Idea Transcript
PERSON-ENVOR NM ENT CiONGRUENCE, B SATISFACTION, AND JOB INVOLVEMENT by
CH ARLENE C LEW
DISSERTATION submitted in fulfilment of the requirements for the degree
MASTER OF ARTS in
PSYCHOLOGY in the
FACULTY OF ARTS • at the RAND AFRIKAANS UNIVERSITY
SUPERVISOR: PROFESSOR Z.J. NEL CO-SUPERVISOR: MR. G.P. [DE BRUIN OCTOBER 1997
FIERSOH-EHMOHNEHT COHOMEHCE, ,,03 SKIISFAC710H, AHD JOB HVO,VENEHT
C. C. LEW
i
ACKNOWLEDGEMENTS
It is with sincere gratitude that I wish to thank the following persons and institutions who each performed a vital role in the composition and completion of this dissertation:
Prof. Z.J. Nel, my supervisor, for continual support, guidance and advice throughout the course of this study, and Mr. G.P. De Bruin, my cosupervisor, for assistance. The RAU for a Merit Bursary. Dr. J.L. Van Wyk and Me. Elsabe Smit of the Statistical Consulting Services (STATCON) at the RAU for advice and for statistically interpreting the data. The Human Sciences Research Council for once again contributing to science through financial support. Please note that the opinions and conclusions drawn from this study are not necessarily those of the HSRC. The Psychological Society of South Africa (PsySSA) and The Optometrist Association of South Africa for providing contact numbers of the samples. Each psychologist and optometrist who so willingly participated by completing questionnaires. My loving husband, Eugene, for encouragement, prayer and a helping hand with our precious baby boy Chaen. My father for spiritual, emotional and financial support, and my mother especially for sacrificing so much to look after Chaen. All praise and thanks to my Lord Jesus Christ who is the Provider of every good and perfect gift.
ii
ABSTRACT Questions regarding the possible interrelations of work-related constructs of person-environment congruence, job satisfaction and job involvement were posed in this dissertation. These constructs were defined in terms of three separate theoretical frameworks, namely Holland's (1985b) vocational choice theory, the Theory of Work Adjustment (Lofquist & Dawis, 1984), and Kanungo's (1982b) conceptualisation of job involvement, respectively. In accordance with these theories, the Self-Directed Search questionnaire, the Minnesota Satisfaction Questionnaire, and the Job Involvement Questionnaire were used to operationalise these variables for the 114 psychologists and 44 optometrists who participated in the study. The person-environment congruence scores were computed by means of the Congruence Index (Brown & Gore, 1994). A secondary aim of the study was to evaluate the accuracy of environmental codes (one of the components of person-environment congruence) as have been provided by The South African Dictionary of Occupations (1987) for samples of psychologists and optometrists in the South African context. Different research designs were used to accommodate the diversity of the research objectives and hypotheses. A factorial research design was used to determine the potentiality of influences of person-environment congruence and job involvement on overall job satisfaction, intrinsic job satisfaction and extrinsic job satisfaction. Possible differences between the samples of psychologists and optometrists, and males and females were built into the general linear models encompassed by this design. Similarly, this design was used to determine whether job involvement is a function of personenvironment congruence, overall job satisfaction, intrinsic job satisfaction, extrinsic job satisfaction, gender and occupational group. Possible divergent effects of the samples and gender were further probed by means of generalised linear models (regression). Gender and sample-specific differences in terms of the variables of the study separately were examined through inferential statistics (t-tests).
iii The possibility of a significant effect of job involvement on the relationship between person-environment congruence and job satisfaction assumed by Holland (1985b), was investigated by means of correlational analyses to conclude the examination of the relations among these variables. For the secondary aim of this study, the Environment Assessment Technique (Holland, 1985b) was used to calculate the respective environment codes of this sample of optometrists, and samples from four categories of registration of psychologists, namely clinical, counselling, educational and industrial psychologists, and then compared to the codes for these vocations listed in The South African Dictionary of Occupations (1987). The results of this study revealed significant effects of job involvement on overall job satisfaction and intrinsic job satisfaction, but nonsignificant effects of occupational group (or sample) and gender. Of the independent variables, only person-environment congruence had moderately significant effects on extrinsic job satisfaction. When job involvement served as dependent variable, only overall job satisfaction or intrinsic job satisfaction and the occupational group had any significant influence on it. The effect of the psychologist sample on job involvement was greater than that of the optometrist sample. No gender differences were found in terms of these linear models, or in terms of the variables of the study separately. Further findings revealed that job involvement does not have any significant effect on the hypothesised congruence-job satisfaction, congruence-intrinsic job satisfaction, or congruence-extrinsic job satisfaction relationships. When the environmental codes of the various samples were compared to the proposed Holland ISE (investigative-social-enterprising) environment code, vast discrepancies were found. A predominantly social environment code was obtained for clinical, counselling and educational psychologists, but an enterprising code for industrial psychologists. Although the optometrists obtained a predominately investigative code, their environment's subtypes could not be characterised as social and enterprising, but rather as enterprising and realistic. Suggestions were made that research employing sample-specific environmental codes in studies of person-environment congruence and its
iv
possible covariates is warranted. A need for empirical examinations of environmental codes of other vocational populations in South Africa was also identified. It was further recommended that the job involvement construct should be included in theories describing the antecedents, correlates and consequences of job satisfaction. This study was then evaluated in terms of criterion validity and external validity requirements, and the conclusion was drawn that within the limitations of the study, the research questions had been answered.
v
OPSONNORIG Vrae ten opsigte van die moontlike onderlinge verbande van die konstrukte persoon-omgewingskongruensie, werkstevredenheid en werksbetrokkenheid is in hierdie verhandeling gestel. Hierdie konstrukte word gedefinieer in terme van drie aparte teoretiese raamwerke, naamlik Holland (1985b) se beroepskeuseteorie, the werksaanpassingsteorie (Lofquist & Dawis, 1984), en Kanungo (1982b) se konseptualisering van werksbetrokkenheid, onderskeidelik. In ooreenstemming met hierdie teoried is die Self-Directed Search vraelys, die Minnesota Satisfaction Questionnaire, en die Job Involvement Questionnaire gebruik om hierdie veranderlikes vir 114 sielkundiges en 44 optometriste, wat aan die studie deelgeneem het, te operasionaliseer. Die kongruensietellings was bepaal deur middel van die Congruence Index (Brown & Gore, 1994). Die sekondere doelwit van die studie was om die akkuraatheid van omgewingskodes (een van die elemente van persoonomgewingskongruensie) wat in The South African Dictionary of Occupations (1987) voorkom, te evalueer vir steekproewe sielkundiges en optometriste in die Suid-Afrikaanse konteks. Verskillende navorsingsontwerpe is gebruik ten einde aan te pas by die diversiteit van die navorsingsoogmerke en hipoteses. `n Faktoriale navorsingsontwerp was gebruik om die moontlikheid van invloede van persoon-omgewingskongruensie en werksbetrokkenheid op algehele werksbevrediging, intrinsieke werksbevrediging en ekstrinsieke werksbevrediging te bepaal. Moontlike verskille tussen die steekproewe van optometriste en sielkundiges, asook mans en vrouens was by die algemene liniere modelle van hierdie navorsingsontwerp ingesluit. Eweneens is hierdie navorsingsontwerp gebruik om te bepaal of werksbetrokkenheid 'n funksie van kongruensie, algehele werksbevrediging, intrinsieke werksbevrediging, ekstrinsieke werksbevrediging, geslag en die steekproewe was. Die moontlikheid van verskille in die effekte van die steekproewe en geslag was verder ondersoek deur middel van veralgemeende liniere modelle (regressie). Verskille ten opsigte van geslag en steekproewe vir elk van die veranderlikes van die studie afsonderlik was deur middel van inferensiele statistiek (t-toetse) ondersoek.
vi
Korrelasies was bereken om die moontlikheid van 'n betekenisvolle effek van werksbetrokkenheid op die verhouding tussen kongruensie en werksbevrediging wat deur Holland (1985b) voorveronderstel word, te bestudeer, ten einde die studie van die onderlinge verbande van hierdie veranderlikes of te sluit. Vir die sekondere oogmerk van die studie is die Environment Assessment Technique (Holland, 1985b) gebruik om die onderskeie omgewingskodes van die steekproef optometriste, en die steekproewe van vier registrasiekatgoried van sielkundiges, naamlik kliniese, voorligting-, opvoedkundige en bedryfsielkundiges, te bepaal, en om die verkree kodes te vergelyk met die kodes wat vir hierdie beroepe in The South African Dictionary of Occupations (1987) verskaf word. Die resultate van die studie het betekenisvolle invloede van werksbetrokkenheid op algehele werksbevrediging en intrinsieke werksbevrediging aangetoon, maar nie-betekenisvolle invloede van die beroepsklassifiseringsveranderlike (steekproef) en geslag. Van die onafhanklike veranderlikes het slegs kongruensie 'n matige betekenisvolle invloed op eksterne werksbevrediging uitgeoefen. Met die gebruik van werksbetrokkenheid as afhanklike veranderlike, het slegs algehele werksbevrediging of ook intrinsieke werksbevrediging en die steekproef enige betekenisvolle invloed daarop uitgeoefen. Die effek van die sielkundige steekproef was groter as die van die optometriste. Geen verskille ten opsigte van die rol van geslag kon gevind word in terme van hierdie liniere modelle, of in terme van die veranderlikes apart beskou nie. Verdere bevindinge het aangetoon dat werksbetrokkenheid nie 'n betekenisvolle determinant van die kongruensie-werksbevrediging, kongruensie-intrinsieke werksbevrediging of kongruensie-ekstrinsieke werksbevreding verbande was nie. Deur die vergelyking van die omgewingskodes van die verskeie steekproewe met die voorgestelde Holland ISE (ondersoekend-sosiaalondernemend) omgewingskode het baie teenstrydighede sigbaar geword. 'n Oorwegend sosiale omgewingskode was verkry vir kliniese, voorligting- en opvoedkundige sielkundiges, maar 'n ondernemende kode vir
vi' bedryfsielkundiges. Selfs al was die kode wat vir die optometriste verkry is wel oorwegend ondersoekend van acrd, het hulle omgewing nie eienskappe van sosiale en ondernemende tipes getoon nie, maar van ondernemende en realistiese omgewingstipes. As gevolg van die voorafgaande bevindinge word daar aanbeveel dat navorsing wat gebruik mask van omgewingskodes wat spesifiek vir die betrokke steekproef bepaal word in studies van kongruensie en die moontlike kovariante daarvan, uitgevoer behoort te word. Die belang van verdere empiriese ondersoeke na omgewingskodes van ander beroepe in Suid-Afrika is ook geIdentifisieer. Dit was verder aanbeveel dat die werksbetrokkenheidskonstruk in teoretiese beskrywings van die oorsake, korrelate en gevolge van werksbevrediging ingesluit behoort te word. Die studie is Iaastens gedvalueer in terme van vereistes van kriteriumgeldigheid en eksterne geldigheid in navorsing, en daar is tot die slotsom gekom dat ongeag die beperkinge van die studie word daar nogtans voldoen aan die vereiste om antwoorde te bied vir die navorsingsprobleem wat gestel was.
TABLE OF CORITEMTS PAGE
ACKNOWLEDGEMENTS
i
ABSTRACT (ENGLISH)
ii
SUMMARY (AFRIKAANS)
v
LIST OF FIGURES
xviii
LIST OF TABLES
xix
LIST OF ANNEXURES
xxiii
CH F1\ PTE
:
ONTRODUCTOON
1.1
Introduction
1
1.2
The research problem
6
1.3
Purpose of the study
10
1.4
Core concept description
10
1.4.1
Person-environment congruence
10
1.4.2
Job satisfaction
11
1.4.3
Job involvement
12
1.5
Dissertation structure
12
CHAPTER 2:
HOLLAND'S CA EER CHOICE THEORY
2.1
Introduction
14
2.2
Background principles
15
2.3
Assumptions of Holland's theory
16
2.3.1
Assumption 1
16
2.3.1.1
The development of personality types
17
ix
PAGE
2.3.1.2
Models of dominant personality types
18
2.3.1.3
Subtypes and personality patterns
18
2.3.1.4
Typological relationships
20
2.3.1.5
Behaviour
21
2.3.2
Assumption 2
22
2.3.2.1
The nature of environmental models
22
2.3.2.2
Influences of the environmental models
22
2.3.2.3
Consistency, differentiation, identity, calculus
24
2.3.3
Assumption 3
26
2.3.3.1
Demands and opportunities
26
2.3.3.2
Activities, competencies and achievements
27
2.3.3.3
Vocational choice
27
2.3.3.4
Vocational beliefs and strategies
27
2.3.4
Assumption 4
28
2.3.4.1
The level of person-environment interaction
28
2.3.4.2
A structural model of person-environment interactions
32
2.4
Person-environment congruence and careers
33
2.4.1
Career choice, stability and change
34
2.4.2
Career development
35
2.4.3
Coping with career problems
36
2.5
Measures and indices relating to Holland's theory
37
2.5.1
Assessment of personality types
38
2.5.2
Assessment of the environment
39
2.5.3
Indices of congruence
40
2.5.4
The classification system
44
x
PAGE 2.6
Additional dimensions
45
2.6.1
Data/ideas and things/people dimensions
46
2.6.2
The interest circle and the prestige dimension
48
2.6.3
Gati's (1991) hierarchical structure
51
2.7
Demographic factors and Holland's theory
54
2.7.1
Gender-related issues
54
2.7.2
Age-related and developmental issues
55
2.7.3
Racial issues
55
2.8
Conclusion
56
CHAPTER 3:
RESEARCH O HOLLAND'S THEO Y
3.1
Introduction
61
3.2
Conditions of person-environment congruence
61
3.3
Congruence and job satisfaction
63
3.3.1
Support of the congruence-satisfaction hypothesis
63
Mediating variables of the congruence hypothesis
64
Lack of support for the congruence-satisfaction hypothesis
69
3.4
Congruence and other outcomes
71
3.4.1
Congruence and vocational stability
71
3.4.2
Congruence, achievement and performance
71
3.4.3
Congruence and perceived social climate
72
3.4.4
Congruence and well-being
73
3.3.2 3.3.3
xi PAGE
3.4.5
Congruence and coping strategies
73
3.4.6
Incongruence and negative work outcomes
74
3.5
Demographic factors
75
3.5.1
Gender differences
75
3.5.2
Cross-cultural and racial applicability
77
3.6
Additional dimension
81
3.6.1
Data/ideas and things/people
81
3.6.2
Validation of the circumplex structure
83
3.7
Indices of congruence
85
3.8
The Self-Directed Search
88
3.8.1
Validity of the SDS
89
3.8.1.1
Construct validity
89
3.8.1.2
Predictive validity
90
3.8.1.3
Concurrent validity
91
3.8.2
The effects and usefulness of the SDS
92
3.8.3
Raw versus normed scores
93
3.9
Conclusion
94
CHAPTER 4:
A THEORETICAL PERSPECTIVE ON WORK ADJUSTMENT AND JO INVOLVE% 1 ENT
4.1
Introduction
97
4.2
The Theory of Work Adjustment
97
4.2.1
The impetus of the theory
98
4.2.2
The structural component
100
4.2.2.1
Personality
100
4.2.2.2
The work personality
101
xii
PAGE 4.2.2.3
The work environment
103
4.2.3
The systems component
105
4.2.3.1
Assumptions of the theory
105
4.2.3.2
The work personality and work environment style
106
4.2.4
The prediction component
108
4.2.5
Implications and applications of the Theory of Work Adjustment
110
4.2.6
Assessment in the Theory of Work Adjustment
111
4.2.6.1
Satisfaction '
111
4.2.6.2
Satisfactoriness
111
4.2.6.3
Needs and reinforcers
112
4.2.6.4
Abilities
113
4.2.6.5
An integration of the various measures
113
4.2.7
Comments and elaborations of the Theory of Work Adjustment
114
4.2.8
Summary and evaluation
118
4.3
A theoretical perspective on job involvement
120
4.3.1
Historical perspectives on job involvement
121
4.3.1.1
Conceptualisations of job involvement
121
4.3.1.2
The sociological perspective of alienation
123
4.3.1.3
The psychological perspective of involvement
123
4.3.2
Kanungo's motivational perspective
125
4.3.2.1
Important premises of the motivational approach
125
4.3.2.2
Kanungo's definition of job involvement
126
4.3.2.3
The need satisfaction model
127
4.3.3
Differences from the other two perspectives
130
XIII
PAGE 4.3.4
Antecedents, correlates and consequences of job involvement
130
4.3.5
Work involvement
132
4.3.6
Instruments developed by Kanungo
133
4.3.7
Summary and evaluation
134
4.4
Conclusion
135
CHAPTE 5:
EMPOROCAL SUPPORT FOR THE THEORETOC L PE SPECTOVES ON WORK ADJUSTMENT ND JOB ONVOLVEMENT
5.1
Introduction
137
5.2
Research on the work adjustment theory
137
5.2.1
Propositions of the Theory of Work Adjustment
138
5.2.2
Needs, values and reinforcers
140
5.2.3
The interrelations of correspondence and job satisfaction
141
5.2.4
Job satisfaction and other variables
143
5.2.4.1
Job satisfaction and various occupations
143
5.2.4.2
Job satisfaction and performance
145
5.2.4.3
Job satisfaction and turnover
147
5.2.4.4
Job satisfaction and work goals
149
5.2.4.5
Job satisfaction and genetic factors
149
5.2.4.6
Job satisfaction and age
150
5.2.4.7
Job satisfaction and life satisfaction
151
5.2.4.8
Job satisfaction and view on money
151
5.2.5
Validity of the job satisfaction construct
151
5.2.6
Summary and evaluation
152
xiv
PAGE 5.3
Research on Kanungo's theory
154
5.3.1
The construct of job involvement
154
5.3.2
The distinctiveness of the job involvement construct
155
The interrelations of job involvement and other work-related constructs
158
5.3.3.1
Job involvement and need saliency
159
5.3.3.2
Job involvement and job satisfaction
160
5.3.3.3
Job involvement and tenure
162
5.3.3.4.
Job involvement and performance
163
5.3.3.5
Job involvement and career orientations
163
5.3.3.6
Other antecedents, correlates and consequences of job involvement
163
5.3.4
Job involvement and non-work domains
166
5.3.4.1
Job involvement and person and family characteristics
166
5.3.4.2
Job involvement and gender
167
5.3.4.3
Job involvement and Type A personality
168
5.3.4.4
Job involvement and health
169
5.3.4.5.
Job involvement and locus of control
170
5.3.4.6
Job involvement and avocational affiliations
170
5.3.4.7
Job involvement and machiavellianism
171
5.3.5
Validity of the job involvement construct
171
5.3.6
Summary and evaluation
173
5.4
Conclusion
173
5.3.3
XV
CHAPTER 6:
RESEARCH METHOD
PAGE 6.1
Introduction
175
6.2
Purpose and objective of the study
175
6.2.1
Objective 1
175
6.2.2
Objective 2
176
6.2.3
Objective 3
176
6.2.4
Objective 4
177
6.3
Research design
177
6.3.1
Variables relating to objectives one and two
177
6.3.2
Variables relating to objective three
178
6.3.3
Variables relating to objective four
178
6.3.4
Control of variables
178
6.4
Sampling procedures, participants and response ratios
179
6.5
Research hypotheses
183
6.5.1
Hypotheses with regard to interrelations of the variables
183
Hypotheses with regard to the influence of job involvement
185
Hypotheses with regard to environmental codes of psychologists and optometrists
186
6.5.4
General linear models of the variables
187
6.6
Measuring instruments
188
6.6.1
Measuring and computing person-environment congruence
188
6.6.1.1
Description of the Self-Directed Search
188
6.6.1.2
Statistical properties of the SDS
189
6.5.2 6.5.3
xvi
PAGE
6.6.1.3
Environmental codes
189
6.6.1.4
Computing person-environment congruence
190
6.6.2
Measuring job satisfaction
191
6.6.2.1
Description of the Minnesota Satisfaction Questionnaire (1967)
191
6.6.2.2
Statistical properties of the MSQ
192
6.6.3
Measuring job involvement
193
6.6.3.1
Description of the Job Involvement Questionnaire
193
6.6.3.2
Statistical properties of the JIQ
193
6.6.4
Personal-demographic variables
194
6.7
Procedure and statistical analyses
194
6.8
Summary
196
CHAPTER 7:
EV-PUS AND FllNEINGS
7.1
Introduction
197
7.2
Results of the factorial research design
198
7.2.1
Job satisfaction as dependent variable
199
7.2.2
Intrinsic job satisfaction as dependent variable
204
7.2.3
Extrinsic job satisfaction as dependent variable
208
7.2.4
Job involvement as dependent variable
211
7.2.5
A direct comparison in terms of gender and occupational group
218
7.3
Results of the correlational research design
221
7.3.1
Correlations of the variables for the entire sample
222
xvii
PAGE
7.3.2
Correlations of the variables for high and low job involvement groups
7.4
224
Results of the frequency distributions and descriptive statistics
227
7.4.1
Environmental codes of psychologists
227
7.4.2
Environmental codes of optometrists
231
7.5
Summary of hypotheses and conclusions
233
7.5.1
Null hypotheses
233
7.5.2
Alternative hypotheses
235
7.6
Conclusion
236
CHAPTER :
OMPUC Ak TOONS AND RECOMMENDATOONS
8.1
Introduction
237
8.2
Summary
238
8.3
Implications and recommendations
244
8.3.1
Theoretical implications and recommendations
244
8.3.2
Practical implications and recommendations
246
8.4
Evaluation of the study
248
8.5
Conclusion
249
REFERENCES
250
xv iii
LOST OF FO UGC
FIGURE
PAGE
2.1
The hexagonal configuration of types
20
2.2
The data/ideas and things/people dimensions
46
2.3
The spherical representation of interests
49
2.4
Gati's hierarchical model
52
2.5
Integration of circular-hexagonal and hierarchical models
53
3.1
Men's structure of interests
76
3.2
Women's structure of interests
76
4.1
Personality and environmental interaction
104
4.2
Operational aspects of the Theory of Work Adjustment
113
Hierarchical representation of the satisfaction construct
117
Hierarchical representation of the satisfactoriness construct
117
The motivational approach to involvementalienation
127
The percentage of psychologists of the sample in the various categories of registration and the academic field
182
Distribution in percentages of the population and sample of psychologists according to category of registration
182
Frequency distributions of the first letter personality codes of psychologists (N= 114)
228
Frequency distributions of the first letter personality codes of optometrists (N= 44)
232
Findings on the interrelations of the variables
242
4.3 4.4 4.5 6.1
6.2
7.1 7.2 8.1
xix
LOST OF TABLES TABLE
2.1
PAGE A structural model of person-environment interactions
32
6.1
Demographic variables of the participants
181
6.2
Overall, intrinsic, and extrinsic job satisfaction items of the MSQ
191
General linear model of the dependent variable job satisfaction and the independent variables combined (1)
199
General linear model of the dependent variable job satisfaction and the independent variable job involvement
200
General linear model of the dependent variable job satisfaction and the independent variables (excluding job involvement) combined (1)
201
General linear model of the dependent variable job satisfaction and the independent variables combined (2)
202
General linear model of the dependent variable job satisfaction and the independent variables (excluding job involvement) combined (2)
202
General linear model of the dependent variable job satisfaction and the independent variable person-environment congruence (2)
203
General linear model of the dependent variable intrinsic job satisfaction and the independent variables combined (1)
204
General linear model of the dependent variable intrinsic job satisfaction person-environment congruence and job involvement
205
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
)0(
PAGE
TABLE
7.9
7.10
7.11
7.12
7.13
7.14
7.15
7.16
7.17
7.18
General linear model of the dependent variable intrinsic job satisfaction and the independent variable job involvement
206
General linear model of the dependent variable intrinsic job satisfaction and the independent variables combined (2)
207
General linear model of the dependent variable intrinsic job satisfaction and the independent variables person-environment congruence (2) and job involvement
207
General linear model of the dependent variable extrinsic job satisfaction and the independent variables combined (1)
208
General linear model of the dependent variable extrinsic job satisfaction and the independent variable person-environment congruence (1)
209
General linear model of the dependent variable extrinsic job satisfaction and the independent variables combined (2)
210
General linear model of the dependent variable extrinsic job satisfaction and the independent variable person-environment congruence (2)
211
General linear model of the dependent variable job involvement and the independent variables combined (1)
212
General linear model of the dependent variable job involvement and the independent variables combined (2)
212
General linear model of the dependent variable job involvement and the independent variables job satisfaction and occupational group
213
xxi
PAGE
TABLE 7.19
7.20
7.21
7.22
7.23 7.24 7.25 7.26 7.27
7.28
Parameter estimates of the dependent variable job involvement in terms of occupational group differences and overall job satisfaction
214
General linear model of the dependent variable job involvement and the independent variables person-environment congruence, intrinsic job satisfaction, extrinsic job satisfaction, occupational group and gender
215
Parameter estimates of the dependent variable job involvement in terms of occupational group differences and intrinsic job satisfaction
216
General linear model of the dependent variable job involvement and the independent variables intrinsic job satisfaction and occupational group
217
Descriptive statistics of the variables in terms of the different samples
219
Results of the t-test of the equality of means of the variables for the occupational groups
219
Descriptive statistics of the variables in terms of gender
220
Results of the t-test of the equality of means of the variables for the two sexes
221
Pearson correlations of the variables personenvironment congruence, overall job satisfaction, extrinsic job satisfaction, intrinsic job satisfaction, and job involvement (N = 158)
223
Pearson correlations of the variables personenvironment congruence, overall job satisfaction, extrinsic job satisfaction, intrinsic job satisfaction, and job involvement for the group with low levels of job involvement (n = 42)
224
)ocii PAGE
TABLE
7.29
7.30
7.31 7.32
Pearson correlations of the variables personenvironment congruence, overall job satisfaction, extrinsic job satisfaction, intrinsic job satisfaction, and job involvement for the group with high levels of job involvement (n = 44)
225
Two-sided significance of the differences between the high and low job involvement groups in terms of the congruence-satisfaction relationships
226
Medians and modes of the three-letter codes of the psychologists (N = 114)
228
Medians and modes of the personality codes of the psychologists in the various categories of registration
229
Environmental codes for psychology categories of registration based on first and alternative first personality code letters
230
Medians and modes of the personality codes of the optometrists (N = 44)
232
7.35
Retained null hypotheses and conclusions
233
7.36
Retained alternative hypotheses and conclusions
235
7.33
7.34
LOST OF ANNEXURES:
PAGE
A
Characteristics of the Holland personality types
227
8
Influences of the Holland environmental types
281
C
Propositions and corollaries of the Theory of Work Adjustment
282
Additional propositions of the Theory of Work Adjustment
284
D
1
CHAPTER 1 OMTRODUCTOOM
1.1 INTRODUCTION "Moreover, when God gives any man wealth and possessions, and enables him to enjoy them, to accept his lot and to be happy in his work — this is a gift of God." (Ecclesiastes 5:20) The significance of work in the life of individuals and societies has been contemplated through the centuries, and philosophers and psychologists of the past few generations have probed the meaning of work. According to the psychologist Erikson, all adults need to accomplish their task of generativity, or stated differently, adults need to perpetuate themselves by adding something meaningful and enduring to the world, and this can be done through work (Craig, 1986). One can also assume that without work, a person can not even satisfy basic needs according to Maslow's hierarchy of needs, and will therefore not be able to actualise him or herself (Gerdes, Moore, Ochse, & Van Ede., 1988). If one considers that the greatest part of adult life in our culture is spent working, the importance of selecting a satisfying work environment, and adjusting to one's chosen occupation and specific job is only too apparent. It is therefore important to try and understand the conditions under which an individual will be suited to his particular work environment and the essence of what is needed for a person to be happy and satisfied in his occupation. This investigation into the work life of an adult should explore not only the person and his job, but also his attitudes within and towards the position. There are three main perspectives on the interaction between a person and the environment in which he performs. According to the effect congruency
2 model both personal and situational characteristics influence behaviour and outcomes independently and in a linear way. Main personal and environmental variables interact to result in outcomes such as satisfaction. According to general congruency models "congruency exists when conceptually similar dimensions of persons and situations are correspondingly high or low" (Joyce, Slocum & Von Glinow, 1982, p. 267). In other words, when person and sitivational dimensions are alike, the interaction effect of these results in satisfaction. According to functional congruency, either personal or situational variables can result in satisfaction or performance (Joyce et al., 1982). Walsh and Betz (1995) distinguish between an interactional perspective, a person perspective and an environmental perspective on person-environment psychology. One theory describing the interaction between personality and environment is that of Holland (1973, 1985b). Typically, this theory is used to assist vocational choice (Gati, Garty & Fassa, 1996). It has become a very effectual instrument in vocational interventions and its classification system has proved useful in the social sciences and in the educational, business and industrial fields. No wonder that its constructs and hypotheses have captured the attention of many researchers over the past few decades. Holland's primary concern in explaining vocational behaviour is to facilitate job selection or change, producing job satisfaction. Person-environment congruence is one of the principal constructs of the theory. It has been the focus of much research, and has also been studied in relation to job satisfaction. Research results seem to confirm the hypothesised relationship between person-environment congruence and job satisfaction (heir, Keinan, & Segal, 1986; Smart, Elton, & McLaughlin, 1986). Another theory dealing with the fit between person and environment is the Theory of Work Adjustment (Lofquist & Dawis, 1969). Rather than facilitating career selection, this theory primarily serves as an aid in the selection of a specific job and the adjustment to that particular job. The theory not only describes personality and environments and the systems that link the two, but also clarifies conditions according to which predictions of work outcomes and attitudes such as satisfaction and tenure can be made. Both Holland's (1973, 1985b) theory and the Theory of Work Adjustment can be seen as interactional perspectives on person-environment psychology,
3 according to which an assessment of both the person and environment is required in order to understand development (Walsh & Betz, 1995). Together, these two theories maintain that when a person's abilities, interests and personal traits match the requirements, rewards and interpersonal relations of a work environment, satisfaction and success follow (Gottfredson & Holland, 1990). Although Holland focuses on interests and Lofquist and Dawis (1969) on abilities and values, they share the assumption that a good fit between a person and environment leads to higher levels of well-being, including satisfaction with one's job (Gati, Garty, & Fassa, 1996). Another factor shared by these theories is the role of reinforcement in the work place. According to Holland, members of an environment will give positive reinforcement to a person who acts in accordance to group expectations and the nature of the environment, and negative reinforcement for incongruent behaviour (heir et al.; 1986; Gati et al., 1996). A positive interaction between a person's needs and the reinforcers of the environment results in satisfaction according to Dawis and Lofquist (1984). Supposing accordingly that job satisfaction is a function of the quality of the relationship between a person and his work environment, the history of research on job satisfaction can now briefly be considered. Shortly after the turn of this century, field studies performed by Taylor (1916) already led to the supposition that job satisfaction and financial prosperity codetermine each other. In the late twenties, findings from the Hawthorne studies suggested that job satisfaction was a consequence of social recognition at work. Subsequently, more and more variables had been empirically associated with job satisfaction, for instance steady employment and working conditions in the thirties (Berry & Houston, 1993). There is still uncertainty about the determinants and consequences of job satisfaction. Much of the early research on job satisfaction examined the relationship between job satisfaction and both personal and job characteristics. Satisfaction was therefore seen as a function of both the person and the environment (Lawler, 1973). Locke (1969), reviewing the historical development of the concept of job satisfaction, reports that job satisfaction is caused by challenging jobs, high and equitable remuneration, opportunities for promotion, and good work conditions. In its turn, job satisfaction results in life satisfaction, physical and mental health, less absenteeism and less turnover (in Mitchell, 1979). The
4 importance of job satisfaction in understanding work behaviour is evident from these early research findings. Apart from a characteristic interaction between a person and his job, resulting in a meaningful and satisfying experience of work, a person's own attitudes can govern this impression of work. The attitude of job involvement has been seen as central to work motivation (Kanungo, Misra & Dayal, 1975). A person's level of job involvement gives an indication of the degree of his career development and vocational maturity (Holland & Gottfredson (1976). This attitude is a result of the satisfaction of a person's most salient or important needs (Kanungo, 1982b). It seems that both attitudes of job satisfaction and job involvement hold implications for the meaningfulness of work. In order to extend our knowledge of these phenomena in the life of professionals in the South African context, research in terms of the of person-environment congruence and job satisfaction can be governed by Holland's and Lofquist and Dawis' theories. An attitude of job involvement or lack thereof in the participants can be brought into relation with the participant's level of agreement with his environment and job satisfaction. These associations of work-related variables can be investigated in any given occupation. For the purposes of the present study, the professions of Psychology and Optometry in the RSA can be considered. There are currently 4411 registered psychologists in South Africa, according to the total on the registers of the Interim National Medical and Dental Council (INMDC) on 31 December 1996. Approximately 35% of these are clinical psychologists, 22% are counselling psychologists, 20% are educational psychologists, 17% are industrial psychologists, and the remaining 6% are research psychologists. In the changing South Africa these professionals are faced with ever increasing challenges. Kriek (1996) discusses the challenges that the changing socio-political dispensation in South Africa presents to industrial psychologists, and concludes that it is the task of industrial psychologists to make full use of their knowledge by influencing the policies and formation of the new South Africa. Suggestions in literature concerning the role of psychologists in the post-apartheid South Africa also propound amongst
5 other things organisation development within schools by educational psychologists (Druker & De Jong, 1996), counselling of victims of apartheid (Kagee & Price, 1995) and the rebuilding of the South African society through psychotherapy (Vontress & Naiker, 1995). Likewise one can argue that development in political, social, economic and technological fields is leading to a greater need for the services of psychologists in all spheres of the profession. Apart from operating in private practices, major corporations, businesses, help or research centres, hospitals, government departments, schools and other educational institutes to name but a few, psychologists also impact the changing South Africa through community services and their voice in the media. With such diversity of environments in which psychologists operate, and considering the importance of the tasks psychologists need to perform in our society, it is essential for them to interact in a satisfying way with their work environments. The 1633 registered South African optometrists (INMDC, December, 1996) function in a more homogeneous environment than psychologists. They operate in private practices, but also in group and clinical practices, and are involved in educational institutes and in research. The health service they provide to the community at large is also quite obvious. Consequently, knowledge gained on their interaction with their characteristic environment, the level of satisfaction they may derive from work and their attitude to the job is not redundant. Understanding the nature of the relationship among the constructs of person-environment congruence, job satisfaction and job involvement can greatly increase our knowledge of vocational behaviour, uniting aspects of the theories of Holland (1985b), Lofquist and Dawis (1969) and Kanungo (1982b). Studying these variables in terms of professionals such as psychologists and optometrists in the South African context will enlarge our understanding of the significance of work to people in a changing society and the role of congruence in job selection, satisfaction in adjustment to work and job involvement in these processes.
6 1.2 THE RESEARCH PROBLEM Hackett, Lent and Greenhaus (1991) reviewed vocational behaviour theory and research of the seventies and eighties, including Holland's (1985b) theory and the Theory of Work Adjustment as two of the dominant career development perspectives of this period. They identified the following need for future research on vocational behaviour: "Within the counseling literature, specifically, direct tests of competing models of career development are vital to furthering our understanding of career phenomena and establishing the explanatory power of the various models, and refining or abandoning those that do not fit the data well." . (Hackett et al., 1991, p. 28). Attempts at theoretical integration should include the uniting of conceptually related constructs, more distinct explanations of career outcomes such as job satisfaction, descriptions of relations among disparate constructs such as interests and needs and the identification of key constructs of career development theory (Hackett et al., 1991). Holland (1973, 1985b), Lofquist and Dawis (1969), and Kanungo (1982b) respectively focus on interests, values and needs relating to careers. Elements taken from their theories, namely person-environment congruence, job satisfaction and job involvement will be combined in this study. The theory of person-environment congruence and research supporting the hypotheses thereof form a reliable conceptual framework and basis for further research in terms of variables relating to it. The basic postulate of the congruence hypothesis reads: "Vocational satisfaction, stability and achievement depend on the congruence between one's personality and the environment (composed largely of other people) in which one works" (Holland, 1973, p. 9). In a monograph reviewing research on the congruence hypothesis, Spokane (1985) concludes that research consistently indicates a positive relationship between person-environment congruence and specific variables, including job satisfaction. However, Meir et al. (1986) contest that some of the
7 reviewed studies find only partial or no support for the congruence hypothesis. Contradictory research findings such as these could suggest the influence of other important variables in the complex relationship between congruence and its consequences, such as job satisfaction. Furthermore, Spokane (1985) states that correlational studies give a distorted view of Holland's theory, and is only useful at the initial stages of the development of a theory, yet evidence seem to suggest that findings on Holland's theory can not simply be applied cross-culturally (Glidden-Tracey & Parraga, 1996; Rounds & Tracey, 1996). Our knowledge is incomplete about the relationship of congruence and job satisfaction in the South African context. Clearly there is a need for research that will elucidate the nature of the variables responsible for discrepancies in congruence-job satisfaction research results and for broadening our understanding of these career facets in our own country. This study will follow two directions in response to this need: A more precise definition of the construct of job satisfaction will be utilised; Smart, Elton and McLaughlin (1986) advise researchers to incorporate multiple dimensions of job satisfaction in their studies since the findings of their study indicate that the person-environment fit hypothesis is not equally applicable to all dimensions of job satisfaction. Smart et al. (1986) distinguish between satisfaction of intrinsic and extrinsic needs. The variable of job involvement will be introduced. Organisational psychologists have been progressively studying the role of work in a person's definition of job involvement (Lefkowitz, Somers & Weinberg, 1984). Blau (1987) studying job involvement in terms of a person-environment fit model, suggests that job involvement is partially an interactive function of person and environmental factors, and can be predicted by a person-environment fit model. Blau also underlines the need for further research on the construct of job involvement. Since these findings suggest a relationship between congruence and job involvement, and
8 several studies support the relationship between congruence and job satisfaction, the question can be asked whether there is also a relationship between job involvement and job satisfaction. A possible answer to this question is given by Gruneberg (1979), who concludes that job involvement is a necessary prerequisite of job satisfaction, but is no guarantee of job satisfaction. According to Gruneberg, an individual can be job involved, without being satisfied in his job. However, the level of job involvement is predictive of the level of job satisfaction or dissatisfaction. Kanungo echoes this by arguing that "satisfaction of intrinsic needs at work may be a sufficient but not a necessary condition for work involvement" (Jaeger & Kanungo, 1990, p. 197). Several hypotheses have appeared for the relationship between vocational incongruence and negative outcomes such as a decrease in job involvement, but substantial evidence to prove this is still lacking (Gottfredson & Holland, 1990). According to Rabinowitz and Hall (1977) it is logical to assume a relationship between satisfaction and high job involvement. In literature there are often findings that suggest a moderate positive relationship between job satisfaction and job involvement (Mortimer & Lorence, 1989). There are diverse views in social psychology literature on probable causal relationships between job satisfaction and job involvement. Job satisfaction is seen to either increase or decrease job involvement, or vice versa. Others believe that there is a reciprocal relationship between these two variables, or even indirect influences between the two and yet others suppose that both job satisfaction and job involvement are caused by another variable (Mortimer & Lorence, 1989). These findings propose the possibility of interrelations among these constructs. Consequently one may ask what the nature of the relationship between job involvement and job satisfaction would be, should the dimensionality of job satisfaction be taken into account. Several practical questions also result in terms of the possible interrelations of the constructs of person-environment congruence, job satisfaction and job involvement: If a person works in a congruent job environment, will he also feel psychologically close to his job?
