home
-0.09
0.08
-0.09
Life Stress
0.25*
-0.15**
0.41*
*£<.0001 **£<.001
***£< 01

50
Although none of the above correlations were high enough to suggest that
the investigated independent variables would have an impact on the analysis to test
Hypothesis 2, the three highest correlates with the HPSI-T, BOS and/or SAT scores
(i.e., perceived life stress, income, and years in the profession ) were entered as
covariates in the analysis to be certain that the gender differences found could not
be attributed to these covariates. Clearly, if these covariates proved to be
nonsignificant as expected, the others with lower correlations with the dependent
variables of interest would also be non-significant. Since age and years in the
current job were only siightly less correlated with the dependent variables than were
years in the profession, it was logical to use only one as a covariate (i.e., years in
the profession).
Preliminary chi-squared analyses were performed to determine if there were
gender differences in the independent variables that were categorical (i.e. marital
status, wage earning status, and board certification status, all of which are serving
as dependent variables in the chi-square analyses). Significant gender differences
were found for marital status (chi-square = 50.4, df = 5, £ = .001) and wage earning
status (chi-square = 39.67, df = 2, £ = .001) but not for board certification status.
Since gender differences were found in the above specified variables, three
MANOVAs were performed: one each with job satisfaction, job stress, and burnout
as the dependent variable, gender as the independent variable and marital status,

51
wage earning status, board certification status and their interactions as covariates.
No significant main effects were found for job satisfaction F (32,441) 1.62, p >.02,
or burnout F (32,441) = 1.46, p > .05. The main effect found in the model with job
stress as the dependent variable F (32,440), p > .0008 was attributed solely to the
independent variable, gender. Thus, it was not necessary to include marital status,
wage earning status, and board certification status in the analyses to test
Flypothesis 2.
Flypothesis 2 states that, there will be no gender differences in the overall
levels of job stress, job satisfaction and burnout in veterinarians in private, equine
practice. To examine this hypothesis, a MANCOVA was performed with the
measures of job stress, job satisfaction and burnout as the dependent variables,
gender as the independent variable, and respondents years in the profession,
income, and perceived life stress as the covariates.
The MANCOVA (Wilks Lambda) revealed significant main effects for years
in the profession F(3, 446) = 10.5, p = .0001, perceived life stress F (3, 446) = 29.1,
P = .0001, income (F3, 446) = 7.3, p = .0001, and gender F(3,446) = 6.09, p =
.0005. All three follow-up univariate ANCOVAs for the dependent variables (using
years in the profession, income, and life stress as co-variates) were significant: Job
stress F(4, 451) = 4.79, p = .03, Job satisfaction F(4,447) = 10.01, p = .0017, and
Burnout F(4, 451) = 5.52, p = .02. Results revealed significant gender differences

52
in job stress (£< .0001), job satisfaction (£ < .0001), and burnout (£ < .0001) in
equine veterinarians in private practice. These differences were shown not to be
accounted for by years in the profession, income, or perceived life stress. Table
4.2 shows the least squares means of the scores of male and female veterinarians
on the three measures of job stress, job satisfaction and burnout. As shown in this
table, the women, as compared to the men in this study, had significantly higher job
stress and burnout scores as well as significantly higher job satisfaction scores.
Thus, Hypothesis 2 was not supported. (It is noteworthy that if the variance in job
satisfaction due to income was not removed from the model, female veterinarians
were not significantly more satisfied with their jobs then were male veterinarians.)
Table 4.2
Least Squares Means comparing male and female scores on the HPSI-T, SAT,
and BOS.
Gender
HPSI-T
SAT
BOS
Female
7.16
3.87
3.38
Male
6.81
3.52
3.18
Because the covariate income was unexpectedly significant when it
appeared in the model which investigated gender differences in job satisfaction,

53
R-square procedures were run to be certain that the variables chosen for use as
covariates in the MANCOVA (i.e., life stress, income, years in the profession)
were the most appropriate covariates. Thus, three R-square procedures with
job stress, job satisfaction or burnout as the dependent variable in each and the
15 possible covariates and their combinations as the predictor variables in each
were run to determine the best reduced model for each of the dependent
variables. Results revealed that the covariates chosen (years in the profession,
income, and life stress) were indeed the most appropriate covariates for the
MANCOVA to test Hypothesis 2.
Hypothesis 3
Hypothesis 3 stated that gender differences in the severity and frequency of
specific job stressors would be found. Although gender differences in the overall
measure of job stress were examined using the covariates of number of years in the
profession, income, and perceived life stress, it was concluded that gender
differences in each individual stressor would be examined without using covariates.
Using seventy separate ANCOVAs would diminish the power of the statistic so
much so that potentially significant gender differences would not be identified.
Thus, Hypothesis 3 was tested using two-tailed T tests. Since a total of 35
stressors were to be compared on both severity and frequency of occurrence, a

54
particularly stringent test was necessary to avoid finding significant differences
which would occur due to chance alone. Adjustments were made, therefore, using
permutations of p, a more rigorous criteria than the Bonferronis adjustment.
Significant differences (p > .01) were found between the means of 20 of 70
items (16 of 35 on the severity scale and 4 of 35 on the frequency scale). In all of
these instances, the women found the particular problem to be significantly more
stressful than did the men. In fact, although not significant at the p = .01 level,
women found all but 6 of the 70 items to be more stressful than did the men. And,
the differences between the male and female scores on four of the six items which
the men ranked as more stressful were of the magnitude of .03 or less. Notably,
both sexes agreed relatively well on the frequency scale, but not well on the severity
scale. Thus, Hypothesis 3 was confirmed: there were gender differences in the
frequency and severity of specific job stressors reported by equine veterinarians in
private practice. See Table 4.3.
Hypothesis 4
To test Hypothesis 4 that there will be no gender differences in coping styles
reported by equine veterinarians, a MANCOVA was performed with the three valid
coping scales as the dependent variables, gender as the independent variable, and
years in the profession, perceived life stress and income as the covariates. These
covariates were included given that they were used as covariates in the analyses

55
to test Hypothesis 2 which also was concerned with gender differences.
The MANCOVA (Wilks Lambda) revealed significant main effects for years
in the profession F(4, 440) = 5.4, £ = .0003; perceived life stress F(4, 440) = 13.23,
£ .0001; income F(4, 440) = 2.96, £ = .020; and gender F(4,440) = 3.98, £ =
.0035. The models of all three of the follow-up univariate ANCOVAs for the specific
types of coping were significant: Rational coping F(3, 447) = 11.13, £ = .0001;
Emotional coping F(4, 447) = 22.17, £ = .0001; and Detached coping F(4, 447) =
23.89, £ ~ .0001. Table 4.4 lists the least squares means of each gender for each
coping style. These means show that men as compared to women used
significantly more rational and detached coping styles, while women as compared
to men used significantly more emotional coping. Thus, Hypothesis 4 was not
supported.
Table 4.4
Least Squares Means for Coping Styles of Male and Female Veterinarians*
Gender Rational Coping Emotional Coping Detached Coping
Females 1.96 3.20 2.55
Males 1.86 3.30 2.42
*L.ower scores signify higher usage of each particular styie of coping.

56
Hypothesis 5
Hypothesis 5 stated that the relationship between job stress and burnout will
be mediated by coping style in equine veterinarians in private practice and that
rational and detached coping styles will decrease the impact of job stress on
burnout while avoidant and emotional coping styles would increase the impact of job
stress on burnout. Recall that, due to the results of the factor analysis performed
on the Coping Styles Questionnaire (CSQ), avoidant coping was omitted from this
comparison.
To test this hypothesis, a multiple regression analysis was used with burnout
score (BOS) as the dependent variable and job stress (HPSI-T), frequency of use
of each of the three coping styles, and the interactions of job stress with each
coping style use frequency as the predictor variables. It was found that 67% of the
variance in BOS could be accounted for by job stress and coping styles use
frequency and the interactions between job stress and frequency of use of each of
the coping styles (F (9, 454) = 100.44, £ = .0001).
Typical of multiple regression analysis, the order in which each predictor
variable and its interaction is entered into the overall model will affect the outcome.
Thus, three separate multiple regression analyses with BOS as the dependent
variable in each and with one of the coping style use frequency variables, job stress
(HPSI-T) and their interactions as independent variables in each were performed

Table 4.3:
57
Means of Ratings of Health Professions Stress Inventory (HPSI) Items By
Gender
PROBLEMATIC
FREQUENT
Female Male
Female Male
Having so much work to do that everything cannot be done well
3.50**2.96
3.53
3.27
Experiencing conflicts with supervisor's and/or administrators
2.09** 1.51
1.90*
1.52
Feeling ultimately responsible for patient outcomes
3.46 3.26
3.58
3.55
Not receiving the respect or recognition that you deserve from the public
2.63* 2.24
2.45
2.27
Being uncertain about what to tell an owner about the patients condition
. 2.29 2.01
2.23
2.15
Caring for the emotional needs of owners
2.67 2.46
3.07* 2.66
Disagreeing with other professionals concerning the treatment of a patient 2.53* 2.23
2.30
2.22
Not having opportunities to share feelings and experiences with colleagues 2.56* 2.18
2.35
2.15
Experiencing conflicts with coworkers
2.24 1.92
2.12
1.99
Having job duties which conflict with family responsibilities
3.69 3.56
3.64
3.61
Allowing personal feelings/emotions to interfere with the care of patients
2.26** 1.88
2.07
1.93
Keeping up with new developments to maintain professional competence
2.86 2.60
2.63
2.38
Feeling that opportunities for advancement on the job are poor
2.43** 1.94
2.21
1.91
Trying to meet societys expectations for high quality veterinary care
2.29 2.13
2.15
2.12
Supervising the performance of coworkers
2.05 2.14
2.11
2.24
Dealing with difficult. owners
3.31* 2.97
2.83
2.86
Not being recognized as a true health professional by other health
2.27** 1.74
1.99** 1.63
professionals
Being inadequately p repared to meet the needs of patients
2.45** 1.92
2.04
1.82
Possessing inadequate information regarding a patients medical condition 2.65** 2.17
2.22
2.09
Not receiving adequate feedback on your job performance
2.48** 1.89
2.22** 1.86
Not having enough staff to adequately provide necessary services
2.95** 2.33
2.85*
2.41
Having nonhealth professionals determine the way you must practice
2.34 2.13
1.95
2.05
Not knowing what type of job performance is expected
1.85 1.56
1.65
1.53
**Perm-p > .01 *Perm-^> >.05

Table 4.3 (continued)
58
PROBLEMATIC FREQUENT
Female Male
Female Male
Being interrupted by phone calls or people while performing job duties
3.22**2.75
3.
55 3,i
Not being allowed to participate in making decisions about your job
1.96** 1.39
1.74*
1.39
Not being challenged by your work
1.95
1.75
1.80
1.87
Feeling that you are inadequately paid as a health professional
3.76**2.83
3.65**2.92
Caring for terminally ill patients
2.45**
12.00
2.26
2.01
Not being able to use your abilities to the fullest extent on the job
2.71
2.40
2.42
2.26
Fearing that a mistake will be made in the treatment of a patient
2.85**
: 2.17
2.49**2.13
Having owners whose values and beliefs about care differ greatly from
2.80*
2.43
2.38
2.39
yours
Problems related to business and management of veterinary practice
3.15
2.83
3.06
2.96
Worrying about the possibility of zoonotic illness or injury to self or staff
2.12**
1.74
1.96
1.78
Having to make difficult ethical or moral choices
2.41**
1.99
2.15
2.01
**Perm-p > .01
*Perm-p >.05

59
to further and more dearly address this research hypothesis.
The multiple regression model with burnout as measured by BOS as the
dependent variable and with job stress as measured by HPSI-T, frequency of use
of rational coping (RAT), and the interaction of (RAT) with HPSI-T as the predictor
variables was significant, F (3, 453) = 108.8, £ = .0001, r squared = 42%. In this
model, HPSI-T (t = 4.4, £ = .001) and rational coping (t = 4.8, £ = .001) but not the
interaction term were significant predictors of burnout.
The multiple regression model with BOS as the dependent variable and
HPSI-T, emotional coping (EM), and the interaction of EM with HPSI-T as the
predictor variables was significant, F (3, 454) = 272.5, £ = .0001, r squared = 65%.
In this model, HPSI-T (t=-2.0, £ = .05), EM (t = -11.9, £ = .0001), and the
interaction between HPSI-T and frequency of use of EM (t= 3.2, £ = .0015) were all
significant predictors of burnout. Data plots of the interaction between EM and
HPSI-T indicate that as frequency of use of EM increased, the position association
between stress and BOS strengthened.
The multiple regression model with BOS as the dependent variable and
HPSI-T, frequency of use of detached coping (DET) and the interaction between
HPSI-T arid frequency of use of DET as predictor variables was significant, F(3,
454) = 97.3, £ = .0001, r squared = 39%. In this model, HPSI-T (t= 2.6, £ = .009)
and frequency of use of DET (t = 3.5, £ = .0005) were significant predictors of

60
burnout. The interaction of job stress with frequency of use of detached coping style
was not significant.
Coping style was shown to have an impact on burnout in this study.
Specifically, higher usage of rational and detached coping styles was associated
with decreased amounts of burnout, while higher usage of emotional coping (as
well as job stress) was associated with higher amounts of burnout. Hypothesis 5,
however, suggested that coping style wouid mediate (amplify or moderate) the
impact of job stress and burnout. This proved to be true only for the frequency of
use of the emotional style of coping, which strengthened the positive association of
job stress and burnout.
Although Hypothesis 5 did not specify that gender differences would be
found, two post-hoc multiple regression analyses identical to the original analysis
for this hypothesis were performed on male and female data separately. Results
were identical except that in women, the interaction between emotional coping and
job stress was not a significant predictor of burnout. Rational and detached coping
neither amplified nor moderated the effect of job stress on burnout in males or
females; however, more frequent usage of emotional coping strengthened the
positive association between job stress and burnout in men but not in women.
Thus, Hypothesis 5 was only partially supported.

61
Summary of Results
As is readily predictable, among equine veterinarians in private practice, job
stress was found to be negatively correlated with job satisfaction and positively
correlated with burnout. Veterinarians who are women reported significantly more
job stress and more burnout than did men, while at the same time reporting
significantly higher job satisfaction. Although male and female equine practitioners
seemed to agree on the frequency with which most job stresses occurred, women
found 45.7% of these stressors significantly more stressful than did their male
counterparts. Male and female equine practitioners reported using various coping
styles with different frequencies: women used significantly more emotional coping
than did men, who used significantly more detached and rational coping than did
the females. Emotional coping style was the only style shown to interact (in men
only) with job stress so as to increase the positive association between job stress
and burnout.

