Job burnout among critical care nurses

MISSING IMAGE

Material Information

Title:
Job burnout among critical care nurses
Physical Description:
xi, 177 leaves : ill. ; 28 cm.
Language:
English
Creator:
Stechmiller, Joyce K., 1947-
Publication Date:

Subjects

Subjects / Keywords:
Nurses -- Job stress -- Florida   ( lcsh )
Intensive care nursing -- Psychological aspects   ( lcsh )
Burn out (Psychology)   ( lcsh )
Burn out (Psychology)   ( fast )
Intensive care nursing -- Psychological aspects   ( fast )
Nurses -- Job stress   ( fast )
Florida   ( fast )
Foundations of Education thesis Ph. D
Dissertations, Academic -- Foundations of Education -- UF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 168-176).
Additional Physical Form:
Also available online.
Statement of Responsibility:
by Joyce K. Stechmiller.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 025055720
oclc - 24229079
System ID:
AA00025752:00001

Full Text












JOB Bl'.YOl'T A1iT!,( CRITICAL CARE NURSES


By


JOYCE K. STECHMILLER




































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


1990

























Copyright 1990

by

Joyce Kolbek Stechmiller
















ACKNOWLEPCGEME- iTS

I wish to express my deepest love and appreciation to my husband,

Bruce, and my children, Stephanie and Christopher, for their support,

understanding and constant love. Also to my mother and father and

Kate Stechmiller I extend my thanks for their moral support and

encouragement.

I would also like to thank the faculty of the University of Florida,

with special thanks to Dr. Hannelore Wass, chairman of my supervisory

committee, and supervisory committee members, Drs. Linda Crocker, Robert

Ziller, and Barry Guinagh, who contributed greatly to the completion of

the dissertation. Special appreciation goes to Dr. Hussein Yurandi for

his assistance with the statistical analyses.

My special thanks go to Donna Hall and Dan O'Brien, who were always

there when I needed them, and to my research assistants, without whom

these data could not have been collected. I also wish to express

gratitude to a special group of friends, MTG, INC., and to my students

for their support and encouragement when I needed them.

This research was supported by a Post-Baccalaureate Faculty

Fellowship from the Department of Health and Human Services. I am

grateful to the governmental agency for easing the burden of doctoral

education.
















TABLE OF CONTENTS


Page

ACKNOWLEDGEMENTS . . . . . . . . . . . . iii

LIST OF TABLES . . . . . . . . ........ vii

LIST OF FIGURES . . . . . . . . . . . . ix

ABSTRACT . . . . . . . . . . . . . . x

CHAPTERS

I INTRODUCTION . . . . . . . . . . . 1

Background of the Problem . . . . . . . . 1
Statement of the Problem . . . .. . . . . 3
Purpose of the Study . . . . . . . . . 5
Significance of the Study . . . . . . . 10
Limitations . . . e o . . . . . . . 12
Assumptions .. .o. .o .a .* * a o a o o * & o 13

II REVIEW OF THE LITERATURE ................. 14

Derivation of the Concept Burnout . .. . . . . 14
Definitions of Burnout . . . . . . . . 15
Stages in Development of Burnout . . . . . . 18
Causes of Burnout . . . . . . . . 19
Conceptual Models of Burnout . . . . . .. 21
Burnout of Nurses . . . . . . . . 24
Personal Characteristics . . . . . . .. 25
Demographics . . . . . . . . * 25
Personal Stressors . . . . . . . .. 26
Personality Characteristics . . . . . .. 26
Social Support . . . . . . . . . .. 27
Coping Behavior . . . . . . . . .. 29
Organizational Characteristics . . . . . .. 30
Work stress . . .... o *...... . . . * 30
Work Environment . .. .. a. .... . . .... 30
Job Satisfaction . . . . . . . . . 32
Theoretical Framework . . . . . . . .. 34

III METHODOLOGY . . . . . . . . . . 39

Sample of Respondents . . . . . . . . . 39
Study Design and Procedures . . . . . . . . 43










Page


Instrumentation . . * o
Demographic Questionaire . . .* . . .
Daily Hassles Scale . . . *.. . . .
Psychometric Properties of Daily Hassles Scale .
Job Diagnostic Survey (JDS) . . . . ...
Psychometric Properties of JDS . . . .
Psychological Hardiness Test . . .. . .
Psychometric Properties of Psychological Hardiness


Test . . . . . o .
Maslach Burnout Instrument (MBI) .
Psychometric Properties of MBI .
Commitment and Workload . ...
Statistical Analysis . . . .


* 9 9 9 9
* 9 9 9
* 9 9 9 9
* 9 9 9
* 9 9 9


IV STUDY FINDINGS . . . . . . . ......


Full Model and Statistical Analysis .
Exogenous Variables . . . .
Endogenous Variables . . . .
Question 1 . . . . . ..
Question 2 . . . . ...
Question 3 . . . ...
Question 4 . .. . . ...
Question 5 * 9 e e o .....
Reduced Model and Path Analysis . ..
Contributions to Situational Stress .
Contributions of Perceived Job Stress
Contributions to Job Satisfaction .


Contributions to Burnout-Emotional Exh
Goodness-of-Fit . . . . ...


0 0 0 0 0P






9austion .
9 9 9 9 9 9


V DISCUSSION, INTERPRETATIONS, CONCLUSIONS AND
FECClt1F'" DATIONS . . .... . . ..

Overview . . . . . . . .
Causes of Burnout . . . . ..
Causes of Job Dissatisfaction .*... ..
Comparison of Job Dissatisfaction and Burnout
Causes of Job Stress . 9 . . ..
Limitations and Recommendations . .. ..
Conclusions . . . . . e .


APPENDICES


A INVITATION AND CONSENT FOR PARTICIPATION FROM
CRITICAL CARE NURSES . . . . . . .


9 0 9 9 9 9


B PROCEDURES FOR ADMINISTRATION ...............

C DEMOGRAPHIC SHELT AND WORK SURVEY . . . . . . .

D DAILY HASSLES SCALE DIRECTIONS AND SAMPLE ITEMS . . .

E JOB DIAGNOSTIC SURVEY DIRECTIONS AND SAMPLE ITEMS . ..


. 0 0 0 0









Page

F WORKLOAD Al-,M STAFF SIZE . . . . . . . . 132

G HARDINESS TEST DIRECTIONS AND SAMPLE ITES . . . . 134

H COMIIT'Et.T TO CAREER . . . . . . . . . 136

I MASLACH BURNOUT INVFNT'lTRY DIRECTIONS AND SAMPLE ITEMS . 138

J MEAS, STALDAPRD DEVIATIONS AND CORRELATIONAL ANALYSIS . 140

K MULTIPLE REGRESSION OF THE FULL MODEL . . . . .. 158

REFERENCES . . . . . . . . .. . . . 168

BIOGRAPHICAL SKETCH . . . . . . . . . . . . 177
















LIST OF TABLES


Table Page

1-1 Variables Found to be Significantly
Related to Burnout . * . o ... . 6

3-1 Characteristics of Critical Care Staff Nurses:
Sex, Age, and Education (N = 300) . . . . . . 41

3-2 Duration as Staff Nurse in Critical
Care (N'= 300) . . . . . . . . . . 42
3-3 Years of Experience as a Critical Care Staff
Nurse at the Present Institution . . . . . 44

3-4 Work Hours Per Week (N = 300) . . . . . . . 45

3-5 Psychometric Properties of Variables: Ability
and Time . . . . . . . 9 * * * 47

3-6 Psychometric Properties of Variables: Daily
Hassles Scale . . . .. . . 0 . . 51

3-7 Job Diagnostic Survey Subscales, Related
Variables and Sample Items . . . . . . . 53

3-8 Psychometric Properties of Variables: Job
Diagnostic Survey . . . . . . * * * 57

3-9 Psychometric Properties of Variables: Commitment
and Workload . . . . . . . . . 68

3-10 Instrumentation: Personal, Organizational
Variables, Situation Conducive to Stress,
Job Stress, Burnout, Job Satisfaction, and
Internal Work Motivation . . . . . . . 69

4-1 Multiple Regression of Degree Situation
Conducive to Stress on the Variables Within
Each Block of Characteristics . . . . . . 78

4-2 Multiple Regression of Job Stress on the
Variables Within Each Block of
Characteristics and Situation Conducive
to Stress * . e e e * . * * a o 80

4-3 Multiple Regression of Job Satisfaction on the
Variables Within Each Block of Characteristics,
Job Stress and Situation Conducive to Stress . . .. 82









Table


Multiple Regression of Internal Job Motivation
on the Variables Within Each Block of
Characteristics, Job Stress and Situation
Conducive to Stress . . . . .* * * *..


Multiple Regression of Burnout-Emotional
Exhaustion on the Variables Within Each Block
of Characteristics, Job Satisfaction, Internal
Job Motivation, Job Stress and Situation
Conducive to Stress . . . . .

Magnitudes of Direct and Indirect Effects on
Degree Situation Conducive to Stress, Level
of Perceived Job Stress, General Job
Satisfaction and Burnout . * o . .

Intercorrelations of Exogenous Variables in
the Path Analytic Model of Burnout Among
Critical Care Nurses . o . . . .


* * * 86





* * 88



* * 98


viii


Page





84
















LIST OF FIGURES


Figure Page

1-1 A Perceptual-Feedback Stress Paradigm . . . . 4

1-2 Full Model of Burnout . . . . . . . . 7

4-1 Reduced Model of Burnout . . . . . . . 93
















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


JOB BL_'.1i''I AMONG CRITICAL CARE NURSES

By

Joyce K. Stechmiller

December, 1990

Chairperson: Hannelore Wass
Major Department: Foundations of Education

This study was designed to test a path-analytical model of the

theoretical conceptualization of burnout for critical care nurses. Three

hundred critical care nurses in Florida were administered a demographic

questionnaire and Work Survey Instrument, Daily Hassles Instrument,

Psychological Hardiness Test, Job Diagnostic Inventory, and the Maslach

Burnout Inventory. Personal, organizational, situational stress, job

stress, job satisfaction, and burnout variables were examined.

Family demands, health difficulties and a low psychological

hardiness and a high degree of skill variety and a dissatisfaction with

job security are linked to situational stress, contributing 70 percent of

explained variance. This situational stress is causally linked to

perceived job stress along with the direct effects of family demands,

health difficulties, dissatisfaction with nursing supervision, pay, and

job security. These variables account for 71 percent of explained

variance in job stress. Job stress directly affects job satisfaction.










In addition, job dissatisfaction is also directly affected by high job

expectations, meaninglessness of work, low knowledge of work results,

health, high task identity, a high frequency of dealing with others on

the job, dissatisfaction with opportunities for advancement, pay,

supervision, and satisfaction with job security. Together these account

for 62 percent of the variance in general job dissatisfaction. This

situation will then lead to burnout-emotional exhaustion. Burnout-

emotional exhaustion is directly affected by job dissatisfaction, as well

as low commitment to the career, health difficulties, low psychological

hardiness, high workload, a high degree of dealing with others on the

job, and dissatisfaction with job security; however, only 38 percent of

the variance in burnout-emotional exhaustion was explained by these

variables.

The study results provide a better understanding of the factors

relevant to the development of burnout. This research indicates that

there is a causal progression of situational stress, job stress, job

dissatisfaction resulting in burnout.
















CHAPTER I
INTRODUCTION

Background of the Problem

The empirical study of job stress in the helping professional,

frequently termed "burnout" (Freudenberger, 1974; Maslach, 1976; Maslach

& Jackson, 1986), has provided valuable information and insight into this

phenomenon over the last 15 years. Freudenberger (1974) first described

burnout as a state of physical and emotional depletion resulting from

conditions at work. Maslach (1976) claimed that "burned out"

professionals "lose all emotional feelings for the persons they work with

and come to treat them in detached or even dehumanized ways" (p. 16).

Freudenberger and Richelson (1980) later described burnout as a "sense of

fatigue or frustration brought about by devotion to a cause, a way of

life, or relationship that failed to produce the expected reward"

(p. 13). Edelwich and Brodsky (1980) further defined it as a

"progressive loss of idealism, energy, purpose, and concern as a result

of conditions of work" (p. 14), and Pines, Aronson, and Kafry (1981)

added that it is "characterized by physical depletion, by feelings of

helplessness and hopelessness, by emotional drain, and by the development

of negative self-concepts and negative attitudes toward work, life, and

other people. . It is a sense of distress, discontent, and failure in

the quest for ideals" (p. 15). Thus, although the concept of burnout has

been defined in many ways, there is general consensus that the symptoms

include attitudinal, emotional, and physical components (Freudenberger,

1974; Maslach, 1976; Maslach & Jackson, 1986).









Various researchers (Berkeley Planning Association, 1977; Maslach &

Jackson, 1981; Perlman & Hartman, 1982) have conceptualized burnout as a

syndrome of emotional exhaustion that proceeds to depersonalization, and

results in reduced personal accomplishment that may occur with persons

who work with others in some manner. The term "emotional exhaustion"

describes the first stage of burnout, which includes feelings of being

emotionally drained and overextended by one's exposure to other people.

When emotional resources are over-used, one feels unable to give of

oneself to others. The second stage of burnout is depersonalization,

which includes an "unfeeling and callous response toward people, often

the recipient of one's service of care" (Maslach, 1982b, p. 30). This

negative attitude may be manifested in rude, inappropriate, or

insensitive behavior toward clients, as well as withdrawal from them.

Reduced personal accomplishment is the third stage of burnout, which is

characterized as a decline in one's sense of competence and perception oJ

successful achievement in one's work with people resulting in feelings ol

inadequacy, failure, loss of self-esteem, and even depression (Maslach,

1987).