9
If a person identifies psychologically with his job, will the person necessarily feel satisfied that both his intrinsic and extrinsic needs are being met? If a person works in a congruent job environment, will he be equally satisfied with the reinforcement of his intrinsic and extrinsic needs? If people work in congruent environments, and are satisfied with their jobs, will they be more involved in their jobs than individuals who are in congruent jobs, but are not satisfied with their jobs? Will people in different professions also differ in terms of their level of job satisfaction, job involvement and congruency with their jobs? Will males and females differ in terms of the degree to which they are satisfied with their jobs, involved in their jobs and experience congruence with their jobs? In order to find solutions to these questions, this study will seek to confirm hypotheses surrounding the nature of the relationships between personenvironment congruence and job involvement, job involvement and intrinsic versus extrinsic job satisfaction, person-environment congruence and intrinsic versus extrinsic job satisfaction, as well as job satisfaction and different levels of job involvement. These variables will also be compared in terms of the two different professional groups and gender differences. The hypotheses will be elaborated in Chapter Six and will serve to clarify the relationships among these variables. Primarily it aims to show that a distinction between intrinsic and extrinsic aspects of job satisfaction, as well as the variable of job involvement, can determine whether an individual will experience satisfaction in congruent job environments or not. However, it does not imply that job involvement is the only determinant in this complex relationship between job satisfaction and person-environment congruence. With reference and complementary to Holland's theory, knowledge gained from the study could have invaluable implications for career guidance purposes and could provide solutions to vocational problems, such as career indecisiveness.
10
1.3 PURPOSE OF THE STUDY Following from the above, the purpose of this study is to test Holland's congruence hypothesis in the South African context, taking the dimensional aspects of the job satisfaction construct into account. The relation of job involvement to the constructs of person-environment congruence and job satisfaction will also be examined in order to partially account for contradictory findings eminent in literature in terms of the congruence hypothesis. Differences between males and females, and optometrists and psychologists in terms of these variables will also be examined. The validity of one element of person-environment congruence, namely the environment codes of the participants, will be investigated.
1.4 CORE CONCEPTS DESCRIPTION In this section clarity is provided on the specific connotations and denotations of the constructs of person-environment congruence, job satisfaction and job involvement as employed in the present study.
1.4.1 Person-environment congruence A person is "an individual human being" (Collins English ttictionary). Relevant to this study, person refers to the personality type as measured by The person will be described according to the Self-Directed Search. Holland's (1973, 1985b) typology presented in Chapter Four. It can therefore be seen as being predominantly either realistic, investigative, artistic, social, enterprising, or conventional, based on preferred activities, personal competencies, interest in occupations, and self-rated abilities and skills, with secondary and tertiary competencies, interests abilities and skills making up the personality pattern of the person.
11 Environment is defined by the Collins English Dictionary as "external conditions or surroundings, esp. those in which people live or work." In this study only the work environment will be included, and more specifically, the occupational environments of either psychologists or optometrists, according to the number of different personality types in each of these environments (Walsh & Betz, 1995). Congruence, according to the Collins English Dictionary, is "the quality or state of corresponding, agreeing, or being congruent." The term is often used to denote the similarity of shapes with all parts corresponding. The hexagonal shape described in Chapter Four will be utilised in this respect. Person-environment congruence hence refers to the similarity between an individual's personality type and the type of environment in which the individual functions (Holland, 1973, 1985b). It denotes complementary links between the person and his or her work environment (Walsh & Betz, 1995). This similarity or complementariness is indicated by one of the indices of congruence that relates to Holland's theory.
1.4.2 Job satisfaction A "job" in the context of this study only entails the specific and current job in which a person functions. Traditionally Locke's (1969) definition of job satisfaction is employed in research literature. According to Locke, job satisfaction is defined as "an emotional state resulting from the appraisal of one's job as achieving or facilitating the achievement of one's job values" (cited in Saleh, 1981) or as a "positive emotional state resulting from the appraisal of one's job or job experience" (cited in Dunnette, 1976, p. 1300). Whereas Locke emphasises affective components, others regard job satisfaction as a job attitude with cognitive, emotional and behavioural elements that affect various aspects of work behaviour (Berry & Houston, 1993). However, since the construct will be operationally defined by the Minnesota Satisfaction Questionnaire (short form) in the present study, and since the MSC) deals with feelings of satisfaction with job aspects, affective aspects of
12 job satisfaction is emphasised in the current conceptualisation, and includes both intrinsic and extrinsic aspects of the concept. Job satisfaction is defined as "a function of the correspondence between the reinforcer pattern of the work environment and the individual's values" (Dawis & Lofquist, 1984, p. 64), or as the correspondence between an individual's vocational needs and the reinforcers available in the work environment (Thompson & Blain, 1992). The concepts of correspondence, reinforcer pattern, values, needs and reinforcers will be explicated within their conceptual framework in Chapter Three.
1.4.3 Job Involvement Job involvement is defined as "a specifiC cognitive belief state of psychological identification ... with work" (Kanungo, 1982b, p. 70) (ellipse added). It refers to a person's work attitude and "psychological closeness" that an individual experiences towards his current job (Blau, 1987). The construct will be operationalised by Kanungo's Job Involvement Questionnaire (see Chapter Three). It is a conceptually distinct construct from job satisfaction. Whereas job satisfaction refers to an emotional state, job involvement entails a cognitive or belief state, even though both constructs refer to a specific job (Brooke, Russell & Price, 1988). It also differs from job satisfaction, as job involvement can be regarded as a stable attitude and job satisfaction as a circumstantial attitude (Saleh, 1981).
1.5 DISSERTATION STRUCTURE Chapter One provides the rationale of the present study, based on the research problem it presents. The core constructs that guide the study, namely person-environment congruence, job satisfaction and job involvement are defined. Holland's (1973, 1985b) theory as foundation of the present study and background to the constructs of person-environment congruence and job satisfaction is described in Chapter Two. In Chapter Three empirical evidence of the validity and utility of Holland's theory is given.
13
The description of the Theory of Work Adjustment in Chapter Four serves as framework to the assessment of job satisfaction and along with Kanungo's conceptualisation of job involvement further explicate the key constructs of this study. Research relevant to each of these theoretical perspectives is described in Chapter Five. Chapter Six deals with the research design and methodology based on the specific purpose and hypotheses of the study. The findings of the study and explication thereof are given in Chapter Seven. Consequential conclusions and recommendations in Chapter Eight demonstrate the significance of the study.
14
CHAPTER 2 HOLLAMO'S CAREER CHOICE THEORY
2.1 INTRODUCTION Person-environment psychology describes behaviour as a function of the person, as well as the psychological and/or physical environment. Some theorists in this field, such as Moos, and Barker, place emphasis on the environmental influence on behaviour. Others stress the effect of personality on behaviour. The theoretical approaches of psychoanalysts, and Roe's person-environment theory are examples of the latter group. Holland's theory and construct of person-environment congruence belong to a third group, along with theorists such as Pervin and Stern, whose interactional perspectives bear both personality and environmental influences on behaviour in mind (Walsh & Betz, 1995). Holland's (1973, 1985b) theory has been chosen for the purposes of the present study as theoretical foundation of the reciprocal relationship of personality and environment, and particularly environments as represented by the various occupational groups. At the end of this theoretical discussion, the theory will be evaluated in terms of its ability to describe, explain and predict the relationship of personalities with occupational environments. Research findings in relation to the theory will be presented in the following chapter to further elucidate the value of Holland's theory as a person-environment fit model. Holland terms his career choice theory, a "typological-interactive theory" that defines certain types of personality and types of environments, and explores the interaction and fit between persons and environments. It describes vocational behaviour with the purpose of facilitating career choice and change in order to promote job satisfaction. Unlike most career choice theories, it emphasises the role of personality (Costa, McCrae & Holland, 1984). It is one of the major counselling psychology theories that has emerged since the early seventies. The theory was originally formulated in the fifties. It continued to develop and important differences can be seen between the 1973 revision thereof and its refinement in 1985, which included
15 the construct of identity, less emphasis on the construct of consistency, and an in depth look at demographic influences and the development of types (Hackett, Lent & Greenhaus, 1991). This discussion on Holland's theory will start with a brief outline of its foundational principles. Secondly, the person, the environment, the interaction between these, and the predicted behaviour will be described in accordance with the four assumptions of the theory. The five main constructs that operationalise the theory will be emphasised throughout. Thirdly, for the purpose of the present study, the concept of congruence in terms of careers will be highlighted. Fourthly, the measures and indices that operationalise the theory will be explored, as well as the classification system that adds to the utility of the theory. Lastly, due to its relevance to the congruence construct, criticism of the hexagonal model as well as of issues of gender, age and culture will be supplied.
2.2 BACKGROUND PRPNCOPLES Holland (1985b) uses six principles as foundation to the theory. The first principle embodies the notion that a person's vocational choice is an expression of his personality. Therefore, vocational interest should not be seen a separate entity from vocational personality, but is instead an expression of personality. Secondly, interest questionnaires are regarded as personality inventories. Not only is this a logical inference of the previous principle, but this has also become evident through the validation of personality inventories such as the Vocational Preference Inventory (1977). The fact that vocational stereotypes have reliable and important meaning, both psychologically and sociologically, constitutes his third principle. Hence the validity of interest inventories is based on the validity of popular perceptions of various occupations, such as the concept that scientists are unsociable or sales-people are persuasive.
16 According to the fourth principle one can expect similar histories of personal development for members of the same vocation. Since people with similar personalities are attracted to the same occupations, they form groups with members that are similar in many ways. Following from the fourth principle, Holland (1985b) states that due to the similar personalities in a given vocational group, responses of group members to many situations or problems will be similar. These personalities will also create interpersonal environments that are characteristic to them. Indirect evidence of this hypothesis is found in studies which indicate that the overall feel of a milieu is a reflection of the most prominent personality type in that environment (Eberhardt & Muchinsky, 1982). The final principle on which the theory rests is most crucial to the present study. According to this principle, congruence between personality and work environment is the determining factor for vocational satisfaction, stability and achievement (Holland, 1985b; Smart et al., 1986).
2.3 ASSUMPTOONS OF HOLLAND'S THEORY Holland's theory (1973, 1985b) revolves around four primary assumptions on personality, environment, the interaction between personality and environment and behaviour resulting from this interaction. Each of these assumptions will be cited to introduce the relevant theory that will follow. The secondary assumptions based on the principles of consistency, differentiation, congruence and calculus operationalise the theory and will be highlighted throughout the discussion of the four primary assumptions.
2.3.1 Assumption 1
In our culture, most persons can be categorised as one of six types: realistic, investigative, artistic, social, enterprising, or conventional.
17 Holland (1985b) describes these types as being theoretical models against which real people can be examined. Each of these typological models represents different categories of attitudes displayed by people. A person is hence categorised in terms of his resemblance to one of these types. However, a person often displays attitudes of more than one of these types, and this constitutes his subtypes and personality pattern (Holland, 1995). This assumption will subsequently be amplified by examining the development and the nature of the various types and subtypes, and how an individual's different attitudes constitute a specific personality pattern. The interrelation of the types will then be discussed as well as the behaviour ensuing from it.
2.3.1.1
The development of personality types
A child's biological predisposition, potential and early experience give rise to preference of (or aversion to) certain activities. This matures into the differentiation of specific interests and values, which in turn develops specialised personal competencies. From this, the person's specific disposition or personality type evolves, resulting in related personality traits, attitudes and behavioural patterns. These traits, attitudes and behaviours form a person's repertoire of skills and coping mechanisms (Eagan & Walsh, 1995). Environment plays an important role in this development of personality in as far as it provides opportunities and reinforcement of preferences, interests, competencies and values in accordance with the type which dominate a given environment (Holland, 1985b). Traits are developed by a child's exposure to an environment with particular characteristics, and by participation in activities relating to that specific type of environment. These activities engender related interests, competencies and dispositions. According to Holland (1996) research attests to the fact that these personality dispositions are a result of both life history and biological disposition.
18 2.3.1.2
Models of dominant personality types
Holland (1973, 1985b) identifies six dominant personality types, each with its distinctive character traits. The letters of the acronym RIASEC are each representative of one of these types. As an example, the personality types of the present study's participants, namely psychologists and optometrists may be considered. The psychologist's and optometrist's personality types are both predominantly investigative (Taljaard & Von Mollendorf, 1987). The investigative personality shows preference for investigative situations and occupations, which entail activities of investigating physical, biological or cultural phenomena. On the other hand, investigative personalities show aversion to enterprising situations, i.e. activities involving persuasive, social or repetitive activities. Accordingly, they have developed competencies to solve scientific and mathematical problems. They see themselves as being scholarly, intellectual, mathematical and as having scientific ability. Therefore they also value scientific activities. Actual traits associated with them include being intellectual, curious, precise, unpopular and so forth (Holland 1985b). It is important to note that people exhibit the personality traits and behaviours of a particular type in different levels of proximity to that type (Holland, 1995). For a full summary of the predisposed preferences, competencies, setf-perceptions, values and personality traits of the six dominant types, see Annexure A.
2.3.1.3
Subtypes and personality patterns
A person usually displays characteristics of more than one of the dominant types. A profile of a person's personality can be obtained by means of the The Self-Directed Search or the Strong Vocational Interest Blank. personality pattern, consisting of two to six variables, defines the subtype to which a person belongs. Examples of personality patterns derived from the SDS include OAS, REA, etc. According to The South African Dictionary of Occupations (1987) both optometrists and psychologists are high level workers with a "medical" easy code, and an ISE personality pattern. In other words, although psychologists and optometrists are predominantly investigative personalities, they also have preferences, competencies, setfperceptions, values and traits of social personalities, and to a lesser degree of enterprising personalities. This is with the exception of psychologists in
19 the research category, who have an ISC personality pattern. Holland (1985b) classified psychologists under the OAS code. However, since personality patterns may differ, one may find psychologists with patterns differing from the norm, for instance with ISA personality patterns and so forth. The constructs of consistency, differentiation and identity further define subtypes. Consistency in personality patterns refers to personality types of which related elements have mutual characteristics. For example in the subtype ROA, the realistic and investigative types share traits such as unsociability and an orientation towards things, and it is therefore a consistent pattern. The realistic and artistic pattern is less consistent however, since it displays opposing traits such as the need for explicit, ordered or systematic activities (realistic) versus preference for ambiguous, free and unsystematised activities (artistic). The following patterns are inconsistent or have very low consistency: RS, 1E, AC, ST, El, CA (Holland, 1985b). From this, one could infer that the ISE pattern of psychologists and optometrists have low consistency. Differentiation is operationally defined as the "numerical difference between and individual's highest and lowest scores on scales measuring Holland's six types" (Swanson & Hansen, 1986). It refers to how clearly a personality is defined, or the degree of variation within a personality profile (Alvi, Khan & Kirkwood, 1990). Two seemingly identical personality patterns could differ in terms of the differentiation thereof. For example, the personality pattern SAIERC of one person can differ from the SAIERC pattern of another person, if the value of the difference between the S and C scores differs. An undifferentiated personality pattern will display little difference between the type a person is most like and the type he resembles the least (Holland, 1973, 1985b). Identity applies to the clarity and stability of a person's vocational goals and self-perceptions in terms of talents and interests. Identity is measured by means of an eighteen-item Identity Scale of the My Vocational Situation. When a person has only a few occupational goals, that relate to a few main occupational categories, his personality type has strong identity, whereas a
20 person who has diverse and several different occupational aims has weak subtype identity (Holland, 1985b). The level of a person's vocational identity is predictive of his career stability or instability. It is representative of several of a person's vocational beliefs and strategies, and relates positively to job satisfaction and psychological health (Holland, 1996). In short, people do not only belong in various levels of proximity to one of six dominant types, but also differ in terms of the degrees of consistency, differentiation and identity of their subtypes.
2.3.1.4
Typological relationships
The interactions among personality types are depicted on the following hexagonal model (Figure 2.1), from which the consistency and congruence of types and behaviours can be calculated in accordance with their spatial relations on it: REALISTIC
INVESTIGATIVE
CONVENTIONAL
ARTISTIC
ENTERPRISING
SOCIAL
Figure 2.1: The hexagonal configuration of types (Holland, 1985b, p. 29). Two types that are geometrically close are consistent with each other and hence similar, whereas opposite types on the hexagon are inconsistent types (Holland, 1985b). Person-environment congruence can also be
21 calculated by means of the hexagon (see point 2.5.3). Research findings concerning the hexagonal model will be described in the following chapter.
2.3.1.5
Behaviour
Participation in activities develops the personality type of and individual. Likewise, the personality type determines subsequent behaviour, including vocational choice. The consistency, differentiation and identity of a personality pattern determine the degree of predictability of behaviour. Holland (1985b) formulated several hypotheses relating to expected behaviour when the personality pattern is known. Some of the relevant predictions in regard to career related behaviour, educational behaviour and social behaviour will now be briefly discussed. The personality pattern is predictive of primary career choice in terms of the dominant personality type and further career choice in terms of the subtype of a person, occupational performance, stability, aspirations, achievement, creativity and level of accomplishment. Congruence of personality patterns and occupations relates positively to career involvement and satisfaction. A person's ability to cope with job change and unemployment relates closely to the specific personality type of the person and proximity to the type (Holland, 1985b). Aspiration, achievement and responsiveness in educational behaviour are functions of personality pattern. Social behaviour is affected by personality pattern in as far as personality pattern is responsible for the level of participation in leadership, choice of avocational activities, responsiveness to others and interpersonal competence. On general the level of a person's consistency and differentiation is predictive of the person's responsiveness to environmental demands and reaction to environmental stressors (Holland, 1985b). Holland's typological model thus describes six dominant personality types, and a specific personality pattern that is partially inherited and partially a result of personal history. This personality pattern can be represented on the hexagonal model. A person's personality type is influenced by the level
22 of consistency, differentiation and identity of his subtype. personality pattern, behaviour is predicted.
From the
2.3.2 Assumption 2
There are six model environments: realistic, investigative, artistic, social, enterprising, or conventional. In accordance with this assumption regarding environments a discussion on the six environmental models will now follow, as well as information on the effects of these models, and the characterisation of these models in terms of • levels of consistency, differentiation, identity and calculus.
The nature of environmental models
2.3.2.1
Holland (1985b) defines an environmental model as the situation or atmosphere produced by the people who dominate a given environment (1985b). The six environmental types exist parallel to the six personality types, reflecting the traits of its members. Each environmental type is the source of correspondent environmental experiences that reinforce the traits of the environment within its members. A person in a social environment for example could develop his traits of being friendly, persuasive and kind.
2.3.2.2
Influences of the environmental models
The environment consists of the specific climate characterising the nature of its members. Although it is stated that "types flourish in congruent environments" (Holland, 1985b, p. 35) the particular influence of an environment on a person will vary according to the following factors: the influence of the specific subenvironmental unit in which a person functions; the person's perception of the environment;
23
a distinction between institutional influence versus the influence of personal relationships; the distribution of power in a given environment; and the size and complexity of the environment. In other words, just as the members of an environment do not fully display the traits of only one personality type, the environment is also never homogeneous, but can consist of subenvironments and different psychological fields, which should be taken into consideration when examining the reciprocal influences of persons and environments. The environmental types are models that facilitate our understanding of the interaction between people and various environments and in so doing facilitates research and interpretation (Holland, 1985b). Each of the environments has certain primary influences on its members: It stimulates its members to perform activities that relate to their preferences; It encourages the competencies and achievements of its characteristic members; It fashions the self-perception of the members as well as their perception of the world; It rewards its members for type-related values (Holland, 1985). For example, an investigative environment encourages its members, for instance psychologists and optometrists, to occupy themselves with investigative activities and to develop investigative competencies such as scientific and mathematical problem solving skills. The investigative environment also stimulates its members to see themselves as scholarly and intellectual and as having mathematical and/or scientific skills and to see the world in complex, abstract, independent and original ways. Due to the investigative environment, its members value science.
24
2.3.2.3
Consistency, differentiation, identity and calculus
Each environmental model is further defined by the constructs of consistency, differentiation, identity and calculus. Consistency in terms of the environment refers to the proximity of relations of environments. Environments that are geometrically close on the hexagon (Figure 2.1) are consistent with each other and could have similar activities, competencies and rewards. Examining the hexagonal model, it is evident that social and artistic environments for instance are consistent with each other, but artistic and conventional environments are not. When by way of illustration one examines the influences of these environments respectively, it is also clear that the personal, emotional and imaginative influences of the artistic environment are inconsistent with the materialistic influences of the conventional environment. Holland hypothesises that the consistency of an environmental pattern will promote stability of vocational choice as well as stability and level of vocational achievement (Holland, 1985b). Differentiation pertains to the degree to which and environment is defined. Ot can be calculated by finding the percentage difference between the most and the least type-related personality types within a given environment. In other words, if there are substantially more conventional personality types than social personality types in a conventional environment, one could say that that conventional environment is well differentiated. Differentiation of an environment encourages the stability of vocational choice and level of vocational achievement (Holland, 1985b). Identity involves the presence of clear, integrated and stable goals, tasks and rewards of an environment. Clear environmental identity is evident in an environment that encompasses only a few occupations, whereas identity of an environment can be very diffused when there are many occupations represented by that one environment. The number of distinct three-letter code occupations at a specific level of education in an environment constitutes the number of different behaviour settings. An environment displays clearer identity when it encompasses only a few behaviour settings. In an environment where there are only a small amount of classifiable
25 occupations at a specific level of education, it is assumed that its members will display stability in their vocational aspirations and careers and stability in their level of vocational achievement. Calculus concerns the similarity of environmental types as defined by their proximity on the hexagonal model. The role of this in determining consistency of environmental types has already been discussed. The hexagonal relations furthermore provide information on the similarity as well as discrepancies and internal relations of environments. By means of it comparisons can be drawn between the environments (Holland, 1985b). The levels of identity, differentiation and consistency of environments determine the nature of the influence of the environment. More specifically, the higher the level of identity, differentiation and consistency of an environment, the greater will be its influence on the vocational, educational, social and organisational behaviour as well as personal effectiveness of its members. Relevant to the present study, the following prediction can be made: "The consistency, differentiation, and optimal staffing of an environment promote involvement, satisfaction, and stability of vocational choice" (Holland, 1985b, p. 44). In this statement, "optimum staffing" relates to environmental identity. It is evident from the above that there are six environmental models that are parallel to the six personality types. Not only are the environments determined by the characteristic personality type of its members, but the environment also controls the behaviour, and even organisational behaviour of its members, in accordance with the nature of a given environment. From the above one could infer that differences in the level of satisfaction of the psychologists and optometrists of the present study are partially brought about by the nature of the environments of which they form part.
26 2.3.3 Assumption 3
People search for environments that will let them exercise their skills and abilities, express their attitudes and values, and take on agreeable problems and roles. Assumption three demonstrates the reciprocal role of a person and his environment. The discussion of the theory relating to this assumption will firstly deal with the demands and opportunities presented by the environment; secondly, the environmental response to the activities, competencies and achievements of its members will be described; thirdly, the role of the social environment in vocational choice will be presented; lastly, the effect or vocational beliefs and strategies in the personenvironment interaction will be mentioned.
2.3.3.1
Demands and opportunities
Since each of the six environmental types places characteristic demands on its members, and presents opportunities in accordance with its specific nature, it is understandable that a person will strive to be in an environment where the demands of the environment are met by the competencies, selfperceptions and personality traits of the same type of person. This assumption implies for instance that an investigative type person will seek to belong to an investigative environment. Also, the specific opportunities provided by each of the environmental types will be matched with preferences and values of the same type person (see Annexure B). For example, an enterprising environment's requirements and opportunities entail the manipulation of others to attain organisational and personal goals and an enterprising person displays preference for manipulation of others for organisational goals or economic gain, and values political and economic achievement (Holland, 1973, 1985b).
27 2.3.3.2
Activities, competencies and achievements
Each environment stimulates activities that relate to it, and places demands on its members to have type-related competencies. These competencies result in certain achievements for which the environment rewards its members. A realistic environment, for example, encourages its members to perform realistic activities, such as using machines and tools, and rewards them for valuing money, power and possessions. This is the motivation for seeking out corresponding environments (Holland, 1973, 1985b). Similarly, each environmental type has specific requirements, presents specific opportunities and rewards its members for specific activities, abilities and achievements, and thus reinforces the choice of environment (Holland, 1985b).
2.3.3.3
Vocational choice
Holland (1985b) hypothesises that an individual's dominant personality type is predictive of his primary direction of vocational choice. The second and third letter of an individual's subtype point to secondary and tertiary directions of vocational choice. Whereas the dominant personality type reflects which one of the six environmental types the individual's occupation can be categorised as, the subtype specifies the subgroups of occupations to which the individual will aspire. To a lesser degree, the members of an environment will also seek out similar people to become part of that environment.
2.3.3.4
Vocational beliefs and strategies
In order to enhance the explanatory power of his theory, Holland (1996) has incorporated the constructs of vocational beliefs and strategies, which further depict the interaction between personality and environments. Each personality type has different beliefs that can be assessed by means of the Career Attitudes and Strategies Inventory. Similarly, each environment encourages different belief systems. For example, in the occupation of
28 psychology there is a belief in the complexity of human behaviour. Consequently, psychologists have an analytical attitude. According to this assumption of the interaction between personality and environment, people will seek environments in which they will belong and in which they will be rewarded. However, it is evident from this discussion that the environment itself also fulfils an important role in this selection process. It is therefore rather an interaction between persons and environments that leads to the maintaining of the homogeneity of the environment (Holland, 1985b).
2.3.4 Assumption 4
Behaviour is determined by an interaction between personality and environment. Under point 2.3.1.5 the influence of personality on behaviour has been discussed, and under point 2.3.3.1 the influence of the environment on behaviour was expounded. In this section behaviour as a result of the interaction between personality and environment will be described. The importance of the level of interaction will be explained, with specific emphasis on congruency and its role in the person-environment interaction and the outcomes of this interaction. Certain behavioural predictions made by Holland will be mentioned. Thereafter the structural model depicting several facets of the interaction between person and environment, and the outcomes of these interactions will be described.
2.3.4.1
The level of person-environment interaction Both vocational and avocational behaviour are results of the interaction between a person's specific personality pattern and the environment and subenvironments in which he functions. The predictability of the behaviour is dependent on the level of the interaction, and is also based on several secondary assumptions. The level of interaction between personality and environment is defined in terms of congruence, differentiation and
29 consistency of personality and environmental types, and further characterised by degrees of identity and level of education (Holland, 1973, 1985b).
(a)
Degrees of congruence
Congruence is regarded by Holland as the most influential variable in the person-environment relationship. It can be defined as the level of similarity between personality type and environmental type (Meir, 1989). In vocational psychology it is defined as the degree of fit between personality and jobs (Thompson, Flynn & Griffith, 1994), or the degree of fit between personality type and occupational choice (Meir, 1994). In terms of the hexagonal model, the highest degree of congruence occurs when personality and environmental type correspond. The second highest degree of congruence is seen between two adjacent personality and environmental types. Personality and environmental types opposing each other on the hexagonal model are incongruent (Holland, 1985b). In practical terms, congruency implies that a congruent environment encourages a person's favourite behaviour repertoires in familiar and compatible conditions. As previously indicated, congruence of an environment and personality type involves that the person is encouraged by the environment to perform activities for which the person shows preference, to use the skills at his command, and to view himself and the world in a way he already does. Considering the participants of the present study, this will entail that an optometrist with a predominantly investigative personality will feel at home in an investigative environment, due to the activities, preferences, skills and perceptions nourished by the environment. The environment will also encourage the proper amount of interpersonal contact. Conversely, an enterprising personality type, for example a marketing manager, will have inappropriate skills, preferences, personality traits and values for the investigative environment. The degree of congruence not only influences the interaction between person and environment, but also affects the consequences of the interaction. Meir (1989), examining the development of the conceptual meaning of congruence, states that congruence does not only refer to the
30 agreement of personality and environmental type, but congruence has become the independent variable in the mapping sentence: Level of congruence
(=>
Well-being.
A distinction can be made between person-environment congruence with consequences of job satisfaction, achievement and stability and personalityaspirational congruence, which is correlated to stable aspirations and less need for career assistance (Thompson, Flynn & Griffith, 1994). Muchinsky and Monahan (1987) differentiate between supplementary and complementary person-environment congruence. Supplementary congruence refers to the fit between an individual and the people who constitute an environment, and is indicated by individual satisfaction, performance and tenure. Complementary congruence on the other hand refers to the fit between an individual's talents and the corresponding environmental needs, as indicated by organisational level variables. Supplementary models define the environment in terms of its members, but the definition of the environment according to the complementary model is independent of the characteristics of its members. Congruence according to Holland can therefore be seen as supplementary congruence.
(b)
Degrees of consistency
The degree of consistency influences the interaction between personality and environment to a much lesser degree than the other factors. Consistent environmental models manifest an integration of similar requirements and rewards, and consistency influences the predictability of the behaviour that members of an environment will exhibit. Consistent persons and environments result in predictable outcomes and greater reciprocal influences of persons and environments (Holland, 1973; 1985b). According to this, the environment of an optometrist, being predominantly investigative, is consistent if there are few artistic requirements for instance. Consequently, the behaviour and level of influence of the environment on the optometrists will readily be predicted.
31 Degrees of differentiation Differentiation or definability, in combination with congruence, consistency and identity, affects the interaction of personality and environmental types, since identical personality and environmental patterns can have different degrees of differentiation. When both the personality type and environmental type are well defined, predictability of the nature and extent of type-related behaviour is enhanced. In a practical sense, an organisation that is differentiated will not place diverse or even competing demands on an individual, so that the individual will know what is expected of him or her. Similarly, a person with differentiated interests will not be swayed by every environmental demand that is placed on him or her (Holland, 1973; 1985b). In terms of differentiation of interests (the degree to which and individual's interests are clearly defined), Swanson and Hansen (1986) suggest that one should also distinguish between strong interests that are undifferentiated (high score undifferentiated profiles), and weak interests that are undifferentiated (low score undifferentiated profiles). They explain that people with strong and undifferentiated interests are more likely to display predictable behaviour such as staying on in college and achieving better academic results than those with weak and undifferentiated interests.
Degrees of identity Identity also plays a role in the predictability of the person-environment interaction (Holland, 1985b). Identity, with regard to personality type pertains to the clarity and stability of talents, interests and goals, and with regard to environment refers to the clarity, integration and stability of environmental goals, for example occupational goals. As is the case with the other factors, high degrees of identity also increase the predictability of behaviour.
(e)
Levels of education (GED)
The average level of education of people in an environment, and the level of education of an individual is a further factor influencing the person-
32 environment interaction. Outcomes of the person-environment interaction, such as satisfaction and achievement can also be the result of a high degree of approximation of environmental and personal levels of education. The GED is correlated to the variables of prestige, intelligence, educational level and social class (Holland, 1985b).
2.3.4.2
A structural model of person-environment interactions
The structural model presented in Table 2.1 is a summary of the relevant information to be gathered when examining the congruence of persons and environments. Since positive outcomes are directly related to congruence, it is a useful tool in career counselling: Table 2.1 A structural model of person-environment interactions PERSON
>
ENVIRONMENT -->
OUTCOMES
(BEHAVIOUR SITUATIONS) Activities
Opportunities
Interests and Satisfaction
Competencies and GED
GED and Special Skill Demands
Achievement and
Encouragement
Stability
Encouragement
Outlook
Disposition
Perceptions
Satisfaction
Self World Values
Values
Traits
Encouragement
Personality
Encouragement
Repertoires Resources (Economic, Educational, Social)
Traits
Repertoires Sufficiency and Flexibility
Increase of all above outcomes
Holland (1985b, p. 52).
33
From this model, relevant information on the role of congruence in relation to job satisfaction can be deduced. It is apparent that congruence between activities and opportunities results in satisfaction. Congruence between the skills demands of an environment as well as the general educational level of its members on the one hand, and the educational level of a person as well as his competencies on the other hand, has outcomes of satisfaction and achievement (Elton & Smart, 1988; Sutherland, Fogarty & Pithers, 1995). Vocational satisfaction is encouraged by the congruence of a person's work environment (Holland, 1985b). Interestingly, this model also recognises the impact of a person's economic, educational and social resources on the outcome of satisfaction as a result of specific congruence. Stability is a result of a person's disposition and the encouragement from the environment. Congruency thus increases the probability of staying in an environment, but also to continue performing well in the environment (heir et al., 1986). From these considerations of the interaction between personality and environment it is apparent that the behaviour resulting from the interaction is moderated by the degrees of congruence, consistency, differentiation, identity and levels of education of the personality and social environment. Of these variables, congruence seems to be the most influential factor culminating in positive behaviour and work outcomes. In addition to this, particular interactions between personal qualities and behaviour result in specific outcomes, including satisfaction, stability and so forth.
2.4 PERSON-ENVIRONMENT CONGRUENCE AND CAREERS Having discussed the background principles and assumptions of Holland's (1973, 1985b) theory, the construct of congruence and consequent outcomes such as job satisfaction has already been made clear. Since the present study deals specifically with person-environment congruence with regard to occupations, it is fitting to discuss the particular effect of congruence in terms of career choice, stability, and change, and well as career development and coping mechanisms. Knowledge of the level of person-environment interaction provides the means for predicting development, vocational and educational behaviour and development, as
34 well as personal development, effectiveness and responsiveness to environmental influences (Holland, 1985b).
2.4.1 Career choice, stability and change Personal development plays an important role in vocational choice, work involvement and satisfaction. The environment, composed of parents, schools, and the neighbourhood reinforces certain activities, interests, selfestimates and competencies (behaviour) in a person. This develops a typological disposition, which in turn results in a cluster of personality traits. A person selects an occupation in accordance with this predisposition (Holland & Gottfredson, 1976). Growing up with congruent parents leads to congruent career choices and ultimately to achievement, positive reinforcement and satisfaction. Incongruent parents convey inaccurate self and world perceptions, and inconsistent interest, values, competencies and finally poor functioning in environments (Holland, 1985b). Satisfaction and success in the chosen occupation are results of the congruency of the person and the work environment. An incongruent environment, or insufficiently defined and inconsistent personality pattern result in lack of work involvement. Vocational success again can be increased by personal resources such as money and good health, whereas factors such as gender and poverty could interfere with success (Holland & Gottfredson, 1976). Occupational choice is influenced by the variables of economic position, race and gender that affect the development of personality. Career choice is also dependant on the decision-making ability of people. High levels of consistency and differentiation of personality go hand in hand with positive decision-making ability. Moreover, different personality types also differ in terms of their decision-making capacity, in as far as investigative personalities are the best decision makers and conventional personalities are the worst (Holland & Gottfredson, 1976). Throughout the course of a person's life and career, a person develops behavioural and personality repertoires due to the person-environment interactions. When an environment is reinforcing and satisfying due to
35 complementary environmental and personality patterns, stability is induced, and change is a result of incongruent interactions. In the event of incongruent interactions, a person will either seek a new congruent environment, try to change his present environment or his own personal behaviour and perceptions (Holland, 1985b). Congruent environments engender stability in the direction and level of a person's career choice as well as stability in the level of occupational or educational achievement (Holland, 1985b). Career stability (staying in the same kind and level of work) is also due to the fact that the work environment requires specific experience from employees and there is a high cost in energy, money and so forth when career change takes place (Holland & Gottfredson, 1976). Alluding to research findings that are supportive to his theory, Holland (1996) concludes that both aspirations and interests have an effect on the stability of work histories. Research results indicate that even when career change takes place, people tend to remain in the same occupational category with subsequent jobs. Moreover, vocational aspirations or intentions also manifest this continuity. Besides, occupations are chosen in accordance with a person's dominant interests. Research findings also confirm that career stability increases with age. People nevertheless change careers. Dissatisfaction, personal incompetencies, pressure from others to leave and other personal and environmental reasons such as opportunities to increase salary, or congruency, can all be cited as reasons for career change. Sometimes factors outside the scope of the theory can cause career change, for instance the closing of a company. Subsequent job search however is still based on a search for congruency (Holland & Gottfredson, 1976).
2.4.2 Career development Although Holland's theory does not have a developmental approach on careers, this typological model nonetheless underlines certain concepts of developmental processes. The following factors of career development theory can be accounted for by means of Holland's model:
36
Consistency of career choice can be explained by means of the classification system and the hexagonal model; Realism of career choices (an agreement between interests, preferences, abilities, social class on the one hand and career choice on the other) is seen in the differentiation of the SDS profiles, as well as the consistency of profiles which can indicate the predictability of vocational choices; Career choice competencies (decision-making ability) is evident from evaluating the match between current vocational aspirations and interests, competencies and self-estimates as measure by means of the SDS; Career choice attitudes are seen in the consistency and differentiation of SDS profiles (Holland & Gottfredson, 1976). In other words, Holland's theory accounts for the degree of career development for which the summary codes of aspirations or occupations are calculated (Holland & Gottfredson, 1976). Self-expressed vocational aspirations and preferences, and in particular, a hierarchy of vocational aspirations, are predictive of future career choice (Holland, Gottfredson & Baker, 1990). Stable aspirations and less need for career assistance are results of congruence between personality and occupational aspirations. One can also predict that a person will have congruent and stable employment and aspirations if there are high levels of similarity among past and current jobs (employment coherence) and similarity between occupational aspirations and preferred occupations (aspirational coherence), (Thompson, Flynn & Griffith, 1994).
2.4.3 Coping with career problems The consistency and differentiation of personality and environmental types determine the coping style of a person. Although age is related to important life events and can clarify vocational predictions, people tend to experience similar vocational problems regardless of age. Congruency and other
37 vocational aspects can also be assessed regardless of a person's age. Research results indicate that person-job congruencies and job satisfaction increase with age and there exists substantial evidence for the stability of interests (Holland & Gottfredson, 1976; Holland, 1996). The degree of change in careers can be determined by means of this model. The least amount of change occurs when a person works in a congruent situation. When the environment changes in a positive direction (more opportunities) or a negative direction (e.g. loss of a job) the personality type is the determining factor of the coping behaviour that will follow (Holland & Gottfredson, 1976). As a result of congruence, effective coping with the environment leads to personal and vocational adjustment, integration, vocational maturity and `planfulness' (Eagan & Walsh, 1995). From this early work of Holland and Gottfredson (1976), the behaviours predicted by Holland (1985b) and Holland's (1996) summary of pertinent research findings the usefulness of the theory in describing vocational choice, development and coping skills is clear. Once again the relationship between congruence and job satisfaction, and also congruence and work involvement comes to the fore. Evidently a variety of demographic variables govern career stability and choice in addition to personality type. However, congruence, consistency and differentiation remain the predominant factors affecting career choice, development and coping mechanisms.
2.5 MEASURES AND INDICES RELATING TO HOLLAND'S THEORY In the preceding sections of this chapter personality and environmental models, and the hypothesised interactions between these in accordance with variable such as congruence, consistency, differentiation and so forth, have been described. A basic requirement of person-environment fit models is that it must be quantifiable. Stated differently, a good person-environment interaction model necessitates accurate measurement of both persons and environments, as well as the association between these (Osipow, 1987). The heuristic value of Holland's person-environment fit model will now be demonstrated in the description of the associated ways of measuring personality and environments. Most importantly too, a list of various indices
38 of congruence that have been developed to calculate the quality of the interaction of persons and environment in accordance with Holland's model will be provided, before the classification system is described. Measures and indices of this theory seem to provide the means for assessing personality, environment and the quality of a match between them. This contemplation firstly gives a short description of measures for assessing personality according to Holland's typology; secondly, a method for measuring the environmental type is mentioned; thirdly, a list of indices of congruence and their strengths are given as methods of quantifying the associations between personality and environmental types; lastly, Holland's classification system is discussed, as it adds to the heuristic value of the theory.