CHAPTER 5
DISCUSSION
Four-hundred and ninety-two questionnaires were returned from an initial
sample size of 968. This represents a response rate of 51%, a rather remarkable
return rate following receipt of a single four-page questionnaire by a group of health
care professionals. Single survey mailouts to the general public are typically
followed by a 20% response rate (Bailey, 1982).
This response rate compares favorably with previous results obtained from
surveys of veterinarians. The American Association of Equine Practitioners typically
achieves response rates of between 20-30% from its membership with single
mailouts but no reminder postcard. A single mailout (of a considerably shorter
survey than the present survey) to 1000 female veterinarians (Elkins & Kearney,
1992) achieved a 61% response rate and a single mailout to 1000 mostly male
veterinarians achieved a response rate of 33% (Elkins & Elkins, 1987). Other health
care professionals are notoriously difficult to survey. One of the largest studies sent
over 3000 questionnaires to nurses, pharmacists and physicians and achieved a
42% response rate with a second, repeat mailing (Wolfgang, 1988).
The response rate of 51% following a single mailout is particularly
high in view of the fact that the survey was mailed in May, the month that
62

63
falls in the middle of the breeding, racing and showing seasons, the busiest times
in the lives of equine practitioners. So why might the response of these
veterinarians have been so high? It might be because the researcher was a
colleague, and a former academic equine practitioner whose name may have been
known to some of the participants. Further, the personal appeal made by the
researcher on the follow-up postcard was likely influential. This high response rate,
however, may have mostly been the result of interest in the topic of the research by
the target participants. Indeed, equine veterinarians might be interested in job
stress, coping and burnout in their profession--a profession noted for its stagnant,
relatively low wages, increasing debt burdens and long working hours (Bovee,
1997).
The response rate of the women (57%) was much higher than that of the
men (43%) in this study. This is in accord with the two previous surveys of
veterinarians cited (Elkins & Elkins, 1987; Elkins & Kearney, 1992). Why might
female veterinarians be so much more willing to respond than men? For one thing,
three recent research efforts have substantiated significant gaps in earning between
male and female veterinarians which cannot be explained by demographic variables
(Bird, 1992; Smith, 1996; Frazier & Howell, 1991). For another, the women in this
study were indeed more stressed on the job and more burned out than were the
men, and reported significantly higher levels of general life stress. Thus, these

64
topics may be more salient to female veterinarians.
As in other areas of the workforce, women equine practitioners were
significantly different from their male counterparts. Not surprisingly, there was only
one variable assessed in which women were not different from male practitioners,
and that was the size of the practice in which they worked. This is in contrast to a
previous survey (Smith, 1996) in which it was found that female veterinarians (who
cared for various species, not solely the horse) tended to work in smaller practices.
The significance of this disparity is that it has been shown (Smith, 1996) that
practitioners who work in larger hospitals (determined by the number of doctors)
tend to earn more than those who work at smaller clinics. The mean number of
equine veterinarians per practice in this survey was 1.7, which is typical of most
large animal, non-referral practices. In other words, most equine practitioners
(regardless of gender) work in small practices.
As in the previous studies cited (Bird, 1992; Smith, 1996), these female
practitioners were about 10 years younger and had practiced veterinary medicine
for 8 fewer years than the men. As noted in the introduction, this is because of the
late entry of women into the veterinary profession, which, until the 1990s was
entirely male-dominated. For example, in 1970, 5.3% of the profession was female;
this increased to 13.3% in 1980, and 22% in 1990 (Bird, 1992; Elkins & Kearney,
1992).
As in the previous studies, the women in this study worked about 10% fewer
hours per week than did the men. Equal percentages of men and women were
practice owners, although a greater percentage of women worked as associates in

65
practices. The average female in this study earned about $50,000 whereas the
average male reported earnings in the range of $70,000 to $90,000. This finding
has also been replicated previously. In the Bird (1992) study, a wage gap of $8000
was found between men and women which could not be explained by number of
degrees, board certification, years of experience, hours worked per year, or
ownership or partnership status. Smith (1996) found that even when earnings ratios
were adjusted for years of experience, number of hours worked per week, number
of patients seen per hour, type of practice, hospital size and geographical region,
female veterinarians earned only 65% of the salary of their male counterparts.
Hypothesis one stated that among equine veterinarians in private practice,
job stress is negatively correlated with job satisfaction and positively correlated with
burnout. This hypothesis was supported. The moderate correlations between job
stress and burnout (.54) and job stress and job satisfaction (-.39) are in line with
previous research completed on female veterinarians (r = .38 between stress and
burnout), pharmacists (r = -.46 between stress and satisfaction), and nurses (r = -
.47 between stress and satisfaction) (Elkins & Kearney, 1992; Wolfgang, 1995).
Hypothesis two stated that among equine veterinarians in private practice,
there are no gender differences in the overall amount of job stress reported, and the
degree of job satisfaction and burnout experienced. This hypothesis was not
supported. Indeed, there were significant gender differences in the amount of job
satisfaction reported, and the degree of burnout and job stress experienced. Female
equine practitioners experienced significantly more job stress and burnout than men
but reported significantly higher job satisfaction when the variance due to years in

66
the profession, life stress and income was taken into account.
The finding of significant gender differences in job satisfaction with women
more satisfied than men is not congruent with most of the recent research which
found no gender differences in overall job satisfaction (Cassidy & Warren, 1991;
Martin & Hanson, 1985; Mortimer, Finch, & Maruyama, 1988). However, it is
generally acknowledged that men and women most likely derive different amounts
of satisfaction from differing aspects of their work. Unfortunately, due to space
limitations the components of job satisfaction were not studied in this sample.
These components (which are rarely enumerated and studied) very likely do vary
with gender (as compared to a measure of overall job satisfaction). In general, in
the few studies which have looked at the various components of satisfaction,
women have defined job satisfaction by hours worked per week, travel time to and
from work, job pressures, work hours, job exhaustion, satisfaction with time with
their families, job complexity, and the equality of social relationships at work. Men,
on the other hand, find such factors as leadership opportunities, advancement
opportunities, ability to make decisions, work setting factors, autonomy, earning
potential, and prestige as the things which make jobs most satisfying to them.
It is intriguing that female equine practitioners (who earn less than their male
counterparts and indicated that they feel inadequately paid) report higher levels of
job satisfaction (or equal levels when dissatisfaction due to income is not removed
from the model) than do male equine practitioners in spite of feeling more burned
out and stressed than male equine practitioners. Apparently, some aspects of
veterinary medicine are more satisfying to women than to men. Again, this study

67
was unable to determine the specific nature of male and female veterinarians
satisfaction.
Contrary to Hypothesis two, female equine practitioners report being
significantly more stressed and burned out on the job than men. The literature on
job stress has produced inconsistent findings when assessing gender differences.
A large meta-analysis (Martocchio & OLeary, 1989) found no gender differences
in job stress. Jick and Mitz (1985) suggest that failure to identify gender-related
differences in the occupational stress literature may be due to sampling problems,
with men being over-represented in management positions and women in clerical
and service jobs. This study, in which the men and women held very similar
professional positions, may support the suggestion of Jick & Mitz (1985) that gender
related differences in occupational stress could possibly exist and might be found
in research that eliminates job position as a confounding variable when examining
these gender related differences.
It is actually not surprising that female veterinarians report significantly more
burnout than male veterinarians. A number of burnout studies which have looked
for gender differences report that women feel more burned out. Henning and
Jardim (cited by Pines & Aronson, 1988), based on their interviews of over 100
female executives, believe that women see a career as a source of personal growth
and self-fulfillment, and they seek in it the satisfaction of doing what they want to do.
These perceptions, according to the authors theory, may cause women to have
higher expectations of their career than men, and if these expectations are not met,
women may burn out. This did not prove to be true for the equine veterinarians in

68
the present research. When asked if their current job measured up to their initial
expectations both male and female responses to this question were positive and
were not significantly different from one another.
Role conflict is a common and logical explanation for female burnout. It has
frequently been shown that the more successful a woman manager is in her career,
the less successful she feels about her home life. For men, success at work has
not been related to success or failure at home (Pines & Aronson, 1988). In this
study, men and women reported equal amounts of stress due to job duties which
conflict with family responsibilities. This item was rated the most stressful Item for
men, and the second most stressful (behind inadequate pay) for women. Yet,
women reported significantly more life stress than did men. Nonetheless, even
when the variance due to life stress was removed, women still reported significantly
more job stress and burnout than did men.
Women probably do pay a higher personal price for their careers in terms of
family life, however. This was very much in evidence in this sample. Although the
average female in the study was in her thirties (the average male was in his forties),
46% of these women had never married or were divorced and currently single; this
compares to only 19.5% of the men. While clearly related to marital status, these
female veterinarians had far fewer children than did their male counterparts and yet
they report higher overall life stress.
Womens values may make ambition and competition secondary to satisfying
work relationships as well as fulfilling relationships at home. Good personal work
relationships have been found to be highly negatively correlated with burnout in

69
women but not in men in one study (Pines & Aronson, 1988). Since the average
male and female veterinarian in this study came from a very small practice, it can
be assumed that both men and women had few supportive work relationships. This
relative isolation may well have affected women more than men. In fact, women
rated Not having opportunities to share feelings and experiences with colleagues
as more problematic than did men.
Hypothesis three stated that there are gender differences in the frequency
and severity of specific job stressors reported by equine veterinarians in private
practice. This hypothesis was supported. Perhaps the most unexpected finding of
this study was the fact that female equine practitioners rated 64 of 70 job stressors
as more stressful or more frequent in occurrence than did their male colleagues.
The 20 items rated as significantly more stressful by women appeared to fall into
four categories: 1. conflicts 2. lack of respect, recognition, fair pay, and opportunity
3. inability to deliver optimal patient care, and 4. emotional issues.
Hypothesis four stated that there are no gender differences in coping styles
reported by equine veterinarians in private practice. This hypothesis was not
supported. Female equine practitioners used emotional coping methods more
frequently than did men while men used detached and rational coping methods
more frequently than did women. The literature on this matter has been
contradictory, although these findings further the observation of Parasuraman and
Cleek (1984) who found a tendency (but not a statistically significant difference) for
male managers to use more detached and rational and less emotional coping styles
than female managers. Long and Kahn (1993) attributed these confusing gender

70
differences to three causes: 1. women experience more and different work stressors
than do men. (This was perceived to be true by the female equine practitioners in
the present research) 2. women are disadvantaged with regard to coping if they lack
workplace coping resources such as power, perceived competence, and financial
advantages. (This was also shown to be true for the women in this sample.) 3.
gender role stereotypes (Hansen & OLeary, 1983; Deaux & Emswiller, 1977) affect
male and female perceptions of what is appropriate coping (such as disclosing to
friends and colleagues, for example).
Hypothesis five stated that the relationship between job stress and burnout
is mediated by coping style in equine veterinarians in private practice. This
hypothesis was only partially supported. Emotional coping increased the positive
association of job stress with burnout, but only in men. Neither rational nor
detached coping mediated the relationship between job stress and burnout. Thus,
this hypothesis was only partially supported. In women, no type of coping style
mediated the relationship between job stress and burnout.
The coping aspect of this research was based on Lazarus transactional
model of stress and coping which was discussed in Chapter Two. Lazarus early
work (1966) demonstrated that the way people evaluate an event (as it impacts
upon their well-being) and how' they cope with that event influences the existence
and intensity of stress. Thus, if we are interested in studying the impact of stress
in the work environment, we need to assess coping style as well. The present study
attempted to recognize the impact of stress and coping on the outcome measure,
burnout. It was hypothesized that method of coping would mediate the effect of job

71
stress on burnout, and specifically that rational and detached coping would
moderate the impact, while emotional coping would amplify the effect. This seems
to be a logical extension of Lazarus' theory, and is in agreement with the conception
(and findings) of Billings and Moos (1981).
In this study, no interaction between coping style and job stress on the
dependent variable, burnout, was observed with the exception of the interaction
between job stress and the emotional style of coping in men. Relatedly, Wolfgang
(1995) found no interaction between coping style and job stress when the
dependent variable was job dissatisfaction. The findings of the current research
were in contrast to what might be intuited and to what was hypothesized by Lazarus
and his colleagues. Lazarus and Launier (1978), noting the importance of coping
by itself, stated that the ways people cope with stress are even more important.
. than the frequency and severity of episodes of stress themselves (p. 308). In the
present study, it was seen that 67% of the variance in burnout was accounted for
by job stress, coping and their interactions.
Osipow and Davis (1988) conducted a study to evaluate the relationship of
coping resources to occupational stress and strain. By coincidence, the subjects
were 213 (of 600) veterinary students who responded to a mail questionnaire. The
researchers assessed the impact of four types of persona! coping resources
(recreation, self-care, social supports, and use of rational coping) on occupational
stress and strain. All types of coping resources were effective in reducing job
stress, with high levels of social support reducing the impact of all types of
stressors. Least in importance was cognitive/rational coping which moderated only

72
the strain of responsibility and role overload.
it is certainly conceivable that the lack of interaction found between job stress
and coping in this study (with the exception of emotional coping in men) may relate
to the fact that the above types of coping resources were not assessed.
Implications of the Research Findings
This study fails to confirm the findings of the past decade of job satisfaction
literature by revealing significant gender differences in job satisfaction among
veterinarians in private equine practice. (However, as noted earlier, many of these
studies did not control for variables which might have obscured such differences.)
Due to financial constraints and survey space limitations, the specific components
of job satisfaction in these equine veterinarians were not measured. The current
study corrected the usual methodological limitations of failing to consider the impact
of potentially confounding demographic variables when assessing gender
differences in the work place. Age, years in the profession, years in the current job,
marital status, life stress experienced, type of practice, board certification status,
number of children (and whether or not they experienced significant disciplinary,
academic or emotional problems), salary, family wage earning status, and practice
ownership were assessed for gender differences. Although many gender
differences in these variables were detected, none were shown to significantly
Impact the assessment of job stress, job satisfaction or burnout in this population
with the exception of income, which impacted job satisfaction.
This study found significant gender differences in job stress and stressors,
which is in contrast to three of four studies cited previously which were conducted

73
on physicians. This finding is a!so in contrast to much, of the current literature on job
stress in the general population as well, Linn, Yager, Cope and Leaks (1985)
observation of the inverse relationship between age and job stress was confirmed.
In the current study, years in the profession and years on the job exhibited a slightly
stronger inverse correlation with job stress than did age. As in their study, a positive
relationship was observed in the present study between age and job satisfaction,
but this was a weak correlation only.
The current research confirms Miller and Kirsch (1987) summary of their
review of the gender and coping literature which concluded that the different sexes
typically cope in different ways, in this study the styles of emotion-focused,
detached, and rational coping all yielded significant gender differences. Further, it
seems that these female veterinarians appraised threatening events as more
stressful than did male veterinarians.
This study profoundly illustrates the problem noted in Spielberger and
Reheisers (1995) observation that most commonly used job stress measures tend
to confound the frequency of occurrence of a stressful event with its perceived
severity. The HP-SI, as used in most of the health professions studies of stress,
merely asked about the frequency of occurrence of a particular job stressor. In this
study of veterinar ians, men and women generally agreed upon the frequency of job
stressor occurrence, but evaluated the severity of such occurrences very differently.
Had the HPSI not been expanded (by the inclusion of the problematic scale in
addition to the frequency scale), a great deal of information would have been lost,
and gender differences in assessment of the severity of job stressors would have