V Recently, a number of researchers have posited transactional models

of burnout to aid understanding of its etiology (Cherniss, 1980a;

Chiriboga & Bailey, 1986; Cooper, 1986; Courage & Williams, 1987; Cox &

Mackay, 1981; Fletcher & Payne, 1980; Golembiewski et al., 1986;

Harrison, 1983). According to Handy (1988), the major theme of

transactional models is that stress and burnout result from the

transaction between individual worker needs, personal resources, and the

demands, constraints, limitations, and/or facilitators within the work

environment. Key to this conceptualization is the emphasis on the










individual worker's subjective perception of stressors as opposed to

conditions that actually exist in the work setting (Golembiewski et al.,

1986; Leiter & Maslach, 1988; Maslach & Jackson, 1986).

Perlman and Hartman (1982) have proposed a transactional model (see

Figure 1-1) of burnout that is of particular interest because of its

breadth and because it includes groups of personal-psychological and

organizational-work variables that have been empirically studied

individually, or in combination, since 1977.

Statement of the Problem

In their model, Perlman and Hartman (1982) represented burnout as a

function of personal characteristics, organizational-work

characteristics, and stages of stress including the degree to which a

situation is conducive to stress, perception of organizational job

stress, and outcomes of the stress such as job satisfaction and

psychological response. Although this conceptualization of burnout

seems to be fairly comprehensive in terms of variables included, there

are limitations with respect to the posited relationship among the

components of the Perlman and Hartman (1982) model. One problem is that

this model is not totally consistent with empirical findings reported in

the literature. For example, according to the literature, job

satisfaction has a direct relationship to pay, dealing with others at

work, opportunity for advancement, and economic market conditions

(Jayaratne & Chess, 1983; Oldham, Hackman, & Pearce, 1976; Paredes,

1982), but from Figure 1-1 it would appear that burnout has a direct

relationship to pay, support from others, and economic market conditions

and that job stress has a relationship with opportunity for advancement.

Also, Perlman and Hartman (1982) seemed to suggest that burnout precedes

job satisfaction, but other formulations of burnout would equate it with

what Perlman and Hartman (1982) defined as coping. Another limitation is












that Perlman and Hartman (1982) conceptualized each variable as affecting

only the next continuous variable in the model. They did not allow for

the possibility of some variables having direct or indirect causal

relationships to more than one outcome variable. From their discussion

of the burnout model, it seems likely that Perlman and Hartman (14,)

were unaware of how systems of linear regression equations can be

specified using path analysis (or causal modeling) to test a theoretical

model. Such an empirical test of a transactional model of burnout seemed

to call for obtaining measures on a fairly wide array of personal and

organizational variables entering them into multiple regression and path

analysis to permit the assessment of both direct and indirect effects of

independent variables on outcome variables.

Purpose of the Study

The purpose of this study was to collect data to develop and refine

a theoretical model of burnout for critical care nurses using variables

identified from research literature. The first stage of the study

involved an extensive literature review to identify variables that had

been studied by others in relationship to burnout. Table 1-I contains i

summary of variables that were identified in the literature review for

this study that are posited to have a meaningful relationship in the

process that leads to job burnout. As will be indicated in Chapters II

and III, multiple indicators have been used to operationalize the

variables labelled as Personal Work Needs and Work Group Norms in

Table 1-1.

The second stage of the study required depicting the relationships

among the variables in a path diagram. The path diagram in Figure 1-2

graphically displays the pattern of hypothesized relationships among a










Table 1-1

Variables Reported to be Significantly Related to Burnout


Predictor Predictor
Personal Organizational Outcome
Variables Variables Variables


Ability

Time

Family demands

Job expectations

Personal work
needs

Physical health

Commitment to
career

Psychological
hardiness


Workload

Expected role
performance and
role ambiguity

Work group norms

Dealing with
colleagues

Opportunity for
advancement

Pay

Support from
others

Organizational
climate and
economic/market
conditions


Situation conducive
to stress

Level of perceived
job stress

Job satisfaction

Internal job
motivation


Burnout


Supervision





































oe





--- '- --'--p/ a
/ M t0
/ / ^i E2



0 M 0

2 U~
0 -4 2
I a /

_____ ^ a. M
2 va --n


a)

w~ 0 Ch w




(0 0 44-4 cc

0 w
U) 4.j 0 U
C.i toi 0 -C




Q) 0
m o .-i 3
;T. 11 7- ac: 2 u
- 01 C~qQ 00tU4-T SMO


"U

u
C,
<<


cc fac

0 U 4- c 0 0


4 E >b u 4-4
i 44 r- e w 04- 1 -- 1
.,- 0. 0J wC 0

Z-I >% w r- 4 3 .0
0 a) $4 > 14 E "c z- 4,j o u!!

P-4 "- w- 0iiu > a 4 1 ^ E W)
z U~2 2$ CW-- 00 W
o a) a o 4" o w
w. .( CO (n 4-) a. CO&.> Q. 44 .0
el 0 0 m; =a zC w 3 Q.B = o
0 3 M iE--4E-1 < cn =OCLU)









set of personal exogenous variables, a set of organizational exogenous

variables, and a series of endogenous outcome variables. An exogenous

variable is defined as a variable that has variability due to causes

outside the model, when the analysis of the cause of an exogenous

variable is not under consideration in the model (Asher, 1976). It is

not the purpose of a path analysis to explain the variability of an

exogenous variable or its relations with other exogenous variables.

An endogenous variable is a variable that has variance contributed

to it by exogenous or other endogenous variables in the model (Asher,

1976). The endogenous variables under study in the path analytical model

include the degree to which a situation is conducive to stress, job

stress, job satisfaction, internal work motivation, and burnout.

Unidirectional arrows or paths have been drawn from the exogenous causes

to the endogenous effects. The causal flow in the burnout model under

study is recursive or unidirectional. Some endogenous variables are

treated as dependent variables in one set of analyses and independent

variables in relation to other variables. For example, the degree to

which a situation is conducive to stress is a dependent variable for one

set of variables and an independent variable for job stress. To simplify

the visual presentation, the personal and organizational variables have

been grouped in boxes, and arrows have been drawn from the boxes to

represent direct effects of each exogenous variable in the boxes in the

left column on the respective endogenous variables. Typically curved

arrows are used to indicate correlations among exogenous variables. In

Figure 1-2, these curved arrows were not shown to allow the model to be

depicted more clearly.









The third stage of the study was to answer the following questions

using multiple regression analyses:

1. Is there a significant relationship between the situational

stress variable and (a) the weighted linear combination of a set of

personal variables and (b) a set of organizational variables? Which

variables in the model contribute significantly to the variance in

situational stress?

2. Is there a significant relationship between job stress and the

combination of (a) situational stress and (b) the preceding exogenous

variables? Which variables in the model contribute significantly to the

variance in job stress?

3. Is there a significant relationship between job satisfaction and

the combination of (a) job stress, (b) situational stress, and (c) the

preceding exogenous variables? Which variables in the model contribute

significantly to the variance in job satisfaction?

4. Is there a significant relationship between internal job

motivation and the combination of (a) job stress, (b) situational stress,

and (c) the preceding exogenous variables? Which variables in the model

contribute significantly to the variance in internal job motivation?

5. Is there a significant relationship between burnout-emotional

exhaustion and the combination of (a) job satisfaction, (b) internal job

motivation, (c) job stress, (d) situational stress, and (d) the preceding

exogenous variables? Which variables in the model contribute

significantly to the variance in burnout-emotional exhaustion?

The final stage of the study was to refine the purposed model on the

basis of the empirical results obtained from stage three in an









exploratory fashion. Each of the independent variables that made a

significant contribution to the variance in one or more dependent

variables was considered a candidate for inclusion in a reduced model of

burnout. Path analysis was used to estimate the direct and indirect

effects of exogenous variables on endogenous outcome variables in a

reduced model of burnout. Results of the analyses of the full model and

a suggested reduced model will be presented in Chapter IV.

Significance of the Study

This study has practical value for the critical care staff nurses

themselves, the patients cared for by the critical care nurses, the

institutions in which the critical care nurses work, and the educational

preparation of nurses taught by nursing faculty. Maslach (1982) stated

that the burnout syndrome stems from a social interaction between helpers

and helpees, when in certain circumstances, helpers become unduly

involved emotionally with the recipients, overextend themselves, and

demonstrate emotional exhaustion. Nurses, as well as teachers, are

helping professionals and share the burdens and frustrations in the

helping relationship. Maslach (1982) described the progression of

burnout from emotional exhaustion to depersonalization in which helpers

decreased contact with clients and showed a callous, detached, and

dehumanizing response towards helpees. He proposed that clients

receiving care from helping professionals in the stage of

depersonalization do not receive adequate care because of ineffective

interpersonal helping relationships. Because nursing activities require

frequent contact with patients in an effective interpersonal









relationship, patients cared for by nurses in the stage of emotional

exhaustion or depersonalization may suffer from pathological detachment

on the part of the nurse. Critical care nurses who avoid their patients

and families lose opportunities essential for timely and potentially

useful interventions. This results in an unfortunate physiological loss

to the patient and a psychological loss to the patient, family, and

nurses and other members of the health team.

Burnout is costly as well to the institutions in which the helping

professionals work because of poor performance, job dissatisfaction,

absenteeism, job turnover, and illness of workers, all of which have been

related to this phenomenon. Knowledge about the causes of burnout and

the organizational work variables that critical care nurses perceive as

contributing to burnout could allow nursing administrators to identify

strengths and weaknesses in the hospital organization in reference to

prevention, recognition, and management of burnout. Such information

might give insight into the control of burnout. A variety of strategies

by those concerned with professional development and nursing management

could be employed to help alleviate nursing burnout, including redesign

of jobs, changing organizational policies, establishing flexible

schedules and support services, improving training for staff, and

designing explicit programs for more decisionmaking, emotional support,

and recognition.

Knowledge of results of this study could assist nursing educators in

curricular development. By recognizing burnout as a legitimate problem

with specific causes, greater efforts may be made to deal effectively

with it in academic institutions.

Clinical nursing faculty are also in an excellent position to teach

students how to cope with a stressful job as well as how to make the job

less stressful, but they are hampered by a lack of information about









sources of emotional stress on the job and the risk of burnout in many

nursing baccalaureate programs. It is important for nurse educators to

understand the causes of burnout among helping professionals. Students

should have more accurate information about the work they are undertaking

before they actually start. If this were accomplished, they might have

fewer surprises that destroy their professional ideals or lead them to

leave the profession. The consequences of attrition and turnover may be

greatly reduced by providing realistic job expectations. Prior knowledge

about the causes of burnout may enable nurses to recognize it in its

early stages, whether in themselves or in others. Greater awareness of

the risk of burnout can lead nurses to be better prepared for it in

advance. They can anticipate causes of emotional stress before they

occur and develop definite plans to deal with them. The results of this

study will allow them to have a clearer understanding of the personal and

organizational variables that may enable them to be successful at their

work.

Limitations

The following limitations of the study must be taken into

consideration when interpreting the findings:

1. The range of the study was limited to a geographical region

within the state of Florida that included Melbourne, Orlando,

Jacksonville, Ocala, Tallahassee, and Gainesville.

2. The sample was limited to a cross-section of critical care

nurses who work in critical care settings, including Medical and Surgical

Intensive Care Units.

3. The sample may be biased because individuals in advanced stages

of burnout may have been so apathetic that they did not participate in

the study.









4. Several of the constructs were measured by self-report

questionnaires that included items containing content of a sensitive

nature; some respondents may have reported inaccurately due to their

inability or unwillingness to recognize symptoms in themselves.

5. In this study, only the first stage of burnout (i.e. emotional

exhaustion) was examined.

6. Variables examined were limited to those that could be logically

viewed as representing constructs in Perlman and Hartman's (1982)

conceptual model of burnout.

Assumptions

There are three assumptions upon which the present study is

grounded:

1. Burnout in a biopsycho-social concept related to personal,

psychological, and organizational variables as a result of a stress

response.

2. The strength of relationship between burnout and these

contributing factors can be estimated through application of a linear

model.

3. The independent variables in the model are all those

contributing importantly to burnout.
















CHAPTER II
REVIEW OF THE LITERATURE

Derivation of the Concept Burnout

The concept "burnout" originated from stress research. There are

many publications reviewing varied definitions of the concept of stress

(Antonovsky, 1979; Cannon, 1932; Lazarus, 1966; Mason, 1975; Selye, 1956;

Selye, 1974; Sharit & Salvendry, 1982; Vachon, 1987). In an early

definition, Cannon (1932) reflected upon the physiological aspect of

stress and related it to society and job organizations. Selye (1956)

broadened the physiological concept of stress even further.

Selye's (1956) work was based upon earlier research of Cannon that

included homeostasis adaptations. His identification of body hormones in

the "fight or flight" mechanism for maintenance of homeostasis focused on

protective maneuvers in physical, social, or chemical situations that

elicited a physiological response. He defined stress as "the

non-specific response of the body to any demand made upon it" (Selye,

1974, p. 141) and stressors as any stimuli that cause physiological

adaptation. If stress results in positive effects, it is identified as

eustress; stress associated with negative results is distress.

Furthermore, he classified physiological adaptation to stress into three

stages--alarm, resistance, and exhaustion--and referred to this process

as the general adaptation syndrome. The alarm phase consists of

neurohormonal mechanisms that prepare the body for defense against an

adaptation caused by stressors. The second phase, the resistance phase,









consists of adaptive mechanisms that are instituted to destroy the

stressor and restore homeostasis. Exhaustion, which occurs if the body's

adaptive resources are depleted, is irreversible and results in death.

Once the stressor is perceived as a genuine stressor, the condition is

recognized as stress and the general adaptation syndrome.

The concept of coping as human adaptation to stress has been studied

by many psychologists. French, Rodgers, and Cobb (1974) proposed a model

for adaptation to stress that considers the interaction of

characteristics of the individual and environment. A cognitive appraisal

analysis of psychological stress has been addressed by Lazarus (1977),

who defined coping as problem solving attempts to deal with threatening

conditions. Coping, in this approach, depends on how individuals

appraise their encounters with the environment with regard to the

potential of those encounters as threatening, harmful, or challenging.