2.5.1 Assessment of personality types Both quantitative and qualitative measures are applied for the assessment of personality types. Qualitative assessments are methods such as selfreporting of interests (Brown et al., 1984). Holland has developed quantitative measures that are used either to verify and validate his theory, or as instruments in career counselling (Shari, 1992). The Vocational Preference Inventory (VPI; 1977) gives an indication of interest or lack of interest in 160 occupational titles. Results from the VPI should be interpreted in combination with demographic variables such as age, gender, GED and so forth (Shari, 1992). The Self-Directed Search (SDS; 1979; 1985a; 1985c) yields a summary code of a personality type. The personality code can be linked to the occupational code that can be obtained by means of The Occupations Finder (1985). The Occupations Finder comprises the 500 most common US occupations categorised according to threeletter codes (Shari, 1992). The goals and uses of the SDS include that it provides feedback on the individual's current vocational status, gives reassurance in terms of occupations desired, stimulates thinking and action by means of The
39 Occupations Finder, and increases occupational knowledge (Brown et al., 1984). The Vocational Exploration and Insight Kit (VEIK; 1980) is a comprehensive package with the purpose of increasing the amount of future occupations a person would consider. It provides information on personal requirements for jobs and careers, draws relations between past experience and present vocational goals, and clarifies the individual's present position and immediate action steps to be taken (Brown et al., 1984). As its name implies, the Career Attitudes and Strategies Inventory measures attitudes and strategies that may affect career change, work performance and job satisfaction. Its subscales measure job satisfaction, work involvement, skill development, dominant style, career worries, interpersonal abuse, family commitment, risk-taking style, and geographical barriers. This inventory can increase the informative capacity of the typology (Holland, 1996).
2.5.2 Assessment of the environment The Environmental Assessment Technique (EAT) developed by Astin and Holland (1961) provides the means for categorising environments according to environmental patterns by assessing its population (Walsh & Betz, 1995). More specifically, the frequency of careers and preferences of population is calculated and according to the results, the personality types of the population are ordered from the most dominant type to the least significant type. These results are then converted to percentile scores to yield an environmental code. (Holland, 1985b). One important result from this has been the development of the Dictionary of Holland Occupational Codes (Gottfredson, Holland & Ogawa, 1982), that organises 12 099 occupations according to the hexagonal model and level of education.
40 2.5.3 Indices of congruence Several indices have been developed to support the five constructs that operationalise Holland's theory, namely consistency, differentiation, identity, congruence and calculus. These indices provide the means for calculating the level of interaction between personality and environmental types (Spokane, 1985). Focus will be placed only on some of the indices of congruence, for the purposes of the present study. These indices are not interchangeable, but differ in terms of measurement sensitivity and discriminatory abilities (Brown & Gore, 1994). They depict the utility of the construct of congruence, and allow for the comparison of personality and educational, vocational or other environments. According to these indices the interrelation of types can be classified as either being congruent or incongruent (Holland, 1985b). The earliest index of congruence, the Dichotomous first-letter agreement assigned congruence scores of 1 (congruent) or 0 (incongruent) based on a comparison of an individual's first letter code and the psychology code (Holland, 1963). Holland described another measure of congruence in the SDS manual (1972). Congruency, according to this measure occurs when there are identical occupational and personality codes or identical first two letters and the same order of codes or same but out-of-order three letters. Two types are partially congruent when the codes have identical first letters only or when the first two letters of one code are the same but in a different position from the second code. Any other two sets of codes represent incongruent types. First-letter agreement refers to an assigning of scores ranging from 1 (low congruence) to 4 (high congruence) according to the highest orientation themes of people (Holland, 1973). Two-letter code agreements yield scores of 1 (lower congruence), 2, or 3 (higher congruence) according to the agreement of the first two letters of the personality and environmental codes (Healy & Mourton, 1983). Moreover, three-letter code agreements yield scores of 0 (incongruent), 1 (partially congruent), and 2 (congruent) according to all three letters of the code (Wolfe & Betz, 1981).
41 The Z-S Index assigns scores ranging from 0 to 6, with highest congruence at 6 for three-letter codes (Zener & Schnuelle, 1976). The Rank Comparison Congruence Scale (RCCS) and the Compatibility Index (CI) are both expansions of the Z-S Index. The RCCS assigns scores ranging from 0 to 10 for three-letter codes (Robbins, Thomas, Harvey & Kandefer, 1978) and the CI gives scores ranging from 0 to 8 according the weighted first, second and third letters of codes (Wiggins & Moody, 1981). Kwak and Palvino (1982) combined information concerning the hexagonal model (depicting the degree of interrelations of types) and three-letter code personality patterns (the order of the letters represents the relative strength of the characteristics) to develop their index of congruence. They identified three levels of person-environment congruency: High congruency:
The first letter of the environmental code is the same as the first letter of the personality pattern;
Moderate congruency: The first letter of the environmental code is adjacent to the first letter of the personality pattern; Low congruency:
The first letter of the environmental code is neither adjacent nor similar to the first letter of the personality pattern.
Their aim was to develop an index that would not only take into account the first letters, but would include all three personality or environment letters. This led to the development of their mathematical model of congruency for use with three-letter codes. This index allows for the comparison of any two sets of codes, for example comparing two personality codes at two different times. X = (W1 + W2 + W3)-1 (wiAD + W2BE + W3CF)
A, B and C respectively represent the first, second and third letters of the personality code and D, E and F respectively represent the first, second and third letters of the environmental code. Wi , W2 , and W3 are the arbitrary weights of 4, 2 and 1 that have been assigned to the dominant, secondary
42 and tertiary code letters (Kwak & Palvino, 1982). This model can indicate approximately 14 000 levels of congruence (Spokane, 1985). When compared to the Z-S Index, this mathematical model presents greater sensitivity, discriminates among all comparisons, for example, same but outof-order codes, and mathematically emphasises first-letter identity between codes. On the negative side, it is more complex than the Z-S Index and requires more calculations (Kwak & Palvino, 1982). Holland (1985b) endorses the Zener-Schnuelle and lachan measures of congruence. lachan (1984) developed an index (the M-Index) to determine congruence among various aspects of the classification system, and specifically between the SDS personality code and self-reported occupational preferences. Scores of this mathematical index are weighted according to matching positions of letters in codes and range from 0 to 28. When compared to the Z-S Index the M-Index displays finer discrimination between same but out-of-order codes and is more accurate than the Z-S Index. Compared to the K-P Index, it has more vocational and avocational applications, and whereas tabulation of the K-P Index would require 14,400 entries (all possible three-letter codes) the M-Index would only require 120 entries. Also, the K-P Index requires knowledge of all paired correlations of the six types and can only measure corresponding positions in two codes. The M-Index on the other hand is easily calculated, can match all letters of the codes, and can be used for any ranked or ordinal data (lachan, 1984). lachan (1990) has extended his congruence index so that when ties occur between codes, congruence can be calculated as the average of the index values of the different codes. The refined index can also be used when only the two highest letters of codes are available. Tracey and Rounds (1992) criticised the Z-S and M indices for not taking the relationships among RIASEC and environmental types as per the circumplex model into account as their measure does. Swaney and Prediger (1985) used theory on occupational locations on the hexagon as foundation to their Hexagon Congruence Index, and in particular the data/ideas and people/things dimensions of the hexagon (see point 2.6.1). This index measures person-occupation congruence. It can be
43 used with profiles consisting of six, three or two-letter codes as well as highpoint codes. The Hexagon Congruence Index is operationally defined as "the absolute difference between the angles associated with any two locations on the hexagon (e.g., the locations of occupations, persons, theory-based Holland types, or any combinations of these)" (Prediger & Vansickle, 1992, p. 119). These hexagonal locations can be computed by dividing the arc tangent of the location's data/ideas score, by the location's things/people score, and angles can subsequently be computed according to the geometry of the hexagon. Accordingly, the Hexagon Congruence Index is obtained when the personality type's angle is subtracted from the occupation's hexagonal angle. Zero percent angles suggest maximum congruence with minimum congruence (angle of 180°) occurring between opposite types. This index is superior to others in many ways: It can be used with all measures of Holland's types, or combinations of procedures such as six-score profiles with three-letter codes; -
It is based on actual scores and not on ranks. Two different sets of ranks may yield the same three-letter code; Both the hexagonal proximities of RIASEC types, and the order of these types are taken into account; The scores obtained can easily be related to the hexagonal positions;
-
It enables the researcher to evaluate whether theory-based locations of occupations are reasonable, or to compare results with that of previous research designs (Prediger & Vansickle, 1992).
According to Holland and Gottfredson (1992) however, Prediger and Vansickle's angular congruence index is not totally accurate as it ignores the length of the vectors in two-dimensional space. As a result persons or occupations that are not congruent may seem congruent. Planar congruence should rather be calculated that describe congruence both in terms of dimensions and space.
44
Lastly, the complex Sb Index of Gati (1985) makes use of information concerning the dominant scales in person or environmental profiles, the correspondence of scales and the proximity structure of scales to calculate congruence (Camp & Chatrand, 1992). An index developed by Brown and Gore (1994) that are used in the present study will be described in Chapter Six.
2.5.4 The classification system Over and above the measures and indices that have just been discussed, Holland's classification system can be seen as another indispensable tool that enhances the utility of the theory. It is the arrangement of personality, environmental and occupational categories in a purposeful and functional way. The six main RIASEC categories consist of five to sixteen subcategories such as SAO, SEA and so forth. The categories are further divided into levels of general educational development (GED). These categories are arranged according to the hexagonal model discussed before, providing a measure of the degree of differences and similarities of various categories (Holland, 1985b). A chronological observation of the development of the system portrays the expansive nature thereof. The preliminary classification consisted of the six RIASEC categories, as well as the classification of occupational titles as measured by the Vocational Preference Inventory (1977) in accordance with these categories, with separate occupational classifications for men and women. The intermediate classifications arranged all common USA occupations according to these categories and subclasses of occupations in accordance with the level of general educational development. Further intermediate classification work involved the verification and revising of intermediate classifications. Currently the occupations of the SDS arrange 501 occupational titles and as seen before the Dictionary of Holland Occupational Codes (1987) arranges 12 099 occupational titles, both according to the hexagonal model and level of general educational development. The usefulness and verification of the classification system are supported by a strong basis of research (Holland, 1985b; 1996). The
45 South African Dictionary of Occupations classifies South African occupations according to a three-letter Holland code that gives an approximation of the characteristics of the dominant personality type in an environment (Taljaard & Von Mollendorf, 1987). The classification system greatly enhances the applicability of the theory: It organises occupational information obtained by means of the VPI or SDS; In the analysis of work histories, it shows similarities between various jobs. Calculation of a cumulative summary of the last three vocations are possible, amongst other things for rehabilitation purposes; It can also be used in the development of occupational exploration plans, that serve as a tool and framework to systematically explore careers (Brown et al., 1984). In summary, Holland's person-environment theory is very useful to the researcher, since instruments have been developed to measure both the person and the environment as well as the degree of fit between the person and the environment. The various indices of congruence have applicability to a wide range of personality or environmental profiles. For the present study, information can be taken from the classification system concerning the occupations of psychology and optometry, and in particular from The South African Dictionary of Occupations (1987), and be brought into relation with the personality profiles of psychologists and optometrists by means of an index of congruence.
2.6 ADDITIONAL DIMENSIONS Holland's (1973, 1985b) hexagon stands at the centre of the construct of congruence as seen in previous sections. As a two dimensional structure it summarises vast amounts of information concerning types and their interrelations. Should it come to the light that this structure is incorrect or incomplete, basic assumptions of Holland's construct could be questionable. In this section, expansions and modifications of this structure will be
46 examined, as it holds implications for comprehension of the construct of congruence. Research relevant to these dimensions will be discussed in Chapter Three. In this section, these observations can be regarded as criticism of Holland's theory and hexagon. Three dimensional expansions are contemplated, namely Prediger's data/ideas and things/people dimensions, Tracey and Rounds' prestige dimension, and Gati's hierarchical model.
2.6.1 Data/Ideas and people/things dimensions Prediger (1976, 1982) explained the nature of the vocational interest dimension underlying Holland's theory. Where Holland focused on the spatial configuration of types, Prediger took it one step further by describing the dimensions of Holland's circular structure. Two basic dimensions underlie this structure, namely a data/ideas and a people/things dimension.
Enterprising (E)
DATA 2 Conventional (C)
Artistic (A) -2
Investigative (I)
IDEAS
Figure 2.2: The data/ideas and things/people dimensions (Prediger, 1982, p. 260). From this evidently two bipolar dimensions, namely a data/ideas and a things/people dimension distinguish interest categories related to different personality or environmental models. For instance, an orientation towards
47 ideas can be found in artistic and investigative environments, whereas social individuals are orientated towards people. These dimensions provide a basis for efficient summarising of data obtained from the five instruments assessing Holland's types, including the SDS (Lamb & Prediger, 1981). The four poles of the two dimensions can be represented according to basic work task categories. Data tasks are impersonal, relate to facts, records, files, numerical and systematic procedures, and are found in occupations such as accountancy. Ideas tasks on the other hand are intrapersonal, and are associated with abstractions, theories, knowledge, insights and new ways of expressing things, as would be the task of scientists and philosophers. Things tasks are nonpersonal, for instance working with machines, materials, tools or biological mechanisms and are performed by technicians, bricklayers and so forth. Finally, people tasks can be described as caring, persuading, entertaining or directing activities, done by teachers, psychologists and the like (Prediger, 1982). The essence of each occupation can be characterised by the predominant one of these tasks. The positions of these tasks in terms of the Holland types are illustrated in Figure 2.2. It portrays the relative relationships of the six Holland types in terms of these two independent dimensions or factors. The work task preferences representative of the six types are as follows: Realistic: Investigative: Artistic: Social: Enterprising: Conventional:
things; ideas and to a lesser degree things; ideas and to a lesser degree people; people; data and to a lesser degree people; data and to a lesser degree things.
Suggestions of the bipolar aspects of the types can also be seen in Holland's own work, because pairs of types have opposite qualities for instance the realistic and social types (Holland, 1985b). Nevertheless, according to Holland and Gottfredson (1992) information is lost with the use of Prediger and Vansickle's two-dimensional distillation as opposed to the use of all six scales.
48
2.6.2 The interest circle and the prestige dimension Tracey and Rounds (1995) rejected the idea of six distinct interest categories as implied by the RIASEC model, and argued that interests are ordered in a continuous circular structure. Their proposed concentric circle model of vocational interests demonstrates the dimensional relationships between Prediger's (1982) four interest dimensions, Holland's hexagon, Roe's eight occupational fields and various occupations. Interestingly, the occupation of psychology is positioned in the Helping occupational field, and contrary to the Dictionary of Holland Occupational Codes (1982), between the social and artistic types, and in relation to the people and ideas dimensions. In a controversial article Tracey and Rounds (1996a) proposed an additional dimension to vocational interest, namely an occupational prestige dimension according to which "the relations of RIASEC interest scales to each other vary as a function of prestige" (Tracey & Rounds, 1996a, p. 6). Furthermore, the researchers reiterated that the amount of interest types is arbitrary and could be depicted on a circumplex model. In order to represent their elaboration on a three dimensional model instead of on a two dimensional model, they proposed a spherical model of vocational interest that incorporates the arbitrary interest types, the data/ideas and things/people dimensions of Lamb and Prediger (1981) as well as the new prestige dimension (see Figure 2.3). The prestige dimension is represented on the north pole—south pole axis of the sphere, with greater differentiation of interests evident in the centre or "equador". The RIASEC types around the axis vary as a function of prestige. The importance of the data/ideas and things/people dimensions diminishes when prestige increases or decreases (Tracey & Rounds, 1996a). In Figure 2.3, the bottom figure represents the lower prestige half of the sphere and vice versa.
49
(Helping 3) Helping I
Life Sciences 1
Business Contact I
(Lite Sciences 2
(Business Contact 2)
Business Detail t
Technical I (Mechical & Techincal 3)
(Helping 3) Helping I
Business Contact 1 (Business Contact 2)
(Lite Sciences 2)
Business Detail 1
Mechanical 1
Technical I (Mechanical & Technical 3(
Figure 2.3: The spherical representation of interests (Tracey and Rounds, 1996a, p. 27). Invited comments on Tracey and Rounds spherical model ranged from support for the heuristic and theoretical value thereof (Gottfredson, 1996; Harmon, 1996; Prediger, 1996) and praise for the sophisticated research methodology (Gottfredson, 1996) to criticism on the lack of utility due to the complexity thereof (Borgen & Donnay, 1996; Hansen, 1996) as well as comments on the lack of a perfect sphere and methodology used in support
50 of their model (Borgen & Donnay, 1996). Some of the relevant criticism will now briefly be discussed: Borgen and Donnay (1996) criticised the lack of interpretability of the octant scales proposed by Tracey and Rounds and the inconsistency of their proposal with well-established interest measures. They view it as "reinventing the wheel" when Tracey and Rounds propose a "Prestige" scale, as the VPI already allows for the measuring of status. According to them the proposed spherical model would be an insufficient replacement of the hexagonal model. Although they acknowledge the possibility of a third dimension of the model, they state that a prestige dimension is not theoretically, statistically or practically justified. Gottfredson (1996) valued the fact that the Tracey and Rounds' model is capable of simultaneously locating a person's interests in both the prestige and occupational fields. He mentioned the practical importance of considering a status level in job selection. For example, one could expect dissatisfaction when a desired status level is not matched by the available status level. He also commended them on several aspects of their research design. Yet, Gottfredson argued that actual data can only produce a "gnarled glob" and never a perfect sphere. Most importantly for the present study Gottfredson said that "Tracey and Rounds have given us some new items for the tool box for exploring person-job match, satisfaction, vocational adjustment and career attainment" (Gottfredson, 1996, p. 71). Harmon (1996) argued that Tracey and Rounds' work presented a wellconstructed measure of interests. Harmon suggested that further spatial dimensions should be explored, that more differentiated measures of prestige should be used, and that the stability of the spherical shape should be examined. Tracey and Round's extension of Holland's hexagonal model could advance the field of vocational psychology beyond the status quo of the past 25 years, which was dominated by the Holland's hexagon according to Prediger (1996). Prediger argued that prestige (or social status) is a job value rather than an occupational interest, and that the term "occupational attribute preference" is preferred to "vocational interest dimension", and "occupational aspiration level" is preferred to "prestige".
51
In response to the aforementioned reactions to their model, Tracey and Rounds (1996b) summarised the contribution of their proposal as follows: Vocational interests are presented as a continuous circular structure, instead of several distinct types, for example RIASEC; The concept of prestige explains certain variation in interest response that, through its representation on a spherical model, can be applied both to individual interests and occupations; A spherical representation of vocational interests is not meant as a measure of interests, but is useful in several ways, such as for the graphical depiction of scores, the use of prestige as a scoring variable, and the reporting of scores as sections of a sphere instead of separate values. Two further responses by Tracey and Rounds (1996b) pertain to the present study: Due to the plateau in congruence research reported by Spokane (1985), congruence research has almost disappeared from literature, but the prestige dimension opens doors for further research; It is possible that the spherical model can bridge the gap between the two career development orientations of values (Dawes & Lofquist, 1984) and interests (Holland, 1985b).
2.6.3 Gati's (1991) hierarchical structure Vocational psychology has presented various structures as models for depicting occupational interests. Gati (1991) compared Roe's circular model and Holland's hexagonal structure on which the congruence concept is based. Although several studies support the hexagonal arrangement of interests, Gati (1991) questioned the two-dimensional nature of the structure. Gati argued that the bipolar dimension of people/things and data/ideas could just as well be unipolar factors, since some occupations
52 necessitate both poles of the dimensions, and since there is often no significant variance between the poles. He also indicated that the order of the RIASEC types is not always supported in research, for example, findings of some studies strongly suggest the interchangeability of the realistic and investigative fields for women. Likewise, although the hexagon suggests symmetrically equal distances between adjacent types, this can not be replicated empirically. Furthermore, some data reveal stronger correlations between non-adjacent fields (artistic and enterprising) than between adjacent fields (artistic and investigative). As alternative, Gati (1991) claimed that a hierarchical model for the structure of interests is superior to the hexagon. People rate occupations according to the attributes thereof and some occupations have similar attributes. These attributes can be represented on a hierarchical tree structure of occupational interests. Figure 2.4 represents the structure as applied to Holland's theory.
RIA S E
C
Figure 2.4: Gati's hierarchical model (Gati, 1991, p. 313). The predictions resulting from this model, imply that correlations between fields of the same cluster is higher than those between fields of different clusters. A proposed integrated structure of Holland and Gati's model appears in Figure 2.5.
53
Figure 2.5: Integration of circular-hexagonal and hierarchical models (Gati, 1991, p. 318). Tracey and Rounds (1993) argued that Holland's structure of vocational personalities and work environments is superior to Gati's (1991) model. Both the order model (the calculus assumption) and the circumplex model (the hexagon) of Holland's portrayal of the relations among RIASEC types are preferred to Gati's model. Whereas Gati's model only accounts for 34% of possible order predictions, the order model yields 72 order predictions and the circumplex model accounts for all possible RIASEC relations. Tracey and Rounds refuted Gati's criticism of Holland's structure, by criticising Gati's use of only certain studies and the combination of Roe and Holland's models. They also pointed out that there should be a "hole in the middle" or no types represented between the bipolar dimensions, for the structure to be correct. From this section one may conclude that, although attempts have been made to expand or even replace Holland's hexagon, new models are not without criticism themselves. Nonetheless, these proposed additional dimensions emphasise the position of Holland's model within the paradigm of vocational interests, and provide means of combining information from various theories. The work tasks described by Prediger are especially useful in this regard. If one should accept that the spatial configuration of RIASEC types is always a function of prestige, as proposed by Tracey and Rounds, measures of congruence probably need to be revised.
54
2.7 DEMOGRAPHIC FACTORS AND HOLLAND'S THEORY Is Holland's theory equally applicable to people of both sexes and of different ages or cultural backgrounds? In this section criticism that has been brought against Holland's model based on these demographic factors, as well as Holland's response to such criticism will be briefly mentioned. Specifically, issues of gender, age, development and race will be addressed.
2.7.1 Gender-related issues Holland's model and the Self-Directed Search have been criticised as being sex-biased. When Prediger (1980) reviewed studies that compare raw versus normed scores for the Holland types, he came to the conclusion that standard scores yielded a more accurate description of the occupational types. Prediger still argued that normed scores should be used. He found that data suggests that the use of normed scores results in the suggestion of more balanced occupational options to women, than only social, artistic and business related occupations (Prediger, 1981). To this Holland (1982) responded that, regardless of whether raw or normed scores are employed, the SDS increases the occupational options considered, self-understanding and assurance for both genders. Gottfredson (1982) also responded to Prediger's criticism of the use of raw SDS scores for females, by pointing out that Prediger had used ambiguous methods of analysis to come to his conclusion, and that Prediger also falsely assumed that there are no real gender differences in vocational interests. Prediger (1982) insisted that although gender differences may be found in vocational interests, genderdivergent raw scores of interest are still not useful in career counselling. The validity of predictions of occupational aspirations and assessments also tends to be higher for women than for men (Shari, 1992). Holland (1985b) refuted the use of normed scores as an artificial and dishonesj method for measuring and interpreting interest scores. Men and women are actually different in terms of interests and should not be led to believe otherwise just for the purpose of exploring new occupational fields (Brown et al., 1984).
55
2.7.2 Age-related and developmental issues Apart from gender issues, Holland's theory also does not make a distinction in terms of age or age-groups. Holland (cited in Brown et al., 1984) argues that people of all ages are analytical in terms of their interests and not only during certain developmental stages. Holland has also been criticised for giving insufficient information on the development and change of personality. However, he does provide some explanation on how a life history of activities, skills and so forth result in certain personality orientations (Brown et al., 1984). Holland (1996) recognises a weakness of his theory in not explaining development, but reasons that the Career Attitudes and Strategies Inventory will provide the necessary information concerning the processes of change.
2.7.3 Racial issues Although Holland's theory, concepts and instruments have been applied internationally, Shari (1992) quotes several research findings indicating differences in terms of race. One can argue that this is not a negative aspect of the theory, but instead shows its ability to indicate social differences. Such differences are evident in studies that yielded higher scores for African Americans than for Whites on social, conventional and enterprising subscales, or from results indicating that more Hispanic and African American men and women than white men and women were in realistic occupations and more white men and women were in enterprising occupations (Holland, 1985b). The discussion of research findings on racial issues will be elaborated under point 3.5.2. Holland and Gottfredson (1976) provide three reasons why this model does not distinguish between different groups based on gender, race and other group factors: -
People belonging to different groups are all part of the same species and only differ in terms of personal development in as far as their personal histories differ;
56
There is no empirical evidence of different psychological processes for different groups, but instead there are both similarities and differences in terms of groups belonging to all of the personality types; It is practical to use the same typology and measures for different groups, as it enhances the assessment of current status, results in more advantageous career development planning, and improves the design of job-seeking strategies. It seems that issues of gender, age, development, and race do not decrease the value of Holland's theory. Instead, the theory is useful in clarifying individual differences based on demographic variables. Research findings on these issues will be further contemplated in Chapter Three.
2.8 CONCLUSION In summary, Holland's theory can be seen as a classification system that categorises personality and environments in terms of types and subtypes. These classifications range from the basic RIASEC categories, to complex interactions of types and subtypes as assessed by means of the hexagonal model and portrayed for example in the Dictionary of Holland Occupational Codes (1982). The four assumptions discussed constitute the basis of this classification system. The concept of congruence is therefore understood in terms of the interaction and compatibility of types and job satisfaction is seen as a consequence of congruent interactions. This theory is very effective as a person-environment fit model. It is more balanced than theories that overemphasise either personality and inner motivation or environment, biological predisposition and prior socialisation in behaviour. The principles of consistency, differentiation, identity, beliefs, strategies and especially congruence characterise the level and quality of such interactions. There are however questions that remain unanswered in terms of the processes involved in for instance the interaction between a person and his occupational environment. Although the theory postulates that the demands and opportunities of the environment must be matched by the activities, competencies and achievements of the person, as is the case
57 with cognate personality and environmental types, there are still uncertainties as to the underlying processes of a successful match. The theory of Lofquist and Dawis (see Chapter Four) provides greater clarity on the mechanisms of such an interaction in describing the necessity of a match between specific needs and relating rewards for correspondence to take place, or the theory of Stern (in Walsh & Betz, 1995) describes the necessity of an interaction between personal needs and environmental presses. Heidema (1995) criticises Holland amongst other things for seemingly basing the environmental models on the predetermined personality types. One could propose as an alternative, a description of environments based not only on the predominant personality type but also on the specific behaviour patterns of the members of the environment, for example their degrees of involvement, supportiveness and openness, as suggested by Moos (in Walsh & Betz, 1995). In other words, the environment's social climate, and not only typological classification could be considered in order to make predictions of the behaviour of its members. A further concern about Holland's theory lies in its description of the environment. Whereas Holland has provided a static occupational classification, one can argue that environments are constantly changing and also never homogeneous. In fact, the social environment presented by an occupation may change from day to day, since a person's interaction may not be primarily with his colleagues, but instead with a diverse group of people such as represented by clients and other workers in the environment. Instead of evaluating the fit between a personality and a fixed environment, one could instead compare interactions in terms of various situational episodes as suggested by Pervin (in Walsh & Betz, 1995). Taljaard and Von Mollendorf (1987) hold that occupational codes that are based on the local context are superior to nationally determined codes to define the characteristics of a vocational environment. When determining the interaction of persons and environments, Holland's constructs of differentiation, consistency, identity and calculus should not be underemphasised. Congruence of personality type and occupational code in itself is an incomplete portrayal of the degrees of interaction between a person and his environment.
58 Nonetheless, the parallel models of environmental and personality themes are very useful in explaining and comparing personalities and environments and to make predictions of resultant behaviour. As Holland (1996) states, the value of the typology of personality and environments lies in its ability to organise vast amounts of information in ways that are both economical and accessible. The personality-environment interaction according to Holland focuses on a match between an assessed personality type and previously determined occupational classes. The importance of human cognition is understated in this formula. As alternative theorists such as Stern and Moos (in Walsh & Betz, 1995) emphasise the interaction between actual personality and perceived environment as opposed to actual environment. Although Holland recognises the role of perception and factors such as the size and complexity of an environment, he does not present an adequate way of assessing such environmental variables when investigating personenvironment interactions. The fit between persons and environments deals specifically with a person and his social environment, i.e. the members of an environment such as an occupation. Holland (1985b) recognises the role of external resources in affecting satisfaction and other work attitudes, but these environmental factors are insufficiently represented in the theory. Likewise, Holland's allusion to biological predisposition in the development of personality types is inadequately substantiated in the theory. The result of a mismatch between person and occupation or of incongruence involves either a change of personality attributes, or changing environments, such as finding a new job. The theory does not provide sufficient explanation for the adjustment processes that may take place and could be complemented by the work adjustment theory (see Chapter Four) that gives clarity on the adjustment styles of personalities and environments. Over and above cognition, one may also expect an affective component to regulate behaviour. Pervin (in Walsh & Betz, 1995) recognises the role of affect in person-environment interactions and stresses the importance of goal-setting in adjustment to and between occupations and other environments. Holland's (1996) incorporation of concepts of beliefs and
59 strategies is a definite improvement as it introduces the role of cognition and affect in the theory. Thus inner motivation and goal-setting can also play a role in career choice and change, and not only a match or mismatch of types and subtypes. One of the most valuable aspects of this theory lies in the fact that not only is the level of congruence between persons and environments quantifiable, but the information obtained by means of such indices of congruence are practically applicable when it is associated with information from the classification system. Other relevant points of criticism that has been raised against Holland's theory in literature include that it provides an inadequate description of personality development, that the personality and environmental typology is too static, that it is too simplistic, and that it does not adequately account for social and economic influence on career choice and change. Some of the cited virtues of the theory include its cross-cultural applicability, and the vast amount of empirical support for the theory (Heidema, 1995). Despite all the criticism that has been cited above against Holland's theory, its heuristic value lies in the construct of congruence of which the personality and environmental elements can effectively be depicted on the hexagonal model. In addition to this, both personality and environment can be reliably and validly assessed with the relative measures, and results obtained can be meaningfully interpreted in relation to the ever expanding classification system. The theory will be further evaluated in response to the research findings presented in Chapter Three. In summary, Holland's theory provides an integral description and explanation of the relationship between persons and environments or personality and occupations. Its concept of congruence has been substantiated by various indices and the theory is readily tested due to its useful instruments and classification system. Although its applicability, and recently its hexagonal model has been challenged, it still proves to be a key theory in vocational psychologists' understanding of occupational interests. It also has great predictive value in that both vocational and avocational behaviour can be predicted, for instance certain behaviours associated with each personality type, or within each environmental type, and the likelihood
60 of accurate predictions of behaviour can be seen as a function of levels of congruence, consistency, differentiation and identity of persons and environments.
61
CHAPTER 3 RESEARCH ON HOLLAND'S THEORY
3.1 INTRODUCTION The popularity of Holland's theory (1973, 1985b) is evident from the vast amount of research that has been conducted, testing the hypotheses thereof. In this chapter, the discussion of research on Holland's theory focuses on person-environment congruence, and controversial issues of the theory, to set the arena for evaluating the validity and utility thereof. Firstly, a few examples of research relating to the conditions of person-environment congruence are mentioned. Secondly, empirical support for the relationship between congruence and job satisfaction is sought. Research on the relationship between congruence and other work-related outcomes forms the third part of this chapter. In order to illustrate the applicability of the theory, research on gender and cultural issues will be referred to, as well as studies of the dimensional aspects of the theory. Lastly, the usefulness of the theory will be shown in studies dealing with indices of congruence and the SelfDirected Search (1979, 1985a, 1985c).
3.2 CONDITIONS OF PERSON-ENVIRONMENT CONGRUENCE From the theoretical discussion of the theory it is apparent that congruence occurs when personality type and environmental type match. A simple similarity between types may be an insufficient qualification for the congruence to occur. The following studies highlight other conditions that need to be present for a congruent interaction between a person and his occupational environment. Raphael and Gorman (1986) examined the effect of occupational selfknowledge and also subjective occupational structure on congruence. In other words, they studied both whether a knowledge of personal work interests, as well as the participants' ability to compare pairs of occupational environments in terms of the degree of similarity between them, are
62 correlated to congruence. The participating female college students assigned personality codes to themselves after reading a brief description of each of the personality types. The "self-knowledge" scores were obtained by comparing the personality types of the participants as assessed by means of the Strong-Campbell Interest Inventory (1985) to their selfassigned personality codes. The similarity of fifteen randomly ordered pairs of six occupations was rated on a nine-point scale. Results indicated that accurate occupational perceptions and an understanding of own work interests were positively correlated with congruence. Thompson, Flynn and Griffith (1994) in a change study, investigated the following four diagnostic signs with reference to congruent employment outcomes: employment congruence: the degree of fit between personality and an occupation; aspirational congruence: the degree of fit between personality and occupational aspirations; employment coherence: the degree of similarity between previous and current jobs; and -
aspirational coherence:
the degree of similarity between occupations aspired to and preferred occupations.
The results indicated that employment congruence and, to a lesser degree, employment coherence were predictive of future person-environment congruence and that aspirational congruence was correlated with employment coherence. Congruence and coherence are thus predictive of future person-environment congruence. The aptness and importance of the constructs of congruence and career interests as defined by Holland typology norms are illustrated by the following study. Considering the problem of unemployment in the USA, Downes and Kroeck (1996) compared existing positions in various occupations to normative interests (as measured by the SDS) in those
63 occupations. Different normative interests were evident for the high school students and adult samples. The results indicated an excess in certain occupational positions but with normatively low interests in those positions. Conventional and enterprising occupations for both adults and high school students fell into this category as well as realistic occupations for high school students. Contrarily, normative interests in other occupations, such as enterprising-social occupations (managerial positions) were high, but there were only a few existing positions available in the USA. Likewise interests in social, artistic and investigative positions were high for high school students and investigative and realistic positions for adults, while there was a greater supply of conventional and enterprising occupations. These results hold implications for the evaluation of training needs and the development of career orientations that are suitable to current job market requirements. From the cited examples it seems that congruence depends on a knowledge of personal work interests and of occupations and future congruence depends on current employment congruence and coherence. The correct training can develop the appropriate interests to ensure that congruent career choices are made.
3.3 CONGRUENCE AND JOB SATISFACTION One of the significant consequences of person-environment congruence, namely job satisfaction is also often the subject of research. Some of the results support the congruence-satisfaction hypothesis, others specify the mediating variables that need to be present for the relationship to be significant, while some researchers even fail to find support for the relationship between congruence and satisfaction:
3.3.1 Support of the congruence-satisfaction hypothesis Spokane (1985) reviewing research on person-environment congruence for the period of 1959-1983, distinguishes two trends in person-environment congruence research designs. On the one hand there are several correlational studies comparing congruent and incongruent subjects in terms
64 of a set of criterion measures, and on the other hand, time series designs examine change in person-environment congruence. Findings indicate that correlational studies consistently show positive relationships between congruence and the variables of, amongst others, job satisfaction, academic performance, persistence, etc., but non-significant relationships between congruence and self-concept or sociability. In fact, it seems that correlational studies cannot surpass the "magic .30 correlational plateau" and is unlikely to add to our understanding of congruence. He advises that more studies that are indicative of the change in congruence over time, rather than correlational studies are needed. A sample of school counsellors was used in a study by Wiggins (1984) that confirmed the hypothesised positive relationship between congruence and job satisfaction. Congruence was defined in terms of the subjects' VPI scores and Holland's (1977) coding of the school counsellor occupation. This interest-occupation congruence as well as job satisfaction also positively related to the variable of differentiation. The congruencesatisfaction hypothesis has also been studied in terms of disabled people. Jagger, Neukrug and McAuliffe (1992) found support for the personalityenvironment congruence hypothesis and for the congruence-vocational satisfaction hypothesis for 72 vocational rehabilitation clients who had been successfully rehabilitated. In the title of their research report, Witt and Handal (1984) asked: "Is satisfaction predicted by congruency, environment, or personality?" (Witt & Handal, 1984, p. 503). In their study, congruency was defined in terms of discrepancy between personality and environmental factor scores. Their results showed that congruency, environment and personality were all equally predictive of satisfaction.
3.3.2 Mediating variables of the congruence hypothesis The following examples indicate that other variables such as personality type, environmental type, consistency, tenure, group importance, personal environment reference and gender can influence the congruencesatisfaction relationship.