74
gone undetected.
Another finding of pertinence to survey researchers is the confirmation of the
observation of Calvert (1996) who suggested that a larger initial sample size with
a single follow-up postcard would result in a comparable response size (albeit a
lower response rate) to the mailing of a second questionnaire to non-respondents
Savings in manual labor, postage and printing costs, and response time were
considerable in this study, and a more than adequate response rate (51%) was
obtained.
A significant implication of this study for the practice of counseling and
psychotherapy is that frequent use of emotional coping by male practitioners
strengthened the relationship between job stress and burnout. Further, for the men
and women in this study, more frequent use of rational and detached coping
methods weakened the relationship between job stress and burnout. While many
times the constraints inherent in the work environment appear to limit the possibility
of constructive action by individuals, how the individual copes will have a huge
impact on how he or she feels. This study suggests that for male practitioners when
rational coping methods are difficult to enact, detached coping should be attempted,
and emotional coping should be avoided. It is clear that the veterinary school
curriculum and professional conferences would profit from the addition of seminars
which teach students and practitioners about coping styles, enable them to
understand how they personally cope with stress, and help them to change their
coping styles if necessary.
In terms of veterinary practice, this study makes it clear that female equine

75
veterinarians feel inadequately paid, and earn less than male equine practitioners.
These findings confirm the results of several other studies which have concluded
that female veterinarians earn considerably less than men even when a number of
demographic factors such as number of hours worked, years in the profession etc.
are considered (Bird, 1992; Frazier & Howell, 1991; Smith, 1996; Verdn, 1997).
Perhaps the women charge less for their services, see fewer patients per hour
[although Smith (1996) showed this not to be true of his sample], or work fewer
weeks per year. Maybe they give away more of their time to charity cases than do
male veterinarians. Possibly they do not negotiate as well as men when buying a
practice or buying equipment and drugs. Or, perhaps the general public
discriminates against female equine practitioners by preferring to work with male
veterinarians. It is clear that, for whatever reasons gender-related salary
differences persist, female practitioners find these differences to be very stressful,
and these salary differences need to be addressed and studied further by the
profession.
Limitations of This Study
There are several methodological limitations of this study which relate to the
fact that it was a mail survey study. A frequently cited problem of mail survey
research is the issue of non-response bias. Is it accurate to assume that the
respondents ideas uniformly represent those of the non-respondents? In this
study, a very good response rate (51%) was attained, comparable to that expected
from the genera! public following four mailings (Dillman, 1983)! If the respondents
were the veterinarians most likely to suffer from job stress, then the analysis

76
benefits from a having a rigorous sample, If the respondents were those who
found the time to respond because they were experiencing less stress, then
significant results would be less likely and again the analysis would be more
rigorous.
Barling, Kryl, and Bluen (1990) conducted an organizational survey and
hypothesized that the time individuals took to complete and return their
questionnaires would be associated with specific work attitudes, experiences and
individual differences. Their results revealed that none of the organizational factors
nor any demographic variable predicted when respondents would return their
questionnaires. In some survey research, it is assumed that late responders fairly
represent non-responders, in this study, while a rigorous review of the 25
questionnaires which were received beyond the due date was not conducted,
perusal of the responses revealed that these Individuals appeared to be no more
stressed, dissatisfied or burned out than their more promptly responsive colleagues.
A second methodological limitation of the study is related to history. If the
practitioners received the survey at the busiest time of their practice (which is
precisely v/hat occurred), they might report less satisfaction and more stress than
is typical for them. In general, these veterinarians were fairly unstressed and
satisfied, and if, indeed, this was their most stressful time of year, then it provided
the best opportunity to study their coping behavior.
Another limitation of mail survey research and thus this study is that
individuals may easily misread or misunderstand the questions and/or their intent.
The impact of this problem was reduced by first administering the survey to several

77
veterinarians and personally interviewing them after they completed the survey.
Areas of consistent confusion were subsequently clarified and edited in the survey
administered in the study.
In regard to coping, all researchers agree that coping is not a static process,
although a single questionnaire attempts to portray it as such. Coping styles may
change with situations or maturity. St is also likely that individual recollections of
how they actually cope might be incorrect or biased.
A significant limitation of the coping portion of this study is that only three
coping styles were assessed. Other positive means of dealing with life or work
stress ( arid therefore means of coping) should be assessed in future research.
These could include self-care (eating habits, vitamin and mineral ingestion,
meditation), recreation (appropriate amounts of exercising, time spent in enjoyable
family activities), and social supports (the degree to which a respondent feels and
requests support from co-workers, family, and friends.)
The generalizability of this study of equine practitioners to other veterinarians
is, of course, questionable. The respondents were all individuals who work
predominantly with horses. It is this authors feeling that of all private practice
situations, equine practice could be the most stressful. The equine veterinarian
spends a large part of the day, usually alone, driving in a truck from place to place.
He or she routinely must cover his or her own emergency duty, since emergency
clinics such as those commonly established for small animals do not typically exist.
These individuals work outside, in extreme temperatures, with animals capable of
killing them with a single blow to the head. But these doctors were not

78
overwhelmingly stressed, and again, use of this group may, therefore have made
for a more rigorous study, or at least one in which coping was more readily studied.
Conversely, equine veterinarians may have fewer client contact hours than
do small animal veterinarians who typically see between three and four patients and
their owners per hour, versus equine practitioners who may only average one to two
patients and their owners per hour. Since burnout relates highly to hours of direct
client contact, small animal practitioners may, therefore suffer from more burnout
than equine or other large animal practitioners.
it is definitely inappropriate to compare this population of private practitioners
with veterinarians who are not in private practice. This authors earlier work with
academic veterinarians (Welsch, 1991) did reveal gender differences in job
satisfaction, much like those which have been shown for other university professors
(Sandler, 1986; Olsen, 1991; Plascak-Craig & Bean, 1989; Etaugh, 1984).
Future Research
Future research in this area should compare veterinarians in other types of
practice with one another. These results could aid veterinary students making
decisions regarding further education and training by revealing which types of
practice are more stressful or more satisfying than others. The components of job
satisfaction should be studied in an effort to better compare male and female
practitioners. An important question which needs to be assessed is whether a
veterinarians job is important to him or her most as a source of personal fulfillment
or as a source of income. Another important consideration which was neglected in
this study is the impact of physical limitations (specifically female size) on job stress

79
in those who must handle large animals.
it truly does appear that a comprehensive understanding of why veterinarians
who are female earn significantly less than male veterinarians is needed, since the
women in this research study found the salary situation to be more stressful than
any other stressor assessed. This should not be a particularly difficult question to
answer if enough of the necessary information is obtained (e.g. hours worked per
week, weeks worked per year, patients seen per hour, years in the profession, etc.)
For future research on coping styles, the Coping Styles Questionnaire (CSQ)
of Roger, Jarvis, and Najaran (1993) was shown to be a reasonable, though
imperfect, instrument by the results obtained in this research. It can effectively be
shortened from 60 to 30 questions (as was done in this study). The avoidance
coping scale, which had the weakest coefficient alpha of the four scales in the
authors study (.69) (Roger, Jarvis, & Najaran, 1993) also had very poor factor
loadings in the present study. In this author's opinion, the face validity of the scale
is quite weak, and methods which people use to moderate the impact of job stress
on the mind or body (such as drug or alcohol usage, TV watching, exercising to an
extreme) should be included.
Further, the finding that the interaction between emotional coping and job
stress increases burnout in men but not women should be confirmed and then
studied further. For example, perhaps higher frequency of use of emotional coping
in male veterinarians (since it appears to be more of a female veterinarian coping
preference) lowers self-esteem in men more so than in women. Or perhaps the use
of emotional coping has a greater negative impact on employees or clients if used

80
by a male than if used by a female, thus leading to increased job stress.
Conclusions
It is clear from this study that equine practice is a satisfying profession for
both male and female practitioners and that these veterinarians do not exhibit high
job stress or burnout. Further, despite having higher job stress and burnout than
male practitioners (and lower income), the female practitioners studied were as
satisfied as their male counterparts. It appears that coping style had a significant
impact upon the difference in burnout noted between men and women. The quality
of the male practitioners work lives could most likely be significantly improved if
they could learn to use fewer emotional coping strategies and more rational and
detached coping strategies. In addition, it is likely that a more comprehensive
study of coping as described earlier could readily identify other ways in which male
and female veterinarians could improve the quality of their work lives
Since salary typically impacts job satisfaction, research into the
organizational and management factors, work pattern factors and other factors that
may be contributing to the lower salaries earned by these female practitioners could
be invaluable in understanding (and then correcting if possible) the income disparity
noted between male and female equine practitioners. Female equine practitioners
job satisfaction would then most likely increase even further.
in conclusion, further comprehension of the complex processes of job
satisfaction, stress, burnout and coping will have implications for the profession as
it continues to become feminized. A full understanding of these complicated issues
can prove important to professors and counselors of veterinarians as well as to the

SI
practitioners themselves. Such research and associated interventions may
ultimately insure long-term tenures and healthy quality of lives among equine
practitioners, especially women, who may indeed be satisfied with their career
choice but are at higher risk of stress and burnout.

APPENDIX A
SURVEY

Instructions: Please circle one number in each of the columns that best describes how
problematic each situation is and how frequently each condition occurs. Use the keys below.
How problematic is each condition for you (when it happens, even if rarely) in your current work situation?
l=not at ali problematic; 2=not very problematic; 3=somewhat problematic; 4= problematic; 5=very
problematic.
.How frequently did you experience stress ora each of these conditions in the past year?
l=never; 2=occasionally; 3=about once each month; 4=about once each week; 5=nearly every day
HOW PROBLEMATIC? FREQUENT?
Having so much work to do that everything cannot be done well 12345 12345
Experiencing conflicts with supervisors and/or administrators 12345 S 2345
Feeling ultimately responsible for patient outcomes 12345 12345
Not receiving the respect or recognition that you deserve from the public 12345 12345
Being uncertain about what to tell an owner about fee patients condition 12345 12 3 45
Caring for the emotional needs of owners 12345 12345
Disagreeing with other professionals concerning the treatment of a patient 1 2 345 1 2 345
Not having opportunities to share feelings and experiences with colleagues 12345 12345
Experiencing conflicts with coworkers 1 2345 12345
Having job duties which conflict with family responsibilities 1 2. 345 12345
Allowing personal feelings/emotions to interfere vrith the care of patients 12345 12345
Keeping up with new developments to maintain professional competence 1 2 3 4 5 12345
Feeling that opportunities for advancement or) the job are poor 12345 12345
Trying to meet societys expectations for high quality veterinary care 1234 5 12345
Supervising the performance of coworkers 1 2 345 12345
Dealing with difficult owners 12345 12345
Not being recognized as a true health professional by other health professional 12345 1 2345
Being inadequately prepared to meet the needs of patients 12345 1234 5
Possessing inadequate information regarding a patient's medical, condition 12345 12345
Not receiving adequate feedback on your job performance 12345 12345
Not having enough staff to adequately provide necessary services 12345 12345
Having nonhealth professionals determine the way you must practice 1 2345 12345
Not knowing what type of job performance is expected 12345 123 4 5
Being interrupted by phone calls or people while performing job duties 12345 12345
Not being allowed to participate in making decisions about your job 12345 12345
83

84
HOW PROBLEMATIC? FREQUENT?
Not. being challenged by your work
1 2 3 4 5
1 2 3 4 5
Feeling that you are inadequately paid as a health professional
1 2 3 4 5
1 2 3 4 5
Caring for terminally ill patients
1 2 3 4 5
1 2 3 4 5
Not being able to use your abilities to tire fullest extort on the job
1 2 3 4 5
1 2 3 45
Fearing that a mistake will be made in tre treatment of a patient
1 2 3 4 5
1 2 3 45
Having owners whose values and beliefs about care differ greatly from yours
1 2 3 4 5
1 2 3 4 5
Problems related to business and management of veterinary practice
1 2 3 4 5
1 2 3 4 5
Worrying about the possibility of zoonotic, illness or injury to self or staff
1 2 3 4 5
1 2 3 4 5
Having to make difficult ethical or moral choices
1 2 3 4 5
1 2 3 4 5
instructions: Circle the appropriate number beside each of the next five statements.
Strongly disagree
Strongly Agree
I am satisfied with my current job.
1 2
3
4
5
I cannot think of any jobs for which would exchange.
1 2
3
4
5
If I liad the choice to make a,gain, I would choose the veterinary profession.
1 2
3
4
5
I do NOT intend to lea ve this job anytime in the near future.
1 2
3
4
5
This job measures up to my initial expectations
1 2
3
4
5
Instructions: Although people may react in different ways to different situations, we all tend to
have a characteristic way of dealing with things which upset us. How would you describe the
way you typically react to the stress you encounter during the practice of veterinary medicine?
A=always 0=often S=somerimes N=never
See the situation for what it actually is and nothing more A O S N
See the problem as something separate from myself so I can deal with it A O S N
Become miserable or depressed A O S N
Feel that no-one understands A O S N
Become lonely or isolated A O S N
Take action to change things A O S N
Avoid family or friends in general A O S N
Feel helpless-heres nothing you can do about it A O S N
Try to find out more information, to help make a decision about things A O S N
Keep tilings to myself and not let others know how bad things are for me A O S N
Feel independent of the circumstances A O S N
Cry, or feel like crying A O S N
Pretend theres nothing the matter, even if people ask whats bothering me A O S N

85
A=always Q=ofte S=sometimes N=never
Get things into proportion-nothing is really that important A 0
Feel that time will sort things out A 0
Feel completely clear-h eaded about the whole thing A 0
Believe that. I can cope with most things with the minimum of fuss A 0
Try not to let my heart rule my head A 0
Try to find a logical way of explaining the problem A 0
Decide its useless to get upset: and just get on with things A 0
Feel worthless and unimportant A 0
Trust in fate-things have a way of working out for the best A 0
Use my past experience to try to deal with the situation A 0
Just take nothing personally A O
Criticize or blame myself A 0
Talk about it as little as possible A 0
Feel completely calm in the face of any adversity A 0
See the thing as a challenge that must be met A 0
Be realistic in my approach to the situation. A 0
Try to think about or do something else A O
Instructions: How often do you have any of the following experiences?
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
l=Never 2=Once in a great while 3=Rarely 4=Sometimes 5=Often 6=UsuaIly 7=Always
Feeling depressed
1 2
3
4
5
6
7
Ha ving a good day
1 2
3
4
5
6
7
Being physically exhausted
1 2
3
4
5
6
7
Being emotionally exhausted
1 2
3
4
5
6
7
Being tired
1 2
3
4
5
6
7
Being happy
1 2
3
4
5
6
7
Being '"'wiped out
1 2
3
4
5
6
7
Feeling burned out
1 2
3
4
5
6
7
Being unhappy
1 2
3
4
5
6
7
Feeling rundown
1 2
3
4
5
6
7

86
I=Never 2-Once in a great while 3=Rarely 4=Sometimes 5=0ften 6=Usually 7=Aiways
Feeling trapped
Feeling worthless
Being weary
Being troubled
Feeling disillusioned and resentful about people
reeling energetic
Feeling way
Feeling hopeless
Feeling rejected
Feeling optimistic
Feeling anxious
instructions: Please circle the appropriate answer below.
Please answer ail the items!
Age 25-29 30-39 40-49
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
This is a crucial part of the study.
50-59 60-69 7(H
Gender male female
Years in the profession of veterinary medicine
Years in the current job
0-5
6-10
11-15
16-20
>20
0-5
6-10
11-15
16-20
>20
Circle one: Practice owner(sole owner) Practice partner Employee
Yearly income <$30,000 30,00049,999 50,000-69,999 70,000-89,999 90,000-109,999 >110,000
Hours worked per week <20 2040 41-50 51-60 >60
Practice type small animal mixed animal large animal other (state type)
Number of veterinarians in practice 1 24 5-7 >7
Board certification? no yes (state type)
Marital status never married married divorced & single divorced & re-married remarried>l widowed
Family wage earning status: primary (or sole) wage earner secondary wage earner equal wage earner
Number of children living at your home under age 5 between 6-18 between 19-25
Have any of your childr en had (circle if applicable) emotional disciplinary significant academic problems?
On a scale of 1-10, how stressful would you rate your life (excluding your worklife) this past year ?
Not at all stressful Average for an American Very stressful
123 456789 10
END OF SURVEY.
PLEASE ENCLOSE IN ENVELOPE PROVIDED.