Burnout has been identified as a syndrome resulting from a negative

response to occupational stress. Researchers in a variety of

specialities, including educational psychology, clinical and social

psychology, psychiatry, sociology, cultural anthropology, nursing,

occupational medicine, and personnel management have provided an

important volume of literature which constitutes the background for the

present study. The following review of literature is concentrated on

job-related personal, psychological, and organizational factors that for,

the basis for burnout.

Definitions of Burnout

Freudenberger (1974) first used the term "burnout" to describe an

unfavorable response among helping professionals working in psychiatric









settings. He observed a pattern of behaviors and attitudes that he also

experienced. He identified a helping professional experiencing burnout

as one who becomes exhausted from excessive demands on strength, energy,

or resources or who is worn out. Freudenberger and Richelson (1980)

later described burnout as a "sense of fatigue or frustration brought

about by devotion to a cause, way of life, or relationship that failed to

produce the expected reward" (p. 13). At the First National Conference

on Burnout, Maslach (1982b) presented the following definitions of

burnout that were reported in the literature:

A syndrome of emotional exhaustion, depersonalization, and reduced
personal accomplishment that can occur among individuals who do
people-work of some kind.

A progressive loss of idealism, energy, and purpose experienced by
people in the helping professions as a result of the conditions of
their work.

A state of physical, emotional, and mental exhaustion marked by
physical depletion and chronic fatigue, feelings of helplessness and
hopelessness, and the development of a negative self-concept and
negative attitudes toward work, life, and other people.

A syndrome of inappropriate attitudes toward clients and self, often
associated with uncomfortable physical and emotional symptoms.

A state of exhaustion, irritability, and fatigue that markedly
decreases the worker's effectiveness and capability.

To deplete oneself. To exhaust one's physical and mental resources.
To wear oneself out by excessively striving to reach some
unrealistic expectations imposed by oneself or by the values of
society.

To wear oneself out doing what one does to do so. An inability to
cope adequately with the stresses of work or personal life.

A malaise of the spirit. A loss of wellbeing. An inability to
mobilize interests and capabilities.









To become debilitated, weakened, because of extreme demands on one's
physical and/or mental energy.

An accumulation of intense negative feelings that is so debilitating
that a person withdraws from the situation in which those feelings
are generated.

A pervasive mood of anxiety giving way to depression and despair.

A process in which a professional's attitudes and behavior change in
negative ways in response to job stress.

An inadequate coping mechanism used consistently by an individual to
reduce stress.

A condition produced by working too hard for too long in a
high-pressure environment.

A debilitating psychological condition resulting from work-related
frustrations, which results in lower employee productivity and
morals. (pp. 30-31)

Maslach (1982b) analyzed the problem of defining burnout and

concluded that some definitions are broad, others narrow; some include

emotional and physical behaviors, others include psychological and

cognitive terms. In addition, some describe a process while others

present a process; some relate causes while others relate consequences.

One commonality in the definitions is that the burnout syndrome presents

a psychological process affecting individual attitudes, motives,

feelings, and expectations. The individual perceives the syndrome as

unfavorable, dealing with distress, problems, malaise, and/or negative

outcomes. Furthermore, Maslach pointed out that there is agreement on

the component of exhaustion as a loss of energy and debilitation,

physiologically and psychologically, a loss of trust and apathy, with

loss of feeling, concern, and spirit. Another component includes a

disparaging response to others, with depersonalization, inappropriate









attitudes toward clients, loss of ideals, and irritability. A third

component is characterized by unfavorable responses toward oneself and

one's personal achievements, with depression, withdrawal, low morale,

lowered production, and a decrease in effective coping.

Stages in Development of Burnout

Maslach (1982a) described a progression of stages in burnout leading

from emotional exhaustion to depersonalization and finally ending with

decreased personal accomplishment. Helping professionals who have

emotional exhaustion feel drained, used up, and repleted without

resources. As a protective mechanism, some helping professionals acquire

a cold indifference to others' needs, become detached to close

relationships, and acquire a callous disregard for the feelings of

others. The occurrence of this dehumanizing attitude heralds in the

second component of the burnout syndrome, depersonalization, which

consists of the expression of poor opinions about clients, actively

disliking them, expecting the worst from them, ignoring their requests,

and giving inappropriate help and care. The third aspect of the burnout

syndrome is decreased personal accomplishment, which consists of helping

professionals developing a sense of inadequacy in dealing with their

clients. The helping professional may even develop a poor self-esteem

and perceive of himself as a failure. As a result, depression,

absenteeism, poor job performance, and changing jobs may occur.

Similar developmental models have been suggested by other

researchers. For example, Edelwich and Brodsky (1980) designed a

four-phase developmental process: (a) enthusiasm with high energy, high

hopes, and unrealistic job expectations; (b) stagnation, in which the job

is no longer perceived as the central force of a worker's life;









(c) frustration, in which a worker directs personal confidence and value

of the job; and (d) apathy, a defense mechanism against job frustration

necessary for survival. Similarly, Veninger and Spradley (1981)

presented a developmental process that consisted of five stages: (a)

honeymoon, (b) fuel shortage, (c) chronic symptoms, (d) crisis, and (e)

hitting the wall.

Costello and Zalkind (1963) and Daley (1979) posited stages of

burnout specifically for the nursing profession using the framework of

Selye's general adaptation syndrome. The first stage is an alarm stage,

characterized by an emergency mobilization of the body's defense

maneuvers used to maintain successful performance or to prevent an

internal conflict leading to frustration. The second stage is the

resistance stage, in which there are continued attempts to manage the

stress. If stress management is not achieved, exhaustion occurs. Shubin

(1978) described nurses as high-risk victims for burnout and delineated

the stages as beginning with physical fatigue and emotional drain,

followed by dehumanization of patients and guilt for not caring any

longer, and finally disgust for oneself and others.

Causes of Burnout

Freudenberger (1975), Freudenberger and Richelson (1980), and Pines,

Aronson, and Kafry (1981) have contended that personal variables

including psychological stress are the major precipitating cause of

burnout. Freudenberger (1974) studied ego analysis and believed that a

committed and over-dedicated personality type was susceptible to burnout.

He and Richelson (1980) also suggested another personality trait that

included a restricted social life with all meaning and gratification









achieved from the job. Edelwich and Brodsky (1980) discovered an

association of burnout with the young, enthusiastic, overcommitted

helping professionals. Other specific psychological factors associated

with burnout in the care provider include competence, intolerance in

confronting obstacles, lack of self-confidence, nonassertiveness in

dealing with people (Gann, 1979), lack of psychological hardiness (Keane,

Ducrete, & Adler, 1985; McCranie, Lambert, & Lambert, 1987), life events

(Chiriboga & Bailey, 1986), ineffective coping skills, and inadequate

social support (Chiriboga & Bailey, 1986; Cronin-Stubbs & Rooks, 1985;

Numerof & Abrams, 1984).

Evidence to date suggests that not all individuals are equally at

risk to develop burnout (Cherniss, 1980a-b). Cherniss stated that

organizational factors that include availability of resources and power

of the helping professional to apply them, autonomy, the stimulation and

challenge of the job, the rewards at work, and the degree of structural

support all share in the occurrence of burnout. Role conflict, role

overload, lower socioeconomic status, and job dissatisfaction (Oldham,

Hackman, & Pearce, 1976) also appear to promote burnout (Kahn, 1978).

Burnout has also been related with a greater number of hours spent in

direct patient contact (Lewiston, Conley, & Blessey-Moore, 1981; Maslach

& Jackson, 1982), more difficult client problems (Meadow, 1981; Pines &

Maslach, 1978), caseload (Berkeley Planning Associates, 1977;

Freudenberger & Richelson, 1980; Larson, Gilbertson, & Powell, 1978;

Maslach & Jackson, 1984a-b; Maslach & Pines, 1977; Perlman & Hartman,

1982; Solomon, 1979), a low degree of peer support (Burk et al., 1984;

Jackson, Schwab, & Schuler, 1986; Leiter & Maslach, 1986; Maslach &

Jackson, 1982), a low degree of commitment to the organization (Leiter &









Maslach, 1988), and organizational components of leadership,

communication, supervision and responsibility. Leiter and Maslach (1988)

and Gains and Jermier (1983) indicated that dealing with co-workers

was identified as the strongest source of job stress and burnout.

Burnout has also been linked to social and physical isolation in the work

setting (Larson, Gilbertson, & Powell, 1978). Maslach and her colleagues

(Maslach & Jackson, 1981a; Maslach & Pines, 1977) indicated that the

basis 6f burnout is interpersonal contact that includes intense

commitment and personal care when helping others, especially those with

severe problems.

Conceptual Models of Burnout

Our understanding of the burnout syndrome has grown over the last

15 years. No longer are conceptual models restricted to intrapsychic

factors. There is considerable acceptance that burnout is not a simple,

unidimensional syndrome with easily identified causes. Rather, it is

considered a complex problem, generic to intrapsychic, interpersonal,

social, occupational, and organizational components.

Fischer (1983) posited a psychoanalytic model of burnout in which

workers are driven to maintain a high self-esteem by working even harder

despite unrealistic expectations. These workers "idealize their work"

and relate to a "compensatory illusion of grandiosity." He also stated

that workers who are exhausted may be inappropriately identified as

burned out and use the label of burnout as an "excuse for poor

performance and as a justification for both easier working conditions and

higher pay" (p. 41). Fischer added to the knowledge of the function of

the intrapsychic factors of burnout and focused on the importance of

self-esteem as a mediator variable in the burnout process.









Harrison (1983), Farber (1983), and Heifetz and Bersoni (1983)

conceptualized burnout as a process resulting when helping professionals

do not receive positive and/or accurate feedback from their work

environment regarding their performance efforts. Harrison (1983) viewed

burnout as inversely related to the helping professionals' perceptions of

competence. The likelihood of the helping professional feeling good

about his/her work performance may be affected by lack of institutional

support, excessive workload, and inadequate professional skills.

Furthermore, he stated that if success is rarely experienced by the

helping professional and if failure is usually experienced as a result of

his or her job performance, burnout is likely. Farber (1983)

conceptualized work related stresses and burnout as due to feelings of

"inconsequentiality" by the helping professional. Like learned

helplessness, Farber hypothesized that burnout results when helping

professionals perceive that their efforts do not matter, and, as a

result, their efforts cease. The concept of cybernetics was used by

Heifetz and Bersoni (1983) in their model to describe the phenomenon of

burnout. The helping professional's perception of growth in themselves

and their recipients of care is strongest when goals are realized. In

order to be successful in realizing goals, requirements in the cybernetic

process must be identified: identification of the task, clearly defined

goals, short-term progress reports, and plans for modification of one's

goals. According to Heifetz and Bersoni (1983), burnout occurs when

one's pursuit of goals is interrupted because there is an absence of one

or more of the requirements in the cybernetic process.

Some authors have attempted to describe the burnout syndrome using a

job deficit model which proposes that burnout is due to the absence of









job motivations rather than the occurrence of job stressors. Jayaratne

and Chess (1983) reported that role conflict, excessive workload, and

role ambiguity are not significant predictors of burnout in social work.

They identified job challenge, financial rewards, and promotions as the

strongest predictors of burnout and job satisfaction. Eisenstat and

Felner (1983) isolated the variables of job motivations and job stressors

and reported that while job stressors are related to emotional

exhaustion, job enrichers, including autonomy, task significance, and

skill variety are related to job motivation in helping professionals.

Cherniss and Krantz (1983) also focused on the relationship between

burnout and lack of job motivation. Specific variables in their

conceptualization included long hours and the absence of meaning in work.

In addition, Fibkins (1983) and lanni and Reuss-Ianni (1983) emphasized

that the crucial deficit variable in the burnout model is lack of

organizational support. According to Fibkins, burnout in teachers occurs

when the school organization is not responsive to the complex, intensive

nature of teachers' work. lanni and Reuss-Ianni (1983) suggested that

individual and social factors may contribute to work stress; however,

burnout is more likely to be caused by a deficit within the

organizational structure.

Several other authors have advanced the stress model, incorporating

essential elements of modern stress theory and drawing on past research

(Beehr & Neuman, 1978; Chiriboga & Bailey, 1986; House & Wells, 1978;

Matteson & Ivanovich, 1979; Perlman & Hartman, 1982). In the stress

model, antecedent conditions such as sociodemographic characteristics,

are seen as laying the ground work for stress that results in conditions

such as burnout. Sociostructural conditions in the workplace constitute









a second element that helps shape the stress context. Specific stressful

work situations and chronic strain are seen as a third level of

predictor. Social support and coping strategies in turn are seen as

additional predictors that may act as mediators between the stress

context and stress response. Finally, burnout is viewed as the stress

response. An outcome of a maladaptive stress response is low

productivity; other sequelae include absenteeism, job dissatisfaction,

and excessive job turnover. Although exploratory testing of the stress

model of burnout has been accomplished, further comprehensive analysis is

needed with larger samples.

Burnout of Nurses

Consistent with recent findings in other professions (Chiriboga &

Bailey, 1986; Jenkins & Ostchega, 1986; Stone, Jebsen, Walk and Belsham,

1984), nurses who experienced more frequent work-related stress reported

greater burnout. Critical care nursing is one profession where the

burnout phenomenon has been acknowledged because a high rate of patient

mortality, severely ill patients who are emotionally consuming,

inadequate staffing and resources, and a difficult work load are

encountered. According to the literature, critical care nurses are at

risk for burnout when the patient presents with clinical problems in

which the complexity and acuity are beyond the resources of the nurse

(Bailey, Steffan, & Grout, 1980; Bartz & Maloney, 1986; Chiriboga &

Bailey, 1986; Claus & Bailey, 1980; Gray-Toft & Anderson, 1981a-b;

Hinshaw & Atwood, 1984; Kelly & Cross, 1985; Maloney, 1982; McCranie,

Lambert, & Lambert, 1987; Numerof & Abrams, 1984; Stone, Jebsen, Walk, &

Belsham, 1984; Vachon, 1987). In one study, commissioned by the American

Association of Critical-Care Nurses, a national panel of experts









identified both nursing stress and nursing burnout as being among the top

ten research priorities facing the profession (Lewandowski & Kositsky,

1983).