65 One of the important earlier studies testing Holland's congruence-job satisfaction hypothesis is that of Mount and Muchinsky (1978). Congruence was determined by comparing first letter codes of the personality types (SDS) and the occupational environments (Occupations Finder) of the subjects (N = 362). The participants were representative of all but the artistic type categories. Overall satisfaction was measured by means of the Job Descriptive Index (1969), which includes several factors of job satisfaction, namely pay, promotions, co-workers, supervision and work. Both overall satisfaction scores and the specific scores of the JDI work scale were brought into relation with congruence. Results confirmed that the job aspects from the JDI positively correlate with person-environment congruence. Thus, both overall satisfaction and satisfaction with work are positively correlated with congruence. Congruence can therefore be associated with satisfaction with all aspects of work. Interestingly though, although overall job satisfaction was related to congruence of conventional and enterprising participants, investigative participants in congruent situations were not significantly satisfied with the work aspects of job satisfaction, and realistic subjects in congruent situations were only significantly satisfied with Promotions and Overall job satisfaction, when compared to incongruent realistic subjects. Also, the social environment revealed no significant differences between congruent and incongruent individuals, although the differences were in the predicted direction. A valuable aspect of this research design is the generalisability of the results thereof due to the use of several occupations that are representative of the various environmental types. From this study it seems that the hypothesised correlation between congruence and job satisfaction is a function of personality type. In a study representative of five RIASEC areas (five subject matter fields) in the teaching occupation (N = 247), Wiggins, Lederer, Salkowe and Rys (1983) found support for the predictive relationship between personenvironment congruence as well as differentiation and reported job satisfaction. They emphasised the following four aspects of job satisfaction: The degree to which a particular job was liked; The amount of time it was liked; The desire to change jobs;
66 The feelings concerning the job when compared to that of other people. When congruence and differentiation were controlled, results showed that conventional and artistic people were generally more satisfied than the other RIASEC types, and enterprising types were more inclined to have overall lower levels of job satisfaction in the teaching profession. This study once again illustrates the importance of personality type classification in the congruence-satisfaction hypothesis. In Spokane's (1985) review he also quoted studies that have indicated the influence of other variables in the congruence-job satisfaction consequential relationship. For example, a study by Frantz and Walsh (1962) indicated that the joint influence of congruence, consistency and differentiation has predictive value for the subjects in graduate school, whereas congruence or consistency taken alone does not. Another study (Weiner & Klein, 1978) found job tenure to be a variable that determines whether the relationship between congruence and satisfaction is significant. Meir, Keinan and Segal (1986) confirming the correlation of personenvironment congruence and job satisfaction, found that the variable of group importance was an intervening factor that interacted with congruence in determining satisfaction. Another study confirming the congruencesatisfaction relationship is that of Smart, Elton and McLaughlin (1986). In their study they distinguished between extrinsic job satisfaction (including factors such as income, fringe benefits and job security) and intrinsic job satisfaction (including opportunity to contribute to society, to be creative, to use training/schooling, and challenge). Gender-specific differences were apparent in the results. The relationship between congruence and extrinsic job satisfaction was evident for males only, and overall job satisfaction for females only. Assouline and Meir (1987) reviewed 41 studies of the relationship between congruence and certain measures of well-being, indicated by Holland's congruence hypothesis, namely achievement, career stability, and satisfaction by means of meta-analysis. As explanation for the variance in research results on the congruence hypothesis as indicated by Spokane (1985), the researchers proposed the influence of individual environmental
67 reference. These included aspects of the occupation, speciality within an occupation, the personality type predominant in the environment, major studies taken and the specific educational institution as well as educational and vocational interests. On the other hand the measure of congruence used could also account for the variance. The researchers found higher mean correlations than the "Spokane 0,30 correlation coefficient" in studies where satisfaction was the dependent variable, and the level of vocational interest-occupational speciality choice congruence or vocational interest-occupational choice congruence was the independent variable. Low correlations with congruence though was found in studies with stability or achievement as the dependent variable. Moreover low correlations were found between occupational congruence and satisfaction when the congruence measure used compared personality or environmental first letter codes only. In summary, when specific environmental reference and congruence measures were used, the congruence-satisfaction correlation went up from the mean coefficient of 0.21 to more than 0.35. Specifying the method of congruence measurement, the environmental reference (as obtained through the census method instead of the occupational or speciality classification methods) and the specific criterion of well-being (satisfaction, stability or achievement) significantly increased the correlations between congruence and measures of well-being. In a longitudinal study Elton and Smart (1988) compared job aspirations and graduation majors to actual jobs held nine years later, in order to investigate the relationship between congruence and extrinsic job satisfaction. Their findings indicated that job satisfaction was lower in social environments than in any of the other types of environment. Moderate support was found for the relation between congruence and extrinsic job satisfaction. Genderspecific differences in this study revealed that more men than women had lower extrinsic job satisfaction. Heesacker, Elliot and Howe (1988) set out to determine whether the Holland personality code could predict job satisfaction and productivity for their sample of clothing factory workers. These participants who were selected from a predominantly female blue collar environment, between the ages of
68 18 and 66 had been in their present occupation for periods ranging from 1 month to 33 years. The researchers assessed: the predominant Holland code for this environment; the validity of the congruence-satisfaction hypothesis; whether productivity was a function of congruence; whether absenteeism was a function of incongruence; whether job injuries could be predicted by work congruence. The personality code was measure by means of the SOS-Form E. Hourly wage was used as indication of productivity, absenteeism scores were obtained from available information and job injuries scores were taken from health insurance claims. Contrary to expectations, the result did not support the expected conventional-realistic environmental code, but instead a predominantly social environment. Moreover, no support was found for the congruencesatisfaction hypothesis, nor for the predicted relations between congruence and absenteeism or congruence and job injuries. Partial support was found for productivity as a function of congruence. In fact, conventional, realistic and social women were more productive than the others. Job satisfaction was highest for social women, followed by both conventional and realistic subjects. Gottfredson and Holland (1990) performed a longitudinal study of the relationship between interest congruence and job satisfaction among 345 bank tellers. More specifically, they tested the hypothesis that there would be closer correspondence between interest congruence and job satisfaction when workers have clearly defined interests as opposed to those with poorly defined interests. They controlled common variables of the congruenCesatisfaction hypothesis in this study, namely organisational context, occupational level and field of work. They also hypothesised that incongruence between vocational interests and actual jobs would result in dissatisfaction and other negative work outcomes such as lower job involvement. The results offered support for the hypothesis that a person's interests and competencies can be expressed in congruent environments. Little support
69 was found however for attempts to show that person-job incongruence results in negative work behaviour, role ambiguities and role conflict. Results did indicate though that person-job congruence is wholly correlated with job satisfaction in a homogeneous sample. Predictions of subsequent job satisfaction are possible if a personal disposition to be satisfied is prevalent in an individual. A valuable aspect of their research design is their consideration of possible influential factors of the congruence-satisfaction relation and the moderating influence of the variables of differentiation, expected satisfaction and perceived employment alternatives. However, the mediating influence of differentiation for instance did not receive support. On examining evidence that challenges the presupposed relations between congruence and variables such as achievement, stability and satisfaction, Schwartz (1992) suggest a hypothesis that proper clarity of role choice is the intermediary variable in the satisfaction-congruence relationship. In a longitudinal study of congruence hypotheses in both enterprising and social bank environments, Meir and Havon (1992) found positive correlations between congruence and satisfaction and congruence and supervisors' ratings. These correlations were only significant though when congruence was defined in terms of the type of bank branch, illustrating the significance of accurate categorisation of the environment for the validity of the congruence-satisfaction hypothesis. Other results included that congruent responses followed employment and no significant relationship between persistence and congruence was found.
3.3.3 Lack of support for the congruence-satisfaction hypothesis The following studies are significant in that they indicate certain instances of failure to find support for the congruence-satisfaction hypothesis. Their findings may suggest the inadequacy of a vocational interest-occupational congruence assumption. In a meta-analysis of 27 studies attempting to find a congruence-satisfaction correspondence, and more specifically the relationship between interest congruence and job or academic satisfaction, Tranberg, Slane and Ekeberg (1993) found that the mean correlation was not significant at all. When the
70 variables of Holland type, gender or academic versus job setting were considered separately, the mean correlation was still not significant. In fact, even the variable of the congruence measure used had no mediating effect on the significance of the relationship. These results suggest that work outcomes are a function of more than just a match between interests and work environments. Upperman and Church (1995) failed to find support for the usefulness of Holland typological codes in distinguishing occupational specialities in their army sample. They also questioned the validity of the congruencesatisfaction hypothesis. The concept of congruence as such was supported, since the majority of the participants had realistic personality types, congruent to the army environment. Results from studies in three British samples indicated few significant relationships between congruence (and differentiation) and job satisfaction. In one of the studies the managerial group had higher levels of both congruence and job satisfaction than the non-managers. In another study no significant relations were found between congruence and job satisfaction and being recently qualified. A third study showed that having specialised in the nursing field does not increase the levels of either job satisfaction or congruence (Furnham, Toop, Lewis & Fisher, 1995). As alternative to Holland's congruence hypothesis, Gati, Gaily and Fassa (1996) also explored the relationship between person-environment fit and job satisfaction. Instead of looking at congruence in terms of vocational interests (Holland), they contemplated career-related aspects of personenvironment fit. According to them, each occupational environment has certain "core aspects" based on the essential characteristics thereof. In their study they considered twelve occupations according to 36 career-related aspects. They predicted higher correlations between career-related personenvironment fit and occupational satisfaction, than the traditional congruence-satisfaction correlations, but also only obtained the "magic .30" correlation across the occupations. Nevertheless, when only the withinoccupation correlations were considered, the congruence-satisfaction correlation was as high as 0.85 for the core aspects, but as low as 0.27 for all 36 aspects. This shows that career-related aspects of congruence are not superior to vocational interests only in predicting job satisfaction.
71
These findings indicate that the relationship between job satisfaction and congruence is more significant when the mediating effect of variables such as personality and environmental type are clearly defined and controlled. The fact that other variables moderate the correlations between congruence and job satisfaction is indicative of theoretical inadequacies of the congruence hypothesis. Researchers seeking correlations between job satisfaction and congruence should keep the above findings in mind, through precise definitions of the personality type and the specific environment.
3.4 CONGRUENCE AND OTHER OUTCOMES Apart from studies on the hypothesised relationship between congruence and satisfaction, congruence has also been studied in terms of various other variables, such as vocational stability, achievement, performance, perceived social climate, well-being and coping strategies, and between incongruence and failure and frustration, although positive relationships have only been found for some of these variables:
3.4.1 Congruence and vocational stability Results of a study by Salomone and Sheehan (1985) failed to show a positive relationship between person-environment congruence and vocational stability for 240 male and 260 female workers. The researchers inferred that other variables, such as economic, social and cultural factors could influence vocational stability.
3.4.2 Congruence, achievement and performance Schwartz, Andiappan and Nelson (1986) questioned the validity of prior research results that support the congruence-achievement hypothesis (i.e. congruence is positively related to achievement). They argued that previous positive outcomes could be ascribed to the failure on the part of the researchers to recognise the association between certain personality types (such as the investigative, social and artistic types) and a high achievement
72 orientation. Their results showed a negative relationship between achievement as indicated by annual income and the conventional personality type of their accountant participants (N = 1087). Their own study has limitations, for instance the use of "annual income" as indicator of achievement, when it is not representative of total occupational achievement, the comparisons of this variable with previous variables of academic achievement, as well as a low response rate (50%) in their research. However they still conclude that pervious research did not account for naturally low achieving personality types and may therefore not have supported the congruence-achievement hypothesis. Another study provides support for the hypothesised relationship between congruence and academic achievement. Findings of Henry (1989) proved that congruent premedical students achieved significantly higher cumulative grade point averages and science grade point averages than incongruent students. The validity of a positive relationship between congruence and achievement is questioned by Schwartz (1991). According to his review of research supporting the congruence-achievement hypothesis, a variable of "achievement-orientation of personality type" confounds the results of research on this and probably other Holland hypotheses. Ostroff (1993) hypothesised a positive relationship between personenvironment congruence and organisational effectiveness or performance. Person-environment congruence was defined in terms of personal orientations and organisational climates. Results confirmed the congruenceorganisational effectiveness relationship. Different facets of organisational effectiveness related more strongly with different spheres of congruence, and sometimes incongruence yielded more significant relationships.
3.4.3 Congruence and perceived social climate Hildebrand and Walsh (1988) found no relations between personenvironment congruence and the perceived social climate of the work environment for their sample of nurses. Although there were no significant differences of the perception of the work climate in terms of congruence,
73
managers defined their environment as more positive, rewarding, friendly supportive and relationship-orientated than the staff did.
3.4.4 Congruence and well-being Meir and Melamed (1986) studied three aspects of congruence, namely vocational, avocational and skill utilisation congruence in terms of different aspect of well-being, namely occupational satisfaction, work satisfaction, somatic complaints and anxiety. Results indicated that all three aspects of congruence correlated with the different facets of well-being, and especially when all three aspects of congruence were present for their sample of teachers. Congruence was positively related to satisfaction and negatively related to anxiety and somatic complaints. Vocational, leisure and skill utilisation congruence were studied in terms of six outcomes, which included work satisfaction, but also occupational choice satisfaction, and self-esteem on the positive side, and anxiety, burnout and somatic complaints on the negative side (Meir, Melamed & Dinur, 1995). Moderate correlations were found between vocational, leisure and skills utilisation congruence. They found correlations between these three congruence aspects and the six outcomes listed above for their sample of lawyers, psychologists and physicians (RI = 120). Because results indicated that the more encompassing the congruence score was (the more congruence aspects included), the higher the correlation of congruence was to these measures of well-being, they concluded that the positive outcomes such as job satisfaction was a function of diverse aspects of congruence.
3.4.5 Congruence and coping strategies Eagan and Walsh (1995) probed the relationship between congruence and different coping strategies for undergraduate male and female students. Congruence was defined in terms of congruence, incongruence or indecisiveness in college major choice. Gender differences were evident when they found that women used the following coping strategies more than men do: escape avoidance, seeking of social support, as well as environment flexible and environment productive styles. Congruent men
74 differed from the other participants in that they were less prone to use escape avoidance as coping strategy, whereas all women differed from the congruent men in their likelihood of seeking social support.
3.4.6 Incongruence and negative work outcomes Since congruence seems to relate to positive aspects of work, such as job satisfaction, one might expect relationships between incongruence and negative work outcomes. For instance, Schwartz (1991) presumed relations between incongruence and failure. Furnham and Walsh (1991) found some support for this hypothesis in as far as frustration correlated negatively with congruence and consistency for their sample of 46 psychiatric nurses. However, positive correlations were found in some cases between congruence and absenteeism. In a further study involving 65 working adults, Furnham and Walsh (1991) found negative relations between differentiation and stress, but not between congruence or consistency and stress. An important observation by the researchers is that other major determinants should be taken into account for the hypothesised relations between congruence, consistency and differentiation on the one hand and occupational behaviour on the other hand to be valid. Sutherland, Fogarty and Pithers (1995) investigated the relationship between congruence and occupational stress and found that congruence was not predictive of occupational stress when the lachan (M) Index was used or in terms of the discrepancy between the personality pattern and the job environment pattern. In these examples, relationships between congruence and achievement, vocational stability and perception of work climate were not supported, but relationships between congruence and organisational effectiveness, academic achievement and well-being were confirmed. Also, a relationship between incongruence and frustrations was found but not between incongruence and stress. Clearly the validity of the congruenceachievement and congruence-vocational stability assumptions are subject to the presence of other variables that still needs to be theoretically specified and described.
75 3.5 DEMOGRAPHIC FACTORS Under point 2.7 the significance of demographic factors in understanding the RIASEC types have been discussed. Issues of gender and race are often at the centre of evaluations of Holland's theory. In this section, research relating to gender differences in terms of SDS scores and the cross-cultural applicability of the theory will be emphasised.
3.5.1 Gender differences Aronowitz, Bridge and Jones (1985) reported sex bias for the investigative subscale of the SDS. Holland (1985a) argued that these results were based on a version of the SDS that is no longer published. Henry, Bardo and Bryson (1988) found gender differences for the realistic and investigative personality types, suggesting that gender impacts the SDS scores. In the abovementioned Heesacker et al. (1988) study Heesacker argues that when normed scores are not used for the SDS, undifferentiated scores for women are obtained. The SOS usually yield social personality codes for women, which according to Heesacker et al. (1988) contributes to the uselessness of the SDS. Their data suggest that: not all types are equally represented for women; raw scores generate reliable effects on outcomes such as satisfaction and productivity; and normed scores would be a deviation from Holland's theory, that focuses on differences between groups rather than differences within groups. The results of Tracey and Rounds' (1992) study on a circumplex structure as representative of Holland types, brought to the light that gender and educational differences were prevalent in the RIASEC codes distribution.
76 Hansen, Collins, Swanson and Fouad (1993) analysed gender differences in the structure of interests through multidimensional scaling. They showed the importance of using matched-interest samples to determine whether men and women, with different interests, will also have different perceptions of the characteristics of the different types and their interrelations. (Previous research results often indicated that Holland's model did not present as good a fit for women's interests as it does for men's interests according to the researchers. Hansen et al. (1993) ascribed these findings to the failure on behalf of the researchers to ensure sampling of comparable interest domains for males and females, and therefore using data that is not representative of male and female interest types). When they analysed data from the Women-in-General and the Men-in-General reference samples (1985 data consisting of each 300 women and men from 93 occupations) of the Strong Interest Inventory's General Occupational Themes, the following two configurations of the male (Figure 3.1) and female (Figure 3.2) data resulted:
Figure 3.1: Men's structure of interests
Figure 3.2: Women's structure of interests (Hansen et al., 1993, p.205, 206).
These two representations of the six general occupational themes for men and women respectively strongly indicate the evidence of gender differences in the structure of interests. Men regard the artistic type as most unlike the
77 other types, and the configuration of men's interests forms a hexagonal structure. The interests of women on the other hand tend to rather be represented by a circumplex than a hexagon. Women also differentiate less between realistic and investigative themes than men do. "For women these shared characteristics may constitute the defining features of one, rather than two, categories" (Hansen et al., 1993, p. 208). Interestingly women traditionally have little occupational experience in these two fields. The social themes have greater centrality in all six types for the women. The above illustrates that there are gender differences in terms of the perceptions of the various RIASEC categories and the response to the diverse attributes of these types. These researchers concluded that their research results hold implications for the construct of congruence, which might necessitate certain modifications of Holland's theory in order to accommodate gender and ethnic differences.
3.5.2 Cross-cultural and racial applicability In order to learn whether Holland's assumptions apply to both black and white Americans, several researchers have studied the personality types and vocational interests of these groups. Walsh, Woods and Ward (1986) ascertained that black and white women in the same occupations (maids, florists, dancers or teacher aides or rather realistic, artistic and social environments) had more similarities than differences in terms of their personality codes and vocational preferences. A study of the concurrent validity of Holland's theory for black professional men (engineers, physicians, attorneys, coaches, music teachers and accountants) and the influence of race on vocational choice, suggested that the SDS and VPI are moderately appropriate for this population. Specifically, results showed that these two measures could discriminate more sufficiently among the RIASEC environmental classifications than among the six occupational groups (Sheffey, Bingham and Walsh, 1986). Henry, Bardo and Bryson (1988) found no differences in terms of race in the SDS's vocational interest classifications for their two samples of 57 African American and 43 white premedical student samples. Results of a study by
78 Miller, Springer and Wells (1988) indicated that African American youths' first or second letter of their SOS code is likely to be social. The relationship between personality type and educational satisfaction among Native American high school students was studied by Gade, Fuqua and Hurlburt (1988). Investigative and social students exhibited higher levels of educational satisfaction than realistic students. Since this is in the theoretically predicted direction, one can tentatively assume the generalisability of this hypothesis for Native American students. Ryan, Tracey and Rounds (1996) examined issues of ethnicity, gender and socio-economic status of Holland's circular RIASEC structure. Two ethnic groups, namely an African American and a white sample were used. Results confirmed the circular structure, and a similarity of the structure for the two ethnic groups. When the effect of ethnicity and socio-economic status was taken together though, the RIASEC structure of the low socio-economic African Americans had a better structural fit than the high socio-economic participants from the same sample. Although the vocational interests of females had a better structural fit than that of the males, the structure was found to be applicable across gender for both ethnic groups. Their results proved the applicability of the structure regardless of ethnicity, gender or socio-economic status. The applicability and validity of Holland's model in cultures, other than the American culture, have also been studied. Cross-cultural applicability of the SDS is demonstrated in the study of Payne and Sabaroche (1985), who found the six personality classifications to be valid for their sample of 101 Caribbean teenagers. Some deviations from theoretical expectations were apparent though, since the relationship between the realistic and social categories was too close. In another study, the validity of Holland's theory in a non-western context (Pakistan) was examined (Khan, Alvi, Shaukat, & Hussain, 1990). More specifically, the concepts of congruence and consistency were viewed in terms of personality types and career readiness levels of students. The supposition that different personality types need different work environments
79 was confirmed, thus suggesting the utility of Holland's theory in non-western contexts. Rounds and Tracey (1996c) also examined the applicability of Holland's personality constructs to other cultures. This was done by structural metaanalysis of three representations of the RIASEC types in different cultural groups. They explored the relative validity of Holland's circular model, Gati's realistic-investigative, artistic-social and enterprising-conventional partition, and their own partition (namely realistic-investigative; artistic; socialenterprising-conventional). Rounds and Tracey's alternative model proceeds from evidence that suggest that the artistic type is separate from the other Holland types, and not so closely related to the social type. The cross-cultural categories were represented by a US benchmark group, US ethnic participants and an international sample. They failed to find support for any cross-cultural structural equivalence for Holland's circular order model. As a matter of fact, Holland's model did not fit fifteen of the eighteen countries of the international sample. Even data from Canada and Australia, that have economic and occupational structures similar to that of the USA, did not fit the model. This strongly implies that RIASEC research results from the USA cannot simply be applied to other countries. Gati's partition and the alternative model fitted the US benchmark data equally well, and matched the international information better than the circular model. In fact, data for only six of the eighteen international groups did not comply to these two models. None of the three models were representative of the vocational interest structure of the US ethnic sample. Glidden-Tracey and Parraga (1996) applied these three structural models to a sample of Bolivian students, who completed the Spanish version of the SDS. Low correspondences were found between the three models on the one hand and the reported interests of the participants on the other hand. This illustrates once again the caution that needs to be taken to not simply apply vocational interest models cross-culturally. How applicable are the SDS codes in the South African context? Van Der Merwe, Le Roux, Meyer and Van Niekerk (1990) examined the occupational
80 codes of 60 occupations in the South African context according to their RIASEC categories. Moreover, they endeavoured to find out whether the SDS codes could be predicted from the Position Analysis Questionnaire (1977), that classifies occupations in accordance with the work tasks of each. When only one letter of the assigned American occupational code was compared to one letter of the empirically determined South African code, 100% similarity was obtained, yet two-letter codes were only 86,7% alike and three-letter codes only 40%. A mere 11,7% of the codes corresponded when the sequence of the three letters was taken into account. An important conclusion is that American assigned occupational codes are not always applicable to South African occupations. However "the different suboccupation groups can to some extent be classified on the basis of the mean empirically observed SDS scores of the incumbents" (Van Der Merwe et al., 1990, p. 31), thus proving the South African applicability of the RIASEC model. The results also indicated that the personality dimension (SDS scores) could be predicted from the job dimension (PAQ scores). Brand, Van Noordwyk and Hanekom (1994) determined the applicability of the SOS for a group of black South African teenagers (N = 983), and in so doing the non-western cultural applicability of Holland's typology. Their findings proved significant positive correlations between the Vocational Interest Questionnaire, that had previously been standardised for South African populations, and the SOS. They concluded that the SDS was effective in assessing interests in this non-western culture. These recent examples of research confirm previous suppositions that gender differences exist in terms of SDS raw scores and the perceptions of the RIASEC categories. Further evidence suggests that Holland's theory can be applied across diverse ethnic groups. Since findings of American studies are not necessarily valid for other cultures, there is a need for further studies on the validity of Holland's theory in other countries such as South Africa.
81 3.6 ADDITIONAL DIMENSIONS Since the findings of studies on the dimensions of personality and environmental groups hold implications for congruence indices, such as the K-P and HCI indices that are based on the hexagonal shape, it is necessary to look at research on the proposed new dimensions discussed under point 2.6.
3.6.1 Data/ideas and people/things In his study, Prediger (1982) examined how his proposed data/ideas, people/things dimensions fit 27 sets of intercorrelations for the six types. In the initial study three sets of intercorrelations representing the mean scores of career groups and 24 sets of intercorrelations for individuals were used. Results of the analyses of these sets supported the existence of the dimensions both for career groups and individuals. The relationship between the RIASEC types and the dimensions closely resembled the hexagonal structure. In the second study, Prediger considered the heuristic value of the denominations of the dimensions to increase knowledge on the mechanisms of interest inventories. He thus recognised the need for a greater understanding of the concept of "interests" as basis of all interest measurements. Prediger correlated vocational interests (activities people enjoy doing) and occupational work tasks (activities required by specific jobs). Evidence from this study strongly suggested that the proposed dimensions provided support for interest inventories that are based on the theory and that similarities exist between people in specific occupations and the occupational activities they enjoy doing. In a study of 311 adults who were physically, psychiatrically or socially disadvantaged, or who had learning disabilities, Souliere, MacPhee and Flynn (1991) investigated the concurrent and predictive validity of the two dimensions proposed by Prediger. Moderate and consistent support was found both for the concurrent validity in terms of gender, social involvement, years of education and intellectual, clerical, manual and psychomotor skills, and for the predictive validity in terms of economic achievement,
82 occupational qualification requirements, social involvement, personality-job congruence and vocational identity. Rounds and Tracey (1993) performed meta-analysis on Prediger's data/ideas, people/things factors, as well as a general third factor or the "response-bias" factor. The researchers set out to achieve the following goals: to verify the presence of a general factor in the RIASEC data and its relative relation to the various RIASEC types; to employ structural meta-analysis on the RIASEC correlation matrices. This was in order to obtain data that would be representative of a variety of measures of the RIASEC types, while taking the variables of gender and age into account so as to ensure the generalisability of the RIASEC modes. (As seen above, controversy exists in previous research results due to dissimilar measures used and the influence of subject variables such as gender, ethnic, cultural and developmental factors.); and to find a valid description of the two axes of the RIASEC structure. Their results can be summarised under these points: This study accomplished its aim in confirming the presence of Prediger's general factor, yet, although the data/ideas, people/things dimensions were generalisable, they were not superior to other factors. -
The specific orientation of the axes appears to be arbitrary. Results suggest that both the circumplex model, describing the relations among types, and the factor-list model, describing the underlying dimension of the types have merit. The general factor (response-bias) accounted for 50% of the variance, but had equal loadings for all the RIASEC types, implying that it was independent of the RIASEC structure. They postulated that it may represent a third dimension of the structure of interests.
83 -
They found that the circumplex structure could be generalised to a variety of inventories, ages and both genders.
Whilst studying the learning outcomes of the Self-Directed Search (1979), Shivy, Phillips and Koehly (1996) did not find support for the twodimensional RIASEC configuration proposed by Holland, but results of their multidimensional scaling techniques supported the things/people, data/ideas constructs as well as a prestige dimension.
3.6.2 Validation of the circumplex structure According to a factor analysis of the SDS, Tuck and Keeling (1980) discovered that the social and enterprising scales had the same factor for their male and female samples and that only one factor was evident from the investigative and realistic scales for the females. The structural configuration of the factors corresponded closer to the hexagonal model for males than for females. The implication is that care should be taken before calculating consistency according to the hexagon especially for women. Similar gender differences in terms of the hexagon have been reported in the Hansen et al. (1993) study discussed above. Positive support for the hexagonal model was deduced from the responses of African American and white premedical students on the SDS (Henry, 1988). Holland and Gottfredson (1992) also reported that studies by Prediger and Vansickle (1992), Fouad and Dancer (1992) and Swanson (1992) support the hexagonal model. Hyland and Muchinsky (1991) studied the structural validity of Holland's model in the light of job analysis information as obtained from the Position Analysis Questionnaire. Vocational types were differentiated according to job analyses representative of 86 jobs. Vocational type was best predicted by environmental awareness, decision and communication responsibilities, and clerical activities. Using job analysis data, the realistic and conventional jobs could be better predicted than investigative, artistic, social or enterprising jobs. This study illustrates the usefulness of job analysis information to evaluate vocational structure.
84 Tracey and Rounds (1992) further investigated the dimensions underlying Holland's hexagon. They aimed to establish whether there was a match between the circumplex structure and the RIASEC codes. Their data showed that the RIASEC types, both in terms of personality and work environment, were proportionally arranged according to the circumplex structure. The hexagonal model was validified in relation to the constructs of consistency and congruence and a structural test of Holland's occupational classification (how well his model fits the occupational high-point codes). The representation of the types was significant for the two letter high-point code data. (The term "high-point codes" refers to the highest RIASEC codes used to study relations between and within personality types and work environments.) Their so-called quasi-circumplex best represented the frequencies of two letter high-point codes. (The term uquasi-circumplex" refers to a circumplex structure for which base rate differences are used and when only partial matching of frequencies occurs.) Since the base rates in distribution of different high-point codes are all very different, all adjacent types also do not have equal frequencies. Therefore the differences in base rates had to be accounted for, for the circumplex to fit the data. These results have important implications for the present study. Tracey and Rounds (1992) corroborated that the concept of person-environment congruence is supported by the quasi-circumplex. Another implication is that, since it is clear from this study that parallel structures exist for personality and environmental types, comparisons made of these structures are valid. The generalisability of this study is partially limited by the fact that the SDS was the only measure of RIASEC types used, and it is also limited to the specific sample used and the use of two letter high-point codes. Tracey and Rounds (1995) also studied the structural model of interests of high school and college students according to the Vocational Preference Inventory as well as the Inventory of Occupational Preference developed by themselves. A circular distribution of interests was confirmed for both groups of participants and for both measures used. Their conclusion was that vocational interests are uniformly distributed in a circular order, and that any number of scales, and not only the six RIASEC scales would fit this model.
85 Empirical evidence thus seems to support Prediger's (1982) data/ideas, people/things dimensions, as well as the circumplex structure of both personality and environmental types. The hexagon is also valid in terms of congruence, consistency and occupational classifications.
3.7 INDICES OF CONGRUENCE As seen in Chapter Two, one of the most useful consequences of this theory has been the development of indices of congruence that quantifies the relationship between personality and environment, and even between various personalities or interests at different points in time. Several studies deal with the indices of congruence discussed under point 2.5.3. Some of these findings are now discussed, not only to reflect the differences of the various indices, but also the relations among these. lachan (1984) tested his measure of congruence, by applying it to 42 students who were seeking vocational guidance. He found agreement between his measure and the self-reported occupational preferences of the students. Gottfredson and Holland (1990) compared a traditional congruence index with that of lachan. The traditional index is calculated as the degree of proximity of the highest personality score to the occupational score according to the hexagon, where 3 represents the highest proximity and 0 represent the lowest proximity. They found the lachan index to correlate significantly with the traditional index, and that neither of these indices were superior to the other. They however recognised the need for further research to illustrate the usefulness of indices such as the M-Index. Miller (1992) compared correlations among three measures of congruence, and summarises some of the strengths and weaknesses of these indices as follows: The Kwak and Palvino (K-P) Index (1982): This index takes the proximity of structures of types into account and is sensitive to subtle differences in secondary and tertiary code
86 letters. However, knowledge of all the paired correlations of the six types is required for the use thereof. The Zener-Schnuelle (Z-S) Index (1976): This index is less complex and easier to calculate, but is does not discriminate between similar but out-of-order codes. The lachan (M) Index (1984): It is a very precise measure, but does not take the proximity structures of different scales into account. Miller showed preference for the use of the K-P or M-indices for precision calculation (i.e. for research), whilst the Z-S Index could suffice for counselling purposes. Camp and Chatrand (1992) compared 13 indices of congruence and found such variability among these that they concluded that these indices were not interchangeable. The Kwak and Palvino Index, the Rank Comparison Congruence Scale and the lachan M-index displayed the highest correlations with the other scales, but the Primary Interests Congruence Scale hardly correlated with the other scales. These congruence measures were also compared with outcome measures. It is remarkable however that none of the indices correlated significantly with satisfaction with majors, career indecision or academic adjustment, and there were only small correlations between congruence and achievement and aptitude for this sample of 125 female students. The degree of correspondence between congruence measures and the satisfaction with major variable remained consistent regardless of the specific index used. Even dichotomous first letter, first-letter hexagon, two- and three-letter indices and the CAI did not vary from the other indices in terms of satisfaction. Brown and Gore (1994) evaluated indices of congruence in terms of the following three discriminant qualities, namely the ability: to discriminate between ordered and identical codes and others with same, but out-of-order codes;
87
-
to discriminate code order of out-of-order codes, identifying single identical letters in the corresponding place; to discriminate among out-of-order codes for hexigonal distances between non-matching letters in two sets of codes.
The results indicated that only the K-P Index has all three discriminating abilities and is able to fully operationalise congruence. The researchers also presented a new index of congruence - the Congruence (C) Index, which is comparable to the K-P Index, but also facilitates calculation, interpretation and cross-sample comparisons. They showed preference for the use of either the K-P or C Indices in future research. Myors (1996) proposed and tested a method by means of which the complete hexagon can be evaluated based on the psychometric techniques of the Spearman rank-order correlation coefficient. Two characteristics of the interrelations of types were taken into account: Three levels of correlations between personality types and environments exist, namely correlations between adjacent types, alternate types and opposite types; Correlations within each of these levels should be equal if the corresponding lines of the hexagon are of equal length. The proposed method for testing these correlations takes all the 72 possible ordered relations between the types as well as 33 equal types into account, as well as the presence of the circumplex model. In other words, all 105 possible comparisons between types can be tested. The procedure involved the calculation of the rank order of observed correlations and correlating these with the given ranks according to the hexagon. The significances of these correlations are also obtained. The proposed technique seems superior to previous tests, since it is easier to compute and uses an ordinal method with ordain data and also account for all relations of the hexagon.
88 Lent and Lopez (1996) employed the C-index, the Kwak and Palvino mathematical model and the Hexagon Congruence Index to determine the relations among these indices, as well as the relations of these indices to job satisfaction, whilst controlling the variable of organisational mobility. Two different work environment codes were used for each index, namely job titles compared to the Dictionary of Holland Occupational Codes (DHOC; 1982) and the Environmental Assessment Technique (EAT). This eventuated in the formation of six indices of congruence for their research design. All the indices were convergent and strongly interrelated. Interestingly, the method of job coding greatly influenced the correlations, that even negative correlations were found between the same indices with different job coding schemes. An important implication of these findings is that the DHOC codes cannot simply be applied to every occupational situation. Except for the Hexagon Congruence Index in accordance with the DHOC, none of the indices were predictive of job satisfaction. A very important key of this study for future research, is the role of the method of assigning job codes in studies of congruence and job satisfaction. Information from both the DHOC and the EAT should be combined and as many sources of information as possible should be used to compile the environmental code of a specific job. Although Holland (1985b) agreed with the use of the RA-Index of congruence, it seems from these research results that different indices may be applied depending on the level of precision required. There seems to be general agreement that the K-P Index is a very accurate measure. The importance of a thorough and precise assignment of occupational codes, in order to enhance the predictive value of the index used, is most noteworthy from the above.
3.8 THE SELF-DIRECTED SEARCH This discussion on research of the SDS will deal with some findings concerning the concurrent, predictive and construct validity of the SDS. The usefulness of the SOS will be examined, as well as its relation to personality traits. Finally, research that forms the basis of the debate on gender
89 differences in SDS scores will be briefly discussed. More specific aspects of the psychometric properties of this measure will be discussed in Chapter Six.
3.8.1 Validity of the SDS In this section the validity of the SDS will be depicted by an examination of a few studies on its construct validity, its predictive validity, especially in terms of career aspirations, and lastly its concurrent validity.
3.8.1.1
Construct validity Findings of more than four hundred studies confirm the validity of the SDS (Gregory, 1996). Some examples of these follow. Rachman, Amernic and Aranya's (1981) study confirms the validity of the SDS and hexagonal model in as far as the appropriate items measure the relating personality types. However, only the Interest subtest could differentiate among all the personality types, and the activity and ability subtests could not discriminate between the enterprising and social types. In a study referred to in the professional manual of the SDS, Holland (1985c) calculated the correlations between the six personality types for a sample of 175 women and managed to confirm the theoretical premises regarding the relations of types. These findings support the construct validity of the SDS.
The construct validity of the SDS is also illustrated by a study that found cross-cultural invariance of the pattern of intercorrelations of the SDS scales for American and Israeli participants (Peiser, 1984). The validity of the Self-estimates of the SDS (versus objective measures of abilities and competencies) was examined by Lowman and Williams (1987). They found higher correlations between self-ratings of abilities (other validated measures) and objective measures than for self-ratings of competencies (SDS). Moderate correlations were found between self-rating of ability and the objective ratings.
90 Confirmatory factor analysis of the SDS by Khan, Alvi and Kirkwood (1990) found support for the relationships among the RIASEC types in terms of the fields of activities, competencies, occupations and self-estimates for their sample of 155 high school students. Boyle and Fabris (1992) on the other hand questioned the validity of the RIASEC model in the Australian context. This follows their findings of analyses of the SDS that only partially supported the RIASEC typological model. In fact, their confirmatory analysis supported only the artistic factor, and their congeneric factor analysis only tentatively supported five of the six factors. Oosterveld (1994) who reanalysed data from the normative sample in the SDS manuals, found support for the construct validity of the SDS. He found a multitrait model (that includes the six RIASEC factors, as well as activities, competencies, occupational preferences, and self-rating of abilities) matched the data of the structural relations between the subscales better than a six factor model (that includes only the RIASEC types). Dumenci (1995) confirmed the construct validity of the SDS according to six correlated traits and five correlated method factors of several hierarchically nested structural models. The discriminant validity of the SDS was also apparent for the six vocational types and the data also supported the convergent validity of the model.
3.8.1.2
Predictive validity
In an early study of the predictive validity of the SDS, Gottfredson and Holland (1975) found that current vocational choice was the best predictor of later vocational choice. They describe the predictive efficiency of the SDS as either moderate or high. Higher predictive ability were found for older subjects, when there had been shorter intervals between two tests, or for investigative and social personality types. The validity of the personality type codes from the SDS in predictions of career choice is portrayed in a study by McGowan (1982) who found in a follow-up study that 31,4% of the participants made career choices
91 concurrent with their primary summary code determined four years prior and 51,2% of career choices coincided with their secondary codes. The predictive validity of Holland's theory for job search success using the SDS and VPI was assessed by Magnusson and Stewin (1990). Only very limited support was found for the predictive value of these measures in terms of job search success for their sample of technical students.
3.8.1.3
Concurrent validity
Neethling (1986), examining the psychometric properties of the SDS, reports that the concurrent and predictive validity thereof is comparable, if not superior to that of similar interest questionnaires. A few studies proving its concurrent validity are quoted in this section. The concurrent validity of the VPI, Holland Scales and SOS has been shown in early studies. Holland and Narzinger (1975) found correlations in the predicted direction for the SOS and the following scales: the Kuder Preference Record, the Thurstone Temperament Schedule, the Bennett Mechanical Comprehension Test and the Minnesota Paper Form Board. In another study, five of the VPI and Holland Scales and four of the SDS could successfully differentiate the occupational groups (Gaffey & Walsh, 1974). The ability of the SOS and VPI to discriminate among occupational groups has also been proved for a sample of employed black males (O'Brien & Walsh, 1976) and Fishburne and Walsh (1976) showed that two scales of the VPI and four scales of the SDS could successfully differentiate occupational groups for employed males. Neethling (1986) also found support for the ability of the SDS to effectively classify the enterprising and realistic groups based on the Occupations Finder. Costa, McCrae and Holland (1984) sought correlations between the SDS Strong relationships were found between and the NEO Inventory. investigative and artistic interests and a trait of openness to experience. Social and enterprising interests related to extroversion. Thus the investigative, artistic, social and enterprising themes are supported in research. Tokar and Swanson (1995) found gender differences in their study of the correspondence of SDS personality scores and the NEO FiveTraits of openness and extroversion were Factor Inventory (Form S).
92 effective in discriminating among the personality types of males, and openness, extroversion and agreeableness discriminated the personality types of females. Partial correspondence was found between the SDS and the NEOFFI-S.
3.8.2 The effects and usefulness of the SDS From the previous points it is evident that the SDS has adequate psychometric properties in terms of its construct, predictive and concurrent validity. Its usefulness can be seen in the following studies that are examples of research on the effects of this measure on both students and adults. When studying the effects of the SDS on high school students, Zener and Schnuelle (1976) found that it did accomplish its objectives. It increased the amount of vocational alternatives considered, was associated with greater satisfaction with current vocational choice and increased knowledge of Holland's classification scheme. A study by McGowan (1977) showed the effectiveness of the SDS in abating career indecision and therefore its usefulness in educational and vocational planning among high school students. A two month follow-up study of the effect of the SDS on high school girls, revealed that both the number of occupational options presented by the SDS and the use of and instructional booklet can play an important role in increasing the number of occupational options considered (Holland, Rakai, Gottfredson & Hanau, 1978). An abbreviated version of the SDS did not increase career knowledge of students or their abilities to make career decisions, but lowered the students' estimates of the career decision making in a study by Healy and Mourton (1984). O'Neil, Price and Tracey (1979) assessed the stimulus value and treatment effects of the raw scored and the norm-scored SOS as well and the StrongCampbell Interest Inventory (1985) for students, with a follow-up study one month later. From their results it became clear that the SDS had stronger stimulus value than the SCII, but that the SCII gave greater clarity of direction than the SDS. Importantly, no gender differences were found in terms of reaction to testing. In the follow-up test, no gender differences were
93 found in terms of satisfaction with career planning or clarity and certainty of ideas about career planning, or time spent thinking about career planning. Schaefer (1976) indicated that the effectiveness of the SDS as a vocational guidance tool was not contingent on the variables of consistency, differentiation, level of mental ability or level of achievement. Healy and Mourton (1983) found that the accuracy of the sample of 138 students in estimating their own SDS profile correlated significantly with other indices of career maturity. These indices included decision making, knowledge of the world of work in general, and knowledge of one's preferred occupational area from the Career Development Inventory, as well as indices of grade-point average and differentiation. This shows that the Accuracy scale of the SDS is useful in assessing progress in career counselling. Levinson (1990) commented on the usefulness of the SDS for school psychologists. Not only is it a psychometrically sound measure, but it is also time- and cost-efficient in psychoeducational assessments. Conneran and Hartman (1993) again found that the SDS has concurrent validity in identifying chronic career indecision among vocational education students. The accuracy of scoring of the SDS by ninth-grade students was explored by Elliott and Byrd (1985), who found that although 81% of the subjects committed an error, the final codes of 90% were still correct.