APPENDIX B
COVER LETTER
May 9, 1997
Dear Colleague:
! am a former large animal internist (Ohio 80), currently in part-time private
practice while completing my Ph.D. in Counseling Psychology at the University of
Florida. Throughout my 17 years in veterinary medicines I have observed that some
veterinarians are quite happy with our profession and others are dissatisfied. With the
help of the AAEP, S am hoping to begin to understand what veterinarians find stressful
about their jobs, how satisfied they are with the profession as compared to other health
care professionals, and what might distinguish those who are satisfied from those who
are dissatisfied with veterinary medicine.
My funding is very limited, so I urge you to please take the time to fill out the
next four pages. In pilot studies with small animal' practitioners, the survey took an
average of 12 minutes to complete. Your response is very important to the success
of this study, as a high return rate is crucial for the results to be meaningful.
This is a completely ANONYMOUS questionnaire. In no way can you be
identified. Your completion of the questionnaire constitutes your agreement to allow the
University of Florida to use the results for teaching and research purposes. You need
not answer any question you do not wish to answer. I would gladly answer any
questions, as would the Project Supervisor, Dr. Carolyn Tucker. You can reach me at
352-376-1790 (or e-mail bbwelsch@aol.com) and Dr. Tucker at 352-392-1532. I
would welcome any written comments, either via e-mail or returned with the survey.
The results will be presented in December, 1997 at the Phoenix meeting of the AAEP.
Thank you in advance for helping a colleague in need, and for contributing to an
understanding of our changing profession!
Sincerely,
Barbara Brewer Welsch, DVM, EdS Ed Blach, DVM, MS, MBA
ACVIM, ACVECC AAEP Markefview
PLEASE REPLY BY MAY 26, 1997
87

APPENDIX C
FOLLOW UP POSTCARD
May 22, 1997
Last week an AAEP/University of Florida survey
addressing issues of equine practice was mailed to you.
Your name was randomly selected to represent the
membership.
If you have already completed and returned the
survey to us, please accept our sincere thanks. If not,
please do so today. Because it has been sent to a
small but representative sample of AAEP members, it is
extremely important that yours also be included in the
study if the results are to accurately represent the
responses of equine practitioners.
Thank you so very much!
Barbara Brewer Welsch, DVM
Project Director
88

REFERENCES
Andrisani, P. (1978). Job satisfaction among working women. Signs, 3,
558-607.
Bailey, K.D. (1982). Methods of social research. New York: The Free
Press, Macmillan.
Bandura, A. (1977). Self-efficacy : Toward a unifying theory of behavioral
change. Psychological Review. 84, 191-215.
Barling, J., Kryl, I., & Bluen, S. (1990). Finally, all the questionnaires are
back: Bias in the time taken to return survey questionnaires. Personality and
individual differences. 11,177-180.
Barone, D., Caddy, G., Katell, A., Roselione, F., & Hamilton, R. (1988).
The Work Stress Inventory: Organizational stress and job risk. Educational anad
Psychological Measurement. 48, 141-154.
Beehr, T., & Schuler, R. (1980). Stress in organizations. In K. Rowland &
G. Ferris (Eds.). Personnel management (pp. 390-419). Boston: Allyn and
Bacon.
Ben-Porath, Y., & Tellegen, A. (1990). A place for traits in stress research.
Psychological Inquiry,1, 14-17.
Bickel, J. (1988). Women in medical education. A status report. New
England Journal of Medicine. 319. 1579-1584.
Billings, A., & Moos, R. (1981). The role of coping responses and social
resources in attenuating the stress of life events. Journal of Behavioral Medicine.
4YI39--157.
Bird, C. E. (1992). Comparing women's representation in three health
professions: Dentistry, medicine, and veterinary medicine. (Doctoral
dissertation, University of Illinois at Urbana-Champaign, 1992). Dissertation
89

90
Abstracts International. 53 (10A), 3678.
Bobula, J. (1980). Work patterns, practice characteristics, and
incomes of male and female physicians. Journal of Medical Education. 55. 826-
833.
Bovee, K.C. (1997). Stagnant wages, increasing debt burden: Profession
needs a pian. DVM: The News Magazine of Veterinary Medicine. 29(31.1.
Brief, A. P., & George, J.M. (1994). Psychological stress and the
workplace: A brief comment on Lazarus outlook. In P.L. Perrewe & R. Crandall
(Eds), Occupational stress: A handbook (pp, 15-191. Washington. DC: Taylor &
Francis. (Original work published 1991).
Calvert, P. (1996). Gender differences in clinician predictions of working
alliance with borderline personality disordered and posttraumatic stress
disordered clients (Doctoral Dissertation, University of Florida, 1996).
Dissertation Abstracts international, 57(1 OB), 6561.
Carr, P., Friedman, R., Moskowitz, M., & Kazis, L. (1993). Comparing the
status of women and men in academic medicine. Annals of Internal Medicine.
119. 908-913.
Cassidy, M., & Warren, B. (1991). Status consistency and work
satisfaction among professional and managerial women and men. Gender &
Sodetx^5(2), 193-206.
Clark, A. (1996). The real job market. Journal of the American Veterinary
Medical Association. 208, 860-862.
Cohen, J. (1992). A power primer. Psychological Bulletin. 112(1), 155-
159.
Costa, P. & McCrae, R. (1990). Personality: Another hidden factor in
stress research. Psychological Inquiry,1,22-24.
Cox, T. (1978). Stress. Baltimore, MD: University Park Press.
Dawis, R. V., & Lofquist, L. H. (1969). Adjustment to work. New York:
Appleton-Century-Crofts.

91
Dawis, R. V., & Lofquis, L. (1984). A psychological theory of work
adjustment. Minneapolis: University of Minnesota Press.
Deaux, K., & Emswiller, M. (1974). Explanations for successful
performance on sex-linked tasks: what is skill for the male is luck for the female.
Journal of Personality and Social Psychology. 29. 80-85.
Dial, T., Bickel, J., & Lewicki, A. (1989). Sex differences in rank attainment
among radiology and internal medicine Faculty. Academic Medicine. 64.
198-202.
Dillman, D. A. (1983). Mail and other self-administered questionnaires. In
P. Rossi, J. Wright, & A. Anderson (Eds.), Handbook of survey research (pp.
359-377). San Diego: Academic Press, Inc.
DiSalvo, B., Lubbers, C., Rossi, A., & Lewis, J. (1994). The impact of
gender on work-related stress. In P. L. Perrewe & R. Crandall (Eds.),
Occupational stress: A handbook (pp. 39-50). Washington, DC: Taylor & Francis.
Doherty, W., & Baldwin, C. (1985). Shifts and stability in locus of control
during the 1970s: Divergence of the sexes. Journal of Personality and Social
Psychology. 48, 1048-1053.
Elkins, A., & Elkins, J. (1987). Professional burnout among U.S.
veterinarians: How serious a problem? Veterinary Medicine. 82, 1245-1250.
Elkins, A., & Kearney, M. (1992). Professional burnout among female
veterinarians in the United States. Journal of the American Veterinary Medical
Association. 200. 604-608.
Etaugh, C. (1984, Fall). Women faculty and administrators in higher
education: Changes in their status since 1972. Journal of National Association
for Women Deans. Administrators, and Counselors, pp. 21-25.
Folkman, S., & Lazarus, R. (1980). An analysis of coping in a middle-
aged community sample. Journal of Helath and Social Behavior. 21, 219-239.
Folkman, S., & Lazarus, R. (1985). If it changes it must be a process:
Study of emotion and coping during three stages of a college examination.
Journal of Personality and Social Psychology, 48, 150-170.

92
Folkman, S., & Lazarus, R. (1988). Coping as a mediator of emotion.
Journal of Personality and Social Psychology. 54, 466-475.
F'raser, T. (1983). Human stress, work and job satisfaction. A critical
approach. Geneva, Switzerland: International Labour Office.
French, J.R., Caplan, R.D., & Van Harrison, R. (1982). The mechanisms
of job stress and strain. Chichester, England: Wiley.
Glenn, N., & Weaver, C. (1982). Further evidence on education and job
satisfaction. Social Forces. 61,46-55.
Gehrke, B. (1995). Employment, starting salaries, and educational
indebtedness of 1995 graduates of US veterinary medical colleges. Journal of
the American Veterinary Medical Association. 207. 1290-1291.
Gehrke, B. (1996). Gender redistribution in the veterinary medical
profession. Journal of the American Veterinary Medical Association. 208, 1254-
1255.
Goode, W. (1960). A theory of strain. American Sociological Review. 25,
483-496.
Hamilton, S., & Fagot, B. (1988). Chronic stress and coping styles: A
comparison of male and female undergraduates. Journal of Personality and
Social Psychology. 55, 819-823.
Hansen, R., & O'Leary, V. (1983). Actresses and actors: The effects of
sex on causal attributions. Basic and Applied Social Psychology .4. 209-230.
Harris, J.R. (1994). An examination of the transaction approach in
occupational stress research. In P.L. Perrewe & R. Crandall (Eds.), Occupational
stress: A Handbook (pp. 21-28). Washington, DC: Taylor and Francis. (Original
work published 1991).
Havlovic, S. & Keenan, J. (1995). Coping with work stress: The influence
of individual differences. In R. Crandall & P. Perrewe (Eds.), Occupational
stress: A handbook (pp. 179-192). Washington, DC: Taylor & Francis, Ltd.
Herzberg, F., Mausner, B., & Snyderman, B. (1959). The motivation to
work. New York: Wiley.

93
Holt, R R. (1982). Occupational stress. In L. Goldberger & S. Breznitz
(Eds.), Handbook of stress (pp. 419-444). New York: Free Press.
Hopkins, A. (1983). Work and job satisfaction in the public sector. Totowa.
NJ: Rowman & Allanheld.
Hoppock, R. (1935). Job satisfaction. New York: National Occupational
Conference, Harper.
Ivancevich, J. & Matteson, M. (1980). Stress and work: A managerial
perspective. Glenview. IL: Scott, Foresman.
Jackson, S., & Schuler, R. (1985). A meta-analysis and conceptual
critique of research on role ambiguity and role conflict in work settings.
Organizational Behavior and Human Decision Process. 36, 16-78.
Jick, T., & Mitz, L. (1985). Sex differences in work stress. Academy of
Management Review, 10. 408-420.
Jones, J.W. (1981). Diagnosing and treating staff burnout among health
professionals. In J.W. Jones (Ed.), The burnout syndrome (pp.107-126). Park
Ridge, IL: London House Press.
Kevorkian, C., & Tuel, S. (1994). Faculty departure from academic
programs of physical medicine and rehabilitation. American Journal of Physical
Medicine & Rehabilitation. 73. 378-386.
Lazarus, R. (1966). Psychological stress and the coping process. New
York: Oxford University Press.
Lazarus, R. (1991). Psychological stress in the workplace. Journal of
Social Behavior and Personality 6, 1-13.
Lazarus, R. S. (1995). Psychological stress in the workplace. In R.
Crandall & P. Perrewe (Eds.). Occupational stress, (pp. 3-19) Washington, DC:
Taylor and Francis.
Lazarus, R., & Folkman, S. (1984). Stress, appraisal, and coping. New
York: Springer.
Lazarus, R. S., & Launier, R. (1978). Stress-related transactions between
person and environment. In L. A. Pervin & M. Lewis (Eds.), Perspectives in

94
interactional psychology (pp. 287-327). New York: Plenum,
Lefcourt, H. (1982). Locus of control: Current trends in theory and
research. Hillsdale. NJ: Erlbaum.
Lenhart, S., & Evans, C. (1991). Sexual harassment and gender
discrimination: a primer for women physicians. Journal of the American Medical
Womens Association. 46. 77-82.
Lewis, J., Carson, D., Nace, E., Howard, B., & Barnhart, F. D. (1994). The
lives of female physicians. Texas Medicine. 90. 56-61.
Linn, L., Yager, J., Cope, D., & Leake, B. (1985). Health status, job
satisfaction, and life satisfaction among academic and clinical faculty. Journal of
the American Medical Association. 254, 2775-2782.
Locke, E. (1976). The nature and consequences of job satisfaction. In M.
D. Dunnette (Ed.), Handbook of industrial and organizational psychology (pp.
1297-1349). Chicago: Rand-McNally.
Locke, E. & Taylor, M. S. (1990). Stress, coping, and the meaning of
work. In A. Brief & W. Nord (Eds.). Meanings of occupational work (pp.135-170).
Lexington, MA: Lexington Books.
Long, B., & Kahn, S. (1993). A theoretical integration of women, work, and
coping. In B. Long & S. Kahn (Eds.), Women, work and coping (pp. 296-311).
Buffalo: McGill-Queens University Press.
Lowman, R. L. (1993). Counseling and psychotherapy of work
dysfunctions, Washington. DC: American Psychological Association.
Lyons, T. (1971). Role clarity, need for clarity, satisfaction, tension, and
withdrawal. Organizational Behavior and Human Performance, 6, 99-110.
Martin, J., & Hanson, S. (1985). Sex, family wage earning status, and
satisfaction with work. Work and occupations. 12. 91-105.
Martin, J. & Shehan, C. (1989). Education and job satisfaction. The
influences of gender, wage-earning status, and job values. Work and
Occupations. 16. 184-199.

95
Martocchio, J., & OLeary, A. (1989). Sex differences in occupational
stress: A meta-analytic review. Journal of Applied Psychology, 74. 495-501.
Maslach, C. (1986). Stress, burnout, and workaholism. In R. Kilburg, P.
Nathan, & R. Thoresan (Eds.), Professionals in distress. Issues, syndromes and
solutions in psychology, (pp. 51-75) Washington, DC: American Psychological
Association.
Mason, E. S, (1995). Gender differences in job satisfaction. Journal of
Social Psychology, 135, 143-151.
Matteson, M., & Ivancevich, J. (Eds.). (1987). Controlling work stress:
Effective human resource and management strategies. San Francisco: Jossey-
Bass.
Miller, J. (1990). Individual and occupational determinants of job
satisfaction: A focus on gender differences. Sociology of Work and
Occupations.7, 337-366.
Miller, S., & Kirsch, N. (1987). Sex differences in cognitive coping with
stress. In R. Barnett, L, Biener, & G. Baruch (Eds.), Gender and stress (pp. 278-
307). New York: Free Press, Macmillan, Inc.
Mobley, G., Jaret, C., Marsh, K., & Lim, Y. (1994). Mentoring, job
satisfaction, gender, and the legal profession. Sex Roles. 31, 79-98.
Mortimer, J., Finch, M., & Maruyama, G. (1988). Work experience and job
satisfaction: Variation by age and gender. In J. Mortimer & K. Borman (Eds.),
Work experience and psychological development through the life span (pp. 109-
155). Boulder: Westview Press.
Mottaz C. (1986). Gender differences in work satisfaction, work-related
rewards and values, and the determinants of work satisfaction. Fluman
Relations. 39. 359-370.
Murphy, L., & Hurrell, J. (1987). Stress measurement and management in
organizations: Development and current status. In A.Riley & S. Zaccaro (Eds.),
Occupational stress and organizational effectiveness (pp.29-51). New York:
Praeger.
Murray, M. & Atkinson, T. (1981). Gender differences in correlates of job
satisfaction. Canadian Journal of Behavioral Science. 13. 44-52.