Studies were examined related to potential factors contributing to

burnout of nurses. The review was limited only to studies using the

Maslach Burnout Inventory as a criterion measure. In this review, the

studies have been grouped according to findings related to personal

characteristics of the nurse, including demographics (age, marital and

family status, length of employment, income, and education), personal

stressors, personality characteristics, social support, coping behavior

and organizational characteristics including work stress, the work

environment (worker involvement, type of unit, type of hospital,

involvement in decision making), job enhancement (skill variety and new

approaches, autonomy, clarity, physical comfort, work pressure), and job

satisfaction.

Personal Characteristics

Demographics

Demographic variables that have been associated with burnout in

nurses include age, marital and family status, length of employment,

income, and education. Younger nurses are more susceptible to burnout

than their older counterparts (Chiriboga & Bailey, 1986; Dames, 1983)

However, Bartz and Maloney (1986) found a positive association of burnout

with younger intensive care nurses.

Single care providers tend to be at greater risk for burnout than

married care providers (Dames, 1983; Kimmel, 1983). In addition, junior

nurses without the support of a marital partner are indicated as

vulnerable to burnout (Chiriboga, 1986).









Although Pomales (1982) found a negative correlation between

burnout, income, and number of people living at home, Dames (1983) and

Pomales (1982) agreed that income was inversely associated with burnout.

Grutchfield (1982) found no significant difference among nurses'

education in baccalaureate, diploma, or associate degree programs, but

Keane et al. (1985) and Stone et al. (1984) found that nurses with

baccalaureate degrees evidenced somewhat higher levels of burnout than

those with three-year diplomas. Although less education has been

associated with the occurrence of burnout in nurses (Grutchfield, 1982)

less education with fewer years of work experience tends to make the

nurse more vulnerable to burnout (Chiriboga & Bailey, 1986; Stone et al.,

1984). In addition, a negative correlation has been found between

burnout and a desire to remain in the field of nursing (Dames, 1983).

Personal Stressors

As indicated by the findings of Daubney (1980), Otto (1980),

Chiriboga and Bailey (1986), and Stone et al. (1984) personal stressors

may also be pertinent in the burnout process. In these studies,

undesirable personal life change events and personal stress directly

related to burnout. In addition, perception of a good quality of life

five years ago was found to be inversely related to burnout (Dames,

1983).

Personality Characteristics

Grutchfield (1982) studied the relationship between personality

variables measured by the Edwards Preference Schedule, demographic

variables, and the syndrome of burnout among nurses. Her findings

suggested that burnout emotional exhaustion was associated with the









following personality variables: nurturance, abasement, achievement, and

succorance. In addition these variables were found to be negatively

correlated with a dimension of burnout-personal accomplishment.

Dames (1983) explored the relationship between selected personality

characteristics, demographics, and burnout, using the Spielberger Trait

Anxiety Index, the Gough Adjective Checklist, and the Maslach Burnout

Inventory. The personality traits of intraception and nurturance had

negative correlations with burnout; abasement, aggression and anxiety had

positive correlations with burnout. Autonomy had no significant

relationship with burnout. Pomales (1982) reported an inverse

relationship between self-concept and burnout.

Keane et al. (1985) and McCranie et al. (1987) explored the

relationship of psychological hardiness, demographics, and burnout using

the Hardiness Test (Kobasa et al., 1984). They presented data supporting

the hypothesis that hardiness may be an important personality variable

based resistance resource for preventing burnout among hospital nurses.

They compared samples of staff registered nurses working in intensive

care units and non-intensive care units of a large metropolitan hospital.

Burnout consistently was negatively correlated with psychological

hardiness. This finding supports hardiness as a generalized resistance

resource for nurses working in diverse patient care settings.

Social Support

Effective social support systems are consistently recommended as a

means of coping with job-related stress and preventing burnout. In 1982,

Kimmel reported that nurses who perceived greater emotional support than.









demands at home did not experience different levels of burnout than

nurses who experienced greater household demands than emotional support

at home.

Paredes (1982) studied the effects of social support and

psychological resources on the relationship between burnout and job

dissatisfaction. The findings showed a significant negative relationship

between supervision and co-worker support and burnout. Supervisor

support was more significant in burnout reduction than co-worker or off-

the-job support. Nurses who reported high levels of perceived

psychological support appeared to benefit most from social support in

terms of burnout reduction and job satisfaction.

Kanner, Kafry, and Pines (1981) and Pines, Aronson, and Kafry (1981)

found that on-the-job and off-the-job social support were negatively

related to burnout. Constable and Russell (1986) measured supervisor

support and found that this variable was a major predictor of burnout.

However, there was a significant association only with the emotional

exhaustion dimension of burnout. These results indicate that high levels

of support from supervisors can directly decrease feelings of emotional

exhaustion and, therefore, affect the potential for burnout among nurses.

In addition, their findings indicated significant moderating effects of

supervisor support on the relationship between job enhancement and

emotional exhaustion.

Finally, Chiriboga and Bailey (1986) reported that nurses who relied

more on their supervisor for support proved more likely to report

burnout. Additionally, reliance on co-worker support also correlated








with burnout. The correlations between inadequate psychological support

and higher burnout are consistent with the findings of Edelwich and

Brodsky (1983), Freudenberger and Richelson (1980), and Pines, Aronson,

and Kafry (1981).

Coping Behavior

Another important variable in the study of burnout is coping

behaviors used to reduce job stress. Kimmel (1982) measured coping with

the Ways of Coping Scale by Lazarus as well as role conflict, role

ambiguity, household support, demographic characteristics and burnout.

His sample included head nurses, supervisors, staff nurses, licensed

practical nurses, and nurses aids from a metropolitan hospital in New

York. He reported that coping variables were the best predictors of

burnout. Chiriboga and Bailey (1986) also measured coping with the Ways

of Coping Scale, and burnout. Their results indicated that the nurses

most vulnerable for burnout are younger and unmarried, work in critical

care units, have less involvement in work conditions, experience more

work hassles and distractions in the work place, and are more reliant or

nursing supervisor for support. In addition, these nurses used

anticipated coping which was significantly associated with burnout.

However, because only one variable out of nine measures of coping

strategies contributed significantly to burnout in their study, they

concluded that coping variables were not the best predictors of burnout.

In contrast, Stone et al. (1984) found that nurses endorsing a higher

number of effective coping skills suffer less burnout.









Organizational Characteristics

The organization variables include workstress, the work environment,

job enhancement and job satisfaction.

Work stress

Work stress has been identified as another contributing factor of

burnout. It has been consistently reported that (Chiriboga & Bailey,

1986; Jenkins & Ostehya, 1986) nurses who experienced more frequent

work-related stress reported greater burnout. According to Chiriboga and

Bailey (1986), the work stress variables were second only to work

environment measures in their degree of contribution to the explained

variance in burnout. Work hassles and work distractions were the

significant contributors. The more hassled the nurses felt, the more

likely they were to feel burned out. It is important to point out that

these are not the big stressful events; they are the smaller day-to-day

hassles that can exert a cumulative effect. The distraction variable

focused on trivial things: interruptions in the nurse's work, due to

physicians, nursing staff, or visitors. Nurses who reported more hassles

also reported more burnout. Their findings suggest that distracting and

annoying day-to-day stress in the hospital work environment exerted the

major stress on nurses, and that may lead to burnout.

Work Environment

Work environment variables including worker involvement, type of

unit, type of hospital, amount of patient contact and job enhancement

(variety of tasks and new approaches, autonomy, clarity, physical comfort

and work pressure) are strongly associated with burnout. Worker









involvement was the most important variable and explained the greatest

proportion of variance in burnout suggesting that when staff nurses feel

a low involvement in work, they are most at risk of burnout (Chiriboga &

Bailey, 1986). Burnout has been shown to be related to the type of unit.

There is evidence that units that provide fewer demands on the nurse may

potentiate a condition leading to Burnout (Chiriboga & Bailey, 1986). In

addition, nurses who work in the private hospital are more likely than

nurses employed by a district hospital to report burnout (Chiriboga &

Bailey, 1986).

Researchers have found that more hours worked and more hours of

direct patient contact increased the risk of burnout (Dames, 1983; Des,

1981). In addition, they also stated that the more hours working with

patients with grave prognoses, the higher the burnout scores.

The results offered by Constable and Russell (1986) suggested that

job enhancement was significantly correlated with burnout. This finding

indicates that nurses are more susceptible to burnout when working in

environments where there is a lack of variety and new approaches, there

is a lack of encouragement to be self-sufficient, rules and policies are

not clearly communicated, tasks are not clearly understood, and the work

environment is not considered comfortable and attractive. In addition,

work pressure was also positively associated with burnout-emotional

exhaustion. Critical care nurses experiencing burnout perceive the

critical care environment with little support and the events within the

unit as a threat. Consequently, the critical care nurse expects the

worst (Stone et al., 1984).









Job Satisfaction

The study of job satisfaction has evolved since the early 1900s.

The earlier work was done by Taylor (1911) who determined that job

satisfaction was associated with the amount of pay earned. This research

began during the industrial revolution when workers were equated to

machinery and efficiency was the expectation (Slavitt, Stamps, Piedmont,

& Haase, 1978).

In the 1930s a humanistic-sociological approach to job satisfaction

emerged. Work satisfaction was established as an oversatisfaction with

life (Hoppock, 1935). A cross-sectional study of industrial workers with

similar abilities, interests and job preparation revealed that an

employee's adaptability to the situation, the employee's socioeconomic

group identification, ability to relate to others, and the nature of the

work were associated with work satisfaction. The psychological basis for

job satisfaction was established by the Hawthorne studies (Mayo, 1945).

These studies found that group interaction was the major determinant of

job satisfaction (Slavitt et al., 1978).

One theory that is widely used today in studying job satisfaction is

the motivation-hygiene theory (Herzberg, 1968). Motivators was the term

Herzberg used to identify job satisfiers because they were effective in

contributing to the individual functioning at a higher level and resulted

in job satisfaction. Examples of motivators include responsibility,

advancement and recognition. Hygiene factors was the term used to

identify lower level needs because if not met, they added to

dissatisfaction but when met did not contribute to satisfaction.









Examples of hygiene factors include salary, supervision, policy, and

working conditions.

The theoretical conceptualization of job satisfaction by Herzberg

has been studied extensively among nurses (Hinshaw & Atwood, 1984, 1987;

Slavitt, Stamps, Piedmont, & Haase, 1978). Hinshaw and Atwood (1987)

reviewed the job satisfaction literature and identified 19 significant

factors relating to job satisfaction which were divided into two

categories: personal and work-related characteristics. Personal

characteristics included age, gender, intelligence, educational level,

experience as a nurse, tenure and position in the hierarchy. Work

related characteristics included the specialty area, nursing care

delivery model, supervision, tasks, outcomes and pay. These researchers

concluded that although these variables have been identified as

significantly associated with job satisfaction, the relative impact of

the personal and organizational variables acting directly and indirectly

with job satisfaction has not been determined.

The more relevant theory of job satisfaction related to this study

is the person-environment fit model of occupational stress (French,

Rodgers, & Cobb, 1974). This theoretical approach relates occupational

stress to job satisfaction. As a result of the interaction between the

person holding a job and the environment in which he or she is employed

with good person-environment fit, the job provides the necessary needs of

the individual (salary, fringe benefits, social involvement, opportunity

to achieve, and a sense of self worth). On the other hand, if the job is

too strenuous or too demanding, job stress and burnout may result. Job









satisfaction is described as the result of a good person-environment fit

(Vachon, 1987).

There are only two studies to date linking job dissatisfaction and

burnout for nurses. Job dissatisfaction accounted for the largest

variance in burnout in a dissertation study conducted by Parades (1982).

Younger age, low psychological resources, shorter length of time on the

hospital unit and less social support also contributed to the total

variance. In addition, nurses experiencing burnout are found to be

dissatisfied with their jobs because there are limited opportunities for

personal growth on the job (Stone et al., 1984).

Theoretical Framework

The burnout model proposed by the researcher is an adaptation of

the burnout model offered by Perlman and Hartman (1982) which integrates

many theoretical models. It evolved from Selye's (1974) identification

of stimuli and stressors, the definition of stress as the general

adaptation syndrome, as the theory that too much stress may lead to

maladaptation. The model is further grounded in a framework proposed by

House and Wells (1978), Beehr and Neuman (1978) and Matteson and

Ivancevich (1979).

Perlman and Hartman (1982) posited a transactional model of burnout

and this appears to be the trend in recent years (Cherniss, 1980a;

Chiroboga & Bailey, 1986; Cooper, 1986; Courage & Williams, 1987; Cox &

McKay, 1981; Fletcher & Payne, 1980; Golembiewski et al., 1986; Harrison,

1983). The major theme of the transactional model relates to the

conceptualization of stress and burnout as the result of a transaction









between individual worker needs and resources and the demands,

constraints, limitations, and/or facilities within the work environment.

Their model is broad and allows an examination of personal and

organizational-work variables which have been studied individually or in

combination in the burnout literature from 1977 to the present. In this

model personal, as well as organizational-work variables that are

subjectively perceived, are conceptualized.