3.8.3 Raw versus normed scores One principal point of criticism that has been raised against Holland's operationalisation of personality types, is the lack of gender-related norms for the SDS. Aspects of the debate regarding gender and the SDS will now be mentioned. On analysis of American College Testing Interest Inventory scores of 10 578 student participants, Prediger and Hanson (1977) found that the criterionrelated validity of standard scores based on gender norms was equal to or greater than raw scores. An example of research results that favour the use of raw SDS scores is that of Keeling and Tuck (1978) who found that raw
94 scores were superior to same-sex normed scores for first letter congruency of preferred vocations, and females especially preferred occupations that were congruent with their raw score first letter than their normed scores. In summary of the research findings concerning the SOS, it seems that the recent data supports the construct validity of the SDS, although there are some exceptions to this, such as the cited Australian study. Findings also support the predictive validity of the SDS for career choice and give limited support in terms of job search success. Other results illustrate the usefulness of the SDS in career decision making and career counselling and the relations of SDS personality types and other personality traits. Looking at the origin of the raw versus normed scores debate, it seems that there are no real grounds for the use of gender-related normed SDS scores.
3.9 CONCLUSION Holland (1996) believes that there has been an overemphasis of personal characteristics and an underemphasis of environmental characteristics in person-environment interaction research designs. It seems that numerous studies have been conducted on the relationship between personality and occupational environment. Studies on the relationship of personality and other environments, such as the family could prove invaluable. Findings emanating from such studies could for instance shed more light on the development of Holland's personality types. One of the most remarkable conclusions that can be drawn from research on Holland's theory and its hypotheses, is that the congruence-satisfaction relationship is not always supported in research. However, there is still a substantial amount of studies that manage to support the assumed association between congruence and satisfaction. Thus, the need for a greater understanding of influencing variables in the interplay of persons, their occupations, and their attitudes such as job satisfaction, and behaviours, is evident. Some of the variables of this reciprocal relationship that has already been identified in research include personality type, consistency, differentiation, job tenure, group importance, gender, individual environmental reference, and specific definition of job satisfaction. Others still remain to be identified and studied.
95
Research on congruence and job outcomes other than job satisfaction reveals that the theory on the construct of congruence may be too simplistic. Hypothesised relationships between congruence and achievement, or congruence and vocational stability for instance are supported by some researchers and rejected by others. Such findings suggest that several other variables need to be incorporated for Holland's theory to be a more complete description of the association of persons and occupations. These findings also clearly depict the precision that is required in the definitions of research variables. Results relating to Holland's theory must only be interpreted in terms of a very specific delineation of the personality type and environment involved. Broad occupational classifications once again seem much inferior to specific environmental codes of populations in studies of congruence and related variables. A further debatable aspect of Holland's theory is that of gender and genderrelated norms in personality type assessments. Although Holland and Gottfredson (1976) argue that differentiating ability of the theory is enhanced by the absence of group norms, the cited study by Hansen et al. (1993) shows that gender differences in terms of interests can not merely be ignored. The hexagonal model which is central to the construct of congruence and the interaction between personality and environment could be revised to reflect the empirically supported gender differences. One could also argue that similar differences in terms of cultural groups should be mirrored in Holland's theory. In fact, some of the research findings mentioned above indicate that there are limitations in the cross-cultural applicability of Holland's theory and hexagon, whilst other studies provide evidence of the heuristic value of the theory in non-western cultures. Lastly, the hexagon as two-dimensional depiction of the interrelationships of types has also been carefully scrutinised by researchers. As it provides a means for portraying and assessing person-environment interactions, it can be seen as one of the most valuable aspects of Holland's theory. The discussion of research on the hexagon shows that data generally supports the model, although as previously mentioned, group-specific modifications thereof may be required to increase the usefulness and validity of Holland's theory. One elaboration of the dimensional representation of types that has
96 evidently enhanced the theory is the empirically supported data/ideas and things/people dimensions of Prediger. In conclusion, the construct of congruence or person-environment fit fulfils an important role in vocational psychology, as illustrated by the diversity and extensiveness of research dealing with this construct. Although much research has been conducted on the relationship between congruence and job satisfaction, there still seems to be a need for studies on the mediating variables of the congruence-satisfaction relationship, for further clarity on the conceptual framework to which these constructs belong. One may assume that congruence between personality and environment will culminate in positive outcomes in all aspects of work, however this is not always the case, as researchers have failed to find significant relationships between congruence and outcomes such as achievement or vocational stability. The fact that gender differences appear in research on Holland's theory and that previous research findings can not always be applied to other cultures, open the door for further studies that may lead to the refinement of Holland's theory and an assessment of the validity of Holland's theory in cultures such as our own. The fact that the hexagonal structure of types is challenged by researchers also in this instance leads to further confirmation of the interrelationships of types and the importance of the congruence construct. The development of indices of congruence and research relating to the relative usefulness of each index is one of the most valuable facets of this theory, as it provides the researcher with the necessary means to validate and elucidate the theory. Moreover, research supports not only the validity of the Self-Directed Search, but also its effectiveness and usefulness in career choice and counselling.
97
CHAPTER 4 A THEORETOCAL PERSPECTOVE OM WORK ADJUSTMENT AM JOB NVOLVEMERIT
4.1 INTRODUCTION In Chapter Two the concept of person-environment congruence based on typological relationships was introduced. The attitude of job satisfaction was seen as a consequence of the proximity of the relationship between a person and his vocational environment. In this chapter, the Theory of Work Adjustment presents an alternative perspective of the person-environment interaction. Unlike Holland's theory, the Theory of Work Adjustment does not examine relationships between personality types and environmental types, but instead of associations between an individual's abilities and values and the particular ability requirements and reinforcers of the work environment specifically (Davis and Lofquist, 1984). This theory constitutes the theoretical framework of the measure of job satisfaction that will be employed in the present study, namely the Minnesota Satisfaction Questionnaire. Given the need for a greater understanding of the role of other work-related variables on the hypothesised congruence-satisfaction (Holland) or correspondence-satisfaction (Davis and Lofquist) relationship, this chapter also contain a description of the construct of job involvement as defined by Rabindra Kanungo (1982b). This perspective on job involvement is operationalised by the Job Involvement Questionnaire that will be employed in this study. Thus these two perspectives have been included in the discussion due to the measuring instruments that has been employed to fulfil the objectives of the study.
4.2 The Theory of Work Adjustment Lofquist and Dawis' (1969) account of the Theory of Work Adjustment is based on work performed by Dawis, England, Lofquist, Scott and Weiss in 1964 as part of the Work Adjustment Project of the University of Minnesota. Several elaborations of the theory have appeared in literature and therefore
98 Dawis and Lofquist presented a revision thereof in 1984. It addresses the social problems that relate to work such as occupational stress, dissatisfaction with work and burnout. It constitutes the theoretical background to several measures, including the Minnesota Satisfaction Questionnaire (MSQ; 1967). Its value in practice and research consists of the sound structural framework it provides for understanding personality, environment and the interaction between these. Based on this theoretical foundation, specific predictions on, amongst other things, satisfaction, satisfactoriness and job tenure equip the researcher with the necessary hypotheses to empirically determine the theoretical pertinence of the constructs thereof. In addition to this, the systems component of the theory supplies the researcher and practitioner with functional information on the processes of work adjustment, adjustment behaviour, the modes of adjustment and personality styles (Hesketh, 1995). Over and above these apparent advantages of the Theory of Work Adjustment, the job satisfaction measure associated with this theory, i.e. the Minnesota Satisfaction Questionnaire is beneficial to any researcher studying job satisfaction. Furthermore, this work adjustment theory is not limited to adjustment in the rehabilitation process, but regards work adjustment as a goal of all employees for successful employment (Cull & Hardy, 1973). Therefore the discussion on this theory will commence with a brief mention of the impetus of the theory, before the structural, system and predictive components, and the operational aspects thereof will be discussed. Its usefulness to various spheres of life will be discussed under the heading "Implications and applications of the Theory of Work Adjustment". Its applicability is subsequently illustrated in the considerations of the measures that operationalise the theory.
4.2.1 The impetus of the theory As mentioned in the introduction, successful work adjustment is a prerequisite for successful employment. Before one can understand work adjustment however, one needs to discern what the meaning of work is and why the theory was necessary. In this discussion, work will firstly be defined in accordance with the Theory of Work Adjustment. Secondly, the significance and meaning of work will be described. Lastly, the solutions
99 that an understanding of person-environment interactions presents to problems in the work domain are mentioned. As point of departure, work can be defined as "the interaction between individuals and their work environments" (Lofquist & Dawis, 1975, p. 132). Nevertheless, work is not merely a means to an end, but is central to the development of man's way of life and part of his total adjustment in life, and work provides the opportunity for satisfying needs (Lofquist & Dawis, 1969). The following quotation illustrates the need satisfaction potential of work: "Work is the means of maintaining a certain standard of living, a certain level of existence, and also of achieving some higher level or standard. Work is something to do, a way of filling the day or passing the time. Work is a source of self-respect, a way of achieving recognition or respect from others. Work defines one's identity, one's role in the society of which he is a part. Work provides the opportunity for association with others, for building friendships. Work allows for self-expression, provides the opportunity for creativity, for new experiences. Finally, work permits one to be of service to others" (Lofquist & Dawis, 1969, p.10). Work is therefore not only related to the satisfaction of basic human needs, but also the satisfaction of psychological needs such as achievement and creativity. Hence the meaning of work for the worker is the opportunity to meet his needs, and for the work environment the meaning of work entails the achieving of organisational objectives, and in so doing meeting the needs of the work environment (Dawis, 1980). Despite the need satisfying potential of work, the world of work encumbers problems, such as the limited freedom of choice of work, due to cultural, social, personal and hereditary pressures, as well as the complexity of the work domain. Lofquist and Dawis recognised the need for "systematic ways of relating characteristics of individuals to information about occupations" (Lofquist & Dawis, 1969, p. 20). In reponse to this need they developed a theoretical framework of work choice and work adjustment, by describing the work personality, the work environment and correspondence between personality and work factors.
100 From the discussion that now follows, it will become apparent that the theory is successful in achieving this aim through its structural component. The usefulness of the systems component lies in the link it provides between individual characteristics and work environments, based on the assumptions of the theory, as well as the styles dimensions that qualify the associations of persons and environments. Its value is enhanced by the predictions of work behaviour and outcomes, such as satisfaction, satisfactoriness and tenure that results from these assumptions in the predictive segment of the theory.
4.2.2. The structural component As seen above, the purpose of the theory in relating characteristics of people with that of occupations is in the first instance accomplished through its structural component. In this section, personality will firstly be defined and descirbed in terms of its structure and development. Thereafter, the work personality as such will be examined. Lastly, the fundamental aspects of the work environment will be described.
4.2.2.1
Personality The following defintion of personality grounds the theory in the theoretical paradigm to which it belongs. The obseravation regarding the structure and development of personality form the foundation of the theoretical principles that will follow.
(a)
Definition Personality is described from a behavioural perspective, or in terms of an individual's reactions to his environment. Personality is defined as a "stable characterization of an individual as a responding organism" (Lofquist & Dawis, 1969, p. 26). Concepts of ability and needs instead of skills and reinforcement values are at the centre of this definition.
101 (b)
Structure and development The personality embodies two dimensions, namely the broader skills dimension and the more specified ability dimension. Skills are acquired through repetition of response sequences. Accordingly, different individuals have different abilities and are exposed to and respond to different stimulus conditions. Each individual also uses different norms to evaluate these conditions, known as preferences (Lofquist & Dawis, 1969). Each stimulus condition has a different reinforcement value for an individual, depending on the individual's preference for certain stimulus conditions (reinforcers). A stimulus condition with high reinforcement value is known as a need. People develop stable need patterns due to consistent exposure to similar stimulus conditions. These needs and abilities develop, differentiate and then stabilise. Complex interrelationships exist between abilities and needs, which give rise to certain preferences or expressed interests as well as exhibited interests (participation in certain activities). The structure of personality consists therefore of needs and abilities and the interactions between these resulting in interests. Personality style refers instead to the method of utilising abilities and satisfying needs (Lofquist & Dawis, 1969).
4.2.2.2
The work personality
Relevant to the theory, a specific definition and description of the work personality will now be described and subsequently the development of the work personality will be examined. (a)
Definition and structure The abilities and needs that are most relevant to work behaviour, and the functioning of these needs and abilities in the work environment, constitute the work personality (Lofquist & Dawis, 1969). in Dawis and Lofquist's (1984) revised description of the work personality, they assert that the work personality consists of the personality structure and personality style. Structure of personality is defined as "the individual's response capabilities and preferences for stimulus conditions" (Davis & Lofquist, 1984, p. 25), and personality style as "the individual's characteristic style of responding in
102 bringing response capabilities and stimulus preferences into play in the interaction with the environment" (Dawis & Lofquist, 1984, p. 25). Four aspects distinguish the various work personalities, namely abilities, values, interests and personality style. The needs of both employers and employees are determined by the dimensions of abilities, values and style (Davis & Lofquist, 1984). Abilities refer to innate aptitudes and not acquired skills. Work experience leads to exposure to new stimulus conditions and can result in the development of new abilities. Shari (1992) refers to the six values of the work personality that are based on needs, namely achievement, comfort, status, altruism, safety and autonomy. The value of comfort is associated with needs for activity, independence, variety, compensation, security and working conditions. Likewise several needs are associated with each value (Shari, 1992). The value dimension of the work personality structure can be assessed by means of the Minnesota Importance Questionnaire (IVIIQ), (Dawis & Lofquist, 1984). Interests can be seen as an expression of the relationships among abilities and values. Therefore instead of assessing interests, Lofquist and Dawis focus on the measuring of abilities and values (Shari, 1992). (b)
Development Work personality develops as a range of various response capabilities. Initially these responses are aimed at maintaining physiological functioning, and are responses to direct stimulation. Later the individual explores and acts to varying conditions of stimulation. The development of personality is a result of the variety, range and complexity of potentialities for responding. It develops in three stages: - Differentiation: exploration, developing and expanding response abilities in educational environment; - Stability:
crystallisation and stabilisation of a response repertoire in the work environment;
- Decline:
response capabilities affected by ageing.
103
Brown (1993) proposes the application of contemporary psychological science to the Theory of Work Adjustment so that more predictions can be made from the theory. In particular, research on the structure of normal personality and affect and the structure of intellect can make the Theory of Work Adjustment more "psychological". To illustrate this, he hypothesises that the personality factors of Neuroticism (N), Conscientiousness (C), and Openness to Experience (0) are predictive of work adjustment. The N-factor is said to increase the Theory of Work Adjustment's usefulness in predicting job satisfaction, the C-factor in predicting satisfactoriness and the 0-factor should be considered in predictions of tenure or occupational change. Also, since research has shown relations between intelligence and job or academic performance, the usefulness of research findings concerning intelligence to the Theory of Work Adjustment is apparent.
4.2.2.3
The work environment
The second aspect of the person-environment interaction in the Theory of Work Adjustment deals specifically with the vocational environment which will now be contemplated. Work environment is described in terms of the way personality relates, through needs and abilities, to the world of work. Three suppositions underlie the interaction between personality and environment: People select work environments which are compatible with their work personalities; The work environment selects people with adequate work personalities; Different work environments have typical work personalities associated with them. Parallel to the work personality, the work environment can also be divided into the work environment structure with ability requirements and reinforcer factors dimensions, and the work environment style with dimensions of celerity, pace, rhythm and endurance. There are two categories of stimulus conditions that impel responses in the work environment, namely cues, that
104 trigger the correct responses (associated with abilities), and reinforcers that maintain certain responses (associated with needs), (Dawis & Lofquist, 1984). Occupational ability patterns and occupational reinforcer patterns further define work environments. Occupational ability pattern refers to the minimum level of required abilities for satisfactory work performance. Occupational reinforcer pattern refers to the minimum level of needs required in a worker to predict his satisfaction in a work environment. The Minnesota Occupational Classification System ll (MOCS II) provides a taxonomy of occupations in terms of their ability requirement level (high, moderate, average) and need-reinforcer patterns based on the six most salient value reinforcers. The third revision of the MOCS categorises 1769 occupations (Shari, 1992). The following figure depicts the Theory of Work Adjustment and the interaction between the aforementioned concepts:
Operationally defined Work personality
Work adjustment
work environment
Work environment
To infer
I
SKILLS/ABILITIES
ABILITY REQUIREMENTS
INDIVIDUALS
SPECIFIC
WITH SUBSTANTIAL
OCCUPATION
To define
OCCUPATIONAL ABILITY PATTERN
TENURE
NEEDS/VALUES
REINFORCER
To define
CHARACTERISTICS
OCCUPATIONAL REIN FORCER PATTERN
Figure 4.1:
Personality and environmental interaction (Dawis & Lofquist, 1984, p. 59).
105 4.2.3 The systems component The Theory of Work Adjustment does not merely provide a description of the characteristics of both work personality and environments as seen above, but it also integrates information on personality and work adjustment styles, with knowledge on the interaction between the work personality and the work environment, and the resultant adjustment behaviour. Therefore the systems component of the theory will be highlighted firstly by the discussion on the assumptions of the theory. Thereafter the role of the work personality and work environmental styles in the person-environment interaction will be explicated.
4.2.3.1
Assumptions of the theory The basic assumption of the theory reads: "Each individual seeks to achieve and maintain correspondence with his environment' (Lofquist & Dawis, 1969, p. 45). Correspondence refers to a mutually responsive harmonious relationship between personality and environment where work and personality meet their reciprocal requirements. This implies that an individual provides certain skills to the work environment, and receives certain rewards from that environment (Davis & Lofquist, 1984). Lawson (1993) states that the dynamic interaction between person and environments involves "coresponsiveness" that is more than mere "correspondence". The term "corresponsiveness" is used by Dawis and Lofquist (1976) to describe the mutual responsiveness of work personality and work environment.
The environment of work is one of the principal environments with which a person seeks correspondence. Correspondence in the work environment presupposes a degree of fit between the work personality structure and the work environment structure. The work personality structure is made up of ability and values dimension and a specific vocational position consists of several required work skills. Work skills in turn requires several underlying abilities, for example, the skill of fixing something requires among other things visual, tactual and auditory abilities (Dawis & Lofquist, 1984). In
106 terms of this, work adjustment is defined as "the continuous and dynamic process by which the individual seeks to achieve and maintain correspondence with his work environment" (Lofquist & Dawis, 1969, p. 46). The stability of the correspondence is called tenure. In practical terms it can be said that when the person and the environment fail to meet their reciprocal requirements, the person or the environment will end the interaction between them (Davis & Lofquist, 1984). The consequences of correspondence are satisfactoriness and satisfaction. Satisfactoriness occurs when an individual fulfils the requirements of the environment and satisfaction is experienced when the environment fulfils the requirements of an individual. Satisfactoriness and satisfaction are respectively external and internal indicators of correspondence (Dawis & Lofquist, 1984).
4.2.3.2
The work personality and work environment style
The question arises as to what will happen when there is discorrespondence between the work personality and the work environment, for example, when a person is not competent enough for a specific job, or the person's needs are not being met, such as a need to be creative. This depends on the work personality style and the work environment style. Personality style is defined in terms of celerity, pace, rhythm and endurance. Celerity refers to the speed with which a person interacts and responds to the environment (promptness versus procrastination). Pace suggests the level of activity, energy or effort of the responding person. Rhythm indicates the variability of response (constancy of pace) and endurance refers to the time the individual continues to respond (tenure). These dimensions of personality style can be assessed through analyses of past records or through clinical observation (Dawis & Lofquist, 1984). Dawis and Lofquist (1978) also define modes of adjustment of the work personality style in terms of four dimensions of flexibility, activeness, reactiveness and celerity. People with similar personality structures (abilities and values) may differ in the level of discorrepondence they are prepared to tolerate (work personality style). A flexible personality style
107 tolerates discorrespondence to a certain level. According to the activeness dimension however, a person will act on the environment to increase correspondence, whereas the reactive dimension results in the change of personality structure to increase correspondence. The celerity dimension of the personality style refers to an "individual's typical rate of movement toward increased correspondence" (Davis & Lofquist, 1978, p. 56). The process of work adjustment is known as the adjustments style dimensions. Adjustment style is described in terms of flexibility, activeness, reactiveness and perseverance. Although "tenure is a function of correspondence between abilities and requirements and between values and reinforcers" (Corollary Villa), different individuals will tolerate different levels of discorrespondence due to different levels of flexibility. Activeness refers to a style in which an individual responds actively to increase correspondence with the work environment and reactiveness is an adjustment style dimension when the individual changes the expression of personality structure to increase correspondence. Perseverance refers to the length of time of tolerance of discorrespondence and indicates the effectiveness of the individual's way of adjustment (Davis & Lofquist, 1984). Lawson (1993) recognised the need for the operationalisation of the personality constructs and adjustment style dimension of the Theory of Work Adjustment. She developed three scales of the four aspects of adjustment behaviour (flexibility, activeness, reactiveness and perseverance) and the four personality styles (celerity or response latency, pace or reponse intensity, rhythm or response pattern and endurance or response duration). Scales of inflexibility, Activeness and Reactiveness were based on archival data and refined according to the reliability and content of the items. Lawson (1993) then aspired to validate these scales by means of correlations with General Occupational Theme scores on the StrongCampbell Interest Inventory and value scores on the NO. She found that these scales distinguish between different demographically determined groups. The Activeness scale was most valid, but the Reactiveness scale demonstrated context problems in that it indicated neuroticism, and the Inflexibility scale suggested a factor of cognitive inflexibility or lack of creativity.
108 According to Tinsley (1993), Lawson's (1993) findings highlight a possibility of relations between the Theory of Work Adjustment personality and adjustment styles and the "big five" personality factors from the NEO Personality Inventory, namely relations between reactiveness and neuroticism, positive relations between pace style and extroversion and negative relations between pace style and conscientiousness. Tinsley (1993) remarked that the Theory of Work Adjustment's personality and adjustment styles are not fully representative of human cognition and affect in work behaviour. The description of adjustment styles are useful, since links have been established between these adjustment styles and Holland personality types (Hesketh, 1993). Connections between reactiveness and the investigative and artistic personality types, activeness and the social and enterprising personality types and adverse relations between flexibility and the conventional and realistic personality types are suggested by Lawson (1993).
4.2.4 The prediction component In addition to its description of work personalities and work environments, and their relationships, and as a person-environment fit model the Theory of Work Adjustment is also useful in predicting behaviour and attitudes. Predictions concerning work adjustment, tenure, correspondence, satisfaction and satisfactoriness have been made. These predictions are based on the assumptions of the theory. One can predict that work adjustment will take place when there is correspondence between the work personality and the work environment. The outcomes of work adjustment, namely satisfactoriness, satisfaction and tenure are predicted by correspondence (Davis & Lofquist, 1984). When the environment and work personality meet each others reciprocal needs (correspondence), the person will be likely to stay on (tenure), implying satisfaction experienced by the person and satisfactoriness offered by the job, and resulting in the meaningfulness of that position for the person (Cull & Hardy, 1973). Hence, satisfaction occurs when the person's occupational
109 needs correspond with the reinforcer system of a certain job (Davis & Lofquist, 1984). Tenure of individuals can be predicted, when the satisfactoriness and satisfaction levels for a group with considerable tenure are known. In other words, a person will remain in a work environment if the minimum requirements of satisfactoriness and satisfaction are being met. If an individual's work personality falls within the limits of satisfactoriness and satisfaction, one can infer that his work personality is correspondent with the specific work environment. Satisfactoriness and satisfaction can be predicted by work personality-work environment correspondence. Work personality-work environment correspondence can also be used to predict tenure. Prediction of work adjustment is dependent on knowledge of the components of needs, reinforcers, task performance and task requirements (Davis & Lofquist, 1984). Based on these assumptions, Lofquist and Dawis (1969) give nine basic propositions for further research on predictions of work adjustment, which is the formal statement of the theory. Relevant to the present study, emphasis can be placed on Proposition III that defines satisfaction as "a function of the correspondence between the reinforcer system of the work environment and the individual's needs, provided that the individual's abilities correspond with the ability requirements of the work environment" (Lofquist & Dawis, 1969, p. 53). The definition of job satisfaction quoted in Chapter One can be expanded to read: "Satisfaction is a function of the correspondence between the reinforcer pattern of the work environment and the individual's values, provided that the individual's abilities correspond to the ability requirements of the work environment" (Davis & Lofquist, 1984, p. 64). The importance of a need to a person can moderate this prediction of job satisfaction from need-reinforcer correspondence. This applies in as far as more important needs (as indicated by the level of the scale score) are associated with greater importance of the need-reinforcer difference in the prediction of job satisfaction. Dawis and Lofquist (1984) cite research supportive of each of the nine propositions and its corollaries. The entire theory rests on the proposition and corollaries thereof (see Annexure C). In addition to these, they
110 developed seven further propositions based on the work adjustment style (see Annexure D).
4.2.5 Implications and applications of the Theory of Work Adjustment Lofquist and Dawis (1969) postulate that the theory can be widely applied. Some of these presumed uses will now be listed. The Theory of Work Adjustment can provide solutions to social problems, ranging from unemployment, poverty and disability to issues of automation of the work place, and retirement. One example of this, is how the matching of an individual's abilities to specific work environment requirements can lead to greater tenure and satisfaction, and prevent unemployment (Lofquist & Dawis, 1969). Likewise, the theory has several other applications. Knowledge of the principles of this theory are important in order to equip counsellors in all spheres of life. Greater focus is placed on individual personality-work relationships than on categories. Also, instead of attempting to change stable work personalities, people are helped to find compatible work environments (Lofquist & Dawis, 1969). The practical application of the Theory of Work Adjustment and its propositions also extend to the domain of personnel assessment. According to Dawis (1980) job performance as such can not be used as criterion for personnel assessment, but "satisfactoriness" as evaluation of job performance can rather be used. Instead of rating employees on job performance, their capability to perform a job or meet the job skills requirements serves as valid criterion. Dawis and Lofquist (1984) further describe the usefulness of the theory for career counselling, job adjustment counselling, counselling for job change as well as vocational rehabilitation counselling and unemployment counselling. Principles of the Theory of Work Adjustment can also be used in counselling for issues such as marriage, stress, addiction, self-esteem, and dependency. In organisational development, the theory is useful, among other things, for personnel selection, employee training programmes, employee motivation,
111 job analysis and the redesign of work environments. Social development is enhanced through the theory when applied to education and training.
4.2.6 Assessment in the Theory of Work Adjustment Several measures have been developed in association with the Theory of Work Adjustment and the constructs thereof. Emphasis will now be placed on measures of job satisfaction, satisfactoriness, needs and reinforcers and abilities.
4.2.6.1
Satisfaction
The Minnesota Satisfaction Questionnaire (MSQ; 1967) is based on the concept of job satisfaction. Lofquist and Dawis (1975) subdivide the job satisfaction construct into general job satisfaction (being satisfied with the job as a whole) and specific job satisfaction (or satisfaction with the various facets of the job, such as pay, working conditions and so forth). The MSQ is a measure of specific job satisfaction. In the work environment, twenty major sources of job satisfaction (called work reinforcers) can be found. Each of the twenty MSQ-subscales are based on one of these work reinforcers. The twenty work reinforcers encompassed by the MSQ are ability utilisation, achievement, activity, advancement, authority, company policies and practices, compensation, co-workers, creativity, independence, moral values, recognition, responsibility, security, social service, social status, supervision in terms of human relations, technical aspects of supervision, variety and working conditions (Lofquist & Dawis, 1975). The validity and reliability of the MSQ will be discussed in Chapter Six.
4.2.6.2
Satisfactoriness
The Minnesota Satisfactoriness Scales (MSS) measure the performance of an individual in specific job environment. In other words, it measures the satisfactoriness of job performance (Lofquist & Dawis, 1975). Whereas the MSQ measures how benefitial the environment is to the person, the MSS assesses the value that the person has for the environment.
112
4.2.6.3
Needs and reinforcers
The needs that are representative of the six work personality values can be measured by means of the Minnesota Importance Questionnaire (MIQ). More specifically, this measure assesses the vocational needs or preferences for certain work reinforcers. Subjects rate the relative importance of twenty needs by means of the MIC) (Shari, 1992). Reinforcer systems comprise the pattern of reinforcers available in an environment. Occupational Reinforcer Pattern (ORP) is the pattern of reinforcers in a given occupation as assessed by supervisors in that occupation (Lofquist & Dawis, 1975). Commensurate measurement of need systems on the one hand and reinforcer systems on the other is required for the assessment of the satisfaction proposition (Proposition Ill). N eeds are measured by means of the Minnesota Importance Questionnaire , and reinforcer systems by means of the Minnesota Job Description Questionnaire (MJDQ). Occupational Reinforcer Patterns (ORP) are representations of these twenty reinforcer dimensions for occupations. In order for person-occupational fit to be measured, the MIQ and ORP both measure the same twenty reinforcer statements (an identical set of reinforcers), which are parallel, with almost identical factor structures and with equal scale units. The needs and reinforcers are thus commensurate in terms of the concept measured, the structure thereof and the units of the structure (Rounds, Dawis & Lofquist, 1987). The rationale of the reciprocal effects of needs and rewards is a most valuable aspect of the theory. These concepts does not merely allow for vague comparisons of persons and their occupations, but facilitate the evaluation of the fit between very specific work values of an individual and the reinforcers of his specific job.
113 Abilities
4.2.6.4
In order to match abilities and values with Occupational Ability Patterns and Occupational Reinforcer Patterns, the Minnesota Importance Questionnaire, General Aptitude Test Battery (GATB) and MOCS are very useful. The GATB of the US Department of Labor (1979) can be used to assess abilities, measuring the nine specific abilities, namely general learning, verbal, numerical, spatial, form perception, clerical abilities as well as eye hand coordination and finger and manual dexterity (Shari, 1992). The Occupational Aptitude Patterns (OAPs) instrument measures the ability requirements of jobs (Cull and Hardy, 1973). Information from these as well as the MJDQ and a biographical information form provide the necessary information to measure abilities.
An integration of the various measures
4.2.6.5
Figure 4.2 illustrates the function of the MSQ in relation to the other measures pertaining to the Theory of Work Adjustment: (Predicts)
Degree of Correspondence
Occupational Aptitude Patterns
General Aptitude I Test Battery
Minnesota Satisfactoriness Scales
Wo rk
(Predicts )
Personality
Minnesota Importance Questionnaire
Occupational Reinforcer Patterns
Degree of Correspondence
Figure 4.2:
Months or Years on job
Minnesota Satisfaction Questionnaire
(Predicts)
I
Operational aspects of the Theory of Work Adjustment (Lofquist & Dawis, 1969, p. 62).
114
From this figure it is obvious that the degree of correspondence between a person's aptitudes and the aptitude patterns of an occupation is predictive of the level of the job performance and usefulness of that person to the specific job. However, this relationship is moderated by the degree of job satisfaction in terms of twenty aspects of the job as assessed by means of the MSQ. Moreover, the degree of correspondence between a person's needs and the reinforcers contributed by the occupational environment is predictive of the satisfaction the person will draw from his job, albeit that this may be influenced by the satisfactoriness of the person's job performance. The job tenure of the person can then be deduced from the satisfactoriness and satisfaction resulting from the interaction of his work personality and work environment. It is apparent that the Theory of Work Adjustment provides very specific and definite ways of assessing the abilities and needs that relate to the work personality, the abilities required by and the reinforcers offered by the work environment, as well as the resultant levels of satisfactoriness and satisfaction.
4.2.7 Comments and elaborations on the Theory of Work Adjustment The dynamic nature of the Theory of Work Adjustment is illustrated both by elaborations on the theory and recent criticism thereof. A model of Wanous (1980) that has been based on the principles of the theory will firstly be mentioned, and several points of criticism of the theory that appears in literature will be reported. Dawis and Lofquist's (1993) response to these points of criticism will also be included. Wanous (1980) developed a matching model to expand the Theory of Work Adjustment's description of the reciprocal relation of needs and rewards, that results in job satisfaction. Focusing on the transition from school to work, the model deals with the degree of fit between needs and rewards during job selection. According to this model, job satisfaction is not only a result of a match between needs and rewards, but is also influenced by the expectation of reaching intrinsic and extrinsic work goals through a current job rather than through another job. Wanous (1980) also places emphasis on the role
115 of organisational commitment in job satisfaction and turnover behaviour. Furthermore, realistic and accurate job information need to be provided by organisations to their employees to enhance correspondence. This match can then result in job satisfaction and organisational commitment, but will still be moderated by the goal achievement possibilities of the current job as compared to other jobs (in Vandenberg & Scarpello, 1990). Hesketh (1985) propounds the use of the Theory of Work Adjustment as basis for a conceptual framework of theories and a structure for integrating research findings. Such a framework must be able to describe various variables that influence career-related behaviour through the entire life span of people. The Theory of Work Adjustment can serve this purpose, since it incorporates factors similar to those found in other theories, it has a dynamic approach to career development, and it incorporates learning theory principles. Secondly, the Theory of Work Adjustment takes both the decision making capacity of the individual in career development and the role of the environment into account. Lastly, its concept of "active" and "reactive" work adjustment modes encompasses both the structural and process aspects of career development. Hesketh (1993) later compliments the Theory of Work Adjustment for its applicability to several fields of psychology and the world of work. It is also a valuable theory as it encompasses all major categories of work variables, and deals with the results of the interaction between people and environments. Consequently it can serve as a uniting factor of several fields of psychology. However, Hesketh recognises the need for the Theory of Work Adjustment to be updated in accordance with theoretical developments in other spheres of psychology, such as personality or learning theory. There also seems to be a need for uniform construct validation and operalisation of Theory of Work Adjustment constructs, such as "satisfaction" and "satisfactoriness", as there is still much unexplained variance for job satisfaction in research. In order to establish construct validity in the Theory of Work Adjustment, the measuring of its constructs against accepted constructs from other theoretical contexts, and the support of constructs through research should greatly enhance the dynamic nature of the Theory of Work Adjustment (Tenopyr, 1993).
116 Morrow (1993) suggests further ways of improving the Theory of Work Adjustment and research relating to it, because it lacks conceptual precision. This can be done for instance by employing uniform interdisciplinary definitions of concepts like "satisfactoriness". There is also a need for model explication, namely how the Theory of Work Adjustment compares to other person-job fit theories, as well as a need for the operalisation of its constructs. These insufficiencies of the Theory of Work Adjustment could explain the recent lack of popularity of the theory according to Morrow (1993). She states that there is especially a need in Theory of Work Adjustment research for the inclusion of variables other than psychological constructs, for example pay and job design, to further enhance our understanding of variances of the constructs of the Theory of Work Adjustment. Murphy (1993) echoes this need for better operalisation of the Theory of Work Adjustment's constructs and for a multidisciplinary approach towards the understanding of the constructs thereof. This need for the operalisation of the Theory of Work Adjustment's key constructs can be seen in difficulties experienced by researchers in operalisation of personality constructs and the concept of correspondence (Bizot & Goldman, 1993; Breeden, 1993). Dawis and Lofquist (1993) acknowledged that: "Murphy's perceptions are correct: we have barely scratched the surface of the problem of measuring correspondence" (Davis & Lofquist, 1993, p. 119). Research and findings from the domain of Industrial and Organisational Psychology can enhance the Theory of Work Adjustment, for example through its findings on the influence of work design and job characteristics on motivation and satisfaction, its experience in testing moderator hypotheses and its ability to clarify the measurement of the complex ability requirements of jobs. Blau (1993) expresses the need for an integrated person-environment fit model, encompassing the Theory of Work Adjustment and other personenvironment fit models. Dawis and Lofquist (1993), evaluating these comments on the Theory of Work Adjustment, respond by recognising the need for validation and operalisation of Theory of Work Adjustment constructs and for testing the interrelations of Theory of Work Adjustment constructs of abilities and needs and other findings on personality constructs. Hierarchical models of Theory of Work Adjustment constructs, consisting of a general factor, followed by
117 major group factors, minor group factors and on the lowest level specific factors could integrate these variables. Constructs of satisfaction and satisfactoriness could be represented in accordance with such a model, as illustrated in Figure 4.3 and Figure 4.4. General satisfaction
intrinsic satisfaction
extrinsic satisfaction
4,
4,
satisfaction of specific needs, eg. ability utilisation
satisfaction of specific needs, eg. working conditions
4,
4/
ability to use specific skills
conditions of a specific job
Figure 4.3:
Hierarchical representation of the satisfaction construct
general satisfactoriness
performance
conformance
4,
4/
quantity and quality of work
eg. co-worker relations
4,
4,
specific units of work
specific firm practices
Figure 4.4: Hierarchical representation of the satisfactoriness construct Bizot (1993) gives a synopsis of the conclusions from the aforementioned comments. Although the Theory of Work Adjustment can form a solid foundation for future research and expansion of our occupational knowledge, it seems that the Theory of Work Adjustment needs to be improved and refined by including research on the "big five" personality dimensions, by taking into consideration the influence of nonwork environments on satisfaction and work adjustment, by the theoretical and operational
118 integration of Theory of Work Adjustment constructs in terms of constructs in related theories, and by the refinement of measurements of correspondence.
4.2.8 Summary and evaluation This section of the chapter provided a general description of the Theory of Work Adjustment. Its applicability to the present research lies in its systematic theory on the interaction between personality and environment in the field of work, and the testable proposition on job satisfaction, which can be operationalised through the Minnesota Satisfaction Questionnaire (1967). Research relating to this theory will be discussed in Chapter Five. The Theory of Work Adjustment clearly coincides with Holland's theory (see Chapter Two), since both theories are person-environment fit models, describing the world of work, and is applicable to career guidance. The relationship between personality and environmental factors in Holland's theory is called congruence and in Lofquist and Dawis' (1969) theory is called correspondence. Satisfaction is a consequence of a positive interaction between person and environment (congruence or correspondence). The relationship between person-environment interaction and satisfaction is influenced according to both theories by reinforcement. In Holland's theory, the congruent environment rewards a person for attitudes, values, competencies and achievements and according the work adjustment theory a person is also rewarded for abilities. In Holland's theory, a social environmental factor, i.e. working with people with similar personalities and orientations, and in Lofquist and Dawis' theory, an environmental reinforcer factor play a role in establishing the match between a person and his work environment. The MSQ as measure of satisfaction of intrinsic or extrinsic needs therefore seems useful to both theories. The theory is not only useful in describing and explaining the necessary requirements for person-occupational correspondence to occur, but also provides the means for applying these principles in practice. The heuristic value of the Theory of Work Adjustment is reflected in the taxonomy of occupations provided by the Minnesota Occupational Classification System I!, since it provides the means for associating the abilities and needs of individuals with the patterns of ability requirements and reinforcers of
119 occupations. However, a taxonomy combining occupational information on ability requirements and reinforcers, with the typology of the predominant personalities in the environment such as the Dictionary of Holland Occupational Codes, would yield a more complete data basis of the requirements of different jobs. The Theory of Work Adjustment can be employed as a useful addition to Holland's theory. The process of selecting a job and remaining in it is not explained in terms of interests, but the theory accounts for the role of values, preferences and abilities in career choice, adjustment and change. In Holland's theory it is accepted that the mutual values of the social environment reinforce the commensurate values of the individual, and eventuate in job satisfaction. Dawis and Lofquist's theory goes further in describing the specific needs, based on values, that have to be met for satisfaction to occur. Moreover, not only is satisfaction recognised in this theory, but also satisfactoriness. However, it seems limited in its explanation of job satisfaction as a consequence of a match between abilities and ability requirements, or values, needs and reinforcers. One could argue that the role of other variables such as life satisfaction, or even personality disposition need to be described for a better understanding of job satisfaction. Certain facets of the correspondence-satisfaction relationship may not be sufficiently represented in this theory. For instance, variables such as organisational commitment or work goals, as proposed by Wanous (1980) should be included in a description of the requirements of job satisfaction and correspondence. The concepts of satisfaction and satisfactoriness of the Theory of Work Adjustment could also be linked to Holland's theory to further explain job choice. Whereas in Holland's theory the nature of person-environment interaction is determined by the congruence, consistency, differentiation and identity of personality types and environmental types, in Dawis and Lofquist's theory specific style dimensions of personalities and environments set the level and nature of interaction between a person and his occupation. Surmises concerning work personality adjustment style dimensions can be coupled with personality theories to improve the theory. The efficacy of the theory is enhanced by the fact that it is grounded on accepted and testable behaviouristic principles, but also recognises the role
120 of cognition as embodied in the various personality styles. The theory provides insufficient information on the development of values, abilities and interests, although one may assume that learning theory processes would be involved. As the name of this theory suggests, one of the most meritorious aspects thereof is its description and explanation of work adjustment processes. Unlike Holland's theory which recognises the fact that both people and jobs change, yet measures the person-environment interaction in terms of a static typology, the Theory of Work Adjustment acknowledges a continuous process of interaction between a person and his occupation. This holds certain practical implications for satisfactoriness and satisfaction in the work place. There is a need for employees to constantly work on developing the required skills for a particular occupation. From the employers point of view, the job description can be continuously adapted to fulfil the needs and to esteem the same values of the employees with the necessary skills. The theory is also commendable for its predictive value based on the well defined and explicit propositions and the corallaries thereof, which can be assessed by the various specific measures of each of the elements of the adjustment process.