96
Neil, C., & Snizek, W. (1988). Gender as a moderater of job satisfaction.
Work and Occupations. 15. 201-219.
Northwestern National Life. (1991). Employee burnout: Americas newest
epidemic. Minneapolis. MN: Author.
Numerof, R.E., & Abrams, M.N. (1984). Sources of stress among nurses:
An empirical investigation. Journal of Human Stress. 10. 88-100.
Olsen, D. (1991). Women and minority faculty job satisfaction: A structural
model examining the effect of professional role interests. Boston, MA:
Association for the study of Higher Education. (ERIC document Reproduction
Service No. ED 339 323)
Osipow, S., & Davis, A. (1988). The relationship of coping resources to
occupational stress and strain. Journal of Vocational Behavior. 32, 1-15.
Parasuraman, S., & Cleek, M. (1984). Coping behaviors and managers
affective reactions to role stressors. Journal of Vocational Behavior. 24. 179-193.
Parkes, K. (1990). Coping, negative affectivity, and the work environment:
Additive and interactive predictors of mental health. Journal of Applied
Psychology. 75.399-409.
Pearlin, L., & Schooler, C. (1978). The structure of coping. Journal of
Health and Social Behavior. 19, 2-21.
Perrewe, P.L. (Ed.). (1991). Handbook on job stress (Special issues).
Journal of Social Behavior and Personality,6(7).
Pines, A., & Aronson, E. (1988). Career burnout: Causes and cures, (pp.
113-130; p. 219). New York: The Free Press.
Pin, G., Mednick, M., Astin, H., Hall, C., Kenkel, M., Keita, G., Kohout, L.,
& Kelleher, J. (1996). The shifting gender composition of psychology. Trends and
implications for the discipline. American Psychologist, 51, 509-528.
Plascak-Craig, F., & Bean, J. (1989). Education faculty job satisfaction in
major research universities. Atlanta: Association for the Study of Higher
Education. (ERIC Document Reproduction Service No. ED 313 978)

97
Price, J., & Mueller, C. (1981). A causal model of turnover for nurses.
Academy of Management Journal. 24. 543-565.
Prichard, J. T. (1996). Dying too young. Journal of the American
Veterinary Medical Association. 208. 1788.
Quick, J.C., Murphy, L.R., & Hurrell, J.J. (Eds.). (1992) Stress & well
being at work: Assessments and interventions for occupational mental health.
Washington, DC: American Psychological Association.
Quinn, R., Staines, G., & McCullough, M. (1974). Job satisfaction: Is there
a trend? The 1977 quality of employment survey. Ann Arbor: Survey Research
Center, University of Michigan.
Repetti, R., Matthews, K., & Waldron, I. (1989). Employment and womens
health. American Psychologist, 44, 1394-1401.
Richardsen, A., & Burke, R. (1991). Occupational stress and job
satisfaction among physicians: Sex differences. Soc. Sci. Med., 33. 1179-1187.
Roger, D., Jarvis, G., & Najaran, B. (1993). Detachment and coping: The
construction and validation of a new scale for measuring coping strategies.
Personality and Individual Differences. 15. 619-626.
Russ, F., & McNeiliy, K. (1995). Links among satisfaction, commitment,
and turnover intentions: The moderating effect of experience, gender, and
performance. Journal of Business Research. 34. 57-65.
Salant, P., & Dillman, D. (1994). How to conduct vour own survey. New
York: John Wiley & Sons.
Sandler, B. (1986). The campus climate revisited: Chilly for women
faculty, administrators, and graduate students (Project on the Status and
Education of Women). Washington, DC: Association of American Colleges.
Schaufeli, W.B., Maslach, C., & Marek, T. (1993). The burnout measure.
Profession Burnout: Recent developments in theory and research, (pp. 205-207).
Washington, DC: Taylor and Francis.
Schneer, J., & Reitman, F. (1995). The impact of gender as managerial
careers unfold. Journal of Vocational Behavior, 47, 290-315.

98
Schuler, R. (1985). Integrative transactional process model of coping with
stress in organizations. In T. Beehr, & R. Bhagat (Eds.), Human stress and
cognition in organizations (pp. 347-374). New York: Wiley.
Schultz, D. P. & Schultz, S. E. (1990). Psychology and industry today (5th
ed.). New York: Macmillan.
Selye, H. (1950). Stress. Montreal: Acta.
Seybolt, J. (1977). Work satisfaction as a function of the person-
environment interaction. Organizational Behavior and Human Performance. 17.
66-75.
Shapiro, J., & Stern, L. (1975). Job satisfaction: Male and female,
professional and nonprofessional workers. Personnel Journal. 54. 388-389.
Simpson L, & Grant, L. (1991). Sources and magnitude of job stress
among physicians. Journal of Behavioral Medicine. 14, 27-41.
Smith, C. (1993). Evaluations of whats at stake and what I can do. In B.
Long & S. Kahn (Eds.), Women, work and coping (pp.238-265) Buffalo: McGill-
Queens University Press.
Smith, C. S., & Sulsky, L. (1995). An investigation of job-related coping
strategies across multiple stressors and samples. In L. Murphy, J. Hurrell, S.
Sauter, & G. Keita (Eds.), Job stress interventions. Washington, DC: American
Psychological Association.
Smith, D.M. (1996, June). The gender gap: Why do female doctors earn
less? Veterinary Economics, pp. 44-51.
Smith, D., & Plant, W. (1982). Sex differences in job satisfaction of
university professors. Journal of Applied Psychology, 67. 249-251.
Spielberger, C., & Reheiser, E. C. (1995). Measuring occupational stress:
The job stress survey. In R. Crandall & P. Perrewe (Eds.), Occupational stress
(pp. 51-69) Washington, DC: Taylor and Francis.
Summers, T., & DeCotiis, T. (1988). An investigation of sex differences in
job satisfaction. Sex Roles. 18. 679-689.

99
Super, D. (1942). The dynamics of vocational adjustment. New York:
Harper and Brothers.
Takooshlan, H. (1994). The burnout measure, Second Edition. In D.
Keyser & R. Sweetland (Eds.), Test critiques, Volume X (pp. 205-207) Austin,
TX: Pro.Ed.
Tung, R. (1980). Comparative analysis of occupational stress profiles of
male versus female administrators. Journal of Vocational Behavior. 17,344-355.
Weaver, C. (1974). Correlates of job satisfaction: Some evidence from the
national surveys. Academy of Management Journal. 17. 373-375.
Welsch, B. (1991). Gender differences in job satisfaction in veterinarians
in academic practice. Unpublished study presented at the 38th Annual American
Association of Equine Practitioners Convention, Orlando, FL.
Wolfgang, A. (1988). Job stress in the health professions: A study of
physicians, nurses, and pharmacists. Behavioral Medicine, 14, 43-47.
Wolfgang, A. (1995). Job stress, coping, and dissatisfaction in the health
professions: A comparison of nurses and pharmacists. In R. Crandall & P.
Perrewe (Eds.). Occupational stress: A handbook (pp. 193-204). Washington
DC: Taylor & Francis, Ltd.
Zanna, M., Crosby, F., & Loewenstein, G. (1987). Male reference groups
and discontent among female professionals. In B. Gutek & L. Larwood (Eds.),
Womens career development, (pp. 28-41). Newbury Park, CA: Sage
Publications.
Zyzanski, S. J. (1993). Cutting and pasting new measures from old. In M.
Stewart, F. Tudiver, M. Bass, E. Dunn, & P. Norton (Eds.), Tools for primary care
research (pp. 97-111) Newbury Park, CA: Sage Publications.

BIOGRAPHICAL SKETCH
Barbara (Dean) Brewer Welsch was born in Youngstown, Ohio, and grew
up in Pittsburgh, Pennsylvania. She obtained a B.A. in psychology with a
teaching certificate in French, general Science and mathematics from Marietta
College, Marietta, Ohio, in 1972. After teaching in the Columbus Public School
System for three years, she obtained an M.A. degree in science education from
the Ohio State University, followed by a Doctor of Veterinary Medicine degree,
also from the Ohio State University. Barbara interned in large animal medicine,
surgery and theriogenology at the University of Georgia and was a resident in
large animal medicine and neurology at the University of Florida where she later
obtained a faculty position. She is a Diplmate of both the American College of
Veterinary Internal Medicine and the American College of Veterinary Emergency
and Critical Care. She left the University of Florida as a tenured, Associate
Professor of Medicine after nearly 10 years and obtained an M.Ed. and Ed.S.
in student personnel in higher education from the Department of Counselor
Education at the University of Florida. Dr. Welsch will receive her Ph.D.
degree in counseling psychology from the Department of Psychology
100

at the University of Florida in August, 1998, following completion of her
predoctoral internship at the University of Florida Counseling Center. She
continues to work part-time in small animal veterinary practice in Gainesville,
Florida.
101

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
Doctor of Philoso
( 1
DhyT-
A/1
v nfnWtrL v
i 1 /
Professor of Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Ellen S. Amatea
Professor of Counselor Education
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree
Professor Emeritus of Psychology
f Doctor of Philosophy.
t 77'
-Ay
Harry A. Grter
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Dqctor of Philosophy.
/"PC
' Shae Graham Kosch
Professor of Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in^
quality, as a dissertation for the degree Doctor oj^Philosophy.
RobeffC. Ziller
Professor of Ps'

This dissertation was submitted to the Graduate Faculty of the
Department of Psychology in the College of Liberal Arts and Sciences and to
the Graduate School and was accepted as partial fulfillment of the requirements
for the degree of Doctor of Philosophy.
August 1998
Dean, Graduate School



17
health, coronary heart disease has been clearly linked to job stress along with the
Type A behavior pattern (Maslach, 1986). Gastric ulceration has less consistently
been linked to job stress as well. Mortality rates have been found to differ by
occupation, and although unproved, it is reasonable to implicate stress on the job
as one explanatory factor. In a recent letter to the editor of the Journal of the
American Veterinary Medical Association (Prichard, 1996), the author observed that
in the monthly obituary column there appeared to be a disproportionate number of
relatively young veterinarians compared to the obituaries listed in a stockbrokers
journal, in fact, when surveyed, fully 87% of female veterinarians considered their
jobs to be veiy' stressful (Elkins & Kearney, 1992).
Stress in the workplace may well be increasing. The proportion of workers
who reported feeling highly stressed on the job had more than doubled from 1985
to 1990 (Northwestern National Life, 1991). Sixty-nine percent of the 600 workers
surveyed in the Northwestern study reported that their productivity was reduced due
to high stress levels, and one in three say job stress is the single greatest stress
in their lives (1991, p. 2).
Job stress has been linked to impairments in psychological well-being and
implicated in reports of depression, dissatisfaction, anxiety, tension, lowered self
esteem, and alcohol and drug abuse (Maslach, 1986). Absenteeism, employee
turnover, decreased productivity and diminished client-relations are obvious costs
to employers (Spielberger & Reheiser, 1995). According to Matteson and


50
Although none of the above correlations were high enough to suggest that
the investigated independent variables would have an impact on the analysis to test
Hypothesis 2, the three highest correlates with the HPSI-T, BOS and/or SAT scores
(i.e., perceived life stress, income, and years in the profession ) were entered as
covariates in the analysis to be certain that the gender differences found could not
be attributed to these covariates. Clearly, if these covariates proved to be
nonsignificant as expected, the others with lower correlations with the dependent
variables of interest would also be non-significant. Since age and years in the
current job were only siightly less correlated with the dependent variables than were
years in the profession, it was logical to use only one as a covariate (i.e., years in
the profession).
Preliminary chi-squared analyses were performed to determine if there were
gender differences in the independent variables that were categorical (i.e. marital
status, wage earning status, and board certification status, all of which are serving
as dependent variables in the chi-square analyses). Significant gender differences
were found for marital status (chi-square = 50.4, df = 5, £ = .001) and wage earning
status (chi-square = 39.67, df = 2, £ = .001) but not for board certification status.
Since gender differences were found in the above specified variables, three
MANOVAs were performed: one each with job satisfaction, job stress, and burnout
as the dependent variable, gender as the independent variable and marital status,


44
and the principal researcher.
Procedure
The four-page mail survey along with a cover letter was sent by first-class
mail to 968 (503 male and 465 female) veterinarians currently in private,
predominantly equine practice selected by a stratified random procedure from the
AAEP membership list. It was intended that a stratified random sampling procedure
would be used to identify the potential participants so that a fairly equal number of
potential male participants and female participants would be identified. However,
only 465 female members who met the established criteria were identified. Thus,
ail of these women were included, and a similar number of men meeting all the
criteria were randomly chosen.
The survey was mailed in AAEP outer envelopes with a business reply
return envelope included. A target date for the receipt of all returned surveys (May
26, 1997) was listed in the survey cover letter. No surveys were returned as
undeliverable. Of the 25 surveys which were returned after the deadline and were
thus not included in the study, 14 were from women and 11 were from men.
Perusal of these surveys revealed no obvious response differences from the
surveys which were included in the study.
A reminder, an individually addressed and signed postcard, was mailed first-
class to all participants one week following the initial mailout of surveys. The


APPENDIX C
FOLLOW UP POSTCARD
May 22, 1997
Last week an AAEP/University of Florida survey
addressing issues of equine practice was mailed to you.
Your name was randomly selected to represent the
membership.
If you have already completed and returned the
survey to us, please accept our sincere thanks. If not,
please do so today. Because it has been sent to a
small but representative sample of AAEP members, it is
extremely important that yours also be included in the
study if the results are to accurately represent the
responses of equine practitioners.
Thank you so very much!
Barbara Brewer Welsch, DVM
Project Director
88


60
burnout. The interaction of job stress with frequency of use of detached coping style
was not significant.
Coping style was shown to have an impact on burnout in this study.
Specifically, higher usage of rational and detached coping styles was associated
with decreased amounts of burnout, while higher usage of emotional coping (as
well as job stress) was associated with higher amounts of burnout. Hypothesis 5,
however, suggested that coping style wouid mediate (amplify or moderate) the
impact of job stress and burnout. This proved to be true only for the frequency of
use of the emotional style of coping, which strengthened the positive association of
job stress and burnout.
Although Hypothesis 5 did not specify that gender differences would be
found, two post-hoc multiple regression analyses identical to the original analysis
for this hypothesis were performed on male and female data separately. Results
were identical except that in women, the interaction between emotional coping and
job stress was not a significant predictor of burnout. Rational and detached coping
neither amplified nor moderated the effect of job stress on burnout in males or
females; however, more frequent usage of emotional coping strengthened the
positive association between job stress and burnout in men but not in women.
Thus, Hypothesis 5 was only partially supported.