Using Selye's (1974) definition of stress as "The nonspecific

response of the body to any demand made upon it" (p. 41), Perlman and

Hartman (1982) built a theoretical model of psycho-social organizational

mediated adaptation congruent with Selye's initial work. The Perlman and

Hartman (1982) model incorporated three major symptom categories of

stress that are reflected by the three dimensions of burnout:

(a) physiological (focusing on physical symptoms and perceptions of

reduced personal accomplishment); (b) affective cognitive (focusing on

attitudes and feelings and emotional exhaustion); and (c) behavioral

(focusing on symptoms or behavior related to depersonalized care of the

recipient of care). The model, which has a cognitive and perceptual

focus, consists of a linear progression of four stages of stress which

include the degree to which the situation is conducive to stress, the

perception of job stress, response to stress, and outcome of stress.

Groupings of significant personal variables and organizational variables

are related to each stage. In the first stage of stress the degree the

situation is conducive to stress is a function of personal variables of

ability, time, family demands, and job expectations and organizational









variables of workload, expected role performance and role ambiguity. The

combination of the variables age and education are theorized by Perlman

and Hartman (1982) to represent the variable ability. The combination of

the variables marital status and life events is theorized by Perlman and

Hartman (1982) to represent the variable family demands. The variable

job expectation is theoretically viewed by Perlman and Hartman (1982) to

represent perception of growth for helping professionals. In the second

stage of stress the level of perceived job stress is a function of

personal variables of personal work needs, and personality and

organizational variables of supervision, work group norms, opportunity

for advancement and dealing with colleagues at work. In the third stage

of stress the response to stress is a function of physiological,

affective and cognitive responses manifested as burnout. This stage is

related to personal variables including commitment to career, and health

and organizational variables that include pay, support from peers,

supervision, and organizational climate and the economic and market

condition. In the fourth stage, burnout is related to work outcomes

including job satisfaction and psychological response.

The model includes a complete representation of transactional

processes including personal characteristics, organizational

characteristics, the degree to which a situation is likely to be

stressful, perception of organizational job stress, and response to job

stress in understanding the etiology of burnout. Although the

conceptualization of burnout offered by Perlman and Hartman (1982) is

fairly comprehensive there are limitations with respect to the

operationalization of two variables labelled personal work needs and work

group norms. Multiple indicators need to be used to operationalize these









variables that are identified in the literature review for this study

that have a meaningful relationship in the process that leads to job

dissatisfaction and burnout. For example, according to the literature,

meaningfulness of work, responsibility for work outcomes and knowledge of

work outcomes have a direct relationship to job satisfaction (Herzberg,

1968; Hinshaw & Atwood, 1984, 1987; Oldham, Hackman, & Pearce, 1976).

Because job dissatisfaction has been linked to burnout, use of these

multiple indicators of personal work needs seems warranted. In addition,

according to the literature, autonomy, skill variety, task significance,

and task identity have a direct relationship to the process of job

satisfaction and job burnout (Cherniss, 1980a-b; Heifitz & Bersoni,

1983). Use of these variables to operationalize work group norms has

theoretical support and is justified.

Another problem is that the model conceptualized by Perlman and

Hartman (1982) has structural limitations of its components. Each

variable affects only the next continuous variable in the model. They

did not allow for the freedom of some variables having direct and

indirect causal relationships to more than one outcome variable. For

example, according to the literature, job satisfaction has a direct

relationship to pay, dealing with others at work, opportunity for

advancement, and economic market conditions (Jayaratne & Chess, 1983;

Oldham, Hackman, Pearce, 1976; Paredes, 1988) but from Figure 1-1 it

would appear that burnout has a direct relationship to pay, support from

others, and economic market conditions and that job stress has a

relationship with opportunity for advancement. In addition, Perlman and

Hartman (1982) seemed to suggest that burnout precedes job satisfaction,

but other theoretizations of burnout would equate it with what Perlman









and Hartman defined as maladaptation, manifested as an observable form of

ineffective coping.

A more logical approach is the expansion of the model offered by

Perlman and Hartman (1982) to a path-analytical model that will allow for

the analysis of direct and indirect effects of personal and

organizational variables simultaneously across stages. There is a need

to attend to a more comprehensive causal model of stress and burnout that

includes personal and organizational variables as well as stages of

stress, job satisfaction, and psychological response in understanding the

etiology of burnout.
















CHAPTER III
MEET1I''DOLOGY

This study was designed to collect data and to test a path analytical

model of the theoretical conceptualization of burnout for critical care

nurses. The model of interest was an expansion of the model offered by

Perlman and Hartman (1982) that allowed analysis of direct and indirect

effects of exogenous and endogenous variables using path analysis. Five

research study questions were used to guide this study in an exploratory

examination of the full model of burnout.

This chapter is divided into four section. In the first section,

the sample of respondents is described. The study design and procedures

used to collect the data are included in the second section of the

chapter. The instrumentation is described in the third section. The

fourth section includes a description of the statistical analysis of the

five research questions used to guide the study in examining the model.

Sample of Respondents

Selection of research participants. Permission to conduct the stud.

was first obtained through the Institutional Review Board of the

University Health Center. Following approval, a convenience sample of

hospitals was selected. Eligible hospitals (N = 14) were identified fru:

the 1985 Hospital Data Report published by the North Central Florida

Health Planning Council. Permission to conduct the study was then

obtained through the Nursing Research Committees of hospitals located in

North Eastern, North Central and Southern regions of the state of









the state of Florida. Administrators of five of the selected hospitals

declined to participate due to the perceived threatening tone of the Job

Diagnostic Survey Questionnaire and the burnout questionnaire. The final

sample consisted of nine hospitals which are licensed for between 250 and

450 beds, and are located in Northeastern, Northwestern, North central,

and Southern regions of Florida. All staff nurses employed in the nine

hospitals who fit the criteria of working on a critical care unit and

having worked full time for at least three months were invited to

participate in the study. During the months of September-November, 1989,

the burnout test battery was presented to the nurses by the author during

a scheduled staff meeting (Appendix A). Consent for participation was

then procured from the critical care nurse employed in these settings

(Appendix B).

A total of 375 female critical care registered staff nurses were

given a package of materials with a cover letter requesting their

participation. A total of 330 nurses agreed to participate by returning

complete questionnaires (86%). Thirty questionnaires were not fully

completed and were excluded from the data analysis. Three hundred

questionnaires were usable for data analysis.

The demographic characteristics of the critical care staff nurses

(age, marital status, sex, and education) are presented in Table 3-1.

These data revealed that all the critical care nurses in the sample were

female and that 75% of the nurses were between the ages of 20 and 39 with

32% under the age of thirty. The majority of the staff nurses held an

associate degree (50%) in nursing (ADN) and 10% were diploma graduates.

Only 34% were baccalaureate graduates in nursing.

The duration of experience as a staff nurse in critical care is

presented in Table 3-2. The duration ranged from 3 months to 30 years









Table 3-1

Characteristics of Critical Care Staff Nurses:
Sex, Age, and Education (N = 300)


Characteristics Number Percent



Sex:

Female 300 100.0

Marital Status:

Single 71 23.7

Married 188 62.7

Separated 6 2.0

Divorced 35 11.7

Total 300 100.1

Age:

20 29 96 32.0

30 39 129 43.0

40 49 60 20.0

50 59 14 4.7

60 or over I 0.3

Total 300 100.0

Education:

ADN 149 50.0

BSN 102 34.0

Diploma 31 10.3

Some graduate work 18 6.0

Total 300 100.3









Table 3-2

Duration as Staff Nurse in Critical Care (N = 300)


Number of years Number Percent Cumulative %



Duration as staff nurse
in critical care:

3 6 months 60 20 20

7 months 1 year 49 16.3 36.3

13 18 months 11 3.7 40.0

19 months 2 years 2 .7 40.7

2 3 years 15 5.0 45.7

3 4 years 25 8.3 54.0

4 5 years 9 3.0 57.0

5 7 years 34 11.7 68.7

7 10 years 31 10.0 78.7

10 15 years 43 14.3 93.0

15 20 years 14 4.7 97.7

20 25 years 5 1.6 99.3

25 years + 2 .7 100.0

Total 300 100.0 100.0









with a mean of 6.86 years and a standard deviation of 5.53 years. Over

45% had 3 years or less of work experience as a staff nurse in critical

care, and 46% had more than 4 years of experience.

Additional demographic information included the duration of

experience as a staff nurse in critical care at the present institution

and the number of hours working per week. The number of years of

experience as a staff nurse in critical care at the present institution

ranged from 1 month to 25 years with a mean of 4.75 years and a standard

deviation of 4.19 years. Over 67% had 5 years or less of work experience

in critical care staff nursing at the present institution and 8% had more

than 10 years of experience at the present institution (see Table 3-3).

The number of hours working per week ranged from 36 hours to 72 hours

with a mean of 38.16 hours and a standard deviation of 7.95 hours. It

was revealed that over 16% had 42 hours or more per week working time

(Table 3-4).

Study Design and Procedures

The design structure for the path analysis methodology was

correlational, involving selection of a number of psychometric

instruments to operationalize each of the variables specified in the

theoretical framework and administration of these instruments to all

subjects in the study. Subjects were asked to complete five

questionnaires, including the demographic sheet, and Work Survey

Instrument, the Daily Hassles Scale, the Job Diagnostic Survey, the

Hardiness Scale, and the Burnout Inventory. Each nurse self-administered

the burnout test battery during an off-work, scheduled, thirty-minute

period. The test battery included a set of standardized instructions for

each of the measures. All subjects were guaranteed anonymity.









Table 3-3

Years of Experience as a Critical Care Staff Nurse at the
Present Institution




Number of years Number Percent



6 months or less 16 5.3

7 months 1 year 29 9.7

13 months 2 years 60 20.0

2 5 years 97 32.3

5 10 years 74 24.7

10 15 years 16 5.3

15 20 years 7 2.3

20 25 years 1 .3

Total 300 100.0









Table 3-4

Work Hours Per Week (N = 300)


Number of hours Number Percent



40 hours 249 83.0

41 42 hours 3 1.0

43 44 hours 5 1.7

45 48 hours 28 9.4

49 50 hours 8 2.6

51 60 hours 5 1.6

61 70 hours 2 .7

Total 300 100.0









Instrumentation

A demographic questionnaire and work survey devised by the

researcher and four widely used standardized instruments including The

Daily Hassles Scale, Job Diagnostic Survey (JDS), Psychological Hardiness

Test, and Maslach Burnout Instrument were used in this study in order to

assess individual critical care nurses' perceptions of personal and

organizational variables in the work environment, levels of stress

including a situation conducive to stress, perceived job stress, and

burnout and outcomes of the stress response including job satisfaction

and internal job motivation.

Demographic Questionnaire

Description. A six-item questionnaire was constructed in order to

assess personal characteristics of the critical care nurses. The

personal characteristics assessed were age, sex, marital status,

education, duration as a staff nurse in critical care, duration as a

staff nurses in critical care at the present institution, and the number

of working hours per week in critical care nursing. Appendix C contains

this instrument.

Use in this study. The personal variables, ability and time

(tenure), were measured with items from the demographic questionnaire

(Appendix C). The measurement of ability consisted of the sum of the

duration of time as a staff nurse in critical care in months and the sum

of the duration of formal education in nursing months. The measurement

of tenure consisted of the numerical value of duration in months as a

staff nurse in critical care and the duration in months as a staff nurse

at their present institution. The means and standard deviation for the

sample used in the present study are presented in Table 3-5.









Table 3-5

Psychometric Properties of Variables: Ability and Time



Standard
Variable Number Mean deviation



Ability 300 117.82 68.858

Time 300 139.33 103.192









Daily Hassles Scale

Description. The measurement of psychological stress through the

use of the Hassles Scale theoretically approaches the role of cognitive

appraisal in the stress and coping process (Lazarus & Folkman, 1989). In

this approach, coping depends on how individuals appraise their everyday

encounters with the environment with regard to the extent to which those

encounters are perceived as threatening, harmful, or challenging.

Everyday encounters are called hassles and are measured in the Hassles

Scale. The Hassles Scale consists of 117 items used to measure the

frequency and severity of a person's transactions with the environment

that are considered by the individual to be stressful events; e.g., "How

much of a hassle was this for you . overloaded with family

responsibilities?" The scale takes approximately 10 minutes to complete.

The response format is a four-point scale, ranging from "none" or not

applicable to "extremely severe." The Daily Hassles Scale yields two

scores: frequency, which is the number of hassles endorsed by the person

without regard to severity and severity, which is the average severity

rating of all items that have been identified. Eight factor-based

subscales scores are also possible. They include future security, time

pressures, work, household responsibilities, health, inner concerns,

financial responsibilities, and neighborhood/environmental.

Psychometric Properties Of Daily Hassles Scale

Reliability. Kanner et al. (1981) reported a test retest

reliability coefficient of .79 on a sample of 432 college students and a

sample of 448 adults aged 20-60. There is no information available on

the internal consistency of the scale.









Validity. In reference to construct validity, an important

relationship has been explored between hassles and life events scores and

the treatment of both as indicators of psychological stress. Kanner et

al. (1981) provide the strongest evidence for construct validity of the

Daily Hassles Scale in order to explain psychological symptoms and

symptoms of somatic illness and emotional distress. His group found that

the hassles scores were strongly related to both affective distress and

psychological symptoms (.34). The Hopkins Symptom Checklist (HSC1)

(Derogotis et al., 1974) correlated between .5 and .6 with the Daily

Hassles Scale. Delongis et al. (1988) correlated hassles and somatic

health status as dependent variables and found that daily hassles

explained more variance than did life events.

Lazarus and Folkman (1989) performed a factor analysis of the Daily

Hassles Scale, using the principal factor method with oblique rotation

and generated eight factors: future security, time pressure, work,

household responsibilities, health, inner concerns, financial

responsibility, and neighborhood and environmental concern.