4.3 A THEORETICAL PERSPECTIVE ON JOB INVOLVEMENT The work-related attitude of job involvement has been receiving increasing attention in literature. It is in fact a vital element of the world of work and relates to many work attitudes, behaviours and outcomes, such as job satisfaction, organisational commitment and so forth (Blau, 1987; Brooke et al., 1988; Mortimer & Lorence, 1989). An understanding of the construct of job involvement, as applied in the present study, will therefore greatly enhance our understanding of the work attitudes of South African professionals. Several dissimilar and even contradictory definitions of job involvement has however been employed in literature, but Kanungo's (1979, 1981, 1982a, 1982b) conceptualisation thereof has brought many loose ends together (Lefkowitz, Somers, & Weinberg, 1984).
121
The discussion in this section will reflect the ambiguities of the conceptualisations of the job involvement construct, and will then focus on the integrative perspective proposed by Kanungo. A brief discussion on the antecedents, correlates and consequences of job involvement, as well as its relationship to work involvement will place the construct in its right context. Lastly, since the value of a theory is reflected in the usefulness thereof, the measures that operationalise Kanungo's viewpoint will be mentioned.
4.3.1 Historical perspectives on job involvement As introduction to Kanungo's definition of job involvement, traditional conceptualisations of this construct and their shortcomings will be presented. The sociological perspective, followed by the psychological perspective on job involvement will then provide the foundation for Kanungo's formulation of the concept.
4.3.1.1
Conceptualisations of job involvement The construct of job involvement has seen different and various conceptualisations in the past few decades. Weissenberg and Gruenfeld (1968) already commented on the complexity of the definition of this construct. Early definitions of job involvement included that of Dubin (1956) who defined job involvement as the degree to which a job situation is a "central life interest" (in Blau, 1985). In early literature, the construct was used to describe diverse phenomena such as work commitment or ego involvement in the job (Weissenberg & Gruenfeld, 1968). Much research on job involvement has been governed by Lodahl and Kejner's (1965) definition of the construct. They defined job involvement as psychological identification with work, and as the importance of work in a person's total self-image. These definitions were simplified by their description of the person who is not involved in his job: "His interests are elsewhere, and the core of his self-image, the essential part of his identity, is not greatly affected by the kind of work he does or how well he does it." (Lodahl & Kejner, 1965, p. 25). Job involvement is also not a necessary result of 'happiness' with the job or job satisfaction. Preliminary interview
122 studies conducted by Lodahl and Kejner (1965) suggested that job involvement was a distinct job attitude, relatively stable over time, and independent of organisational changes, but related to social proximity of coworkers. As a result of these findings, and in order to establish a more reliable method of research on job involvement, they operationalised their definition with a twenty-item job involvement scale. This scale measured job involvement as a multidimensional attitude (at least three dimensions) with reported adequate reliability and discriminant validity for different groups. Controversies surrounding the dimensionality of the job involvement construct are one of the reasons for the different conceptualisations thereof. Whereas Lodahl and Kejner (1965) recognise both a cognitive and affective component of job involvement, Kanungo (1981, 1982b) postulates that job involvement is a unidimensional construct. Saleh (1981) portrays job involvement as being multidimensional with structural components of cognitive, evaluative and behavioural intentions. According to Blau (1985) the three principal historical conceptualisations of job involvement defines it as: the degree to which a person is participating in his job and has needs, such as prestige and autonomy, satisfied; the degree to which the job is a central life interest; or the degree to which a person's self-esteem is affected by his perceived performance level. Kanungo (1981) identifies several problems with prior conceptualisations of job involvement in research. There has been no integrative approach to understanding job involvement, the concepts of alienation and involvement have been used ambiguously, it has sometimes been confounded with the construct of intrinsic motivation and no distinction has been made between work involvement and job involvement. Moreover, the causes and effects of job involvement have often been confused, and no distinction has been made whether the causation was contemporaneous or historical. Kanungo (1982b) categorises the development of the concept of job involvement according to the sociological and psychological theoretical perspectives, and
123 presented in response to these, the integrative motivational approach to job involvement. In order to understand the development of the construct proposed by Kanungo (1981, 1982b), it is necessary to briefly review the sociological and psychological conceptualisations of alienation or involvement.
4.3.1.2
The sociological perspective of alienation
Kanungo (1981, 1982b) gives an overview of the characteristics of the sociological perspectives on alienation, according to the observations made by Marx, Weber, Durkheim and others. The sociological approach deals with the concept of work alienation rather than work involvement, and describes it in terms of social groups and not individuals. It does not consider behavioural aspects of alienation, but rather examines the byproducts thereof, such as loneliness, isolation and so forth. According to this approach, problems of work alienation can be solved by increasing individual autonomy, control and power over the work environment (extrinsic factors). Alienation is caused by social influence, the nature of the technology and social organisation of a work environment and the frustration of workers' needs and expectations. The construct is used in several different contexts with various conceptual meanings, including that of powerlessness, meaninglessness, formlessness, isolation or self-estrangement.
4.3.1.3
The psychological perspective on involvement
As opposed to the sociological approach, psychologists traditionally focus more on involvement at work than on the construct of job alienation. Kanungo (1981, 1982b) criticises the psychological approach for its ambiguities and discrepancies in defining job involvement and the causes, correlates and effects thereof.
124 The definitions of job involvement in this regard include cognitive, affective and behavioural components. By way of illustration, involvement has been defined as: a performance—self-esteem continguency (Rabinowitz and Hall); a status-seeking motive (Allport); psychological identification with work (Lodahl and Kejner); work commitment behaviour (Wiener and Gechma); the source of a sense of personal worth (Siegel); reinforcement of the self-concept (!!room). Job involvement as 'performance—self-esteem continguency' is the result of need satisfaction. According to this motivational approach, job involvement is dependant on several external conditions, including extrinsic-need satisfaction, decision making opportunity and the fulfilment of achievement-, belonging- and self-esteem needs. In contrast to this, job involvement in terms of 'self-image' is shaped according to early socialisation, personal background, self-evaluation and norms. When these definitions are combined, job involvement can be seen as a function of intrinsic- and extrinsic-need satisfaction based on past socialisation and internalised work values (Kanungo, 1982b). According to this paradigm, job involvement entails the perception of work as being intrinsically rewarding, providing the opportunity for self-expression (Kanungo, Misra & Dayal, 1975). Correlates of job involvement from the psychological perspective include personal characteristics (age, gender, work values, extrinsic/intrinsic-need strength, locus of control), situational variables (job characteristics and outcomes), organisational variables and sociocultural factors. Job involvement results in job satisfaction, positive effort expenditure and performance, as well as lower levels of staff turnover and absenteeism (Kanungo, 1982b). An overview of research results by Kanungo (1982b) confirms positive correlations between intrinsic-need satisfaction and job involvement, but ambiguity in terms of the relationship between extrinsic-need satisfaction (such as salary, supervision and benefits) and job involvement. The ambiguities of the psychological perspective are also evident in the different views on the dimensionality, stability and determination of job involvement.
125 A valuable aspect of this perspective is its premises on the positive relationships between intrinsic-need satisfaction and job involvement, and in the absence of need satisfaction opportunities that result in work alienation.
4.3.2 Kanungo's motivational perspective Since the psychological and sociological approaches seem to have conceptual ambiguities and methodological inadequacies, Kanungo (1982b) formulated a new integrative conceptual framework for the work and job alienation and involvement concepts. The following discussion of this new approach will mainly focus on aspects of job involvement. Firstly the suppositions on which Kanungo's definition rest will be established; secondly, the cognitive definition of job involvment will be furnished; and lastly, Kanungo's model depicting the significance of need satisfaction in the job involvement construct will be unfolded.
4.3.2.1
Important premises of the motivational approach This motivational approach constitutes an integration of psychological and sociological conceptualisations of involvement-alienation. The bipolar constructs of involvement and alienation are opposite poles of the same psychological dimension. The involvement-alienation phenomenon is a cognitive belief state in workers and is therefore different to job satisfaction which is an affective state. (Job satisfaction refers both to intrinsic- and extrinsicneed satisfaction.) It is also distinct from overt behaviour such as worker participation. -
Involvement-alienation is different to its causal factors (early socialisation and contemporary working contexts) and effects (behaviours and attitudes).
126 Involvement-alienation is different to the intrinsic motivation of workers, and involvement-alienation is not dependent on intrinsic-need satisfaction. Distinctions should also be made between job, organisational, occupational and work alienation-involvement. Job alienationinvolvement for example relates to a specific job and work alienationinvolvement relates to work in general (Kanungo, 1982b).
4.3.2.2
Kanungo's definition of job involvement
Considering the preceding premises of the motivational perspective on job involvement, Kanungo (1979; 1982b) defines work and job involvement as follows: Work involvement is "a generalized cognitive (or belief) state of psychological identification with work" (Kanungo, 1979, p.131), (italics added). This applies "insofar as work is perceived to have the potential to satisfy one's salient needs and expectations" (Kanungo, 1982b, p. 79). As seen in Chapter One, job involvement on the other hand is "a specific cognitive belief state of psychological identification ... with work" (Kanungo, 1982b, p. 70) (ellipse and italics added). The focus on cognitive, rather than affective and behavioural aspects is evident in this conceptualisation of job involvement as "psychological closeness" to one's work. Since job involvement is regarded as one pole in the single continuum of involvement-alienation, alienation can be defined as a state of psychological separation from work (Kanungo, 1982b). Taken to the extreme, job involvement can entail workaholism or on the other hand complete work alienation. Workaholics' salient needs may be based on obsessive fears of failure and boredom or perfectionism, and the needs of the work alienated ndividual are not satisfied by work (Roodt, Bester & Boshoff, 1994).
127 4.3.2.3
The need satisfaction model
From the discussion of the definitions of job and work involvement, it is apparent that perceptions of need satisfaction opportunities play an important role in a person's psychological identification with work. Indeed, Kanungo (1979) states that a person will feel involved in his job, if his most salient or prominent needs are satisfied. Job involvement does not depend on whether the salient needs are higher order ("growth") or lower order needs. This notion has been derived from expectancy theories of motivation, which hold that job satisfaction is dependent both on the presence or absence of certain job factors, and a person's cognitive value system in terms of the job factors (Kanungo et al., 1975). The causes, definitions and effects of job involvement-alienation according to this motivational perspective are depicted in Figure 4.5. SOCIALISATION PROCESS: CULTURAL, ORGANISATIONAL AND GROUP NORMS NEED SALIENCY INSTRUMENTAL BEHAVIOUR AND ATTITUDES 0.60). Support for null hypotheses Ho2.1, Ho2.3, Ho2.4 and alternative hypothesis HA2.2 is found, but hypotheses Ho2.2 and HA2.1, HA2.3, and HA2.4 are rejected (see point 6.5.1).
Table 7. General linear model of the dependent variable intrinsic job satisfaction and the independent variables person-environment congruence and job involvement
Source PECi JI ERROR .
SS
clf
MS
F
P
0.262
1
0.262
0.008
0.929
602.621
1
602.621
18.377
0.000
5082.821
155
32.792
Significant at the 1% level of significance
The highly significant effect of job involvement on intrinsic job satisfaction (p < 0.000) once the effects of the possible covariates occupational group classification and gender are controlled can be seen in the findings submitted in Table 7.8. Person-environment congruence of the entire sample (N = 158) certainly does not have an effect on job satisfaction according to this linear model (p > 0.90).
**
206
These findings reinforce the support of the alternative hypothesis HA2.2 and the rejection of the null hypothesis Ho2.1 (see point 6.5.1).
Table 7.9 General linear model of the dependent variable intrinsic job satisfaction and the independent variable job involvement Source JI ERROR .
SS
MS
clg
603.525
1
603.525
5083.083
156
32.584
F 18.522
P 0.000
Significant at the 1% level of significance
When considering the solitary effect of job involvement as source of variance to intrinsic job satisfaction, once again the results show the high significance of the influence of job involvement in this linear model (p < 0.000). Thus the data is consistent with the alternative hypothesis HA2.2 and therefore hypothesis Ho2.2 is rejected (see point 6.5.1). For the participants who obtained an alternative or second personality code, data for the various sources of variation of the variable of intrinsic job satisfaction is presented in Tables 7.10 and 7.11. Once again for this linear model only job involvement has a significant effect (p < 0.000) on intrinsic job satisfaction. The effects of person-environment congruence according to the second personality code, occupational group classification, gender, and the interaction of occupational group and gender were also once again nonsignificant (p > 0.45). The support for the null hypotheses Ho2.1, Ho2.3, Ho2.4 and alternative hypothesis HA2.2, and the rejection of hypotheses Ho2.2 and HA2.1, HA2.3,
"
207
and HA2.4 are therefore also valid for the congruence scores based on the second personality types (see point 6.5.1).
Table 7.10 General linear model of the dependent variable intrinsic job satisfaction and the independent variables combined (2) Source
SS
df
MS
IF
P
18.508
1
18.508
0.559
538.888
1
538.888
16.262
OG
1.783
1
1.783
0.054
0.817
G
3.462
1
3.462
0.104
0.747
OG + G
10.537
1
10.537
0.318
0.674
ERROR
5036.896
152
33.137
PEC2 JI
.
0.456 0.000 **
Significant at the 1% level of significance
fable 7.11 General linear model of the dependent variable intrinsic job satisfaction and the independent variables person-environment congruence (2) and job involvement
Source PEC2 JI ERROR ..*
SS
df
MS
F
18.427
1
18.427
0.564
573.813
1
573.813
17.561
5064.655
155
32.675
Significant at the 1% level of significance
P 0.454 0.000 **
208 The results found in Table 7.10 in terms of the lack of significance of the influence of the alternative person-environment congruence scores, and the high significance of job involvement scores remain the same when the effect of the other possible covariates on intrinsic job satisfaction is controlled, as seen in Table 7.11. Correspondingly hypothesis Ho2.1 find support, but hypothesis Ho2.2 does not (see point 6.5.1). In general, these findings on the linear models of the dependent variable intrinsic job satisfaction indicate the highly significant effect of the independent variable job involvement, but nonsignificant influences of person-environment congruence alone, or jointly with the occupational group classification and gender.
7.2.3 Extrinsic job satisfaction as dependent variable General linear models were also created for the dependent variable extrinsic job satisfaction, of which the results are presented in Tables 7.12 through 7.15. Table 7.12 General linear model of the dependent variable extrinsic job satisfaction and the independent variables combined (1) Source PECi
SS
df
MS
F
P
140.588
1
140.588
4.944
0.028
JI
54.270
1
54.270
1.908
0.169
OG
27.308
1
27.308
0.960
0.329
G
11.168
1
11.168
0.393
0.532
4350.916
153
28.437
ERROR .
Significant at the 5% level of significance
*
209 Table 7.12 portrays the significance or nonsignificance of the effects of person-environment congruence, job involvement, sample and gender on extrinsic job satisfaction. The effect of person-environment congruence on extrinsic job satisfaction is significant (p = 0.028) at the 5% level of significance, and that of the other mentioned variables are nonsignificant (ps > 0.10). As opposed to the significant effects found for job involvement on overall job satisfaction (see Tables 7.1, 7.2, 7.4) and on intrinsic job satisfaction (see Tables 7.7, 7.8, 7.9, 7.10, 7.11), the effect of job involvement on extrinsic job satisfaction is nonsignificant (p = 0.169). Moreover, the cited effect of person-environment congruence on extrinsic job satisfaction was not found for either overall job satisfaction or intrinsic job satisfaction. The null hypotheses that are retained are hypotheses Ho3.2, Ho3.3, and Ho3.4, but hypothesis Ho3.1 is rejected (see point 6.5.1).
TabOe 7.13 General linear model of the dependent variable extrinsic job satisfaction and the independent variable person-environment congruence (1) Source PECi ERROR *
SS
dif
MS
121.456
1
121.456
4443.152
156
28.482
IF
4.264
P
0.041
Significant at the 5% level of significance
The data in Table 7.13 once again reflect the moderate significance of the effect of person-environment congruence (p = 0.041) on extrinsic job satisfaction in the absence of possible effects of job involvement, occupational group classification and gender. The retention of the alternative hypothesis HA3.1 with the rejection of the null hypothesis Ho3.1 is thus justifiable (see point 6.5.1).
*
210
Table 7.14 General linear model of the dependent variable extrinsic job satisfaction and the independent variables combined (2) Source
SS
df
MS
F
P
131.557
1
131.557
4.617
0.033 *
JI
40.175
1
40.175
1.410
0.237
OG
23.505
1
23.505
0.825
0.365
8.710
1
8.710
0.306
0.581
4359.947
153
28.496
PEC2
G ERROR *
Significant at the 5% level of significance
When the analyses in terms of the sources of variance of the dependent variable extrinsic job satisfaction were duplicated for the alternative personenvironment congruence variable based on second personality codes of some participants, the findings presented in Table 7.14 were obtained. As was the case with the initial analyses, person-environment congruence proved to be the only variable that belongs in the linear model with extrinsic job satisfaction as dependent variable. The effect of this alternative source of person-environment congruence was significant at the 5% level of significance (p = 0.033). Job involvement, occupational group and gender had nonsignificant effects (p > 0.20) on extrinsic job satisfaction. These results are consistent with the preceding results that led to the retention of the null hypotheses Ho3.2, Ho3.3, and Ho3.4 and the rejection of hypothesis Ho3.1 (see point 6.5.1).
211 Table 7.15 General linear model of the dependent variable extrinsic job satisfaction and the independent variable person-environment congruence (2)
Source PEC2 ERROR *
SS
df
MS
131.956
1
131.956
4432.651
156
28.414
F 4.644
P 0.033
Significant at the 5% level of significance
Viewed singularly, the effect of the alternative person-environment congruence variable on extrinsic job satisfaction remain significant (p = 0.033). These findings once again gives support to the alternative hypothesis HA3.1.
To summarise, results in Table 7.12 through 7.15 suggest that unlike the combined influence of job involvement, occupational group and gender that have nonsignificant effects on extrinsic job satisfaction, the independent variable person-environment congruence has a significant effect on extrinsic job satisfaction at the 5% level of significance.
7.2.4 Job involvement as dependent variable In the following tables data is presented on the general linear models of the dependent variable job involvement. The specified independent variables in these models are once again person-environment congruence, job satisfaction, as well as intrinsic and extrinsic job satisfaction, occupational group and gender. As was the case with the preceding findings, the F-ratios and P-values will be used to analyse the significance of the effect of the independent variables on job involvement.
*
212 Table 7.16 General linear model of the dependent variable job involvement and the independent variables combined (1) Source
SS
dff
MS
F
P
76.153
1
76.153
0.288
0.592
JS
2765.072
1
2765.072
10.459
0.001
OG
3696.888
1
3696.888
13.984
0.000 '
712.356
1
712.356
2.694
40449.29
153
264.374
PECi
G ERROR .
"
0.103
Significant at the 1% level of significance
Table 7.17 General linear model of the dependent variable job involvement and the independent variables combined (2) Source PEC2
SS
off
MS
F
P
50.587
1
50.587
0.191
0.663
JS
2533.624
1
2533.624
9.577
0.002
OG
3413.901
1
3413.901
12.905
737.878
1
737.878
2.789
40474.86
153
264.542
G ERROR .
"
0.000 " 0.097
Significant at the 1% level of significance
The F-ratio's and p-values of the general linear models in Table 7.16 indicate that overall job satisfaction (p = 0.001) and occupational group
213 (p = 0.000) respectively have highly significant effects on job involvement. On the other hand, the joint influence of person-environment congruence and gender is nonsignificant (p > 0.10).
The implication of these results is that the null hypotheses Ho4.1 and Ho4.4 are retained and hypotheses Ho4.2 and Ho4.3 are rejected (see point 6.5.1). In as far as the alternative person-environment congruence scores are concerned, findings in Table 7.17 hold the same implications in terms of the support or lack of support of the hypotheses regarding job involvement, since job satisfaction (p = 0.002) and the occupational group variable (p = 0.000) significantly affect job involvement, but person-environment congruence and gender do not. Specifically these findings show that both the null hypotheses Ho4.1 and Ho4.4 are retained, which implies that hypotheses Ho4.2 and Ho4.3 are rejected (see point 6.5.1).
Table 7.1 General linear model of the dependent variable job involvement and the independent variables job satisfaction and occupational group Source
SS
off
iViS
F
P
JS
2985.075
1
2985.075
11.214
0.001
"
OG
3117.247
1
3117.247
11.710
0.001
"
ERROR
41261.55
155
266.204
.
Significant at the 1% level of significance
Since differences in the level of significance of the effect of job satisfaction on job involvement, when including either of the person-environment congruence variables in the linear model, were evident in the statistical
214 findings presented in Tables 7.16 and 7.17, the separate effect of only job satisfaction and occupational group together were analysed. The findings thereof are presented in Table 7.18. From these results it is apparent that job involvement does indeed vary as a function of job satisfaction in conjunction with the specific occupational group, and that the effect of these variables on job involvement is significant at the 1% level of significance (p = 0.001). The alternative hypotheses HA4.2 and HA4.3 are supported (see point 6.5.1). As a result of the retention of hypotheses HA4.2 and HA4.3, a generalised linear model of regression was compiled to determine which sample or occupational group had a more significant effect along with overall job satisfaction on job involvement. Table 7.19 presents the parameter estimates on which the calculations of this model is based.
Table 7.19 Parameter estimates of the dependent variable job involvement in terms of occupational group differences and overall job satisfaction Parameter JI Intercept Sample of psychologists Sample of optometrists
Std Error
t
P
16.760
10.536
1.591
0.114
9.911
2.896
3.422
0.001
-
-
-
0.133
3.349
0.001
0* 0.455
JS *
f3
This parameter was set to zero, as it is redundant.
The statistics of this generalised linear model from the data documented in Table 7.19 was calculated as follows: y =1.14- a D 4- 13 x
215 (The symbol p. refers to the value of the intercept on the y-axis. The symbols a and (3 stand for the values on the y-axis (job involvement) of the two samples respectively. A random score of D = 1 has been assigned to the sample of psychologists and of D = 2 to the sample of optometrists). In other words, for the sample of psychologists: y = 16.760 + 9.911 + 0.445 (Overall job satisfaction) y = 26.671 + 0.455 (Overall job satisfaction); and for the sample of optometrists: y = 16.760 + 0.455 (Overall job satisfaction). These results indicate that although the combined effect of overall job satisfaction and occupational group on job involvement applies both to psychologists and optometrists, the effect is significantly higher for psychologists (Intercept = 26.671) than for optometrists (Intercept = 16.760 ). TabOe 7.20 General linear model of the dependent variable job involvement and the independent variables person-environment congruence, intrinsic job satisfaction, extrinsic job satisfaction, occupational group and gender Source PECi
SS
df
RAS
IF
P
4.248
1
4.248
0.017
0.898
3773.174
1
3773.174
14.740
0.000
EJS
87.018
1
87.018
0.340
0.561
OG
3084.511
1
3084.511
12.050
0.001
667.221
1
667.221
2.607
0.108
38908.81
152
255.979
IJS
G ERROR .
Significant at the 1% level of significance
**
**
216
According to the findings presented in Table 7.20 intrinsic job satisfaction (p < 0.000) and occupational group classification (p = 0.001) have highly significant effects on job involvement. Variables of person-environment congruence, extrinsic job satisfaction and gender however do not seem to belong in this linear model, as they have nonsignificant effects on job involvement (p > 0.10). It is interesting to note that job involvement does not seem to vary as a function of extrinsic job satisfaction, but of overall job satisfaction (see Tables 7.16, 7.17, 7.18) and of intrinsic job satisfaction only. These results support the null hypotheses Ho5.1, Ho5.3, Ho5.5 and alternative hypotheses HA5.2 and HA5.4 (see point 6.5.1). Since these results once again indicate that job involvement is a function of intrinsic job satisfaction and occupational group jointly, it was once again meaningful to determine which of the two samples had a more significant effect on job involvement. Table 7.21 presents the parameter estimates that were used to compute the relative influences of the two samples.
Table 7.21 Parameter estimates of the dependent variables job involvement in terms of occupational group differences and intrinsic job satisfaction Parameter JI Intercept Sample of psychologists Sample of optometrists IJS
0
Std Error
t
P
- 1.135
12.586
- 0.090
0.928
9.357
2.845
3.289
0.001
-
-
-
0.213
4.226
0.000
0* 0.898
This parameter was set to zero, as it is redundant.
217 From this generalised linear model of regression it is evident that for the sample of psychologists: y = 8.222 + 0.898 (Intrinsic job satisfaction); and for the sample of optometrists: y = -1.135 + 0.898 (Intrinsic job satisfaction). These results indicate that although the joint effect of intrinsic job satisfaction and the occupational group on job involvement applies both to psychologists and optometrists, the effect is significantly higher for psychologists (Intercept = 8.222) than for optometrists (Intercept = - 1.135).
Table 7.22 General linear model of the dependent variable job involvement and the independent variables intrinsic job satisfaction and occupational group Source
SS
clf
MS
F
P
IJS
4571.251
1
4571.251
17.859
0.000
OG
2769.239
1
2769.239
10.819
0.001
ERROR
39675.38
155
255.970
..
Significant at the 1% level of significance
Findings in Table 7.22 echo the findings and implications of Table 7.20. Since intrinsic job satisfaction (p < 0.000) and occupational group or sample (p = 0.001) significantly affect the job involvement scores when variables of person-environment congruence, extrinsic job satisfaction and gender are omitted from the general linear model of the dependent variable job involvement, one may conclude that the effect of intrinsic job satisfaction and
"
218 occupation group on job involvement is indeed not due to any furtherance provided by the other variables. Alternative hypotheses HA5.2 and HA5.4 thus find support contrary to the relating null hypotheses (see point 6.5.1).
On the whole, when these results on the independent variables of job involvement are examined, neither person-environment congruence and gender together, nor person-environment congruence, gender and extrinsic job satisfaction together have any significant effect on job involvement. However, the joint influence of the specific sample (occupational group classification) and either overall job satisfaction or intrinsic job satisfaction is highly significant. The practical implications of these findings will be discussed in Chapter Eight.
7.2.5 A direct comparison in terms of gender and occupational group Since no really significant influences of gender were found in the results, and only two significant influences of the occupational group was found (relating to the job satisfaction job involvement, and intrinsic job satisfactionjob involvement relationships) it was decided to compare the genders and two occupational groups in terms of job involvement, overall job satisfaction, intrinsic job satisfaction and extrinsic job satisfaction by means of t-tests for independent samples. Table 7.23 presents the group statistics of the four variables and the results of the t-test for the samples of psychologists and optometrists in terms of these variables are indicate in Table 7.24. In terms of possible differences between males and females with regards to these variables, Table 7.25 gives the group statistics in as far as gender is concerned and the findings of the relating t-test can be found in Table 7.26.
219 Table 7.23 Descriptive statistics of the variables in terms of the different samples Variable JI
JS
IJS
EJS
Sample
.
N
Mean
sd
sem
Psychologists
114
61.17
16.08
1.51
Optometrists
44
51.07
18.70
2.82
Psychologists
114
77.54
9.95
0.93
Optometrists
44
77.11
9.51
1.43
Psychologists
114
58.94
5.93
0.56
Optometrists
44
58.11
6.27
0.95
Psychologists
114
18.60
5.42
0.51
Optometrists
44
19.00
5.36
0.81
sd: Standard deviation sem: Standard error mean
Table 7.24 Results of the t-test of the equality of means of the variables for the two occupational groups Variable
t
df
P
and
JI
3.379
156
0.001**
JS
0.242
156
0.809
0.42
IJS
0.771
156
0.442
0.82
- 0.421
156
0.675
- 0.40
EJS t: df: P: md:
t-value Degrees of freedom Two tailed significance Mean differences
10.10
220 It is quite clear from Table 7.24 that there are significant differences between the mean job involvement scores of psychologists and optometrists at the 1% level of significance. There were no significant differences in terms overall job satisfaction, intrinsic job satisfaction and extrinsic job satisfaction between the psychologists and optometrist. The results (mean scores) in Table 7.23 show that it is the psychologists in particular that have higher levels of job involvement than optometrists. The results on Table 7.25 and 7.26 indicate no significant differences between the mean scores of males and females in terms of job involvement, overall job satisfaction, intrinsic job satisfaction, and extrinsic job satisfaction. The results of both t-tests are consistent with the findings of the general linear models, namely that gender did not exert a significant influence in any of the relations, and these results therefore reflect the accuracy of the support of the relevant hypotheses.
Table 7.25 Descriptive statistics of the variables in terms of gender Variable JI
Sample Male Female
JS
Male Female
IJS
Male Female
EJS
Male Female
sd: Standard deviation sem: Standard error mean
N
Mean
scl
sem
58
60.43
16.62
2.18
100
57.15
17.79
1.78
58
78.50
10.16
1.33
100
76.79
9.59
0.96
58
59.33
6.01
0.79
100
58.35
6.02
0.60
58
19.17
5.70
0.75
100
18.44
5.22
0.52
221 Table 7.26 Results of the t-test of the equality of means of the variables for the two sexes !Variable
t
df
P
rind
JI
1.144
156
0.254
3.28
JS
1.057
156
0.292
1.71
IJS
0.984
156
0.327
0.98
EJS
0.822
156
0.412
0.73
t: df: P: md:
t-value Degrees of freedom Two tailed significance Mean differences
7.3 RESULTS OF THE CORRELATIONAL RESEARCH DESIGN In order to either validly support or reject the hypotheses that state that the relationships between person-environment congruence and each overall job satisfaction, intrinsic job satisfaction and extrinsic job satisfaction are mediated by the level of job involvement of the participants, Pearson Product Moment Correlation coefficients were calculated between each of the following variables:
Person-environment congruence 1;* Person-environment congruence 2;* Job satisfaction; Intrinsic job satisfaction; Extrinsic job satisfaction; and Job involvement. * See note under point 7.1
222
Initial analyses involved the correlations of the variables of the study for the combined sample of psychologists and optometrists, and for groups with high levels of job involvement and low levels of job involvement separately. Focus was placed on the degrees of congruence-satisfaction, congruenceintrinsic satisfaction, and congruence-extrinsic satisfaction, and these relationships were then compared in terms of the two different levels of job involvement. Results of these correlation analyses are presented in Tables 7.27, 7.28, and 7.29. Table 7.30 combines the findings from these correlations that are relevant to hypotheses WS, HA6, Ho7, HA7, Ho8, and HA8 (see point 6.5.2).
7.3.1 Correlations of the variables for the entire sample The first correlational analyses encompassed variables of personenvironment congruence, job satisfaction, intrinsic job satisfaction, extrinsic job satisfaction and job involvement for the entire sample of participants. Although these results did not relate directly to the hypotheses of the study, certain interesting findings could be made, and general trends of the interrelations of the variables could be observed to guide the rest of the research.
Table 7.27 reveals two-tailed significant relations between the two sets of values of person-environment congruence (r = 0.899); between overall job satisfaction and each intrinsic job satisfaction (r = 0.876) and extrinsic job satisfaction (r = 0.841); and between intrinsic job satisfaction and job involvement (r = 0.326) at the 1% level of significance. Further significant negative correlations at the 5% level of significance were found between each of the person-environment congruence variables and extrinsic job satisfaction (r = - 0.163 and r = - 0.170 respectively). These findings in Table 7.27 also already indicate the lack of significance of congruencesatisfaction relationships for this sample.
223 Table 7.27 Pearson correlations of the variables person-environment congruence, overall job satisfaction, extrinsic job satisfaction, intrinsic job satisfaction and job involvement (N = 158) PECi
PEC2
PIECi
1.000
0.899**
- 0.099
- 0.014
- 0.163 *
- 0.023
PEC2
0.889**
1.000
- 0.150
- 0.092
- 0.170 *
- 0.109
JS
LOS
EJS
JO
JS
- 0.099
- 0.150
1.000
0.875 "
0.841 '
0.256 "
US
- 0.014
- 0.092
0.875 **
1.000
0.475 **
0.326 "
EJS
- 0.163 *
- 0.170 *
0.841 "
0.475 **
1.000
0.101
- 0.023
- 0.109
0.256 "
0.326 "
0.101
1.000
JO * .
k
Significant at the 5% level of significance (two-sided) Significant at the 1% level of significance (two-sided)
One can infer from these results that the high correlation between the two indications of person-environment congruence suggests the relatedness of the two personality codes obtained by some of the participants. This coincides with the relative similarities found for linear models employing the two different sets of person-environment scores in the preceding factorial design. Since intrinsic, extrinsic and overall satisfaction are three factors measured by the same instrument, viz. the Minnesota Satisfaction Questionnaire the high degree of correlations among these variables are understandable. The significant correlation found between intrinsic job satisfaction and job involvement echoes the evidence emanating from the general linear models that indicate the significant effect of intrinsic job satisfaction on job involvement (see Table 7.20 and 7.22). The fact that there is a moderately significant negative correlation between person-environment congruence and extrinsic job satisfaction implies that
_
224 person-environment congruence increases as extrinsic job satisfaction decreases and vice versa for this sample of psychologists and optometrists. As said before, these findings do not have direct bearing on the present study's hypotheses. The correlations of all the variables for the group with low levels of job involvement and the group with high levels of job involvement will now be examined.
7.3.2 Correlations of the variables for high and low job involvement groups As seen in the description of the research method, the frequency distributions of the job involvement scores were examined on a percentile scale. Participants with job involvement scores within the first 25 percentiles were assigned to the low job involvement group, and those with score equal to or greater than the seventy fifth percentile were assigned to the high job involvement group. Correlations of the variables were then calculated for each of these groups separately, as can be seen in Tables 7.28 and 7.29. Tab0e 7.28 Pearson correlations of the variables person-environment congruence, overall job satisfaction, extrinsic job satisfaction, intrinsic job satisfaction and job involvement for the group with low levels of job involvement (n = 42)
PECi
PEC2
PECi
1.000
0.985"
IPEC2
0.965"
JS
JS
LDS
0.035
0.080
- 0.027
- 0.125
1.000
- 0.005
0.044
- 0.062
- 0.054
0.035
- 0.005
1.000
0.900 "
0.861 "
- 0.064
US
0.080
0.044
0.900 "
1.000
0.554 "
- 0.031
EJS
- 0.027
- 0.062
0.861 "
0.554 "
1.000
- 0.067
Ji
-0.125
-0.064
- 0.256 "
-0.067
1.000
-0.031
Significant at the 1% level of significance (two-sided)
EJS
JO
225
The findings presented in Table 7.28 show that, for the group with low levels of job involvement, job involvement does not correlate significantly with any of the other variables. In fact, significant correlations at the 1% level of significance were found between the two variables of person-environment congruence (r = 0.985) and the three job satisfaction factors (r = 0.900; r = 0.861; and r = 0.664 respectively) of the Minnesota Satisfaction Questionnaire only.
Table 7.29 Pearson correlations of the variables person-environment congruence, overall job satisfaction, extrinsic job satisfaction, intrinsic job satisfaction and job involvement for the group with high levels of job involvement (n = 44)
PEG'
PEC2
PECi
1.000
0.916
PEC2
0.916**
JS
IJS
EJS
JI
- 0.123
0.102
- 0.322 *
0.030
1.000
- 0.131
0.087
- 0.320 *
- 0.051
1.000
0.837 **
0.802 **
0.125
JS
- 0.123
- 0.131
LDS
0.102
0.087
0.837 "*
1.000
0.344 *
0.184
EJS
- 0.322 *
0.320 *
0.802 "
0.344 *
1.000
0.013
0.125
0.184
0.013
1.000
JO * **
0.030
- 0.051
Significant at the 5% level of significance (two-sided) Significant at the 1% level of significance (two-sided)
Results presented in Table 7.29 indicate that, for the group with high levels of job involvement, there were also no significant relations between job involvement and the other variables. Once again there were significant high positive correlations between the two indications of person-environment congruence (r = 0.916) and each of the job satisfaction factors (r = 0.837; r = 0.802; and r = 0.344 respectively). A positive significant correlation at
226 the 5% level of significance can be seen between the second personenvironment congruence variable and extrinsic job satisfaction. These findings in themselves also have no direct implications for the hypotheses of this study. In the following table however, some of these correlations will be compared in terms of the level of job involvement.
Table 7.30 Two-sided significance of the differences between the high and low job involvement groups in terms of the congruence-satisfaction relationships CORRELATOON
LOW JI
HIGH JO
P-VALUE
(n = 44)
(n = 42)
(two-sided)
PECi / JS
0.035
- 0.123
0.4802
PEC2/ JS
- 0.005
- 0.131
0.5725
PEC, / IJS
0.080
0.102
0.9212
PEC2/ IJS
0.044
0.087
0.8474
PECi / EJS
- 0.027
- 0.322
0.1738
PEC2/ EJS
- 0.062
- 0.320
0.2316
No significant differences were found when the correlations between personenvironment congruence and overall job satisfaction, person-environment congruence and intrinsic job satisfaction, and person-environment congruence and extrinsic job satisfaction were compared in terms of the level of job involvement, as can be seen in Table 7.30. This applies for both variables of person-environment congruence. The null hypotheses HoS, Ho7, and Ho8 are all supported by these findings and the alternative hypotheses HA6, HA7, and HA8 are rejected (see point 6.5.2).
227 In summary, the pertinent findings of the correlational research design is illustrative of the nonsignificance of the correlations between job involvement and the other variables, as well as the two person-environment congruence variables and the three job satisfaction dimensions. Moreover, the relationships between person-environment congruence and each overall job satisfaction, intrinsic job satisfaction and extrinsic job satisfaction is not moderated by the level of job involvement.
7.4 RESULTS OF THE FREQUENCY DISTRIBUTIONS AND DESCRIPTIVE STATISTICS Considering the frequencies of the first letter of the personality codes of the participants as well as the distribution of their three-letter codes, a comparison of the resultant environmental codes with the accepted South African codes could be made for these populations. The following data was obtained and conclusions drawn. Firstly, findings on the environmental code for the group of psychologists combined are presented, followed by findings for the different categories of registration of psychologists are highlighted. Secondly, the data reflecting the environmental code of South African optometrists are presented. In the tables reflecting the modes and medians of the participants' personality codes, the acronym RIASEC was encoded as follows to facilitate calculations: R= 1;I=2;A= 3;5=4;E= 5;C= 6.
7.4.1 Environmental codes of psychologists Figure 7.1 depicts the frequency distribution of each of the personality types for the entire psychology sample relative to the first letters of personality codes only. This is followed by Table 7.31 that indicates the mode and median of the frequency distributions of all three letters of the personality codes in identical sequences, as well as a discussion of the results. Thereafter findings on the first letters of personality codes of the various categories of registration is presented in Figure 7.2 and the mode and median of the three-letter codes of these categories can be found in Table 7.32, followed by a discussion of these results.