Table 4.3:
57
Means of Ratings of Health Professions Stress Inventory (HPSI) Items By
Gender
PROBLEMATIC
FREQUENT
Female Male
Female Male
Having so much work to do that everything cannot be done well
3.50**2.96
3.53
3.27
Experiencing conflicts with supervisor's and/or administrators
2.09** 1.51
1.90*
1.52
Feeling ultimately responsible for patient outcomes
3.46 3.26
3.58
3.55
Not receiving the respect or recognition that you deserve from the public
2.63* 2.24
2.45
2.27
Being uncertain about what to tell an owner about the patients condition
. 2.29 2.01
2.23
2.15
Caring for the emotional needs of owners
2.67 2.46
3.07* 2.66
Disagreeing with other professionals concerning the treatment of a patient 2.53* 2.23
2.30
2.22
Not having opportunities to share feelings and experiences with colleagues 2.56* 2.18
2.35
2.15
Experiencing conflicts with coworkers
2.24 1.92
2.12
1.99
Having job duties which conflict with family responsibilities
3.69 3.56
3.64
3.61
Allowing personal feelings/emotions to interfere with the care of patients
2.26** 1.88
2.07
1.93
Keeping up with new developments to maintain professional competence
2.86 2.60
2.63
2.38
Feeling that opportunities for advancement on the job are poor
2.43** 1.94
2.21
1.91
Trying to meet societys expectations for high quality veterinary care
2.29 2.13
2.15
2.12
Supervising the performance of coworkers
2.05 2.14
2.11
2.24
Dealing with difficult. owners
3.31* 2.97
2.83
2.86
Not being recognized as a true health professional by other health
2.27** 1.74
1.99** 1.63
professionals
Being inadequately p repared to meet the needs of patients
2.45** 1.92
2.04
1.82
Possessing inadequate information regarding a patients medical condition 2.65** 2.17
2.22
2.09
Not receiving adequate feedback on your job performance
2.48** 1.89
2.22** 1.86
Not having enough staff to adequately provide necessary services
2.95** 2.33
2.85*
2.41
Having nonhealth professionals determine the way you must practice
2.34 2.13
1.95
2.05
Not knowing what type of job performance is expected
1.85 1.56
1.65
1.53
**Perm-p > .01 *Perm-^> >.05


67
was unable to determine the specific nature of male and female veterinarians
satisfaction.
Contrary to Hypothesis two, female equine practitioners report being
significantly more stressed and burned out on the job than men. The literature on
job stress has produced inconsistent findings when assessing gender differences.
A large meta-analysis (Martocchio & OLeary, 1989) found no gender differences
in job stress. Jick and Mitz (1985) suggest that failure to identify gender-related
differences in the occupational stress literature may be due to sampling problems,
with men being over-represented in management positions and women in clerical
and service jobs. This study, in which the men and women held very similar
professional positions, may support the suggestion of Jick & Mitz (1985) that gender
related differences in occupational stress could possibly exist and might be found
in research that eliminates job position as a confounding variable when examining
these gender related differences.
It is actually not surprising that female veterinarians report significantly more
burnout than male veterinarians. A number of burnout studies which have looked
for gender differences report that women feel more burned out. Henning and
Jardim (cited by Pines & Aronson, 1988), based on their interviews of over 100
female executives, believe that women see a career as a source of personal growth
and self-fulfillment, and they seek in it the satisfaction of doing what they want to do.
These perceptions, according to the authors theory, may cause women to have
higher expectations of their career than men, and if these expectations are not met,
women may burn out. This did not prove to be true for the equine veterinarians in


15
modern concept of stress as a response by discovering that tissue damage could
result from diverse stimuli and was thus a nonspecific response to all noxious
stimuli. It is clear, however, that different stressors culminate in differing response
intensities. Although Selye was looking at purely physiological reactions, the
response definition of stress has grown to include the psychological reactions
exhibited by a person who is under some sort of pressure (Maslach, 1986). Neither
definition is capable of explaining individual differences, however (Cox, 1978).
An important advance in the thinking about stress was supplied by French,
Caplan, and Van Harrison (1982). Their ideas about person-environment fit
combine the separate causal variables of environment and person into an ongoing
relationship characterized by fit or misfit. Occupational stress is defined in terms of
job characteristics that pose a threat to the individual because of a poor match
between the abilities of the employee and the demands of the job. But this
approach falls short, because it is basically a static approach. It emphasizes stable
relationships between a person and the workplace rather than a process during
which stress changes over time and varies within similar contexts. It is clear that
even when a person and his or her environment are a good fit, some stress will
occur. In fact, most work researchers agree that some degree of stress is
beneficial.
Lazarus transactional process theory of stress (1995) distinguishes between
stressful antecedent conditions and how these are perceived and cognitively


80
by a male than if used by a female, thus leading to increased job stress.
Conclusions
It is clear from this study that equine practice is a satisfying profession for
both male and female practitioners and that these veterinarians do not exhibit high
job stress or burnout. Further, despite having higher job stress and burnout than
male practitioners (and lower income), the female practitioners studied were as
satisfied as their male counterparts. It appears that coping style had a significant
impact upon the difference in burnout noted between men and women. The quality
of the male practitioners work lives could most likely be significantly improved if
they could learn to use fewer emotional coping strategies and more rational and
detached coping strategies. In addition, it is likely that a more comprehensive
study of coping as described earlier could readily identify other ways in which male
and female veterinarians could improve the quality of their work lives
Since salary typically impacts job satisfaction, research into the
organizational and management factors, work pattern factors and other factors that
may be contributing to the lower salaries earned by these female practitioners could
be invaluable in understanding (and then correcting if possible) the income disparity
noted between male and female equine practitioners. Female equine practitioners
job satisfaction would then most likely increase even further.
in conclusion, further comprehension of the complex processes of job
satisfaction, stress, burnout and coping will have implications for the profession as
it continues to become feminized. A full understanding of these complicated issues
can prove important to professors and counselors of veterinarians as well as to the


10
of control (Lefcourt, 1982), and causal attributions (Doherty & Baldwin, 1985) may
each play a role in moderating or mediating the effects of job stress and individual
differences on job dissatisfaction, burnout, and commitment. Other variables which
may be relevant include the need for achievement, level of self-esteem, tolerance
of ambiguity, Type A vs. Type B personality (Ivancevich & Matteson, 1980), and
negative affectivity (Parkes, 1990). A very pragmatic avenue of investigation,
however, may be identification of veterinarians coping styles since interventions
aimed at altering a style which is associated with burnout or dissatisfaction could
subsequently be offered.
Lazarus transactional theory of stress (1991), which was cited earlier,
reflects the multidimensional, dynamic, and complex nature of stress, including the
interactions between the individual and the context in which he or she is working.
Two processes, cognitive appraisal and coping, are identified as crucial mediators
of stressful person-environment relationships and their immediate and long-term
outcomes (Lazarus & Folkman, 1984). Personal meanings of an event itself and the
choice and effectiveness of ones coping actions are considered integral to the
stress process.
The existing literature presents a confusing view of the role of coping in the
workplace. Various researchers label the forms of coping differently, but three
general types of coping seem to be cited most frequently: problem-focused,


4
made it difficult to generalize from one study to the next or to be certain that gender
differences are, indeed, nonexistent.
As earlier mentioned, no research on job satisfaction has been conducted in
the veterinary population. Several studies on the job satisfaction and status of
physicians and dentists have been published, however (Bickel, 1988; Carr,
Friedman, Moskowitz, & Kazis, 1993; Dial, Bickel, & Lewicki, 1989; Lenhart &
Evans, 1991; Lewis, Carson, Nace, Howard, & Barnhart, 1994). Richardsen and
Burke (1991) found no differences in overall job satisfaction between male and
female physicians, but noted differences in sources of satisfaction that predicted
overall job satisfaction. For women, satisfaction with time available for family and
personal life was an important predictor of overall satisfaction, whereas for men,
several work setting factors (e.g. own working conditions, access to support staff,
and equipment) were related to job satisfaction.
Kevorkian and Tuel (1994) surveyed 49 medical faculty members who had
recently resigned. While they found no significant differences in reasons for leaving,
there were gender differences noted when the physicians were asked what it would
have taken to induce them to stay in their present position. Men rated more
protected time for academics and more support for research more highly than did
women, and women ranked effective mentors or role models and more support for
patient care more highly than did men.


43
Five job satisfaction questions were used following the lead of Martin &
Shehan, 1989. Job satisfaction (SAT) was a composite measure of the responses
to the five items (1=strongly disagree to 5=strongly agree). The specific items
included were as follows: (1) am satisfied with my current job, (2) I cannot think
of jobs for which I would exchange, (3) If had the choice to make again, I would
choose the veterinary profession, (4) I do not intend to leave this job anytime in the
near future, and (5)This job measures up to my initial expectations. The use of five,
unweighted items remedies the inherent shortcomings of relying on a single-item
measure of the construct (Martin & Shehan, 1989). Values for the scale of
satisfaction range from 1.0 (low satisfaction) to 5.0 (high satisfaction). The job
satisfaction scale exhibited adequate reliability (Cronbachs alpha = .85).
In addition to the four above described instruments, the mail survey included
a short demographics questionnaire (DQ) and a cover letter. The DQ asked
respondents to provide information about their age, gender, years in the profession
and in the current job, income, hours worked per week, etc. They were also asked
to describe their family wage earning status, marital status and number of children.
Finally, they were asked to rate their perceived overall life stress for the past year
(see Appendix A). The cover letter (see Appendix B) made an appeal to colleagues
for assistance. In an attempt to gain maximal cooperation, the letter was typed on
AAEP letterhead stationery and was signed by the marketing director of the AAEP


CHAPTER 5
DISCUSSION
Four-hundred and ninety-two questionnaires were returned from an initial
sample size of 968. This represents a response rate of 51%, a rather remarkable
return rate following receipt of a single four-page questionnaire by a group of health
care professionals. Single survey mailouts to the general public are typically
followed by a 20% response rate (Bailey, 1982).
This response rate compares favorably with previous results obtained from
surveys of veterinarians. The American Association of Equine Practitioners typically
achieves response rates of between 20-30% from its membership with single
mailouts but no reminder postcard. A single mailout (of a considerably shorter
survey than the present survey) to 1000 female veterinarians (Elkins & Kearney,
1992) achieved a 61% response rate and a single mailout to 1000 mostly male
veterinarians achieved a response rate of 33% (Elkins & Elkins, 1987). Other health
care professionals are notoriously difficult to survey. One of the largest studies sent
over 3000 questionnaires to nurses, pharmacists and physicians and achieved a
42% response rate with a second, repeat mailing (Wolfgang, 1988).
The response rate of 51% following a single mailout is particularly
high in view of the fact that the survey was mailed in May, the month that
62


CHAPTER 3
METHOD
Participants
Potential participants were 968 (503 men and 465 women) licensed
veterinarians in private, predominantly equine practice who were selected from the
1997 membership directory of the American Association of Equine Practitioners
(AAEP). The AAEP has approximately 5600 national and international members
of which approximately 25% are female. Participants were eligible for selection for
the study as long as they were currently practicing, were in private practice (vs.
academia, government, or business), lived in the United States, described
themselves as working in a practice which is >50% equine, and had a known
gender.
Four-hundred ninety-two (492) surveys were returned, resulting in a final
response rate of 51%. Twenty-seven surveys arrived after the cut-off date of June
13, and were not included in the study. Eight surveys were unidentifiable by gender
and were also not included. Thus, 457 surveys were usable. Of the usable surveys
45% were from men (n=205) and 55% were from women (n=252).
36


75
veterinarians feel inadequately paid, and earn less than male equine practitioners.
These findings confirm the results of several other studies which have concluded
that female veterinarians earn considerably less than men even when a number of
demographic factors such as number of hours worked, years in the profession etc.
are considered (Bird, 1992; Frazier & Howell, 1991; Smith, 1996; Verdn, 1997).
Perhaps the women charge less for their services, see fewer patients per hour
[although Smith (1996) showed this not to be true of his sample], or work fewer
weeks per year. Maybe they give away more of their time to charity cases than do
male veterinarians. Possibly they do not negotiate as well as men when buying a
practice or buying equipment and drugs. Or, perhaps the general public
discriminates against female equine practitioners by preferring to work with male
veterinarians. It is clear that, for whatever reasons gender-related salary
differences persist, female practitioners find these differences to be very stressful,
and these salary differences need to be addressed and studied further by the
profession.
Limitations of This Study
There are several methodological limitations of this study which relate to the
fact that it was a mail survey study. A frequently cited problem of mail survey
research is the issue of non-response bias. Is it accurate to assume that the
respondents ideas uniformly represent those of the non-respondents? In this
study, a very good response rate (51%) was attained, comparable to that expected
from the genera! public following four mailings (Dillman, 1983)! If the respondents
were the veterinarians most likely to suffer from job stress, then the analysis


70
differences to three causes: 1. women experience more and different work stressors
than do men. (This was perceived to be true by the female equine practitioners in
the present research) 2. women are disadvantaged with regard to coping if they lack
workplace coping resources such as power, perceived competence, and financial
advantages. (This was also shown to be true for the women in this sample.) 3.
gender role stereotypes (Hansen & OLeary, 1983; Deaux & Emswiller, 1977) affect
male and female perceptions of what is appropriate coping (such as disclosing to
friends and colleagues, for example).
Hypothesis five stated that the relationship between job stress and burnout
is mediated by coping style in equine veterinarians in private practice. This
hypothesis was only partially supported. Emotional coping increased the positive
association of job stress with burnout, but only in men. Neither rational nor
detached coping mediated the relationship between job stress and burnout. Thus,
this hypothesis was only partially supported. In women, no type of coping style
mediated the relationship between job stress and burnout.
The coping aspect of this research was based on Lazarus transactional
model of stress and coping which was discussed in Chapter Two. Lazarus early
work (1966) demonstrated that the way people evaluate an event (as it impacts
upon their well-being) and how' they cope with that event influences the existence
and intensity of stress. Thus, if we are interested in studying the impact of stress
in the work environment, we need to assess coping style as well. The present study
attempted to recognize the impact of stress and coping on the outcome measure,
burnout. It was hypothesized that method of coping would mediate the effect of job


20
It has been suggested that failure to identify gender-related differences in
occupational stress may be due to sampling problems (men over-represented in
management positions with women holding clerical and service jobs (Jick & Mitz,
1985). Further, these authors conclude that gender may act not only as a direct
predictor of the source of stress, but also as a moderator affecting how stress is
perceived, what coping skills are called upon, and how stress is manifest (p. 409).
In agreement with the above authors, Spielberger and Reheiser (1995), using the
JSS, noted striking gender differences in the severity or frequency of items identified
as stressful. Men and women rated the same six items as highest in perceived
severity and the same 5 items as lowest in severity, however.
Consequently (although somewhat surprisingly), gender related differences
were not statistically confirmed in the JSS severity, frequency or overall scores
despite the numerous differences reported in the severity and frequency of
occurrence of specific JSS stressors. The authors concluded that gender-related
differences in occupational stress appear to be determined by differences in the
perceived severity of specific stressors and in the frequency that these stressors are
experienced by men and women (p. 64).
There has been some speculation but little research concerning differences
in stress experienced by male and female physicians (Simpson & Grant, 1991).
More research has been carried out with medical students than with physicians in
practice.