Use in the study. The personal variables family demands, health,

the situational stress and job stress variables were measured with the

Daily Hassles Scale on the basis that the scale contained factors that

corresponded to the variables used in this study. Family demands were

measured with the Daily Hassles Scale, items 1-4, 7-15, 19-23, 29, 35-3M,

43, 45, 55, 58-59, 61-64, 70, 71, 73-76, 78-79, 82, 87, 90, 92-93,

103-108, and 110-112 that corresponded to three factors: household

responsibilities, financial responsibility and neighborhood and

environmental concern (Lazarus & Folkman, 1989).









Health was measured with the Daily Hassles Scale, items 16-18, 48,

49, 50, 53, 54, 56, 77, 91, and 98 that corresponded to the health factor

(Lazarus & Folkman, 1989).

The degree to which the situation was conducive to stress was

measured with items 5, 6, 24-26, 39-42, 44, 46-47, 51-52, 57, 60, b7-69,

85, 88-89, 94-97, 99-101, 109, 113-117, of the Hassles Scale that

corresponded to the three factors inner concerns, future security, and

time pressure (Lazarus & Folkman, 1989).

The level of perceived job stress was measured with items 27-28,

30-34, 65, 66, 72, 80-84, 86, and 102 pertaining to work hassles of the

Hassles Scale (Lazarus & Folkman, 1989).

Critical care nursing faculty, critical care graduate nursing

students, and two top nursing management members of two critical care

intensive care units of a regional hospital were asked to rate and

evaluate this scale for use in this study. On the basis of the

reliability and validity data available, use of the Daily Hassles Scale

as a measure of family and life demands, health, situation conducive to

stress, and work stress seemed justified. The means, standard deviations

and coefficient alpha estimates for the variables for the sample used in

the present study are presented in Table 3-6. Appendix D contains

instrument directions and sample items.

Job Diagnostic Survey (JDS)

Description. The JDS is composed of 83 items used to diagnose

existing jobs to determine if and how they might be redesigned to affect

employee motivation and performance and to evaluate the effects of job

changes on employers. The measure is based on a theory of how a job

affects worker motivation and provides measures of objective job

dimensions, individual psychological states related to these dimensions,









Table 3-6

Psychometric Properties of Variables: Daily Hassles Scale




Standard
Variable Number Mean deviation Alpha



Family demands 300 .751 .408 .701

Health 300 .501 .396 .699

Situational stress 300 .717 .387 .703

Job stress 300 .659 .427 .714









affective reactions of employees to the work setting and the job, and the

need of the individual for growth. The scale is appropriate for grades 8

and over. The subjects respond to a seven-point response scale used

throughout the instrument.

The JDS yields scores on 18 subscales. All 18 of these subscales

were used as measures of variables in the present study. Table 3-7

relates JDS subscales to variables and gives sample items. The JDS

provides measures of the five core work dimensions which include skill

variety, task identity, task significance, autonomy and feedback from the

job itself. Two additional measures are included for two supplementary

dimensions which include feedback from agents and dealing with others.

They are defined as follow:

Skill variety is the degree to which a job requires a variety of

different activities in carrying out the work, which involve the use

of a number of different skills and talents of the employee.

Task identity is the degree to which the job requires completion of

a "whole" and identifiable piece of work--i.e., doing a job from

beginning to end and with a visible outcome.

Task significance is the degree to which the job has a substantial

impact on the lives or work of other people--whether in the

immediate organization or on the external environment.

Autonomy is the degree to which the job provides substantial

freedom, independence, and discretion of the employee in scheduling

the work and in determining the procedures to be used in carrying it

out.

Feedback from the job itself is the degree to which carrying out the

work activities required by the job results in the employee










Table 3-7

Job Diagnostic Survey Subscales, Related Variables and
Sample Items


Variable in
Subscale this study Sample Item


Total growth
need strength







Meaningfulness
of work


Responsibility
for work
outcomes


Knowledge of
work results


Feedback from
the job itself


Job expectations








Personal work needs


Expected performance
and role ambiguity


Feedback from
agents on the job


Skill variety


Work group norms


Task identity


"Indicate the degree to
which you would care to
have this characteristic
present in your job; e.g.,
opportunities for personal
growth and development in
my job."

"Most of the things I have
to do on this job seem
useless or trivial to me."

"I feel a very high degree
of personal responsibility
for the work I do on this
job."

"Most people on this job
have trouble figuring out
whether they are doing a
good or a bad job."

"Just doing the work
required by the job
provides may choices for me
to figure out how well I am
doing."

"Supervisors often let me
know how well they think I
am performing on the job."

"The job requires one to
use a number of complex or
high-level skills."

"The job provides me the
chance to completely finish
the pieces of work 1
begin."










Table 3-7--continued


Variable in
Subscale this study Sample Item


Task significance


Autonomy


Dealings with
others at work


Supervision
satisfaction


Satisfaction with
work growth and
advancement

Pay satisfaction


Satisfaction with
peers and
co-workers

Satisfaction with
job security


General job
satisfaction


Internal job
motivation


Work group norm


Work group norm


Dealing with
colleagues


Supervision



Opportunity for
advancement


Support from others
at work


Organizational climate
and the economic-
market condition

Job satisfaction



Psychological response
to job stress


"The job is one where a lot
of the other people can be
affected by how well the
work gets done."

"The job gives me
considerable opportunity
for independence and
freedom in how I do the
work."

"The job requires a lot of
cooperative work with other
people."

"The amount of support and
guidance I receive from my
supervisor.

"The amount of personal
growth and development I
get in doing my job."

"The amount of pay and
fringe benefits I receive.

"The chance to get to knuw
other people on the job."


"How secure things look tn r
me in the future in this
organization."

"I am generally satisfied
with the kind of work I dG
in this job."

"Most people in this job
feel a great sense of
personal satisfaction when
they do the job well."









obtaining direct and clear information about the effectiveness of

his or her performance.

Feedback from agents is the degree to which the employee receives

clear information about his or her performance from supervisors or

from coworkers.

Dealing with others is the degree to which the job requires the

employee to work closely with people in carrying out the work

activities.

The JDS also measures three psychological states that are associated

with the core job dimensions. These include experienced meaningfulness

of the work, experienced responsibility for work outcomes and knowledge

of work results. They are defined as follow:

Experienced meaningfulness of the work. The degree to which the

employee experiences the job as one which is generally meaningful,

valuable, and worthwhile.

Experienced responsibility for work outcomes. The degree to which

the employee feels personally accountable and responsible for the

results of the work he or she does.

Knowledge of results. The degree to which the employee knows and

understands, on a continuous basis, how effectively he or she is

performing the job.

Personal, affective reactions or feelings a person obtains from

performing the job include general satisfaction, internal job motivation,

and specific satisfactions (job security, pay and other compensations),

social satisfaction with peers and coworkers, supervision and opportunity

for personal growth and development on the job. They are defined as

follow:









General satisfaction. An overall measure of the degree to which the

employee is satisfied and happy with the job.

Internal job motivation. The degree to which the employee is self

motivated to perform effectively on the job--i.e., the employee

experiences positive internal feelings when working effectively on

the job, and negative internal feelings when doing poorly.

Specific satisfaction. A number of short scales which provide

separate measures of satisfaction with:

(a) Job security

(b) Pay and other compensation

(c) Peers and coworkers ("social" satisfaction)

(d) Supervision

(e) Opportunities for personal growth and development on the job

("growth" satisfaction)

Finally, the JDS measures the strength of the individual's desire to

obtain growth satisfactions from his or her work. This measure is called

the growth need strength index and is viewed as an individual

characteristic which is predicted to influence how well a worker will

react to a job with a high motivating potential (Hackman & Oldham,

1978).

Psychometric Properties of JDS

Reliability. Internal consistency reliabilities for the eighteen

subscales are reported as generally satisfactory with a range from a high

of .88 to a low of .56. The term scale is used to refer to the summary

score obtained for each variable measured by the JDS. On the basis of

these results the reliability estimates are satisfactory. The

psychometric properties of the variables on this sample are presented in

Table 3-8.










Table 3-8

Psychometric Properties of Variables: Job Diagnostic Survey





Standard
Variable Number Mean deviation Alpha



Job expectations 300 4.103 2.446 .714
(Total growth
strength need)

Meaningfulness of 300 5.750 .844 .669
work

Responsibility for 300 5.750 .653 .680
work outcomes

Knowledge of work 300 5.000 .946 .679
results

Expected role 300 4.668 .898 .670
performance and
role ambiguity
(Feedback from the
job itself and
feedback from
agents on the
job)

Skill variety 300 5.916 .818 .675

Task Identity 300 4.212 1.112 .691

Task 300 6.152 .772 .684
significance

Autonomy 300 5.090 .905 .689

Supervision 300 4.950 1.270 .677

Dealing with others 300 6.116 .707 .700
at work

Opportunity for 300 5.459 .893 .661
advancement


300 3.903


Pay satisfaction


1.608 .693









Table 3-8--continued


Standard
Variable Number Mean deviation Alpha



Social support at 300 5.639 .766 .673
work

Job security 300 5.207 1.224 .695
satisfaction

Job satisfaction 300 4.949 1.110 .675

Internal job 300 5.874 .640 .680
motivation









Validity. Convergent validity was demonstrated by Hackman and

Oldham (1978). Assessments of the specified jobs on the job dimensions

were made by the employers who worked on those jobs as well as by the

supervisors and the researchers. The ratings of each group were averaged

for each job, and then correlations were computed. The median of the

correlations between employees and supervisors is .51; between employers

and observers is .63; and between supervisors and observers is .46. In

general, the ratings of the three groups converge moderately well.

The validity of the JDS has been demonstrated further by data that

confirm hypotheses about the relationship of certain job characteristics

and experienced burnout. Hackman and Oldham (1974) found scores on the

JDS and the Maslach Burnout Inventory (MBI) were closely correlated. In

particular, high scores on the job dimension "feedback from the job

itself" were associated with low scores on emotional exhaustion and

depersonalization, and were high scores on personal accomplishment of the

burnout inventory (Maslach & Jackson, 1986). Another job dimension,

"dealing with others," was found to be weakly correlated with emotional

exhaustion. A third job dimension, "task significance," was highly

positively correlated with personal accomplishment.

Additional validation of the JDS is provided by data that confirm

hypothetical relationships between the JDS measure of "growth

satisfaction" and burnout. Scores on the JDS measure on "growth

satisfaction" were negatively correlated with emotional exhaustion and

depersonalization and positively correlated with personal accomplishment.

In addition, employees scoring low on the JDS subscale of "experienced

meaningfulness of the work" scored higher on depersonalization and lower

on personal accomplishment. With respect to coworkers, employees scoring








low on the JDS subscale of peer and coworker satisfaction' scored high

on emotional exhaustion and depersonalization, and low on personal

accomplishment.

Further evidence of the validity of the JDS has been obtained by

distinguishing its subscales by off diagonal median correlations which

provide an indication of discriminant validity of the items. The range

is from .12 to .28. In addition, distinguishing the JDS job satisfaction

subscale from measures of other psychological constructs including

burnout has been supported. Job satisfaction had a moderate negative

correlation with both emotional exhaustion (-.23) and depersonalization

(-.22), as well as slightly positive correlation with personal

accomplishment (.17).

Finally, a preliminary study has been performed in which factor

analysis was used to determine if the JDS measures a multidimensional

construct. The results of these studies are not enough to eliminate the

conceptualization of a multidimensional construct of job characteristics

measured by the JDS. Additional research is required to determine this

issue. Despite the high intercorrelations among the subscales, Hackman

and Oldham (1978) suggest that separate subscale interpretations are

probably warranted but should be made cautiously.

On the basis of the reliability and validity data available, use of

the JDS as a measure of personal and organizational variables seemed

justified.

Use in this study. The personal variables including job

expectations, and personal work needs and organizational variables,

including expected performance and role ambiguity, work group norms,

dealing with colleagues, supervision, opportunity for advancement, pay,









support from others, organizational climate and the economic-market

condition, general job satisfaction and psychological response to burnout

(internal work motivation) were measured with the JDS on the basis that

the scale contained measures that closely resembled the variables used in

this study. In addition, critical care nursing faculty, critical care

graduate nursing students, and two top nursing management members of two

critical care intensive care units of a regional hospital were asked to

rate and evaluate this scale for use in this study. On the basis of

their judgement, and the decision of this researcher, use of the JDS as a

measure of the personal and organizational variables seemed justified.

Appendix E contains instrument directions and sample items.

Psychological Hardiness Test

Description. Hardiness is defined as a learned ability to cope with

a wide variety of stressful situations in such a way that the stresses

are transformed to a positive outlook, or negative stresses, that cannot

be realistically transformed, are met with a plan to eliminate them.

Another way of describing "hardiness" is transformational coping which is

a learned process. The amount of hardiness one has acts as a buffer to

keep stress from changing into strain, and strain from changing into

illness. A sample item of the Psychological Hardiness Test includes, "I

often wake up eager to take up my life where it left off the day before."

The measurement of hardiness consists of 50 rating-scale items developed

by Kobasa, Maddi, Donner, Merrick, and White (1984). The 50 items were

derived from a factor analysis of six existing personality scales which

have been used to measure commitment, control, and challenge dimensions

of hardiness. The Hardiness Test was standardized on a population of 223

women and 1511 men who for the most part were professionals--business

executives, lawyers, white collar professional.









The hardiness scores is computed by transforming raw subscale scores

into standard scores and adding across the three subjscales to produce a

total score for each subject. Commitment is defined as the ability to

commit to the task or project or relationship. Control is defined as the

realistic knowledge and use of the amount of control or lack of control

that one has in this and other situations. Challenge is defined as the

use of both commitment and control in order to see events, relationships,

problems and opportunities as challenges rather than as trouble.

Psychometric Properties of Psychological Hardiness Test

Reliability. As to reliability, the internal consistency estimates

based on 10,000 subjects over the past four years from all walks of life

and a multitude of circumstances has yielded a coefficient alpha of .92

for total hardiness score a mean of 74.02 and a standard deviation of

9.60. Stability appears to be about .960 over a period of two weeks.

The internal consistency estimate for this sample was .89 with a mean

score of 69.31 and a S.D. of 11.01.