228
80 70 60 50 40
—0— Psychologists
30 20
•
10 0 R
I
A
S
E
C
Figure 7.1: Frequency distribution of the first letter personality codes of psychologists (N =114)
Table 7.31 Medians and modes of the three-letter personality codes of the psychologists (N = 114) PERSON A2 LOTY CODE
MEDOAN
MODE
CODE 1
435.50
435
CODE 2
452.00
453
From Figure 7.1 it is apparent that the sample of psychologists has an environmental code of SCA according to the frequencies of the participants' first letter codes. The mode of their three-letter codes presented in Table 7.31 indicate a personality code of SAC for psychologists, or of SCA for the second personality codes. This clearly does not coincide with the ISE environmental code for psychologists provided in The South African Dictionary of Occupations (1987). A comparison of the SCA and SAC codes
229 with the ISE code by means of the Congruence Index yield congruence scores of 6 and 9 respectively. In the range of 0 to 18 these scores can be understood to indicate low congruence. Hypothesis Ho9.1 is therefore rejected and the alternative hypothesis HA9.1 is retained (see point 6.5.3).
Table 7.32 Medians and modes of the personality codes of the psychologists in the various categories of registration PERSONALITY CODE
'.'EDOAN
MODE
CLINICAL
CODE 1
452.00
435
(n = 46)
CODE 2
452.50
453
COUNSELLING
CODE 1
435.00
435
(n = 31)
CODE 2
436.00
453
EDUCATIONAL
CODE 1
425.00
423
(n = 21)
CODE 2
425.00
425
INDUSTRIAL
CODE 1
489.50
546
(n = 20)
CODE 2
527.50
543*
*
Multiple modes exist.
Table 7.32 summarises the medians and modes obtained of the first and alternative environment codes of the clinical, counselling, educational and industrial psychology fields. These findings refer to the three-letter codes representing the personality patterns of the members of each of the environments. According to these findings, both the clinical and counselling environmental codes are either SAE or SEA, the educational environmental code is SAI or SAE. The industrial environmental code is either ESC or ESA, although it may also be SEA or SEC, as multiple modes exist for their
230 three-letter personality codes. Since these findings suggest a disparity when all three letters of the personality patterns are used to compile an environmental code, it is preferable to only consider the first letters of the personality codes of the members of an environment. Table 7.33 outlines the two environmental codes obtained for each of the categories of registration of psychologists, based on the first and alternative first letters of the relevant personality codes. It also provides information on the frequency of each of the three letters of the eventual code for the various categories of registration.
Table 7.33 Environmental codes for psychology categories of registration based on first and alternative first personality code letters
CATEGORY
f: C:
CODE
If
C
CLINICAL
CODE 1
SAI
30; 5; 5
7
(N = 46)
CODE 2
SAI
30; 5; 5
7
COUNSELLING
CODE 1
SAE
22; 3; 2
10
(N = 31)
CODE 2
SEA
22; 3; 2
9
EDUCATIONAL
CODE 1
SIE
17; 2; 1
8
(N = 21)
CODE 2
SIA
17; 2; 1
6
INDUSTRIAL
CODE 1
ESR
10; 8; 1
7
(N = 20)
CODE 2
ESA
12; 7; 1
7
Frequency Congruence Index (Brown & Gore, 1994)
From the results in Table 7.33 it is evident that none of these codes coincides with the South African Dictionary of Occupations (1987) codes. The educational environment's first code is the only code resembling the
231 SADO code, as it is the same but out-of-order code. It is also interesting that the industrial code is the least consistent with the other codes, as it is predominantly enterprising, as opposed to the social nature of the other categories of registration. Considering the C-Index scores, the counselling psychology environment code is closest to the SADO code in terms of congruence, although all the codes display only average to low congruence with the SADO code. Hypotheses Ho9.2, Ho9.3, Ho9.4, and Ho9.5 are rejected and the alternative hypotheses HA9.2, HA9.3, HA9.4, and HA9.5 are retained (see point 6.5.3).
7.4.2 Environmental codes of optometrists In Figure 7.2 results concerning the first letters of the personality codes of the sample of optometrists can be found. Table 7.34 indicates the modes and medians of the three-letter codes of the optometrists' personality code frequency distributions. Findings presented in Figure 7.2 yield an IER environmental code for optometrists according to their first letter code frequencies. The modes of both their first and second three-letter personality codes presented in Table 7.34 suggest an ISE environmental code for optometrists, although, since multiple modes exist, it may also be an ISC, 1ES, or ECI code. According to the Congruence Index the sample's personality code based on first letter frequencies have high levels of similarities (C = 14) with the SADO code, and their personality codes based on the modes of the frequencies of all three letters are identical to the SADO code, although only in terms of one of these modes.
232
14 12 10
#
•
6
•
—0-- Optometrists
4 2
*
0 R
I
A
S
E
C
Figure 7.2: Frequency distribution of the first letter personality codes of optometrists (N = 44)
Table 7.34 Medians and modes of the personality codes of the optometrists (N = 44) PERSONAUTY CODE
MEDOAN
CODE 1
338.50
CODE 2
425.00
MODE 245 .
245*
,
* Multiple modes exist.
Considering the discussion of the findings presented in Table 7.34 and Figure 7.2, hypothesis Hol 0 is rejected and the alternative hypothesis 10 is retained (see point 6.5.3). In general, the results of these frequency distributions compared to the descriptive data obtained from The South African Dictionary of Occupations suggest strong discrepancies in terms of the environmental codes both of psychologists and optometrists.
233
7.5 SUMMARY OF HYPOTHESES AND CONCLUSIONS Table 7.35 and Table 7.36 contain synopses of the null and alternative hypotheses that were retained. With the exception of hypothesis HA3.1 all the results were significant at the 1% level of significance.
7.5.1 Null hypotheses
Table 7.35 Retained null hypotheses and conclusions NO
Conclusions
hypotheses
1.1
Person-environment congruence alone significantly affect overall job satisfaction.
1.1 + 1.3 + 1.4
Person-environment congruence, occupational group and gender do not influence overall job satisfaction.
2.1
Person-environment congruence alone significant effect on intrinsic job satisfaction.
2.1 ÷ 2.3 + 2.4
Person-environment congruence, occupational group and gender together do not affect intrinsic job satisfaction.
does
has
not
no
234 Table 7.35 (Continued) Null hypotheses
Conclusions
3.2 + 3.3 + 3.4
Job involvement, occupational group and gender together have no significant effect on extrinsic job satisfaction.
4.1 + 4.4
Person-environment congruence and gender jointly have no significant effect on job involvement
5.1 + 5.3 + 5.5
Person-environment congruence, extrinsic job satisfaction and gender together do not significantly alter the job involvement scores.
6
Job involvement has no significant moderating influence on the person-environment congruenceoverall job satisfaction relationship.
7
Job involvement has no significant moderating influence on the person-environment congruenceintrinsic job satisfaction relationship.
8
Job involvement has no significant moderating influence on the person-environment congruenceextrinsic job satisfaction relationship.
235 7.5.2 Alternative hypotheses Table 7.36 Retained alternative hypotheses and conclusions
Alternative
Conclusions
hypotheses
1.2
Job satisfaction is determined by job involvement.
2.2
Intrinsic job satisfaction is influenced by job involvement. There is also a significant correlation between intrinsic job satisfaction and job involvement.
3.1
The influence of person-environment congruence on extrinsic job satisfaction is moderately significant. Likewise a moderately significant negative correlation exist between extrinsic job satisfaction and personenvironment congruence.
4.2 + 4.3
The occupational group and overall job satisfaction together have a significant effect on job involvement.
5.2 + 5.4
The occupational group and intrinsic job satisfaction together have a significant effect on job involvement.
9.1
The environmental code of the entire psychologist sample (SCA and SAC) significantly differs from the generally accepted ISE environmental code.
9.2; 9.3;
The environmental codes of the different categories of registration of the psychologists significantly differs from the accepted ISE environmental code.
9.4; 9.5 10
The ISE environmental code for optometrists is not valid for this sample of optometrists.
236 7.6 CONCLUSION The research results and findings presented in this chapter show that the research objectives have been reached. The significance or nonsignificance of each of the interrelations of the variables of the study have been illustrated. Job involvement was also not found to be a significant mediator of the congruence-job satisfaction relationship, which in itself was nonsignificant. As for the environment code component of personenvironment congruence, the findings showed that the codes of the study's participants were not consistent with The South African Dictionary of Occupations' codes at all. The implications of these results and relevant recommendations will be discussed in Chapter Eight.
237
CHAPTER 8 IMPLOCATDONS AHD RC COMMEHDATMS
8.1 INTRODUCTION The researcher in the field of human sciences aims to attain a valid understanding of phenomena based in reality through objective methods (Mouton & Marais, 1988). In the present study inadequacies in our knowledge regarding the interrelations of certain work-related phenomena were identified. These phenomena have been explained in accordance with theoretical models, namely person-environment congruence which was described according to Holland's (1985b) vocational choice theory, job satisfaction, which was operationalised in terms of principles of the Theory of Work Adjustment (Davis & Lofquist, 1984), and job involvement, which was defined by Kanungo (1982b). A further question about reality that was identified by this study dealt with the nature of the vocational environments of the participants of this study in terms of Holland's theory. Hence the relevant aspects of this research that aims at explaining the relations among variables of congruence, job satisfaction and job involvement, as well as extraneous variables of gender and the sample or occupational group will now be summarised. Findings on the aspects of the personality types of the participants that are pertinent to the construct of congruence will also be discussed. The discussion on these findings will be complemented by a few references to certain aspects of the theories and empirical evidence thereof that were described in this dissertation. Thereafter both theoretical and practical implications resulting from the analyses of the study's data will be presented. Culminating from the entirety of the study and these implications, recommendations will be made regarding theory, future research and vocational guidance practices. Lastly, the study will be evaluated in accordance with general guidelines for valid research.
238 8.2 SUMMARY As a consequence of the need for greater clarity regarding the interrelations of the career phenomena as such that has been identified in literature (Hackett et al., 1991), special attention was paid to the three aforementioned career-related aspects in this study. Person-environment congruence, or the degree of similarity between the personality type of an individual and the predominant personality types within the work environment in which he functions is the first of these phenomena. Job satisfaction, or positive feelings resulting from a match of a person's values and needs and the positive reinforcement thereof by the work environment is the second aspect that was studied. A distinction can further be drawn between satisfaction with intrinsic aspects of a job and extrinsic job satisfaction. The third aspect was job involvement, or the cognitive belief of a person's psychological identification with his or her current job. As seen above, these three career aspects are defined according to and belong to different vocational theories. The need for a theoretical model that will incorporate all these constructs is evident. Questions have therefore been raised regarding the possible interrelations of these constructs both in theoretical and practical terms. There seem to be suggestions both in the theories that have been discussed and the cited research examples of positive associations between personenvironment congruence and job satisfaction, as well as job satisfaction and job involvement, leaving questions unanswered about the possibility of associations between person-environment congruence and job involvement. Moreover, the research findings cited in this dissertation show a need for cross-cultural studies of these variables. The practical applicability of congruence construct in terms of South African samples need to be proved, since principles thereof are widely used by psychologists and human resource practitioners alike, such as the use of personality codes and environmental codes (person-environment congruence) based on Holland's theory, for career guidance purposes. When Holland's theory is used, the validity of such environmental codes for South African populations would be an absolute prerequisite for accurate and propitious intervention in career development strategies.
239 The present research has been designed specifically to meet such needs. To recapitulate, the purpose of this study was to describe and explain interactions among person-environment congruence, job satisfaction (overall, intrinsic and extrinsic) and job involvement. To test the generalisability of possible relations among these variables, extraneous variables of gender and occupational group were built into the research design. Testing the possibility of the role of job involvement as mediator of a congruence-job satisfaction relationship was one of the specific aims of the study. Over and above this descriptive purpose of the study, the study also had the aim of testing the applicability of person-environment congruence construct for samples of optometrists and psychologists in the South African context. Very specifically, the research was designed to validate or discredit environmental codes for these samples listed in The South African Dictionary of Occupations (1987). In order to achieve these objectives data concerning job involvement was obtained by means of the Minnesota Satisfaction Questionnaire, and of job involvement by means of the Job Involvement Questionnaire. Personality scores obtained by means of the Self-Directed Search questionnaire was used to compute the levels of person-environment congruence of the participants. This test group consisted of 114 psychologists and 44 optometrists. Due to the diversity of the research objectives, several different statistical methods had to be incorporated in the research design to test the hypotheses relating to the objectives. A factorial research design was used to create and test general linear models of the dependent and independent variables. Pearson correlation coefficients of these variables were used to test the presumed mediating effect of job involvement on Holland's (1985b) congruence hypothesis (congruence-satisfaction relationship). The significance of differences between males and females and the two occupational groups were assessed by means of inferential statistics (the t-test). By means of the Environment Assessment Technique (Holland, 1985b) and frequency distributions of both the first letters and all three letters of the personality codes of the participants, the environmental codes
240 were determined specifically for these samples and compared to accepted environmental codes of the relevant populations. The findings of the research have been delineated in the previous chapter. The following discussion summarises and evaluates these findings in practical terms. For this sample, evidence showed that whether the participants were operative in job environments similar to that of their personality types or not, did not determine whether they were happy and satisfied with their jobs. This was valid both for males and females, and both for psychologists and optometrists. What did indeed determine feelings of job satisfaction, and need satisfaction through these vocations, was whether the participants believed that they identified psychologically with their jobs. The nonsignificance of the effect of person-environment congruence on job satisfaction in this study in particular is contrary to Holland's theory and congruence hypothesis. However, the cited research examples in this dissertation also reflected discrepancies in terms of the validity of positive associations between congruence and job satisfaction, since some researchers found support for this hypothesis (Mount & Muchinsky, 1978; Wiggins, 1984; Witt & Handal, 1984; Jagger et al., 1992) whilst others failed to find support for it (Tranberg et al., 1993). Interestingly, Mount and Muchinsky (1978) found that in the social environment, congruence had no significant effect on job satisfaction. The job involvement-job satisfaction relationship that was confirmed by the findings has previously been attested in research, that also indicated the distinctiveness of these two constructs (Brooke et al., 1988; Paterson & O'Driscoll, 1990; Mathieu & Farr, 1991). There is however no theoretical description in Kanungo's (1979) model that job satisfaction is a function of job involvement. There are only suggestions that job involvement instead is a result of satisfaction of salient needs through appropriate activities. Kanungo (1982b) differentiates between job satisfaction as an affective state and job involvement as a cognitive condition. The lack of support for the congruence hypothesis applied for intrinsic job satisfaction as well. In other words, satisfaction with intrinsic aspects of the
241 jobs was not influenced by harmony between personality types and the environmental type. Once again this was true for both genders and both samples. On the other hand, believing in one's psychological identification with the job (job involvement) determined satisfaction with intrinsic job aspects, for instance the opportunity to act according to one's moral values. This attitude of job involvement surprisingly did not influence the satisfaction with extrinsic aspects of the job, including amongst other things compensation, recognition and so forth. Kanungo (1982b) also cites research findings that support the significance or relationships between job involvement and intrinsic need satisfaction, but not between job involvement and extrinsic need satisfaction. In the present study there were however moderate indications that person-environment congruence can affect the level of extrinsic satisfaction experienced by the participants. Data obtained by means of the correlational analyses showed moderately significant negative relations between person-environment congruence and extrinsic job satisfaction. Practically this means that those in fitting vocational environments are less likely to be happy with their jobs than those in jobs that are incongruent with their personality types. These findings contradict basic assumptions of Holland's theory and intuitively also seem incorrect. It is noteworthy that some of the participants commented on the inapplicability of some of the extrinsic job satisfaction items of the Minnesota Satisfaction Questionnaire, such as items dealing with supervision, when they were selfemployed, and such responses may well have confounded the research results. (More than 75% of the participants were either self-employed or partially self-employed). The degree of congruence between personality types and vocational types for both males and females did not influence the level of job involvement displayed by the participants. According to Kanungo (1982b) the past socialisation process is the main antecedent of job involvement, although present working conditions may also influence it. The lack of significant effects of person-environment congruence on job involvement raises further questions regarding the conditions necessary for job involvement to occur. The results indicated that job satisfaction is a function of job involvement and the effect of the occupational group. This effect was more significant for the sample of psychologists, for whom being satisfied with their jobs affected
242 their belief of personal identification with work more than for the optometrists. This applied both for overall job satisfaction and intrinsic job satisfaction. The level of satisfaction with extrinsic aspects of the job of both males and females regardless of their level of congruence did not affect their attitude of job involvement. There is empirical support in literature for this reciprocal relationship of job involvement and job satisfaction (Mortimer & Lorence, 1989). The findings regarding the nonsignificance of extrinsic job satisfaction in terms of job involvement raise theoretical questions about the role of extrinsic job satisfaction in career motivation. As mentioned before, the study also assumed the validity of the congruencesatisfaction hypothesis, and postulated that job involvement may be a mediating variable in this relationship. Remarkably, evidence showed that the degree of the association between congruence and job satisfaction was not altered by either low or high levels of job involvement of the participants. In order to complement these findings, comparisons were made between males and females and the samples of optometrists and psychologists regarding their overall job satisfaction, intrinsic job satisfaction, extrinsic job satisfaction and job involvement. The only significant difference found was that psychologists were more job involved than optometrists.
JOB SATISFACTION
4.
intrinsic
extrinsic
— dr
JOB INVOLVEMENT
PERSONENVIRONMENT CONGRUENCE
Figure 8.1: Findings on the interrelations of the variables
243 Figure 8.1 represents a model of the interrelations of these variables based only on these findings of the present research, and not any other theoretical or empirical evidence. The proposed extraneous variable gender does not appear on this model, since no gender differences were evident from the results. Previous research findings (Hesketh et al., 1992) however render support for the correlation of gender and job satisfaction, but also show that a correspondence-satisfaction relationship is not a function of gender. Up to this point the discussion on the results of this research dealt with the possibility of reciprocal determination of the variables. Results dealing specifically with person-environment congruence and the use of environmental codes in the South African context show that psychologists work in a predominantly social environment, as opposed to the presumed investigative environment. This environmental code is highly differentiated social code with undertones of conventional and artistic characteristics. The resultant SCA code is also more consistent than the proposed ISE code. In like manner, the environmental codes for each of the psychology categories of registration differed from the SADO code. The first and alternative environmental codes with the use of three-letter codes were too diverse to be suitable. In terms of the first letter codes it was once again evident that the psychologists were primarily Holland social types. The exception was the industrial psychologists who were predominantly enterprising. Also, each category of registration had a different environmental code. Contrary to the above, the findings of this study confirm that the vocational environment of optometrists is predominantly made up by investigative or scientifically orientated members. When the frequency of only the first letters of the personality codes were considered (the EAT technique), it was evident that although the optometrist environment is predominantly investigative, there are more enterprising individuals, followed by realistic individuals in this environment, which contradicts the SADO code. According to findings incorporating three-letter personality codes of the SDS there were once again too many disparate environmental codes, proving the superiority of environmental codes based only on the first letter personality codes of the members.
244
8.3 IMPLICATIONS AND RECOMMENDATIONS Now that the research problem and research methodology have been summarised, and the results of the study have been discussed in terms of other theoretical and empirical considerations, it is necessary to understand both theoretical and practical implications of these results.
8.3.1 Theoretical implications and recommendations The purpose in discussing theoretical implications of this study is to examine whether the research questions have been adequately answered. Although findings of this study were in some instances not consistent with other research findings or even with theory as discussed above, one important result concerning the relationships of the study's variables emerged, namely the confirmation of the reciprocal relationship both of overall job satisfaction and intrinsic job satisfaction, and job involvement. This holds important theoretical implications, and it is therefore recommended that the description of the significance of the job involvement construct must be amplified within the theoretical paradigm of vocational psychology. In particular, theories of vocational choice and adjustment should incorporate a description of the role of job involvement and not only of job satisfaction. For instance, according to Holland (1985b) congruence between environmental types and personality types brings about job satisfaction. In the Theory of Work Adjustment (Davis & Lofquist, 1984) a correspondence between values and environmental reinforcement, or career interests and actual jobs, or personal needs and environmental rewards result in job satisfaction. The perception of the need satisfying potential of a job is an important antecedent of job involvement (Kanungo, 1982b). Although elements of these theories seem to overlap, the need exists for an integrative description of the relationships of these variables. An explanation is also needed for the lack of a significant effect of extrinsic job satisfaction on job involvement that is evident in the findings of this study. Since job involvement is defined primarily in cognitive terms, the role of
245 cognition in vocational behaviour and outcomes also needs to be described and explained to a greater extend. As there were no significant effect found of job involvement on extrinsic job satisfaction, it may be that possible high levels of work autonomy of the participants may have influenced these findings. It is suggested that the possible moderating effect of work autonomy on these relationships should be empirically examined. The extraneous variable of gender did not affect any of the findings of the study and differences between optometrists and psychologists were only apparent in terms of job involvement. There is a need for more studies on the different levels of job involvement in different vocations as this could lead to a more comprehensive understanding of the antecedents and correlates of job involvement. Moreover, future studies on job satisfaction should incorporate demographic variables that have been empirically associated with it, for example age (Schwab & Heneman, 1977). The lack of significance of job involvement as mediator of the hypothesised congruence-satisfaction relationship leaves the questions about the reasons for discrepancies in research findings on the congruence hypothesis unanswered. It is recommended in accordance with Spokane's (1985) suggestions that this relationship should be studied over time, and not merely be investigated in correlational research designs. Findings regarding the dissimilarity of the sample of each of the samples' environmental codes and the accepted SADO code for these samples ironically hold serious implications for the inferential validity of most of the other results of this study. Since a different personality code has been found for the members of these environments, the person-environment congruence scores, which are based on these personality codes, are not valid for the sample. As suggested by the research results of Assouline and Meir (1987), and Meir and Navon (1992) the census method for obtaining environmental codes is superior to occupational classification methods in as far as the validity of the congruence-satisfaction hypothesis is concerned. Taljaard and Von Mollendorf (1987) themselves recommend the use of codes in local context rather than nationally determined codes in particular instances. The possibility of significant relationships between person-environment
246 congruence and job satisfaction, or job involvement therefore still remains should a sample-specific environmental code be used to calculate congruence. Nonetheless, as a result of these findings it is recommended that the validity of The South African Dictionary of Occupations' environmental codes should also be tested for other vocational populations. The cited study by Van Der Merwe et al. (1990) found similarities between South African and American three-letter environmental codes for 60 occupations to be only 40%, and similarities between three-letter codes in the correct sequence to be as low as 11,7%.
8.3.2 Practical implications and recommendations As is the aim of Holland's (1985b) theory, many counselling psychologists, human sciences practitioners and employers endeavour to help people to make the correct choices regarding career selection, development and change. For this purpose vocational theories and relating instruments can prove invaluable if used correctly. An understanding of constructs such as job satisfaction or job involvement can greatly help in this process, which necessitates constant research of the interrelations of vocational phenomena. The results of the present research imply that an increase in the attitude of job involvement of people will result in greater feelings of job satisfaction. Also, for a person's attitude of job involvement to grow, his job satisfaction has to increase. According to Kanungo's (1982b) theory an increase in job involvement comes about as a result of the changed perceptions when a person's activities are perceived to result in the satisfaction of his salient or important needs. It is therefore vital to determine the relative importance of different needs for different individuals in order to help them meet those needs and in so doing increase their levels of job involvement and job satisfaction. As seen before, results of this study show that job involvement relates specifically to overall job satisfaction and intrinsic job satisfaction. In other words, for a person to feel psychologically close to his job, he should feel satisfied with intrinsic aspects of his vocation such as the opportunity to use his own abilities and to achieve, as opposed to extrinsic aspects such as the
247 need for recognition and compensation. In terms of the two samples, this applies even more for psychologists than it does for optometrists. Because it is evident from previous empirical studies that job involvement negatively relates to undesirable work outcomes such as burnout or the desire to leave (Elloy et al., 1991) to give only one example, it is necessary to strive towards increasing the job satisfaction and job involvement levels of people. The SCA environmental code that was obtained for psychologists practically implies, according to Holland's (1985b) theory, different preferences, aversions, competencies, self-perceptions, values and personality traits for psychologist than does the ISE environmental code. Furthermore, instead of abstract, theoretical and analytical influences of the environment on its members, the psychologist's environment exerts social influences on them. Instead of seeing the world in complex and abstract ways, the psychologist sees the world in flexible ways. They also do not cope with others in rational or indirect ways, but in friendly, helpful and co-operative ways. Likewise, the differences of the sample's environmental codes and the SADO code pertaining to the various categories of registration of psychologists also hold several practical implications in terms of the type of person functioning in those environments as suggested by Holland (1985b). For instance, the predominantly enterprising code of industrial psychologists implies amongst other things that they value economic achievement, and manipulate others for economic and organisational profit, instead of for developing or training others. Contrary to what the SADO code implies, industrial psychologists have aversions to investigative situations according to these findings. Similar implications may be listed concerning the differences between the SADO environment code for optometrists and the environment code of this sample of optometrists. Practically these findings illustrate the need for the accurate understanding of the environment by vocational counsellors and others alike in order to successfully match prospective employees and different vocations. It could be interesting for personnel managers and employers to calculate the specific environmental code of the organisation in order to optimise staffing procedures, to mention but one practical application of the knowledge gained from this research.
248
8.4 EVALUATION OF THE STUDY In evaluation of the above, one may conclude that this study has managed to find solutions to the research problems that were presented in the introductory chapter of the dissertation, although findings were not always consistent with previous research results. Certain inadequacies in the implications of the study have been identified and recommendations have been made to further the aims of this descriptive and explanatory study. This study can now be evaluated in view of the guidelines for research that are presented by Mouton and Marais (1988), namely the necessity of the criterion validity and external validity of research. A few points dealing with the limitations of the study are also identified. In terms of the external validity of the study one may note that the findings, especially in terms of the personality codes, can not be generalised, but applies specifically to optometrists and psychologists in the South African context. Also, certain effects of selection may have occurred, since the researcher had to rely on the voluntary response of the participants, which may affect the generalisability of the study. Moreover, the low response rate could futher decrease the generalisability of the findings. The size of the resultant optometrist sample was also not optimum. It may be that the professionals who did not complete the questionnaires differ from the respondents in systematic ways regarding the variables that were tested, although results concerning the characteristics of this sample were compatible with previous research findings. One may also reason that the extensiveness of the questionnaire may have dissuaded some potential subjects from participating and it is suggested that the length of questionnaires sent by mail should not be as long. Also, further studies on the Holland environment code for different vocations should also include non-professional subjects. Nonetheless, the generalisability of the study was enhanced through random sampling procedures. The possible period effects were also lessened by sending out questionnaires at different intervals. In terms of the criterion validity of the research design, it counts in favour of this study that the measuring instruments by means of which the scores for the variables of the study were obtained, fulfil the necessary criteria of
249 validity and reliability. Furthermore, the theoretical validity of the variables of the study was established in the discussion of the relevant theories and relating research. In the previous section one possible inferential inadequacy of the study was mentioned. It referred to the use of a general environmental code to define the environment component of the person-environment congruence construct, instead of an environmental code calculated specifically for the sample. Further research that replicates the objective of the study to test the interrelations of variables of person-environment congruence, job satisfaction and job involvement, but with the use of a sample-specific environmental code, is warranted. It is also recommended that studies dealing with possible causal relations among these variables should be conducted. In other words, more specific analyses of variance could enhance the explanatory power of the findings, especially in as far as possible causal relations among the variables are concerned.
8.5 CONCLUSION To conclude, although the present study has limitations in terms of the generalisability of the results, important theoretical and practical implications of the results have been identified. It is particularly true in as far as the relationships between job satisfaction and job involvement variables are concerned, as well as in terms of the environmental codes that have been determined for psychologists and optometrists in the South African context. It has also illustrated that future research on vocational attitudes and behaviours should not focus on mere correlations, but should incorporate the complex network of causal relations of such variables, and in so doing combine seemingly disparate elements of various vocational theories.
250
REFEREMCES
Adams, G.A., King, L.A., & King, D.W. (1996). Relationships of job and family involvement, family social support, and work-family conflict with job and life satisfaction. Journal of Applied Psychology, 81(4), 411420. Alvi, S.A., Khan, S.B. & Kirkwood, K.J. (1990). A comparison of various indices of differentiation for Holland's model. Journal of Vocational Behavior, 36, 147-152. Anderson, W.T., Hohenshil, T.H. & Brown, D.T. (1984). Job satisfaction among practicing schools psychologists: A national study. School Psychology Review, 13(2), 225-230. Aronowitz, A., Bridge, R.G. & Jones, P. (1985). Sex bias in the Self-Directed Search Investigative subscale. Journal of Vocational Behavior, 26(2), 146-154. Arvey, R.D., Bouchard, T.J., Segal N.L. & Abraham, L.M. (1989). Job satisfaction: Environmental and genetic components. Journal of Applied Psychology, 74(2), 187-192. Arvey, R.D. & Gross, R.H. (1977). Satisfaction levels and correlates of satisfaction in the homemaker job. Journal of Vocational Behavior, 10, 13-24. Assouline, M. & Meir, E.I. (1987). Meta-analysis of the relationship between congruence and well-being measures. Journal of Vocational Behavior, 31, 319-332. Antecedents of Barrick, Mount, M.K. & Strauss, J.P. (1994). Personnel involuntary turnover due to a reduction in force. Psychology, 47(3), 515-535.
251 Berry, L.M. & Houston, J.P. (1993). Psychology at work: An introduction to industrial and organizational psychology. Madison, Wisconsin: Brown & Benchmark Publishers. Bizot, E.B. (1993) A working Theory of Work Adjustment. Journal of Vocational Behavior, 43, 122-126. Bizot, E.B. & Goldman, S.H. (1993). Prediction of satisfactoriness and satisfaction: An 8-year follow up. Journal of Vocational Behavior, 43, 19-29. Blau, G. (1985). A multiple study investigation of the dimensionality of job involvement. Journal of Vocational Behavior, 27, 19-36. Blau, G.J. (1987). Using a person-environment fit model to predict job involvement and organisational commitment. Journal of Vocational Behavior, 30, 240-257. Blau, G.J. (1993). Work Adjusment Theory: A critique and suggestions for future research and application. Journal of Vocational Behavior, 43, 105-112. Bledsoe, J.C. & Baber, W.C. (1979). Factor invariance in the measurement of job satisfaction. Perceptual and Motor Skills, 48(3), 985- 986. Borgen, F.H. & Donnay, D.A.C. (1996). Slicing the vocational interest pie one more time: Comment on Tracey and Rounds. Journal of Vocational Behavior, 48, 42-52. Boshoff, A.B., Bennett, H.F. & Kellerman, A.M. (1994). Career orientations as predictors of the level of job involvement of professional people. Journal of Industrial Psychology, 20(2), 8-13. Boyle, G.J. & Fabris, S. (1992). LISREL analyses of the RIASEC model: Confirmatory and congeneric factor analyses of Holland's Self-Directed Search. Personality and Individual Differences, 13(10), 1077-1084.
252 Die Brand, H.J., Van Noordwyk, J.S.J. & Hanekom, J.D.M. (1994). toepassing van die Self-Directed Search op 'n groep swart adolessente. South African Journal of Psychology, 24(2), 47-52. Breeden, S.A. (1993). Job and occupational change as a function of occupational correspondence and job satisfaction. Journal of Vocational Behavior, 43, 30-45. Bretz, R.D. & Judge, T.A. (1994). Person-organization fit and the Theory of Work Adjustment: Implications for satisfaction, tenure, and career success. Journal of Vocational Behavior, 44, 32-54. Brooke, P.P., Russell, D.W. & Price, J.L. (1988). Discriminant validation of measures of job satisfaction, job involvement, and organizational commitment. Journal of Applied Psychology, 73(2), 139-145. Brown, S.D. (1993). Contemporary psychological science and the Theory of Work Adjustment: A proposal for integration and a favor returned. Journal of Vocational Behavior, 43, 58-66. Brown, S.P. (1996). A meta-analysis and review of organizational research on job involvement. Psychological Bulletin, 120(2), 235-255. Brown, D., Brooks, L & associates (1984). Career choice and development. San Francisco, Washington, London: Jossey-Bass Publishers. Brown, S.D. & Gore, Jr, P.A. (1994). An evaluation of interest congruence indices: Distribution characteristics and measurement properties. Journal of Vocational Behavior, 45, 310-327. Camp, C.C. & Chartrand, J.M. (1992). A comparison and evaluation of interest congruence indices. Journal of Vocational Behavior, 41, 162182. Cheloha, RS. & Farr, J.L. (1980) Absenteeism, job involvement, and job Journal of Applied satisfaction in an organizational setting. Psychology, 65(4), 467-473.
253 Cohen, A. (1995). An examination of the relationships between work commitment and nonwork domains. Human Relations, 48(3), 239-263. Conneran, J.M. & Hartman, B.W. (1993). The concurrent validity of the SelfDirected Search in identifying chronic career indecision among vocational education students. Journal of Career Development, 19(3), 197-208. Cook, J.D., Hepworth, S.J., Wall, T.D. & Warr, P.B. (1981). The Experience of work: A compendium and review of 249 measures and their use.
London: Academic Press. Costa, P.T., McCrae, R.R. & Holland, J.L. (1984). Personality and vocational interests in an adult sample. Journal of Applied Psychology, 69(3), 390-400. Craig, G.J. (1986). Prentice-Hall.
Human Development. Englewood Cliffs, New Jersey:
Cull, J.G. & Hardy, R.E. (1973). Adjustment to work. Springfield, Illinois, USA: Charles C Thomas Publisher. Cummings, T.G. & Bigelow, J. (1976). Satisfaction, job involvement, and intrinsic motivation: An extension of Lawler and Hall's factor analysis. Journal of Applied Psychology, 61(4), 523-525. Dawis, R.V. (1980). Personnel assessment from the perspective of the Theory of Work Adjustment. Public Personnel Management, 9(4), 268273. Dawis, R.V., Dohm, T.E. & Jackson, C.R.S. (1993). Describing work environments as reinforcer systems: Reinforcement schedules versus reinforcer classes. Journal of Vocational Behavior, 43, 5-18. Dawis, R.V. & Lofquist, L.H. (1978). A note on the dynamics of work adjustment. Journal of Vocational Behavior, 12(1), 76-79.
254 Dawis, R.V. & Lofquist, L.H. (1984). A psychological Theory of Work Adiustment: An individual-differences model and its applications. Minneapolis: University of Minnesota Press. Dawis, R.V. & Lofquist, L.H. (1993). From TWA to PEC. Journal of Vocational Behavior, 43, 113-121. Dipboye, R.L., Zultowski, W.H. Dewhirst, H.D. & Arvey, R.D. (1979). Selfesteem as a moderator of performance-satisfaction relationships. Journal of Vocational Behavior, 15, 193-206. Downes, M. & Kroeck, K.G. (1996). Discrepencies between existing jobs and individual interests: An emprical application of Holland's model. Journal of Vocational Behavior, 48, 107-117. Druker, B. & De Jong, T. (1996). The educational psychologist as organization development consultant in South African schools: A framework for conceptualizing substantive issues. School Psychology International, 17(1), 17-32. Handbook of industrial and organizational Dunnette, M.D. (1976). psychology. Chicago, Rand McNally. Dumenci, L. (1995). Construct validity of the Self-Directed Search using hierarchically nested structural models. Journal of Vocational Behavior, 47, 21-34. Dupree, P.I. & Day, H.D. (1995). Psychotherapists' job satisfaction and job burnout as a function of work setting and percentage of managed care clients. Psychotherapy in Private Practice, 24(2), 77-93. Eagan, A.E. & Walsh, W.B. (1995). Person-environment congruence and coping strategies. Career Development Quarterly, 43(3), 246-256. Eberhardt, B.J. & Muchinsky, P.M. (1984). Structural validation of Holland's hexagonal model: Vocational classification through the use of biodata. Journal of Applied Psychology, 69(1), 174-181.
255 Efraty, D., Sirgy, M.J. & Claiborne, C.B. (1991). The effects of personal alienation on organizational identification: A quality-of-work-life model. Journal of Business and Psychology, 6(1), 57-78. Elliott, T.R. & Byrd, E.K. (1985). Scoring accuracy of the Self-Directed Search with ninth-grade students. Vocational Guidance Quartely, 34(2), 85-90. Elizur, D. & Tziner, A. (1977). Vocational needs, job rewards, and satisfaction: A canonical analysis. Journal of Vocational Behavior, 10, 205-211. Elloy, D.F., Everett, J.E. & Flynn, W.R. (1991). An examination of the correlates of job involvement. Group and Organization Studies, 16(2), 160-177. Elloy, D.F., Everett, J.E. & Flynn, W.R. (1995). Multidimensional mapping of the correlates of job involvement. Canadian Journal of Behavioural Science, 27(1), 79-91. Elloy, D.F. & Terpening, VV.D. (1992). An empirical distinction between job involvement and work involvement: Some additional evidence. Canadian Journal of Behavioural Science, 24(4), 465-478. Elton, C.F. & Smart, J.C. (1988). Extrinsic job satisfaction and personenvironment congruence. Journal of Vocational Behavior, 32, 226-238. Fishburne, F.J. & Walsh, VV.B. (1976). Concurrent validity of Holland's theory for non-college-degreed workers. Journal of Vocational Behavior, 8(1), 77-84. Frone, M.R., Russell, M. & Cooper, M.L. (1995). Job stressors, job involvement and employee health: A test of identity theory. Journal of Occupational and Organizational Psychology, 68, 1-11. Furnham, A., Toop, A., Lewis, C. & Fisher, A. (1995). P-E fit and job satisfaction: A failure to support Holland's theory in three British samples. Personality and Individual Differences, 19(5), 677-690.
256
Furnham, A. & Schaeffer, R. (1984). Person-environment fit, job satisfaction and mental health. Journal of Occupational Psychology, 57(4), 295307. Gable, M. & Dangello, F. (1994). Job involvement, machiavellianism and job performance. Journal of Business and Psychology, 9(2), 159-169. Gade, E., Fuqua, D.R. & Hurlburt, G. (1988). The relationship of Holland's personality types to educational satisfaction with a Native-American high school population. Journal of Counseling Psychology, 35(2), 183186. Gaffey, R.L. & Walsh, W.B. (1974). Concurrent validity of Holland's theory. Journal of Vocational Behavior, 5(1), 41-51. Gati, H. (1985). "(ascription of alternative measures of the concepts of vocational interest: Crystallization, congruence, and coherence. Journal of Vocational Behavior, 27, 37-55. Gati, I. (1991). The structure of vocational interests. 109(2), 309-324.
Psychological Bulletin,
Gati, L, Garty, V. & Fassa, N. (1996). Using career-related aspects to assess person-environment fit. Journal of Counseling Psychology, 43(2), 196-206. George, J.M. & Jones, G.R. (1996). The experience of work and turnover intentions: Interactive effects of value attainment, job satisfaction, and positive mood. Journal of Applied Psychology, 81(3), 318-325. Gerdes, L.C., Moore, C., Ochse, R., Van Ede, D. (1988). Die ontwikkelende volwassene. Tweede Uitgavve. Durban: Butterworths. Glidden-Tracey, C.E. & Parraga, M.I. (1996). Assessing the structure of vocational interests among Bolivian university students. Journal of Vocational Behavior, 48, 96-106.