This dissertation was submitted to the Graduate Faculty of the
Department of Psychology in the College of Liberal Arts and Sciences and to
the Graduate School and was accepted as partial fulfillment of the requirements
for the degree of Doctor of Philosophy.
August 1998
Dean, Graduate School


38
$69:999, 10.9% (n= 50) earned between $70,000 and $89,999, 11.8% (n=54)
earned between $90,00 and $109,999 and 15.0% (n=69) earned over $110,000
per year.
Twenty-three percent of respondents (n=105) had never been married, 54%
(n-247) were currently married, 10.5% (n=48) were divorced and unmarried, and
11.6% (n~53) were divorced and remarried once. Sixty-eight percent (n=311)
described themselves as the sole wage earner, 13.6% (n=62) described themselves
as secondary wager earners, and 18.3% (n_= 84) described themselves as equal
wage earners in their families. Eighteen percent of respondents (n=84) had
between 1 and 3 children under the age of five in the household, 23.6% (n=108)
had between one and three children between the ages of six and 18 living at home,
and 5.9% (n=27) of the respondents had one or two adult children (ages 19-25)
living at home. Seventy-seven percent (n=350) had no children living at home. Six
percent (n-27) of respondents reported having children with significant emotional
(n=15), disciplinary (n_=10), or academic (n=15) problems.
Measures
The following four instruments comprised the major part of the mail survey
used in this study: (1) the Health Professions Stress Inventory (HPSI), (2) the
Coping Styles Questionnaire (CSQ), (3) the Burnout Measure (BM) and (4) Five
questions which formed a composite measure of job satisfaction (SAT). (See


29
10-14 years. A high correlation between reported stress and burnout score was
observed, with client relations being the most stressful aspect of practice cited.
Solo owners of multiveterinarian practices had significantly higher burnout scores
than employees or veterinarians owning solo practices.
The Elkins and Elkins (1987) study was replicated in 1992 in a mail survey
of 1000 female veterinarians. (Elkins & Kearney, 1992). Interestingly, this survey
yielded a 61% response rate, using only one mailing as in the previous survey. The
female respondents reported a burnout score mean of 3.41, with 67% scoring over
3. There was a very strong positive correlation between burnout and hours worked
per week and burnout and stress. A strong negative correlation was evident
between job satisfaction and burnout. Female practice owners had the lowest
burnout scores, although most were owners of solo practices rather than multi
veterinarian practices.
The female veterinarians did exhibit higher burnout scores than did their male
counterparts. Further, a greater percentage of women than men scored in the
clinical range. Conclusions cannot be drawn from such direct comparisons since the
studies were widely spaced in time. Lowman (1993) concludes after reviewing the
general work literature that no inherent gender differences have been found in
research on job-related burnout, but that there may well be differences in the
causes of burnout for men and women and in their coping strategies for handling
burnout symptoms.


86
I=Never 2-Once in a great while 3=Rarely 4=Sometimes 5=0ften 6=Usually 7=Aiways
Feeling trapped
Feeling worthless
Being weary
Being troubled
Feeling disillusioned and resentful about people
reeling energetic
Feeling way
Feeling hopeless
Feeling rejected
Feeling optimistic
Feeling anxious
instructions: Please circle the appropriate answer below.
Please answer ail the items!
Age 25-29 30-39 40-49
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
1 2 3 4 5 6 7
This is a crucial part of the study.
50-59 60-69 7(H
Gender male female
Years in the profession of veterinary medicine
Years in the current job
0-5
6-10
11-15
16-20
>20
0-5
6-10
11-15
16-20
>20
Circle one: Practice owner(sole owner) Practice partner Employee
Yearly income <$30,000 30,00049,999 50,000-69,999 70,000-89,999 90,000-109,999 >110,000
Hours worked per week <20 2040 41-50 51-60 >60
Practice type small animal mixed animal large animal other (state type)
Number of veterinarians in practice 1 24 5-7 >7
Board certification? no yes (state type)
Marital status never married married divorced & single divorced & re-married remarried>l widowed
Family wage earning status: primary (or sole) wage earner secondary wage earner equal wage earner
Number of children living at your home under age 5 between 6-18 between 19-25
Have any of your childr en had (circle if applicable) emotional disciplinary significant academic problems?
On a scale of 1-10, how stressful would you rate your life (excluding your worklife) this past year ?
Not at all stressful Average for an American Very stressful
123 456789 10
END OF SURVEY.
PLEASE ENCLOSE IN ENVELOPE PROVIDED.


41
coping style in the participants. The CSQ was developed by Roger, Jarvis, and
Najaran in 1993. Responses to 78 items (chosen by the authors from a variety of
sources) were subjected to principal factoring and rotated to a four-factor terminal
solution following a scree test. The CSQ utilizes the 60 items with the highest
loadings. Each item exemplifies ways in which individuals react to stressful
situations. Respondents are asked to indicate whether each item illustrates a way
they react to stress by circling always, often, sometimes, or never for each
item. A mean score was calculated for each of the following four coping styles:
rational (RAT), detached (DET), emotional (EM), and avoidance (AV). The CSQ
has test-retest reliabilities of .80, .79, .76, and .70 for each scale respectively, and
it has coefficient alphas of .85, .90, .74 and .69 for each scale respectively.
Because of the excessive length of the original questionnaire, only items which had
factor loadings of .5 or higher in the original research (Roger, Jarvis, & Najaran,
1993) were used in this study. This procedure cut the CSQ to 30 items and is in
accord with accepted means of decreasing lengthy questionnaires while maintaining
reliability of measures (Zyzanski, 1993).
To investigate the validity of the Coping Styles Questionnaire as used in this
study (thirty items vs. the original sixty), the responses to the thirty items were
subjected to principal factoring, and rotated to a four-factor terminal solution
following a scree test. With the exception of five items, all loaded on each of four


79
in those who must handle large animals.
it truly does appear that a comprehensive understanding of why veterinarians
who are female earn significantly less than male veterinarians is needed, since the
women in this research study found the salary situation to be more stressful than
any other stressor assessed. This should not be a particularly difficult question to
answer if enough of the necessary information is obtained (e.g. hours worked per
week, weeks worked per year, patients seen per hour, years in the profession, etc.)
For future research on coping styles, the Coping Styles Questionnaire (CSQ)
of Roger, Jarvis, and Najaran (1993) was shown to be a reasonable, though
imperfect, instrument by the results obtained in this research. It can effectively be
shortened from 60 to 30 questions (as was done in this study). The avoidance
coping scale, which had the weakest coefficient alpha of the four scales in the
authors study (.69) (Roger, Jarvis, & Najaran, 1993) also had very poor factor
loadings in the present study. In this author's opinion, the face validity of the scale
is quite weak, and methods which people use to moderate the impact of job stress
on the mind or body (such as drug or alcohol usage, TV watching, exercising to an
extreme) should be included.
Further, the finding that the interaction between emotional coping and job
stress increases burnout in men but not women should be confirmed and then
studied further. For example, perhaps higher frequency of use of emotional coping
in male veterinarians (since it appears to be more of a female veterinarian coping
preference) lowers self-esteem in men more so than in women. Or perhaps the use
of emotional coping has a greater negative impact on employees or clients if used


?
in education, years of experience, board certification, practice type, investment in
private practice, and hours worked). The remaining 58% ($7,400) could not be
explained, indicating that veterinarians who are female, like their counterparts in
medicine and dentistry, receive substantially lower incomes than male veterinarians
(Bird, 1992).
In summary, veterinary medicine is a profession which requires many years
of education typically culminating in high debt. Yet, its practitioners receive
comparatively low starting salaries and are losing the ability to own their own
businesses. Additionally, an increasing proportion of its practitioners who are
women appear to suffer from salary discrimination which, if perceived, is likely
stressful. Satisfaction gleaned from the work itself, therefore, is presumably a
strong motivator in attracting both males and females to the profession and
preventing their attrition.
With the exception of two studies on burnout (Elkins & Elkins, 1987; Elkins
& Kearney, 1992) and one (unpublished) study on gender differences in job
satisfaction (Welsch, 1991), no research on job satisfaction and stress within the
veterinary profession has been conducted. The purposes of this exploratory study
are (1) to examine the relationships among job stress, job satisfaction, and burnout
in veterinarians in private, predominantly equine, practice and (2) to determine if
relationships among these variables are moderated by gender and mediated by


61
Summary of Results
As is readily predictable, among equine veterinarians in private practice, job
stress was found to be negatively correlated with job satisfaction and positively
correlated with burnout. Veterinarians who are women reported significantly more
job stress and more burnout than did men, while at the same time reporting
significantly higher job satisfaction. Although male and female equine practitioners
seemed to agree on the frequency with which most job stresses occurred, women
found 45.7% of these stressors significantly more stressful than did their male
counterparts. Male and female equine practitioners reported using various coping
styles with different frequencies: women used significantly more emotional coping
than did men, who used significantly more detached and rational coping than did
the females. Emotional coping style was the only style shown to interact (in men
only) with job stress so as to increase the positive association between job stress
and burnout.


93
Holt, R R. (1982). Occupational stress. In L. Goldberger & S. Breznitz
(Eds.), Handbook of stress (pp. 419-444). New York: Free Press.
Hopkins, A. (1983). Work and job satisfaction in the public sector. Totowa.
NJ: Rowman & Allanheld.
Hoppock, R. (1935). Job satisfaction. New York: National Occupational
Conference, Harper.
Ivancevich, J. & Matteson, M. (1980). Stress and work: A managerial
perspective. Glenview. IL: Scott, Foresman.
Jackson, S., & Schuler, R. (1985). A meta-analysis and conceptual
critique of research on role ambiguity and role conflict in work settings.
Organizational Behavior and Human Decision Process. 36, 16-78.
Jick, T., & Mitz, L. (1985). Sex differences in work stress. Academy of
Management Review, 10. 408-420.
Jones, J.W. (1981). Diagnosing and treating staff burnout among health
professionals. In J.W. Jones (Ed.), The burnout syndrome (pp.107-126). Park
Ridge, IL: London House Press.
Kevorkian, C., & Tuel, S. (1994). Faculty departure from academic
programs of physical medicine and rehabilitation. American Journal of Physical
Medicine & Rehabilitation. 73. 378-386.
Lazarus, R. (1966). Psychological stress and the coping process. New
York: Oxford University Press.
Lazarus, R. (1991). Psychological stress in the workplace. Journal of
Social Behavior and Personality 6, 1-13.
Lazarus, R. S. (1995). Psychological stress in the workplace. In R.
Crandall & P. Perrewe (Eds.). Occupational stress, (pp. 3-19) Washington, DC:
Taylor and Francis.
Lazarus, R., & Folkman, S. (1984). Stress, appraisal, and coping. New
York: Springer.
Lazarus, R. S., & Launier, R. (1978). Stress-related transactions between
person and environment. In L. A. Pervin & M. Lewis (Eds.), Perspectives in


52
in job stress (£< .0001), job satisfaction (£ < .0001), and burnout (£ < .0001) in
equine veterinarians in private practice. These differences were shown not to be
accounted for by years in the profession, income, or perceived life stress. Table
4.2 shows the least squares means of the scores of male and female veterinarians
on the three measures of job stress, job satisfaction and burnout. As shown in this
table, the women, as compared to the men in this study, had significantly higher job
stress and burnout scores as well as significantly higher job satisfaction scores.
Thus, Hypothesis 2 was not supported. (It is noteworthy that if the variance in job
satisfaction due to income was not removed from the model, female veterinarians
were not significantly more satisfied with their jobs then were male veterinarians.)
Table 4.2
Least Squares Means comparing male and female scores on the HPSI-T, SAT,
and BOS.
Gender
HPSI-T
SAT
BOS
Female
7.16
3.87
3.38
Male
6.81
3.52
3.18
Because the covariate income was unexpectedly significant when it
appeared in the model which investigated gender differences in job satisfaction,


85
A=always Q=ofte S=sometimes N=never
Get things into proportion-nothing is really that important A 0
Feel that time will sort things out A 0
Feel completely clear-h eaded about the whole thing A 0
Believe that. I can cope with most things with the minimum of fuss A 0
Try not to let my heart rule my head A 0
Try to find a logical way of explaining the problem A 0
Decide its useless to get upset: and just get on with things A 0
Feel worthless and unimportant A 0
Trust in fate-things have a way of working out for the best A 0
Use my past experience to try to deal with the situation A 0
Just take nothing personally A O
Criticize or blame myself A 0
Talk about it as little as possible A 0
Feel completely calm in the face of any adversity A 0
See the thing as a challenge that must be met A 0
Be realistic in my approach to the situation. A 0
Try to think about or do something else A O
Instructions: How often do you have any of the following experiences?
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
S N
l=Never 2=Once in a great while 3=Rarely 4=Sometimes 5=Often 6=UsuaIly 7=Always
Feeling depressed
1 2
3
4
5
6
7
Ha ving a good day
1 2
3
4
5
6
7
Being physically exhausted
1 2
3
4
5
6
7
Being emotionally exhausted
1 2
3
4
5
6
7
Being tired
1 2
3
4
5
6
7
Being happy
1 2
3
4
5
6
7
Being '"'wiped out
1 2
3
4
5
6
7
Feeling burned out
1 2
3
4
5
6
7
Being unhappy
1 2
3
4
5
6
7
Feeling rundown
1 2
3
4
5
6
7


76
benefits from a having a rigorous sample, If the respondents were those who
found the time to respond because they were experiencing less stress, then
significant results would be less likely and again the analysis would be more
rigorous.
Barling, Kryl, and Bluen (1990) conducted an organizational survey and
hypothesized that the time individuals took to complete and return their
questionnaires would be associated with specific work attitudes, experiences and
individual differences. Their results revealed that none of the organizational factors
nor any demographic variable predicted when respondents would return their
questionnaires. In some survey research, it is assumed that late responders fairly
represent non-responders, in this study, while a rigorous review of the 25
questionnaires which were received beyond the due date was not conducted,
perusal of the responses revealed that these Individuals appeared to be no more
stressed, dissatisfied or burned out than their more promptly responsive colleagues.
A second methodological limitation of the study is related to history. If the
practitioners received the survey at the busiest time of their practice (which is
precisely v/hat occurred), they might report less satisfaction and more stress than
is typical for them. In general, these veterinarians were fairly unstressed and
satisfied, and if, indeed, this was their most stressful time of year, then it provided
the best opportunity to study their coping behavior.
Another limitation of mail survey research and thus this study is that
individuals may easily misread or misunderstand the questions and/or their intent.
The impact of this problem was reduced by first administering the survey to several


92
Folkman, S., & Lazarus, R. (1988). Coping as a mediator of emotion.
Journal of Personality and Social Psychology. 54, 466-475.
F'raser, T. (1983). Human stress, work and job satisfaction. A critical
approach. Geneva, Switzerland: International Labour Office.
French, J.R., Caplan, R.D., & Van Harrison, R. (1982). The mechanisms
of job stress and strain. Chichester, England: Wiley.
Glenn, N., & Weaver, C. (1982). Further evidence on education and job
satisfaction. Social Forces. 61,46-55.
Gehrke, B. (1995). Employment, starting salaries, and educational
indebtedness of 1995 graduates of US veterinary medical colleges. Journal of
the American Veterinary Medical Association. 207. 1290-1291.
Gehrke, B. (1996). Gender redistribution in the veterinary medical
profession. Journal of the American Veterinary Medical Association. 208, 1254-
1255.
Goode, W. (1960). A theory of strain. American Sociological Review. 25,
483-496.
Hamilton, S., & Fagot, B. (1988). Chronic stress and coping styles: A
comparison of male and female undergraduates. Journal of Personality and
Social Psychology. 55, 819-823.
Hansen, R., & O'Leary, V. (1983). Actresses and actors: The effects of
sex on causal attributions. Basic and Applied Social Psychology .4. 209-230.
Harris, J.R. (1994). An examination of the transaction approach in
occupational stress research. In P.L. Perrewe & R. Crandall (Eds.), Occupational
stress: A Handbook (pp. 21-28). Washington, DC: Taylor and Francis. (Original
work published 1991).
Havlovic, S. & Keenan, J. (1995). Coping with work stress: The influence
of individual differences. In R. Crandall & P. Perrewe (Eds.), Occupational
stress: A handbook (pp. 179-192). Washington, DC: Taylor & Francis, Ltd.
Herzberg, F., Mausner, B., & Snyderman, B. (1959). The motivation to
work. New York: Wiley.