Validity. The validity of the hardiness test has been demonstrated

by data that confirm hypotheses about the relationship of hardiness to

burnout. Keane et al. (1985) reported that nurses who exhibited less

psychological hardiness reported more burnout. In addition, they

reported that perceived job stress and hardiness were significant

additive predictions of burnout. Hardiness appeared to have beneficial

effects on decreasing burnout.

Kobasa et al. (1982) related commitment and coping in stress

resistance among lawyers and identified that increases in strain are

significantly determined by personality characteristics of alienation

(vs. commitment). In another study, Kobasa (1979) reported discriminant

function analysis with the prediction that high stress/low illness

executives show by comparison with high stress/high illness executives,









more hardiness, that is, have a stronger commitment to self, an attitude

of vigor toward their environment, a sense of meaningfulness, and an

internal locus of control.

This evidence suggests that on the basis of the reliability

estimates and validity data available, the total hardiness scores

provides a measure of general psychological hardiness. However, there is

less empirical support for the interpretation of separate subscale scores

of control, commitment, and challenge.

Uses in this study. The personal variable personality was measured

with the Hardiness Test on the basis that this scale would determine if

hardiness or transfunctional coping acts as a buffer to keep situations

conducive to stress from changing into job stress, and job stress from

changing into burnout among critical care nurses. In addition, critical

care nursing faculty, critical care graduate students and two top nursing

management members of two critical care intensive care units of a

regional hospital were asked to rate and evaluate this scale for use in

this study. On the basis of their judgement and the discretion of this

researcher, use of the Hardiness Test is a measure of the personal

variable of personality seemed justified. Appendix G contains

instrument directions and sample items.

Maslach Burnout Instrument (MBI)

Description. Burnout, according to Maslach and Jackson (1986), is a

syndrome of emotional exhaustion, depersonalization, and reduced personal

accomplishment, which often affects helping professionals (e.g.,

teachers, nurses and therapists). A major aspect of the burnout syndrome

is increased feelings of emotional exhaustion; as emotional reserves are

depleted, helping professionals feel that they are no longer able to give

of themselves at a psychological level. In addition, the development of

depersonalization--i.e., negative, cynical attitudes and feelings about









one's own clients occurs. A third aspect of the burnout syndrome is

reduced personal accomplishment, which refers to the tendency to evaluate

oneself negatively, especially with regard to one's work with patients.

Helping professionals may feel unhappy about themselves and dissatisfied

with their accomplishments on the job. The MBI is designed to measure

the three components of burnout. Each aspect is measured by a separate

subscale. The Emotional Exhaustion subscale assesses feelings of being

emotional, over extended, and exhausted by one's work. The

Depersonalization subscale measures an unfeeling and impersonal response

towards recipients of one's service, care, treatment, or instruction.

The Personal Accomplishment subscale assesses feelings of competence and

successful achievement in one's work with people. Each scale further

includes an expression of frequency using a six-point response format.

The MBI consists of a total of 22 items.

Psychometric Properties of MBI

Reliability. Reliability coefficients of internal consistency

ranged from .71 to .90. Test retest reliability (2-4 weeks apart) ranged

from .82 to .53. Standard errors of measurement ranged from 3.16 to

4.99. Reliability estimates, therefore, appear sufficient. The internal

consistency of the emotional exhaustion subscale on the sample in this

study was .72.

Validity. Convergent validity was determined in a variety of ways.

First, behavioral ratings made by an acquaintance of the individual such

as a coworker or a spouse were correlated with MBI scores. The validity

of the job is demonstrated further in studies confirming hypotheses about

the relationships between various job characteristics and experienced

burnout. In one study, it was predicted that the greater number of

clients one must care for, the higher the burnout scores on the MBI

(Maslach & Pines, 1977; Maslach & Jackson, 1982, 1984a-b).










The JDS and the MBI scores were correlated in research performed by

Pines and Kafry (1978). High scores on the job dimension, feedback from

the job itself, were correlated with low scores on depersonalization and

emotional exhaustion and high scores on personal accomplishment. The

dimension, "dealing with others," relates to the degree to which a job

requires the employee to work closely with people in carrying out the job

activities. High scores on this job dimension were weakly correlated

with emotional exhaustion. Task significance, a third dimension,

assesses the degree to which the job has an impact on the lives of other

people. High scores on this dimension were positively correlated with

personal accomplishment.

Further validation of the MBI has been provided by data that

confirms hypotheses related to experienced burnout and various outcomes

or personal reactions. In a study of nurses, and social service and

mental health workers scores on the JDS measure of "growth satisfaction"

were negatively correlated with depersonalization and emotional

exhaustion and positively correlated with personal accomplishment. Low

scores on the JDS subscale of "experienced meaningfulness of the work"

scored higher on depersonalization and lower on personal accomplishment.

In addition, low scores on the JDS subscale of "knowledge of results"

were correlated with high scores on emotional exhaustion and

depersonalization and with low scores on personal accomplishment.

Support for the hypotheses predicting that burnout would be related

to the desire to leave one's job was supported by Maslach and Jackson

(1979, 1982, 1984a-b). In addition, the impairment of one's

relationships with people in general, both on and off the job was highly

correlated with emotional exhaustion. Low scores on the JDS subscale









peer and coworker satisfaction correlated with high scores on emotional

exhaustion and depersonalization, and low on personal accomplishment.

Additional studies have linked burnout to outcomes of stress

including increased use of alcohol, drugs, and insomnia. Individuals

scoring high on emotional exhaustion were also rated as having problems

with insomnia. Police officers in one study were more likely to describe

having a drink to cope with stress if they had high scores on emotional

exhaust-ion. If they scored low on personal accomplishment they were

likely to report using tranquilizers.

Finally, further research determining the discriminant validity of

the MBI has been performed in order to distinguish burnout from measures

of other psychological constructs that might be assumed to be confounded

with burnout. Scores from the general satisfaction subscale measured by

the JDS and scores of the MBI has a moderate negative correlation, -.23

(emotional exhaustion); -.22 depersonalizationn as well as slightly

positive correlations with personal accomplishment (.17). Low

correlations between burnout subscale scores and other measures of job

satisfaction have been reported in other studies as well.

The MBI appears to be the best scale available in measuring burnout.

It is a well constructed instrument and the validity and reliability data

are sufficient to provide meaning and stability of the construct.

Use in this study. The variable burnout-emotional exhaustion was

measured with the Maslach Burnout Inventory on the basis that burnout was

defined as a syndrome with three dimensions: emotional exhaustion,

depersonalization, and personal accomplishment and this measure is

consistent with that definition. In addition, the use of the Maslach

Burnout Inventory permits some comparability between studies among

helping professionals including nurses because it is widely used.









The cognitive, affective and physiologic response of stress,

referred to as burnout, was measured with the Maslach Burnout Inventory

(Maslach & Jackson, 1986). A sample item of the Maslach Burnout

Inventory for Emotional Exhaustion includes, "I feel emotionally drained

from my work." Appendix I contains instrument directions and sample

items.

Commitment and Workload

Description. Commitment to career was measured with one five point

Likert item (Appendix F). Workload was measured with two five point

Likert items (Appendix F).

Psychometric properties. Psychometric properties of these variables

are presented in Table 3-9.

Use in this study. The personal variable commitment to career and

the organizational variable workload were measured with three items that

were designed by the researcher to closely resemble the variables used in

this study among critical care nurses. Critical care nursing faculty,

critical care graduate nursing students, and two top nursing management

members of two critical care intensive care units of a regional hospital

were asked to rate and evaluate these three items for use in this study.

On the basis of their evaluation and the discretion of the researcher

these items were used to measure commitment to career and workload.

Table 3-10 presents a summary of the instrumentation of personal,

organizational variables, and situation conducive to stress, job stress,

burnout-emotional exhaustion, job satisfaction and internal work

motivation variables used in the study.

Statistical Analysis

The five research questions listed in Chapter I (pg. 9) were used to

guide this study in examination of the full model of burnout-emotional









Table 3-9

Psychometric Properties of Variables: Commitment and Workload





Standard
Variable Number Mean deviation Alpha



Commitment 300 3.537 1.357 NE

Workload 300 4.097 .8038 NE


NE = Not estimated















C N
M *I













N if -'-4
I 0







- r-.~ a
I I

r-') *Z
'. CM~
CC) -iD

C^ 1


I I
.D r






N -





0 0

^. -4
,J u


- .,


Lr ,



00
















^.0


r)











IC
cn
Lr)
uq
m






OCcr

,. ..
m.wC ,-


>>
























u
0)

a)
^ -i

> tcl
$.' CO


o cZ















0 T











ILi
a)

1-1 ~



o a


4L1
a)


a)
E-4


cv











-4-
*0 .
1-4 c















o 0



-4
cvi
1.I >
(0


-0 -3-




C C
0 0


00




























0)

$>



0 0
3


-' -
f -< --I v

- ** 0 t' N





o 0 0 0 0





CM ~ ~ 4 Aj \ C -


u uu u u
2 a) a) a) Q)








m c _
o 0 0 0 0
4J *ri *il Lri -f
U U U U Ut-


















a) a-) a-) a> a)


cc

CO




0 U1
0)
uC
cc




4- co
o >~> ** U
1-4.1,-4 S i>i 4i. -4
cu3 -i a) 1-4 4-<
0.btv 0 -*i-i -4- f


LiO 00 -4 cvtl

U 0 i-'



C1-4 -i 01 E-' E-4

0
























r- -




o 1 0 CM

00 0* 0^ 0 0-


)) f CM) UT U

i-r ra o- -
















0l C/ Cl C/ Cl
0*- 0 0 0 0
W* -, -r) -i-


u

U
t-4


0 0
0


0 >,




1-0



c u








N
r.a)
m-4








o o
f-4 >0

4-o
'-1 JrJ


CO


4t-J
0
ca
E
ca



w
0

0



o
u0
oc


;>

o4 -
CD) 0
~. 4-i





0~CL
c >n
0 0-i
1/1 C>-


h-I
"00 'T
r- I oc
-..r x I




I o' --.




I CO r r
1 -0









-a- h- -'



n I I
I4 r,,. r~







--)

Ic Io
1 0 -o~ m


n Sr oy -a 0


a)
IU
:>

u

0
0

o



S0
0
4-o


a)

a) 1-o

wo


-..
I


CM
N








0?'

































1-4






















,,-J

0




4-I
CM)


















































o
o -I
'-4

0
~ Jj
0
0 E
>


























Pt-'






e~u~


00
.- .,-~

QU

rj~


a-'
0o 0'


CM-


5 5-





































































ca
u 0






























.,q .,n~
S01































cu
0






41
co -^

Q- cv

0.' *r
IV









exhaustion, that included ten personal variables, twelve organizational

variables, the degree to which the situation is conducive to stress, job

stress, job satisfaction, and internal work motivation. Multiple

regression analyses were used to answer the study questions. In a post

hoc analysis, each independent variable that made a significant

contribution to the variance explained in one or more dependent variables

was included in a path analysis. Path analysis, as described by Pedhazur

(1982), was used to estimate the magnitudes of the direct and indirect

effects of these selected variables on perceived levels of stress,

general job satisfaction, and burnout-emotional exhaustion,

respectively.
















CHAPTER TV
STUDY FINDINGS

This study was designed to collect data and to test a path-

analytical model of the theoretical conceptualization of burnout for

critical care nurses. The model of interest tested was an expansion of

the model offered by Perlman and Hartman (1982). Five research questions

were used to guide this study in the examination of the full model of

burnout including personal and organizational variables and situation

conducive to stress, job stress, job satisfaction and internal work

motivation. Each of the independent variables that made a significant

contribution to the variance explained in the dependent variables was

included in the path analysis.

This section is divided into two sections. The first section

describes the statistical analysis of the full model. This section

includes the mean, standard deviation, and zero-order correlation

coefficient between each of the variables included in the study and the

results of regression analyses used to address the five research

questions posed in Chapter I. The second section includes a presentation

of the findings related to estimation of the direct and indirect effects

of the exogenous and endogenous variables obtained from applying path

analysis to a more parsimonious version of the model.

Full Model and Statistical Analysis

Appendix J contains the means, standard deviations, and zero-order

coefficients of correlation between the endogenous variables consisting









of the situation conducive to stress, job stress, job satisfaction,

internal job motivation, and burnout-emotional exhaustion and each of the

exogenous variables in the study. As shown in Appendix J, many of the

variables were found to be significantly associated with the endogenous

measures and indicated promise for explaining and predicting burnout in

terms of the model shown in Figure 1-2 (pg. 7).

Exogenous Variables

Personal characteristics. Personal variables consisted of ability,

time, family demands, job expectations, personal work needs including

meaningfulness of work, responsibility for work outcomes, and knowledge

of work results, health, commitment to career, and personality. Among

the personal characteristics assessed family demands, knowledge of work

results, and health were all significantly associated with the situation

conducive to stress. With the exception of ability, time, and job

expectations, the personal characteristics were all significantly related

to job stress, and burnout-emotional exhaustion. Job expectations was

inversely related to internal job motivation. Personal characteristics

including meaningfulness of work, responsibility for work outcomes,

knowledge of work results and commitment to career were all significantly

associated with job satisfaction and internal job motivation. In

addition, while job expectation was significantly related to job

satisfaction, family demands was significantly associated with internal

job motivation. The personality variable was measured with three

psychological hardiness characteristics of commitment, challenge and

control with the total hardiness score. With the exception of job

satisfaction, the total psychological hardiness score was significantly

associated with the variables situation conducive to stress, job stress,

burnout-emotional exhaustion, and internal job motivation.