CHAPTER 2
REVIEW OF THE RELATED LITERATURE
The literature review presents research which will aid in the understanding
of the background upon which this study was based. There are five sections in this
chapter. These include sections on job stress, job satisfaction, burnout, and coping
and cover theories of each along with gender and other differences which have
been noted. The final section concludes with the five research hypotheses
investigated in this study.
Job Stress
The concept of stress has been viewed from several, evolving perspectives
over the past century. Although many different definitions of stress have been
entertained by researchers, most seem to fall into three general categories. In the
stimulus definition, stress is seen as an external force that causes a reaction of
strain within the individual. Holt (1982) defines 57 different work stressors (which
include work overload, role ambiguity, monotony, and lack of control).
The response definition of stress takes the difference in stressors into
account. The Canadian endocrinologist, Hans Selye (1950), profoundly shaped the
14


94
interactional psychology (pp. 287-327). New York: Plenum,
Lefcourt, H. (1982). Locus of control: Current trends in theory and
research. Hillsdale. NJ: Erlbaum.
Lenhart, S., & Evans, C. (1991). Sexual harassment and gender
discrimination: a primer for women physicians. Journal of the American Medical
Womens Association. 46. 77-82.
Lewis, J., Carson, D., Nace, E., Howard, B., & Barnhart, F. D. (1994). The
lives of female physicians. Texas Medicine. 90. 56-61.
Linn, L., Yager, J., Cope, D., & Leake, B. (1985). Health status, job
satisfaction, and life satisfaction among academic and clinical faculty. Journal of
the American Medical Association. 254, 2775-2782.
Locke, E. (1976). The nature and consequences of job satisfaction. In M.
D. Dunnette (Ed.), Handbook of industrial and organizational psychology (pp.
1297-1349). Chicago: Rand-McNally.
Locke, E. & Taylor, M. S. (1990). Stress, coping, and the meaning of
work. In A. Brief & W. Nord (Eds.). Meanings of occupational work (pp.135-170).
Lexington, MA: Lexington Books.
Long, B., & Kahn, S. (1993). A theoretical integration of women, work, and
coping. In B. Long & S. Kahn (Eds.), Women, work and coping (pp. 296-311).
Buffalo: McGill-Queens University Press.
Lowman, R. L. (1993). Counseling and psychotherapy of work
dysfunctions, Washington. DC: American Psychological Association.
Lyons, T. (1971). Role clarity, need for clarity, satisfaction, tension, and
withdrawal. Organizational Behavior and Human Performance, 6, 99-110.
Martin, J., & Hanson, S. (1985). Sex, family wage earning status, and
satisfaction with work. Work and occupations. 12. 91-105.
Martin, J. & Shehan, C. (1989). Education and job satisfaction. The
influences of gender, wage-earning status, and job values. Work and
Occupations. 16. 184-199.


13
which have become feminized have experienced a decline in salary and prestige
leading to decreased job satisfaction. This research was the first to explore job
stress, job satisfaction, and coping styles of veterinarians of either gender. Findings
from this research will have implications for the future of the profession as it
becomes female dominated. More importantly, findings suggest ideas for potential
interventions at the level of both the veterinary student and practicing veterinarian.
Consequently, the results of this study may enhance job satisfaction and decrease
job turnover, burnout, and professional attrition among veterinarians in private
practice.


99
Super, D. (1942). The dynamics of vocational adjustment. New York:
Harper and Brothers.
Takooshlan, H. (1994). The burnout measure, Second Edition. In D.
Keyser & R. Sweetland (Eds.), Test critiques, Volume X (pp. 205-207) Austin,
TX: Pro.Ed.
Tung, R. (1980). Comparative analysis of occupational stress profiles of
male versus female administrators. Journal of Vocational Behavior. 17,344-355.
Weaver, C. (1974). Correlates of job satisfaction: Some evidence from the
national surveys. Academy of Management Journal. 17. 373-375.
Welsch, B. (1991). Gender differences in job satisfaction in veterinarians
in academic practice. Unpublished study presented at the 38th Annual American
Association of Equine Practitioners Convention, Orlando, FL.
Wolfgang, A. (1988). Job stress in the health professions: A study of
physicians, nurses, and pharmacists. Behavioral Medicine, 14, 43-47.
Wolfgang, A. (1995). Job stress, coping, and dissatisfaction in the health
professions: A comparison of nurses and pharmacists. In R. Crandall & P.
Perrewe (Eds.). Occupational stress: A handbook (pp. 193-204). Washington
DC: Taylor & Francis, Ltd.
Zanna, M., Crosby, F., & Loewenstein, G. (1987). Male reference groups
and discontent among female professionals. In B. Gutek & L. Larwood (Eds.),
Womens career development, (pp. 28-41). Newbury Park, CA: Sage
Publications.
Zyzanski, S. J. (1993). Cutting and pasting new measures from old. In M.
Stewart, F. Tudiver, M. Bass, E. Dunn, & P. Norton (Eds.), Tools for primary care
research (pp. 97-111) Newbury Park, CA: Sage Publications.


Table 4.3 (continued)
58
PROBLEMATIC FREQUENT
Female Male
Female Male
Being interrupted by phone calls or people while performing job duties
3.22**2.75
3.
55 3,i
Not being allowed to participate in making decisions about your job
1.96** 1.39
1.74*
1.39
Not being challenged by your work
1.95
1.75
1.80
1.87
Feeling that you are inadequately paid as a health professional
3.76**2.83
3.65**2.92
Caring for terminally ill patients
2.45**
12.00
2.26
2.01
Not being able to use your abilities to the fullest extent on the job
2.71
2.40
2.42
2.26
Fearing that a mistake will be made in the treatment of a patient
2.85**
: 2.17
2.49**2.13
Having owners whose values and beliefs about care differ greatly from
2.80*
2.43
2.38
2.39
yours
Problems related to business and management of veterinary practice
3.15
2.83
3.06
2.96
Worrying about the possibility of zoonotic illness or injury to self or staff
2.12**
1.74
1.96
1.78
Having to make difficult ethical or moral choices
2.41**
1.99
2.15
2.01
**Perm-p > .01
*Perm-p >.05


63
falls in the middle of the breeding, racing and showing seasons, the busiest times
in the lives of equine practitioners. So why might the response of these
veterinarians have been so high? It might be because the researcher was a
colleague, and a former academic equine practitioner whose name may have been
known to some of the participants. Further, the personal appeal made by the
researcher on the follow-up postcard was likely influential. This high response rate,
however, may have mostly been the result of interest in the topic of the research by
the target participants. Indeed, equine veterinarians might be interested in job
stress, coping and burnout in their profession--a profession noted for its stagnant,
relatively low wages, increasing debt burdens and long working hours (Bovee,
1997).
The response rate of the women (57%) was much higher than that of the
men (43%) in this study. This is in accord with the two previous surveys of
veterinarians cited (Elkins & Elkins, 1987; Elkins & Kearney, 1992). Why might
female veterinarians be so much more willing to respond than men? For one thing,
three recent research efforts have substantiated significant gaps in earning between
male and female veterinarians which cannot be explained by demographic variables
(Bird, 1992; Smith, 1996; Frazier & Howell, 1991). For another, the women in this
study were indeed more stressed on the job and more burned out than were the
men, and reported significantly higher levels of general life stress. Thus, these


54
particularly stringent test was necessary to avoid finding significant differences
which would occur due to chance alone. Adjustments were made, therefore, using
permutations of p, a more rigorous criteria than the Bonferronis adjustment.
Significant differences (p > .01) were found between the means of 20 of 70
items (16 of 35 on the severity scale and 4 of 35 on the frequency scale). In all of
these instances, the women found the particular problem to be significantly more
stressful than did the men. In fact, although not significant at the p = .01 level,
women found all but 6 of the 70 items to be more stressful than did the men. And,
the differences between the male and female scores on four of the six items which
the men ranked as more stressful were of the magnitude of .03 or less. Notably,
both sexes agreed relatively well on the frequency scale, but not well on the severity
scale. Thus, Hypothesis 3 was confirmed: there were gender differences in the
frequency and severity of specific job stressors reported by equine veterinarians in
private practice. See Table 4.3.
Hypothesis 4
To test Hypothesis 4 that there will be no gender differences in coping styles
reported by equine veterinarians, a MANCOVA was performed with the three valid
coping scales as the dependent variables, gender as the independent variable, and
years in the profession, perceived life stress and income as the covariates. These
covariates were included given that they were used as covariates in the analyses


16
appraised by a particular person. The consequent emotional reactions which follow
when a stressor is perceived as threatening and the individual is not able to cope
effectively with it form the basis of stress. Thus, Lazarus looks at the stressful
stimulus and looks at the response, but, in addition, places great emphasis on the
intervening variables of appraisal and coping ability, adding dimensions which
others have not considered.
Although Lazarus theory of stress is the most widely known and accepted
theoretical model, it is not without its critics. The major criticism of this theoretical
model is that it is incomplete since it fails to consider traits or personality factors
(Ben-Porath &Tellegen, 1990; Costa & McCrae, 1990). Further, Brief and George
(1994) criticize Lazarus emphasis on the idiographic nature of occupational stress.
They argue that it is especially important to discover those working conditions that
are likely to adversely affect groups of employees who are exposed to them. In a
similar critique of the model, Harris (1994) has noted that the occupational stressors
associated with the climate and the culture of an organization can have profound
effects on employees, and that these may differ as a function of gender and
individual differences in personality as well as coping skills. Lazarus transactional
model of occupational stress will be discussed in more detail under the section
entitled Coping.
Most of the occupational stress research has centered upon the effects of job
stress on the workers health--both physical and mental. In regard to physical


45
postcard personally appealed for help for the researcher and the organization In
their efforts to get the surveys completed and returned to them. The postcard was
signed with a blue pen by the researcher (see Appendix C). The study was
terminated within four weeks of the initial mailing, and the data were double key
punched by qualified operators.


65
practices. The average female in this study earned about $50,000 whereas the
average male reported earnings in the range of $70,000 to $90,000. This finding
has also been replicated previously. In the Bird (1992) study, a wage gap of $8000
was found between men and women which could not be explained by number of
degrees, board certification, years of experience, hours worked per year, or
ownership or partnership status. Smith (1996) found that even when earnings ratios
were adjusted for years of experience, number of hours worked per week, number
of patients seen per hour, type of practice, hospital size and geographical region,
female veterinarians earned only 65% of the salary of their male counterparts.
Hypothesis one stated that among equine veterinarians in private practice,
job stress is negatively correlated with job satisfaction and positively correlated with
burnout. This hypothesis was supported. The moderate correlations between job
stress and burnout (.54) and job stress and job satisfaction (-.39) are in line with
previous research completed on female veterinarians (r = .38 between stress and
burnout), pharmacists (r = -.46 between stress and satisfaction), and nurses (r = -
.47 between stress and satisfaction) (Elkins & Kearney, 1992; Wolfgang, 1995).
Hypothesis two stated that among equine veterinarians in private practice,
there are no gender differences in the overall amount of job stress reported, and the
degree of job satisfaction and burnout experienced. This hypothesis was not
supported. Indeed, there were significant gender differences in the amount of job
satisfaction reported, and the degree of burnout and job stress experienced. Female
equine practitioners experienced significantly more job stress and burnout than men
but reported significantly higher job satisfaction when the variance due to years in


Hypothesis 4 54
Hypothesis 5 56
Summary of the Resuits 61
CHAPTER 5 DISCUSSION 62
Implications of the Research Findings 72
Limitations of This Study 75
Future Research 78
Conclusions 80
APPENDIX A SURVEY 82
APPENDIX B COVER LETTER 87
APPENDIX C FOLLOW UP POSTCARD 88
REFERENCE LIST 89
BIOGRAPHICAL SKETCH 100
V


68
the present research. When asked if their current job measured up to their initial
expectations both male and female responses to this question were positive and
were not significantly different from one another.
Role conflict is a common and logical explanation for female burnout. It has
frequently been shown that the more successful a woman manager is in her career,
the less successful she feels about her home life. For men, success at work has
not been related to success or failure at home (Pines & Aronson, 1988). In this
study, men and women reported equal amounts of stress due to job duties which
conflict with family responsibilities. This item was rated the most stressful Item for
men, and the second most stressful (behind inadequate pay) for women. Yet,
women reported significantly more life stress than did men. Nonetheless, even
when the variance due to life stress was removed, women still reported significantly
more job stress and burnout than did men.
Women probably do pay a higher personal price for their careers in terms of
family life, however. This was very much in evidence in this sample. Although the
average female in the study was in her thirties (the average male was in his forties),
46% of these women had never married or were divorced and currently single; this
compares to only 19.5% of the men. While clearly related to marital status, these
female veterinarians had far fewer children than did their male counterparts and yet
they report higher overall life stress.
Womens values may make ambition and competition secondary to satisfying
work relationships as well as fulfilling relationships at home. Good personal work
relationships have been found to be highly negatively correlated with burnout in


64
topics may be more salient to female veterinarians.
As in other areas of the workforce, women equine practitioners were
significantly different from their male counterparts. Not surprisingly, there was only
one variable assessed in which women were not different from male practitioners,
and that was the size of the practice in which they worked. This is in contrast to a
previous survey (Smith, 1996) in which it was found that female veterinarians (who
cared for various species, not solely the horse) tended to work in smaller practices.
The significance of this disparity is that it has been shown (Smith, 1996) that
practitioners who work in larger hospitals (determined by the number of doctors)
tend to earn more than those who work at smaller clinics. The mean number of
equine veterinarians per practice in this survey was 1.7, which is typical of most
large animal, non-referral practices. In other words, most equine practitioners
(regardless of gender) work in small practices.
As in the previous studies cited (Bird, 1992; Smith, 1996), these female
practitioners were about 10 years younger and had practiced veterinary medicine
for 8 fewer years than the men. As noted in the introduction, this is because of the
late entry of women into the veterinary profession, which, until the 1990s was
entirely male-dominated. For example, in 1970, 5.3% of the profession was female;
this increased to 13.3% in 1980, and 22% in 1990 (Bird, 1992; Elkins & Kearney,
1992).
As in the previous studies, the women in this study worked about 10% fewer
hours per week than did the men. Equal percentages of men and women were
practice owners, although a greater percentage of women worked as associates in