Organizational work characteristics. Organizational work variables

consisted of workload (workload and staff size), expected role

performance and role ambiguity, work group norms (skill variety, task

identity, task significance, autonomy), dealing with colleagues (dealing

with others at work), supervision, opportunity for advancement, pay,

support from others at work (social support satisfaction), organizational

climate and the economic-market condition (job security) measured with

the Job Diagnostic Survey.

As shown in Appendix J, the zero-order coefficient of correlation

between all the endogenous variables and expected role performance and

role ambiguity, supervision, opportunity for advancement and support from

peers at work were significant. Skill variety was significantly related

to all endogenous measures except situation conducive to stress and

burnout-emotional exhaustion. Task identity was significantly associated

with all the endogenous measures. The correlation of task significance

was significantly related to job satisfaction and internal job

motivation. Autonomy was significantly related to the situation

conducive to stress, job stress, and internal work motivation. Dealing

with others at work was significantly associated with all the dependent

variables. With the exception of the situation conducive to stress, and

internal work motivation, pay was significantly associated with the

endogenous measures. Job security was significantly related inversely to

burnout-emotional exhaustion, and positively related to job

satisfaction.

Endogenous Variables

Endogenous variables consisted of situation conducive to stress, job

stress, job satisfaction, internal job motivation, and burnout-emotional

exhaustion.









As shown in Appendix J, situation conducive to stress was

significantly related to all the endogenous measures while job stress was

significantly correlated with all the endogenous measures. Burnout-

emotional exhaustion was significantly related to job satisfaction. Job

satisfaction was significantly related to internal job motivation.

Question I

Question 1 addressed the significance of the relationship between

situational stress and the weighted linear combination of (a) a set of

personal variables and (b) a set of organizational variables. In

addition, this question also addressed which of the variables in the

model contributed significantly to the variance in situational stress.

Multiple regression analysis yielded an R2 of .72, (F = 31.948;

df = 22, 277; p = .0001), indicating that 72% of the variance on the

situation conducive to stress variable was explained by the variables in

the regression model. Regression results in Table 4-1 show that the

personal variables family demands, health difficulties, and psychological

hardiness, and the organizational work variables, skill variety and job

security, were significant predictors of the situation conducive to

stress.

Personal variables that did not relate significantly to situational

stress were ability, time, job expectations, meaningfulness of work,

responsibility for work outcomes, knowledge of work results, and

commitment to career. Organizational variables that did not relate

significantly to situational stress were workload, expected role

performance and role ambiguity, task identity, task significance,

autonomy, supervision, dealing with others, opportunity for advancement,

pay, and support from peers at work. (See Table K-I in Appendix K for

results of the multiple regression for all variables.)



















* .





- I




-)0 Xr


C-0 >00















0 c-I C4
* 0


00
M 0





00
0





















o -
I-
* S


S
II










C14
4-4
c
II


-4


























II
o:
rz.








C141
w0









Question 2

Question 2 addressed the significant relationship between job stress

and the combination of (a) situation conducive to stress and (b) the

preceding exogenous variables (ten personal variables and twelve

organizational variables). In addition, this question also addressed

which of the variables in the model contributed significantly to the

variance in job stress.

Multiple regression analyses yielded an R2 of .74, (F = 33.222;

df = 23, 276; p = .0001), indicating that 74% of the variance in job

stress was explained by the variables in the regression model.

Regression results in Table 4-2 show that the personal variables

including time (tenure), family demands, and health difficulties, the

organizational variables including workload, pay, supervision and job

security and the stress variable, situation conducive to stress were

significant predictors of job stress.

Personal variables that did not relate significantly to job stress

were ability, job expectations, meaningfulness of work, responsibility

for work outcomes, knowledge of work results, commitment to career and

psychological hardiness. Organizational variables that did not relate

significantly to job stress were expected role performance and role

ambiguity, skill variety, task identity, task significance, autonomy,

dealing with others, opportunity for advancement, and support from peers

at work. (See Table K-2 in Appendix K for results of the multiple

regression for all variables.)

Question 3

Question 3 addressed the significant relationship between job

satisfaction and the combination of (a) job stress, (b) situational












































r--o 0 N CM
C0 C) CM4
cn cn 3
* *





































-0~.
ca





4c
.,4 C13 0)

E-4 L4 =


3- r-I C0f
CO *- .-~ 0
0000











F- 3- '0X
* * *
I I











~ ~FC

m 0oo <3
* * *
I I


0




N w



4J .N -J


ca bO CO

U C U









stress, and (c) the preceding exogenous variables (10 personal variables

and 12 organizational variables). In addition, this question also

addressed which of the variables in the model contributed significantly

to the variance in job satisfaction.

Multiple regression analysis yielded an R2 of .64, (F = 20.413;

df = 24, 275; p = .0001), indicating that 64% of the variance in job

satisfaction was explained by the variables in the regression model.

Regression results in Table 4-3 show that the personal variables

including job expectations, meaningfulness of work, knowledge of work

results, health difficulties and organizational variables including task

identity, supervision, dealing with others, opportunity for advancement,

pay and job security and the stress variable, job stress, were

significant predictors of job satisfaction.

Personal variables that did not relate significantly to job

satisfaction were ability, time, family demands, responsibility for work

outcomes, commitment to career, and psychological hardiness.

Organizational variables that did not relate significantly to job

satisfaction were workload, expected role performance and role ambiguity,

skill variety, task significance, autonomy, and support from peers at

work. The stress variable degree situation conducive to stress did not

relate significantly to job satisfaction. (See Table K-3 in Appendix K

for results of the multiple regression for all variables.)

Question 4

Question 4 addressed the significant relationship between internal

job motivation and the combination of (a) job stress, (b) situation

conducive to stress, and (c) the preceding exogenous variables. In

addition, this question also addressed which of the variables in the















U" -i cri M
0l 0 0r-

0000
* * *
















-*00 00 00
oo ioo





0- Do o ,







0 ooo
-I -
* * *


I0: -,T N CM il)
V)i oq 0Y 1-4
LJ^ 0 0 0 0 N










-000 0 z00
IT 0 Lr0 c0 r 3







*o rM. *n CS oo
4I










I j -Lr) CN 0

0'- r) 00














CA-













-C

0)




0
q








41.1. 0


0. >
W M 0
rn =so -


1-.
0


co .


4~J

4- 1


13 w-
Cc
0aC
Qll -









model contributed significantly to the variance in internal job

motivation.

Multiple regression analysis yielded an R2 of .54, (F = 13.552;

df = 24, 275; p = .0001), indicating that 54% of the variance of internal

job motivation was explained by the variables in the regression model.

Regression results in Table 4-4 show that the personal variables

including family demands, meaningfulness of work, responsibility for work

outcomes, commitment to career and health difficulties and organizational

work variables including autonomy, dealing with others, support from

peers at work and the stress variable situation conducive to stress were

significant predictors of internal work motivation.

Personal variables that did not relate significantly to internal job

motivation were ability, time, job expectation, knowledge of work

results, and psychological hardiness.

Organizational variables that did not relate significantly to

internal job motivation were workload, expected role performance and role

ambiguity, skill variety, task identity, task significance, supervision,

opportunity for advancement, pay, and job security. The stress variable

that did not relate significantly to internal job motivation was job

stress. (See Table K-4 in Appendix K for results of the multiple

regression for all variables.)

Question 5

Question 5 addressed the significant relationship between burnout-

emotional exhaustion and the combination of (a) job satisfaction,

(b) internal work motivation, (c) job stress, (d) situation conducive to

stress, and (d) the preceding exogenous variables. In addition, this

question also addressed which of the variables in the model contributed

significantly to the variance in burnout-emotional exhaustion.







84








3l, 0 0 0 0









:3 C >Lno oo Un -' co
Z:) ON 00 l O's^ rn OT uln

> 0 0

4.1 I I




'4-4
0 4.j

_^; a)

o w u
1-1 M -4 Cli n 0cq fn as a C
MQ -o '"i -37 'M 0 ID' -4if 0
c C rN, 00 V"N

U 4.1 0 * * *
as CIU I I

0
C:
'-4
.C
4j-I





ca 10
., W
a)V
-4 "Cco



> Ei



> w I-

0 u

00 4- t
.,4 0 J) (3


> = o 2- ) o
Ai- (A o "ai
0 4-1 (A 41 0 a


to a- o 01u .-
*r-10 00 441 -1 3

00 00 u l,

0 wQ ZV 02 0
Cl coc )-.400 l-i ^je


3 '4-4C C U 0

^3~~c E- to "- ^u*- ~
C :> w Cu 0o $4- wi > >:o
-40o c c4 C 0 o o
04C ii l OQ ^li 0- CC





C m w a) 0 w ai
4-1 W0 > cn; 5 44Cl -
4.J.






0 N

C 0 44
0l "-i0 4~ 0.-4o Cl,
Ir-I 1- E1 2-l C, ~ iiI

-40




CO OI CO CUHI
( U z W 4 1 m -
4-i TI* r i&





b{~~~. C O^
M Lt')


m0 CO i



'd- ~~ toi CC -i(1 O
I D IU -ia) = > > .
G* 0 Q- *
(U -40 -)- a i (




H U 00 CL|U (I u c a









Multiple regression analysis yielded an R2 of .41, (F = 7.394;

df = 26, 273; p = .0001), indicating that 41% of the variance in

burnout-emotional exhaustion was explained by the variables in the

regression model. Regression results in Table 4-5 show the personal

variables that were significant predictors of burnout-emotional

exhaustion. Specifically, these were health difficulties, commitment to

career, and psychological hardiness. The significant organizational work

variables included workload, dealing with others, and job security and

the job stress outcome/variable job satisfaction.

Personal variables that did not relate significantly to burnout-

emotional exhaustion were ability, time, family demands, job

expectations, meaningfulness of work, responsibility for work outcomes,

and knowledge of work results. Organizational variables that did not

relate significantly to burnout-emotional exhaustion were expected role

performance and role ambiguity, skill variety, task identity, task

significance, autonomy, supervision, opportunity for advancement, pay,

and support from peers at work. The job stress outcome variable that di:

not relate significantly to burnout-emotional exhaustion was internal juo

motivation. The stress variables that did not relate significantly to

burnout-emotional exhaustion were job stress and the degree the situation

was conducive to stress. (See Table K-5 in Appendix K for results of the

multiple regression for all variables.)

In combination, these study questions have provided a test of the

full model of burnout. A major problem in testing the full model was

that the large number of variables found to be significantly associated















cc -< Lr c) c


* * *
a0 ,00 -


c.. o






* .
C- Vi C0i M



oo-3.- (Mm .in1
I I











3.,T r C' r.-.
,, co CM ,-

* 0 0

S4,4





















CO2

CO~
a) -
loi C4















C1) M


u -a

u



C.)


w ) 0 m0
-C uJ 0 0
4.J 6 U~ -1

a) e > 1o


all as


0 0
0 0







L f1








c-)
S0







-o -








* 0














CNI n
0-I
L. o



































0









a
* -4































0


0
N4J1


540









with burnout made parsimonious and interpretable statements of

relationship difficult. To proceed with this exploratory analysis it

seemed desirable to reduce the total number of variables by eliminating

those variables that were not significantly related to the next

endogenous variable on the path in the model. A fairly liberal alpha

level of .10 was chosen for variable inclusion in the path analysis to

guard against "over-trimming" of the model in this early stage of theory

building.

Reduced Model and Path Analysis

Contributions to Situational Stress

Table 4-6 contains the estimates of the direct effects of the

exogenous personal variables, and organizational exogenous variables on

the situation conducive to stress. An NE in the Direct Effects column of

Table 4-6 indicates that the path coefficient was not significantly

greater than .00 in the full-model analyses. An NE in the Indirect

Effects column indicates that the effect could not be estimated because

(a) one of the coefficients along the path had not been significantly

greater than .00 in the full model, and thus the path was omitted in the

reduced model, or (b) there was no indirect path between the particular

independent and dependent variable in the full model show in Figure 1-2.

As evidenced in Table 4-6, family demands had the strongest effect (.637)

on situation conducive to stress; skill variety had the weakest effect

(.062) on situation conducive to stress. This set of variables in the

reduced model explained 70% of variance of the situation conducive to

stress, as compared to 72% of variance that had been explained in the

full model.









Table 4-6

Magnitudes of Direct and Indirect Effects on Situation
Conducive to Stress, Level of Perceived Job Stress,
General Job Satisfaction and Burnout


Effect



Variables Direct Indirect Total


Family demands

Health difficult

Psychological hai

Skill variety

Job security


Degree situation


to stress

Time

Family demands

Health difficulties

Psychological hardiness

Workload

Skill variety

Supervision

Pay

Job security


On Degree Situation Conducive to Stress

.637 NE

.es .201 NE

rdiness -.091 NE

.062 NE

-.095 NE

On Level of Perceived Job Stress

conducive .370 NE


-.034

.332

.127

NE

.042

NE

-.183

-.073

-.095


NE

.236

.074

-.034

NE

.023

NE

NE

-.035


.637

.201

-.091

.062

-.095



.370


-.034

.568

.201

-.034

.042

.023

-.183

-.073

-.130









Table 4-6--continued


Effect




Variables Direct Indirect Total


On

Degree situation conducive
to stress

Level of perceived
job stress

Family demands

Job expectations

Meaningfulness for work

Knowledge of work
results

Health difficulties

Psychological hardiness

Skill variety

Task identity

Supervision

Dealing with others

Opportunity for
advancement

Pay

Job security


General Job

NE


-.106


NE

-.115

.290

.100


.193

NE

NE

.096

.237

-.216

.196


.170

-.099


On Burnout-Emotional

General job satisfaction -.334

Level of perceived job NE
stress


Satisfaction

-.039


NE


-.060

NE

NE

NE


-.021

.004

-.002

NE

.019

NE

NE


-.008

.014

Exhaustion

NE

.035


-.039


-.106


-.060

-.115

.290

.100


.172

.004

-.002

.096

.256

-.216

.196


.162

-.085



-.334

.035