257 Gorn, G.J. & Kanungo, R.N. (1980). Job involvement and motivation: Are Organizational intrinsically motivated managers more job involved? Behavior and Human Performance, 26, 265-277. Gottfredson, L.S. (1982). The sex fairness of unnormed interest inventories. Vocational Guidance Quarterly, 31(2), 128-132. Prestige in vocational interests. Journal of Gottfredson, G.D. (1996). Vocational Behavior, 48, 68-72. Gottfredson, G.D. & Holland, J.L. (1975). Vocational choices of men and women: A comparison of predictors from the Self-Directed Search. Journal of Counseling Psychology, 22(1), 28-34. Gottfredson, G.D. & Holland, J.L. (1990). A longtidunal test of the influence of congruence: Job satisfaction, competency utilization, and counterproductive behavior. Journal of Counseling Psychology, 37(4), 389-398. Dictionary of Gottfredson, G.D., Holland, J.L. & Ogawa, D.K. (1982). Holland occupational codes. Palo Alto, Calif.: Consulting Psychologists Press. Psychological Testing: History, principles, and Gregory, R.J. (1996). applications. (Second Edition). Boston: Allyn and Bacon.
Gruneberg, M.M. (1979). Macmillan Press.
Understanding lob satisfaction. Norfolk: The
Hackett, G., Lent, R.W. & Greenhaus, J.H. (1991). Advances in vocational theory and research: A 20-year retrospective. Journal of Vocational Behavior, 38, 3-33. Hansen, J-I.C. (1996). What goes around, comes around. Journal of Vocational Behavior, 48, 73-76.
258 Hansen, J-I.C., Collins, R.C., Swanson, J.L. & Fouad, N.A. (1993). Gender Journal of Vocational differences in the structure of interests. Behavior, 42, 200-211. Harmon, L.W. (1996). Lost in space: A response to "The spherical representation of vocational interests" by Tracey and Rounds. Journal of Vocational Behavior, 48, 53-58. Harmon, L.W. & Zutowski, D.G. (1980). Reliability of Holland codes across Journal of Counseling interest measures for adult females. Psychology, 27(5), 478-483. Hauber, F.A. & Bruininks, R.H. (1986). Intrinsic and extrinsic job satisfaction among direct-care staff in residential facilities for mentally retarded people. Educational and Psychological Measurement, 46(1), 95-105. Healy, C.C. & Mourton, D.L. (1983). Derivatives of the Self-Sirected Search: Potential clinical and evaluative uses. Journal of Vocational Behavior, 23, 318-328. Healy, C.C. & Mourton, D.L. (1984). The effects of an abbreviated SelfDirected Search on the career decision competencies of community college students. Vocational Quidance Quarterly, 33(1), 55-62. Heaven, P.C.L. (1994). Occupational attrubutional style and attitudes to work: An Australian study. Australian Psychologist, 29(1), 57-61. Heesacker, M., Elliott, T.R., & Howe, L.A. (1988). Does the Holland code predict job satisfaction and productivity in clothing factory workers? Journal of Counseling Psychology, 35(2), 144-148. Heidema, H. (1995).
Die ontwikkeling van 'n Suid-Afrikaanse
loopbaaninligtingstoets.
Unpublished thesis.
Rand Afrikaans
University, Johannesburg. Henry, P. (1988). Holland's hexagonal model applied to nontraditional premedical students. Psychological Reports, 62(2), 399-404.
259 Relationship between academic achievement and Henry, P. (1989). Examination of Holland's theory. measured career interest: Psychological Reports, 64(1), 35-40. Henry, P., Bardo, H. & Bryson, S. (1988). The impact of race and gender on Holland's Self-Directed Search for nontraditional premedical students. College Student Journal, 22(2), 206-212. Hesketh, B. (1985). In search of a conceptual framework for vocational psychology. Journal of Counseling and Development, 64, 26-29. Hesketh, B. (1993). Toward a better adjusted Theory of Work Adjustment. Journal of Vocational Behavior, 43, 75-83. Hesketh, B. (1995). Personality and adjustment styles: A Theory of Work Adjustment approach to career enhancing strategies. Journal of Vocational Behavior, 46, 274-282. Hesketh, B., McLachlan, K. & Gardner, D. (1992). Work adjustment theory: An empirical test using a fuzzy rating scale. Journal of Vocational Behavior, 40, 318-337. Hildebrand, J.O. & Walsh, W.B. (1988). Person-environment congruence and perceived work climate. Journal of Career Development, 15(2), 121-133. Holland, J.L. (1963). Explorations of a theory of vocational choice and achievement: A four year prediction study. Psychological Reports, 12, 547-594. The VPI. Palo Alto, California: Consulting Holland, J.L. (1977). Psychologists Press.
Holland, J.L. (1979). Professional manual for the Self-Directed Search. Palo Alto, California: Consulting Psychologists Press. Holland, J.L. (1973). Making vocational choices: A theory of careers. Englewood Cliffs, NJ: Prentice-Hall.
260
Holland, J.L. (1982). The SDS helps both females and males: A comment. Vocational Guidance Quartely, 30(3), 195-197. Holland, J.L. (1985a). Author biases, errors, and ommisions in an evaluation of the SDS Investigative Summary Scale: A response to Aronowitz, Bridge and Jones (1985). Journal of Vocational Behavior, 27(3), 374376. Holland, J.L. (1985b).
Making vocational choices : A theory of vocational
personalities and work environments (Second Edition). Englewood
Cliffs, NJ: Prentice-Hall. Holland, J.L. (1985c). The Self-Directed Search: A guide to educational and vocational planning. Florida: Psychological Assessment Resources Inc. Holland, J.L. (1985d). The SDS professional manual. Odessa, Florida: Psychological Assessment Resources Inc. Holland, J.L. (1996). Exploring careers with a typology: What we have learned and some new directions. American Psychologist, 51(4), 397406. Holland, J.L. & Gottfredson, G.D. (1976). Using a typology of persons and environments to explain careers: Some extensions and clarifications. The Counseling Psychologist, 6, 20-29. Holland, J.L. & Gottfredson, G.D. (1992). Studies of the hexagonal model: An evaluation (or, the perils of stalking the perfect hexagon). Journal of Vocational Behavior, 40, 158- 170. Holland, J.L., Gottfredson, G.D. & Baker, H.G. (1990). Validity of vocational aspirations and interest inventories: Extended, replicated, and reinterpreted. Journal of Counseling Psycology, 37(3), 337-342.
261 Holland, J.L. & Nafzinger, D.H. (1975). A note on the validity of the SelfDirected Search. Measurement and Evaluation in Guidance, 7(4), 259262. Holland, J.L., Rakai, R., Gottfrdson, G.D. & Hanau, C. (1978). A multivariate analysis of the effects of the Self-Directed Search on high school girls. Journal of Counseling Psycology, 25(5), 384-389. Hyland, A.M. & Muchinsky, P. M. (1991). Assessment of the structural validity of Holland's model with job analysis (PAC)) information. Journal of Applied Psychology, 76(1), 75-80. lachan, R. (1984). A measure of agreement for use with the Holland classification system. Journal of Vocational Behavior, 24, 133-141. lachan, R. (1990). Some extensions of the lachan congruence index. Journal of Vocational Behavior, 36, 176-180. laffaldano, M.T. & Muchinsky, P.M. (1985). Job satisfaction and job performance: A meta-analysis. Psychological Bulletin, 97(2), 251-273. Jaeger, A.M. & Kanungo, R.N. (1990). Management in developing countries. London and New York: Routledge. Jagger, L., Neukrug, E. & McAuliffe, G. (1992). Congruence between personality traits and chosen occupation as a predictor of job Rehabilitation Counseling satisfaction for people with disabilities. Bulletin, 36(1), 53-60. Joyce, W., Slocum, J.W. & Von Glinow, M.A. (1982). Person-situation Journal of Occupational interaction: Competing models of fit. behaviour, 3, 265-280. Kagee, A & Price, J.L. (1995). Apartheid in South Africa: Toward a model of psychological intervention. Journal of Black Studies, 25(6), 737-748. Kanungo, R.N. (1979). The concepts of alienation and involvement revisited. Psychological Bulletin, 86, 119-138.
262
Kanungo, R.N. (1981). Work alienation and involvement: Problems and prospects. International Review of Applied Psychology, 30, 1-15. Kanungo, R.N. (1982a). Measurement of job and work involvement. of Applied Psychology, 67, 341-349. Kanungo, R.N. (1982b). York: Praeger.
Journal
Work alienation: An integrative approach. New
Kanungo, R.N. (1983). Work alienation: A pancultural perspective. International Studies of Management and Organization, XII1(1-2), 119138. Kanungo, R.N., Misra, S.B. & Dayal, I. (1975). Relationship of job involvement to perceived importance and satisfaction of employee needs. International Review of Applied Psychology, 24(1), 49-59. Keeling, B. & Tuck, B.F. (1978). Raw scores versus same-sex normed scores: An experimental study of the validity of Holland's SDS with adolescents of both sexes. Journal of Vocational Behavior, 13(3), 263271. Khan, S.B., Alvi, S.A., Shaukat, N., & Hussain, M.A. (1990). A study of the validity of Holland's theory in a non-Western culture. Journal of Vocational Behavior, 36, 132-146. Knoop, R. (1995). Relationships among job involvement, job satisfaction, and organizational commitment for nurses. Journal of Psychology, 129(6), 643-649. Kriek, H.J. (1996). Bedryfsielkunde in Suid-Afrika: 'n Toekomsperspektief. Journal of Industrial Psychology, 22(1), 7-11. Kumar, KG. & Achamamba, B. (1993). A comparative study of job satisfaction and job involvement among public and private sector employees. Psychological Studies, 38(2), 69-72.
263 Kwak, J.C. & Pulvino, C.J. (1982). A mathematical model for comparing Holland's personality and environmental codes. Journal of Vocational Behavior, 21, 231-241. Lamb, R.R. & Prediger, D.J. (1981). Technical report for the Unisex Edition of the ACT Interest Inventory (UNIACT). Iowa City, IA: American College Testing. Lambert, S.J. (1991). The combined effects of job and family characteristics on the job satisfaction, job involvement, and intrinsic motivation of men and women workers. Journal of Organizational Behavior, 12, 341-363. Lawler, E.E. (1973). Motivation in work organizations. Belmont, California: Brooks/Cole. Lawler, E.E & Hall, D.T. (1970). Relationship of job characteristics to job involvement, satisfaction, and intrinsic motivation. Journal of Applied Psychology, 54(4), 305-312. Lawson, L. (1993). Theory of Work Adjustment personality constructs. Journal of Vocational Behavior, 43, 46-57. Lee, R. & Wilbur, E.R. (1985). Age, education, job tenure, salary, job characteristics, and job satisfaction: A multivariate analysis. Human Relations, 38(8), 781-791. Lefkowitz, J. Somers, M.J. & Weinberg, K. (1984). The role of need level and/or need salience as moderators of the relationship between need satisfaction and work alienation-involvement. Journal of Vocational Behavior, 24, 142-158. Lent, E.B. & Lopez, F.G. (1996). Congruence from many angles: Relations of multiple congruence indices to job satisfaction among adult workers. Journal of Vocational Behavior, 49, 23-37. Levinson, E.M. (1989). Job satisfaction among school psychologists: A replication study. Psychological Reports, 65, 579-584.
264 Levinson, E.M. (1990). Vocational assessment involvement and use of the Psychology in the Self-Directed Search by school psychologists. Schools, 27(3), 217-228. Lodahl, T.M. & Kejner, M. (1965). The definition and measurement of job involvement. Journal of Applied Psychology, 49, 24-33. Lofquist, L.H. & Dawis, R.V. (1969).
Adjustment to work: A psychological
view of man's problems in a work-oriented society. New York:
Appleton-Century-Crofts. Lofquist, L.H. & Dawis, R.V. (1975). Vocational needs, work reinforcers and job satisfaction. The Vocational Guidance Quarterly, 24, 132-139. Lowman, R.L. & Williams, R.E. (1987). Validity of self-ratings of abilities and competencies. Journal of Vocational Behavior, 31, 1-13. Lynn, S.A., Cao, L.T. & Horn, B.C. (1996). The influence of career stage on the work attitudes of male and female accounting professionals. Journal of Organizational Behavior, 17(2), 135-149. Magnusson, K. & Stewin, L.M. (1990). The implications of Holland's constructs in the prediction of employment: A cautionary note. International Journal for the Advancement of Counselling, 13(3), 205217. Martin, T.N. & Hafer, J.C. (1995). The multiplicative interaction effects of job involvement and organizational commitment on the turnover intentions of full- and part-time employees. Journal of Vocational Behavior, 46(3), 310-331. Mathieu, J.E. & Farr, J.L. (1991). Further evidence for the discriminant validity of measures of orgnizational commitment, job involvement, and job satisfaction. Journal of Applied Psychology, 76(1), 127-133. McClelland, J. (1986). Job satisfaction of child care workers: A review. Child Care Quarterly, 15(2), 82-89.
265 McGowan, A.S. (1977). Vocational maturity and anxiety among vocationally undecided and indecisive students: The effectiveness of Holland's Self-Directed Search. Journal of Vocational Behavior, 10(2), 196-204. McGowan, A.S. (1982). The predictive efficiency of Holland's SDS summary codes in terms of career choice: A four-year follow-up. Journal of Vocational Behavior, 20(3), 294-303. Meir, E.I. (1989). Integrative elaboration of the congruence theory. of Vocational Behavior, 35, 219-230.
Journal
Meir, E.I. (1994). Comprehensive interests measurement in counseling for congruence. Career Development Quarterly, 42(4), 314-325. Meir, E.I. & Melamed, S. (1986). The accumulation of person-environment congruences and well-being. Journal of Occupational Behaviour, 7(4), 315-323. Meir, E.I., Melamed, S. & Dinur, C. (1995). The benefits of congruence. The Career Development Quarterly, 43(3), 257-266. Meir, E.I., Keinan, G. & Segal, Z. (1986). Group importance as mediator between personality-environment congruence and satisfaction. Journal of Vocational Behavior, 28, 60-69. Meir, E.I. & Mayon, M. (1992). A longitudinal examination of congruence hypotheses. Journal of Vocational Behavior, 41, 35-47. Miller, M.J. (1992). Correlations among three measures of congruence. Measures and Evaluations in Counseling and Development, 25,113120. Miller, M.J. (1994). A "circuitous" test of Holland's theory. Employment Counseling, 31(3), 137-143.
Journal of
Miller, M.J., Springer, T.P. & Wells, D. (1988). Which occupational environments do Black youths prefer? Extending Holland's typology. School Counselor, 36f2), 103-106.
266
Mishra, P.C. & Gupta, J. (1994). Performance as a function of employees' motivation and job involvement. Psychological Studies, 39(1), 18-20. Misra, S., Kanungo, R.N., Von Rosenstiel, L. & Stuhler, E.A. (1985). The motivational formulation of job and work involvement: A cross-national study. Human Relations, 38(6), 501-518. Mitchell, T.R. (1979). Organizational behavior. Psychology, 30, 243-281.
Annual Reviews of
Moorman, R.H. (1993). The influence of cognitive and affective based job satisfaction measures on the relationship between satisfaction and organizational citizenship behavior. Human Relations, 46(6), 759-776. Morrow, P.C. (1993). Work Adjustment Theory: From a distance. Journal of Vocational Behavior, 43, 90-97. Mortimer, J.T. & Lorence, J. (1989). Satisfaction and involvement: Disentangling a deceptively simple relationship. Social Psychology Quarterly, 52(4), 249-265. Mount, M.K. & Muchinsky, P.M. (1978). Person-environment congruence and employee job satisfaction: A test of Holland's theory. Journal of Vocational Behavior, 13, 84-100. Mouton, J. & Marais, H.C. (1988). geesteswetenskappe: Basies begrippe.
Metodologie van die
Pretoria!: Raad vir
Geesteswetenskaplike Navorsing. Muchinsky, P.M. & Monahan, C.J. (1987). What is person-environment congruence? Supplementary versus complementary models of fit. Journal of Vocational Behavior, 31, 268-277. Murphy, K.R. (1993). Integrating research on work adjustment with research on job performance and behavior in organizations: Perspectives from Industrial/Organizational Psychology. Journal of Vocational Behavior, 43, 98-104.
267
Myors, B. (1996). A simple, exact test for the Holland hexagon. Journal of Vocational Behavior, 48, 339-351. Neethling, F.R.L. (1986).
'n Psigometriese ondersoek na Holland se
beroepskeuseteorie: Unpublished thesis. Potchefstroomse Universiteit
vir Christelike Hoer Onderwys, Potchefstroom. O'Brien, W.F, & Walsh, W.B. (1976). Concurrent validity of Holland's theory for non-college degreed Black working men. Journal of Vocational Behavior, 8(2), 239-246. O'Neil, J.M., Price, G.E. & Tracey, T.J. (1979). The stimulus value, treatment effects, and sex differences when completing the SelfDirected Search and Strong-Campbell Interest Inventory. Journal of Counseling Psychology, 26(1), 45-50. Oosterveld, P. (1994). Confirmatory factor analysis of the Self-Directed Search test: A multitrait-multimethod approach. Personality and Individual Differences, 17(4), 565-569. Orpen, C. (1995). The effects of perceived age discrimination on employee job satisfaction, organizational commitment and job involvement. Psychology: A Journal of Human Behavior, 32(3-4), 55-56. Osipow, S.H. (1987). Applying person-environment theory to vocational behavior. Journal of Vocational Behavior, 31(3), 333-336. Ostroff, C. (1993). Relationships between person-environment congruence and organizational effectiveness. Group and Organization Management, 18(1), 103 - 122. Owen, K. & Taljaard, J.J. (1988).
Handleiding vir die gebruik van
sielkundige en skolastiese toetse van IPEN en die NIPN. Pretoria:
Human Science Research Council.
268 Paterson, J.M. & O'Driscoll, M.P. (1990). An empirical assessment of Kanungo's (1982) concept and measure of job involvement. Applied Psychology: An International Review, 39(3), 293-306. Payne, M.A. & Sabaroch, H.F. (1985). Personality type and occupational preference: Testing Holland's theory in the Caribbean. International Journal for the Advancement of Counselling, 8(2), 147-156. Peiser, C. (1984). The use of Holland's typology and interest inventory in Israel. Journal of Social Psychology, 123(2), 169-178. Phelps, L.H. & VVaskel, S.A. (1994). Explanatory style and its functional relationship to job satisfaction for employed women age forty and over. Psychological Reports, 74, 899-902. Prediger, D.J. (1976). A world of work map for career exploration. The Vocational Guidance Quarterly, 24, 198-208. Prediger, D.J. (1980). On the virtues of raw-scored interest inventories: Reaction to O'Neil, Price, and Tracey (1979). Journal of Counseling Psychology, 27(3), 302-303. Prediger, D.J. (1980). The determination of Holland types characterizing occupational groups. Journal of Vocational Behavior, 16(1), 105-114. Prediger, D.J. (1981). A note on Self-Directed Search validity for females. Vocational Guidance Quarterly, 30(2), 117-129. Prediger, D.J. (1982). Dimensions underlying Holland's hexagon: Missing link between interests and occupations. Journal of Vocational Behavior, 21, 259-287. Prediger, D.J. (1996). Alternative dimensions for the Tracey-Rounds interest sphere. Journal of Vocational Behavior, 48, 59-67. Prediger, D.J. & Hanson, G.R. (1977). Some consequences of using rawscore reports of vocational interests. Journal of Educational Measurement, 14(4), 323-333.
269
Prediger, D.J. & Vansickle, T.R. (1992). Locating occupations on Holland's hexagon: Beyond RIASEC. Journal of Vocational Behavior, 40, 111128. Rabinowitz, S. & Hall, D.T. (1977). Organisational research on job involvement. Psychological Bulletin, 84(2), 265-288. Rachman, D., Amernic, J. & Aranya, N. (1981). A factor-analytic study of the construct validity of Holland's Self-Directed Search test. Educational and Psychological Measurement, 41(2), 425-437. Raphael, K.G. & Gorman, B.S. (1986). College women's Holland-theme congruence: Effects of self-knowledge and subjective occupational structure. Journal of Counseling Psychology, 33(2), 143-147. Riipinen, M. (1994). Occupational needs as moderators between locus of control and job involvement. Psychological Reports, 74, 371-379. Roberson, L. (1990). Prediction of job satisfaction from characteristics of personal work goals. Journal of Organizational Behavior, 11, 29-41. Robbins, P.1., Thomas, L.E., Harvey, D.W. & Kandefer, C. (1978). Career change and congruence of personality types: An examination of DOTderived work environment designations. Journal of Vocational Behavior, 13, 15-25. Roodt, G., Bester, C.L. & Boshoff, A.B. (1994). werkbetrokkenheid: 'n Bipolere kontinuum? Bedryfsielkunde, 20(4), 12-16.
Die grand van Tydskrif vir
Rounds, J.B. (1990). The comparative and combined utility of work value and interest data in career counselling with adults. Journal of Vocational Behavior, 37, 32-45. Rounds, J.B., Dawis, R.V. & Lofquist, L.H. (1987). Measurement of personenvironment fit and prediction of satisfaction in the Theory of Work Adjustment. Journal of Vocational Behavior, 31, 297-318.
270
Rounds, J. & Tracey, T.J. (1993). Prediger's dimensional representation of Holland's RIASEC circumplex. Journal of Applied Psychology, 78(6), 875-890. Rounds, J. & Tracey, T.J. (1996). Cross-cultural structural equivalence of RIASEC models and measures. Journal of Counseling Psychology, 43(4), 310-329. Ryan, J.M., Tracey, T.J.G. & Rounds, J. (1996). Generalizability of Holland's structure of vocational interests across ethnicity, gender, and socioeconomic status. Journal of Counseling Psychology, 43(4), 330337. Saal, F.E. (1981). Empirical and theoretical implications of a purely cognitive definition of job involvement. International Review of Applied Psychology, 30, 103-120. Saleh, S.D. (1981). A structural view of job involvement and its differentiation from satisfaction and motivation. International Review of Applied Psychology, 30, 17-29. Saleh, S.D. & Hosek, J. (1976). Job involvement: Concepts and measurements. Academy of Management Journal, 19(2), 213-224. Salomone, P.R. & Sheehan, M.C. (1985). Vocational stability and congruence: An examination of Holland's proposition. Vocational Guidance Quarterly, 34(2), 91-98. Scarpello, V. & Campbell, J.P. (1983). Job satisfaction and the fit between individual needs and organizational rewards. Journal of Occupational Psychology, 56, 315-328. Schaefer, B.E. (1976). Holland's SDS: Is its effectiveness contingent upon selected variables? Journal of Vocational Behavior, 8(1), 113-123. Schwab, D.P., & Heneman, H.G. (1977). Age and satisfaction with dimensions of work. Journal of Vocational Behavior, 10, 212-220.
271
Schwartz, R.H. (1991). Achievement-orientation of personality type: A variable to consider in tests of Holland's congruence-achievement and other hypotheses. Journal of Vocational Behavior, 38, 225-235. Schwartz, R.H. (1992). Is Holland's theory worthy of so much attention, or should vocational psychology move on? Journal of Vocational Behavior, 40(2), 179-187. Schwartz, R.H., Andiappan, P. & Nelson, M. (1986). Reconsidering the support for Holland's congruence-achievement hypothesis. Journal of Counseling Psychology, 33(4), 425-428. Sekaran, U. & Mowday, R. (1981). A cross-cultural analysis of the influence of individual and job characteristics on job involvement. International Review of Applied Psychology, 30, 51-64. Shari, R.S. (1992). Applying career development theory to counseling. Pacific Grove, CA: Brooks Cole Publishing Company. Sheffey, M.A., Bingham, R.P. & Walsh, W.B. (1986). Concurrent validity of Holland's theory for college-educated Black men. Journal of Multicultural Counseling and Development, 14(4), 149-156. Shivy, V.A., Phillips, S.D., & Koehly, L.M. (1996). Knowledge organization as a factor in career intervention outcome: A multidimensional scaling analysis. Journal of Counseling Psychology, 43(2), 178-186. Shore, T.H. Thornton, G.C. & Shore, L.M. (1990). Distinctiveness of three work attitudes: Job involvement, organizational commitment, and career salience. Psychological Reports, 67, 851-858. Shubsachs, A.P., Rounds, J.B., Dawis, R.V., & Lofquist, L.H. (1978). Perception of work reinforcer systems: Factor structure. Journal of Vocational Behavior, 13(1), 54-62.
272 Singh, P. (1994). Perceptions and reactions to inequality as a function of social comparison referents and hierarchical levels. Journal of Applied Social Psychology, 24(6), 557-565. Smart, J.C., Elton, C.F. & McLaughlin, G.VV. (1986). Person-environment congruence and job satisfaction. Journal of Vocational Behavior, 29, 216-225. Smith, R.A. & Davis, S.F. (1997).
The psychologist as detective: An
introduction to conducting research in psychology. Upper Saddle
River, NJ: Prentice Hall. Solly, D.C. & Hohenshil, T.H. (1986). Job satisfaction of school psychologists in a primarily rural state. School Psychology Review, 15(1), 119-126. Spokane, A.R. (1985). A review of research on person-environment congruence in Holland's theory of careers. Journal of Vocational Behavior, 26, 306-343. Steel, R.P. & Rentsch, J.R. (1995). Influence of cumulation strategies on the long-range prediction of absenteeism. Academy of Management Journal, 38(6), 1616-1634. Steiner, D.D. & Truxillo, D.M. (1987). Another look at the job satisfaction-life satisfaction relationship: A test of the disaggregation hypothesis. Journal of Occupational Behaviour, 8(1), 71-77. Souliere, M.D., MacPhee, D.C. & Flynn, R.J. (1991). Concurrent and predictive validity of Prediger's two work-task preference dimensions. Journal of Vocational Behavior, 39, 226-240. Strahan, R.F. (1987). Measures of consistence for Holland-type codes. Journal of Vocational Behavior, 31, 37-44. Sutherland, L.F., Fogarty, G.J. & Pithers, R.T. (1995). Congruence as a predictor of occupational stress. Journal of Vocational Behavior, 46, 292-309.
273
Swaney, K. & Prediger, D. (1985). The relation between interest-occupation congruence and job satifaction. Journal of Vocational Behavior, 26, 1324. Swanson, J.L. & Hansen, J-I. C. (1986). A clarification of Holland's construct Journal of of differentiation: The importance of score elevation. Vocational Behavior, 28, 163-173. Tang, T.L-P., & Gilbert, P.R. (1995). Attitudes toward money as related to intrinsic and extrinsic job satisfaction, stress and work-related success. Personality and Individual Differences, 19(3), 327-332. Taljaard, J.J. & Von Mollendorf, J.W. (1987). The South African dictionary of occupations. Pretoria: Consensus Publishers. Tenopyr, M.L. (1993). Construct validation needs in vocational behavior theories. Journal of Vocational Behavior, 43, 84-89. Thompson, C.A. & Blau, G. (1993). Moving beyond traditional predictors of job involvement: Exploring the impact of work-family conflict and overload. Journal of Social Behavior and Personality, 8(4), 635-646. Thompson, C.A., Kopelman, R.E. & Schriesheim, C.A. (1992). Putting all one's eggs in the same basket: A comparison of commitment and satisfaction among self- and organizationally employed men. Journal of Applied Psychology, 77(5), 738-743. Thompson, J.M. & Blain, M.D. (1992). Presenting Feedback on the Minnesota Importance Questionnaire and the Minnesota Satisfaction Questionnaire. The Career Development Quarterly, 41, 62-66. Thompson, J.M., Flynn, R.J. & Griffith, S.A. (1994). Congruence and coherence as predictors of congruent employment outcomes. The Career Development Quarterly, 42, 271-281.
274 Tinsley, D.J. (1993). Extensions, elaborations, and construct validation of the Theory of Work Adjustment. Journal of Vocational Behavior, 43, 67-74. Tinsley, H.E.A. & Tinsley, D.J. (1989). Reinforcers of the occupation of homemaker: An analysis of the need-gratifying properties of the homemaker occupation across the stages of the homemaker life cycle. Journal of Counseling Psychology, 36(2), 189-195. Tinsley, D.J., Tinsley, H.E.A., Boone, S. & Shim-Li, C. (1993). Prediction of scientist-practitioner behavior using personality scores obtained during graduate school. Journal of Counseling Psychology, 40(4), 511-517. Tokar, D.M. & Swanson, J.L. (1995). Evaluation of the correspondence between Holland's vocational personality typology and the five-factor model of personality. Journal of Vocational Behavior, 46, 89-108. Tracey, J.G. & Rounds, J. (1992). Evaluating the RIASEC circumplex using high-point codes. Journal of Vocational Behavior, 41, 295-311. Evaluating Holland's and Gati's Tracey, J.G. & Rounds, J. (1993). vocational-interest models: A structural meta-analysis. Psychological Bulletin, 113(2), 229-246. Tracey, J.G. & Rounds, J. (1995). The arbritrary nature of Holland's RIASEC types: A concentric circles structure. Journal of Counseling Psychology, 42, 431-439. Tracey, J.G. & Rounds, J. (1996a). The spherical representation of vocational interests. Journal of Vocational Behavior, 48, 3-41. Contributions of the spherical Tracey, J.G. & Rounds, J. (1996b). representation of vocational interests. Journal of Vocational Behavior, 48, 85-95. Tranberg, M., Slane, S. & Ekeberg, S.E. (1993). The relation between interest congruence and satisfacton: A meta-analysis. Journal of Vocational Behavior, 42(3), 253-264.
275
Tuck, B.F. & Keeling, B. (1980). Sex and cultural differences in the factorial structure of the Self-Directed Search. Journal of Vocational Behavior, 16(1), 105-114. Upperman, P.J. & Church, A.T. (1995). Investigating Holland's typological theory with army occupational specialities. Journal of Vocational Behavior, 47, 61-75. Van der Merwe, R.P., Le Roux, J.A., Meyer, J.C. & Van Niekerk, H.G. (1990). Predicting Holland occupational codes by means of PAQ job dimension scores. Journal of Industrial Psychology, 16(3), 27-31. Vandenberg, R.J. & Scarpello, V. (1990). The matching model: An example of the processes underlying realistic job previews. Journal of Applied Psychology, 75(1), 60-67. Varca, P.E. & James-Valutis, M. (1993). The relationship of ability and satisfaction to job performance. Applied Psychology: An International Review, 42(3), 265-275. Vontress, C.E. & Naiker, K.S. (1995). Counseling in South Africa: Yesterday, today, and tomorrow. Journal of Multicultural Counseling and Development, 23(3), 149-157. Walsh, W.B. & Betz, N.E. (1995). Tests and assessment (Third Edition). Englewood Cliffs, New Jersey: Prentice Hall. Walsh, W.B., Woods, W.J. & Ward, C.M. (1986). Holland's theory and working Black and White women. Journal of Multicultural Counseling and Development, 14(3), 116-123.. Weiss, E.J., Dawis, R.V., England, G.W. & Lofquist, L.H. (1967). Manual for the Minnesota Satisfaction Questionnaire. St. Paul, MN: Work Adjustment Project, Industrial Relations Centre, University of Minnesota.
276 Weissenberg, P. & Gruenfeld, L.W. (1968). Relationship between job satisfaction and job involvement. Journal of Applied Psychology, 52(6), 469-473. Wiener, V., Muczyk, J.P. & Martin, H.J. (1992). Self-esteem and job involvement as moderators of the relationship between work satisfaction and well-being. Journal of Social Behavior and Personality, 7(4), 539-554. Wiggins, J.D. (1984). Personality-environmental factors related to job satisfaction of school counselors. Vocational Guidance Quarterly, 33(2), 169-177. Wiggins, J.D. & Moody, A. (1981). Compatibility index description. Dover, D.E.: Training and Associates. Wiggins, J.D., Lederer, D.A., Salkowe, A. & Rys, G.S. (1983). Job satisfaction related to tested congruence and differentiation. Journal of Vocational Behavior, 23, 112-121. Witt, P.H., & Handal, P.J. (1984). Person-environment fit: Is satisfaction predicted by congruency, environment, or personality? Journal of College student Personnel, 25(6), 503-508. Wolfe, L.K. & Betz, N.E. (1981). Traditionality of choice and sex-role identification as predictors of the congruence of occupational choice in college women. Journal of Vocational Behavior, 18, 43-55. Zener, T.B. & Schnuelle, L. (1976). Effects of the Self-Directed Search on high school students. Journal of Counseling Psychology, 23(4), 353359.
277
ANNEXURE A CHARACTEMSMCS OF THE HOLUM PERSORIAUTY TYPES Preferences, aversions and competencies
Personality Type
Preferences
Aversions
Competencies
Realistic
Realistic situations or occupations (e.g. electrician): Manipulation of objects, tools, machines, animals
Social situations or occupations: Educational and therapeutic activities
Realistic problem solving abilities: Manual, mechanical, agricultural, electrical and technical
Investigative
Investigative situations or occupations (e.g. psychologist): Investigation of physical, biological and cultural phenomena
Enterprising situations or occupations: Persuasive, social and repetitive activities
Investigative problem solving abilities: Scientific and mathematical
Artistic
Artistic situations or occupations (e.g. journalist): Manipulation of physical, verbal and human materials to create art products
Conventional situations or occupations: Explicit, systematic and ordered activities
Artistic problem solving abilities: Language, art, music, drama, writing
Social
Social situations or occupations (e.g. Personnel manager): Manipulation of others to inform, train, develop, cure or enlighten
Realistic situations or occupations: Activities involving materials, tools or machines
Social problem solving abilities: Interpersonal and educational
278
Annexure A (continued)
Personality Type
Preferences
Aversions
Competencies
Enterprising
Enterprising situations or occupations (e.g. manager): Manipulation of others for organisational goals or economic gain
Investigative situations or occupations: Observational, symbolical and systematic activities
Enterprising problem solving abilities: Leadership, interpersonal and persuasive
Conventional
Conventional situations or occupations (e.g. accounting): Manipulation of data (e.g. filing materials, operating business machines)
Artistic situations or occupations: Ambiguous, free, exploratory or unsystematised activities
Conventional problem solving abilities: Clerical, computational and business system
(Holland, 1985b).
279
Annexure A (continued) Self-perceptions, values and traits
Personality Type
Self-perceptions
Personality Traits
Values
Realistic
Mechanical and athletic abilities
The concrete and tangible: Money Power Status
Asocial Conforming Frank Genuine Hard-headed Materialistic Natural Normal
Normal Persistent Practical Self-effacing Inflexible Thrifty Uninsightful Uninvolved
Investigative
Scholarly Intellectual Mathematical and scientific ability
Science
Analytical Cautious Critical Complex Curious Intellectual Pessimistic
Precise Rational Reserved Retiring Unpopular
Artistic
Expressive Original Intuitive Nonconforming Introspective Disorderly Artistic, musical, acting, writing and speaking ability
Aesthetic qualities
Disorderly Emotional Expressive Idealistic Imaginative Impractical Impulsive
Intuitive Nonconforming Original Sensitive Open
280
Annexure A (continued)
Personality Type
Self-perceptions
Personality Traits
Values
Social
Like to help others Understand others Teaching abilities
Social and ethical activities and problems
Ascendant Co-operative Patient Friendly Generous Helpful Idealistic
Emphatic Kind Persuasive Sociable Tactful Warm
Enterprising
Aggressive Popular Self-confident Sociable Leadership and speaking abilities
Political and economic achievement
Acquisitive Agreeable Ambitious Energetic Excitementseeking Extroverted
Flirtatious Optimistic Selfconfident Sociable Talkative
Conventional
Conforming Orderly Clerical and numerical ability
Business and economic achievement
Careful Conforming Defensive Efficient Inflexible Inhibited Methodical Obedient
Orderly Persistent Practical Prudish Thrifty Unimaginative
(Holland, 1985b).
281
AMMEXURE fa IIRIFLUERICES OF HOLUM EMVDROMMEMTAL TYPES
Environmental
Perception of the
Type
World
Influences on members
Attractions
Methods of coping with others
Realistic
Simple, tangible and traditional ways
Pragmatic and traditional
Realistic occupations
Simple, direct, natural — less adept
Investigative
Complex, abstract, independent and original ways
Abstract, theoretical and analytical
Investigative occupations
Rational, analytic and indirect methods
Artistic
Complex, independent, unconventional and flexible ways
Personal, emotional and imaginative
Artistic occupations
Personal, emotional, expressive, unconventional methods
Social
Flexible ways
Social, humanitarian and religious
Social occupations
Friendly, helpful, co-operative
Enterprising
Stereotyped, constricted, dependent and simple ways
Social, emotional and materialistic
Enterprising occupations
Enterprising way, including dominance, talkativeness
Conventional
Conventional, stereotyped, simple and dependent ways
Materialistic influences: money, position, power
Conventional occupations
Controlling, conforming, practical
(Holland, 1985b).
282
JE C PROPOSITOOMS AHD COROLLAMES OF THE THEORY OF WORK ADJUSTMERIT
I
Work adjustment at any time is indicated by an individual's concurrent levels of satisfactoriness and satisfaction.
II
Satisfactoriness is a function of the correspondence between the individual's abilities and the ability requirements of the work environment, provided that the reinforcer pattern of the work environment correspond to the individual's values. Corollary ll a. Knowledge of the abilities of individuals and their satisfactoriness permits determination of the effective ability requirements of the work environment. Corollary ll b. Knowledge of the ability requirements of the work environment and of an individual's satisfactoriness permits the inference of the individual's abilities.
III
Satisfaction is a function of the correspondence between the reinforcer pattern of the work environment and the individual's values, provided that the individual's abilities correspond to the ability requirements of the work environment. Corollary Ill a. Knowledge of the values of individuals and their satisfaction permits the determination of the effective reinforcer pattern of the work environment. Corollary Ill b. Knowledge of the reinforcer pattern of the work environment and of an individual's satisfaction permits the inference of the individual's values.
283 Annexure C (continued)
IV
Satisfaction moderates the functional relationship between satisfactoriness and ability-requirement correspondence.
V
Satisfactoriness moderates the functional relationship between satisfaction and value-reinforcer correspondence.
VI
The probability that an individual will be forced out of the work environment is inversely related to the individual's satisfactoriness.
VII
The probability that an individual will voluntarily leave the work environment is inversely related to the individual's satisfaction.
VIII
Tenure is a joint function of satisfactoriness and satisfaction. Corollary VIII a. Tenure is a function of correspondence between abilities and requirements and between values and reinforcers.
IX
Correspondence between work personality and work environment increases as a function of tenure.
X
The correspondence between work personality style and work environment style moderates the prediction of work adjustment for the correspondence between work personality structure and work environment structure.
(Lofquist & Dawis, 1969, p. 50-53).
284
ANNEXURE D ADDOTDONAL PROPOSOTOORIS OF THE THEORY OF WORK ADJUSTMENT
XI
The flexibility of the work environment moderates the functional relationship between satisfactoriness and ability-requirement correspondence.
XII
The flexibility of the individual moderates the functional relationship between satisfaction and value-reinforcer correspondence.
XIII
The probability that the work environment will use adjustment modes is inversely related to the individual's satisfactoriness.
Corollary XIII a. Knowledge of this probability associated with the individual's satisfactoriness permits the determination of the work environment's flexibility threshold.
XIV
The probability that the individual will use
adjustment modes is inversely
related to the individual's satisfaction.
Corollary XIV a. Knowledge of this probability associated with the individual's satisfaction permits the determination of the work environment's flexibility threshold.
285 Annexure D (continued)
XV
The probability that the work environment will act to remove the individual is inversely related to the perseverance level of that work environment.
Corollary XV a. Knowledge of this probability associated with the work environment's perseverance permits the determination of the work environment's perseverance threshold.
XVI
The probability that the individual will act to leave the environment is inversely related to the perseverance level of that individual.
Corollary XVI a. Knowledge of this probability associated with the individual's perseverance permits the determination of the individual's perseverance threshold.
XVII
Tenure is a function of satisfactoriness, satisfaction, and the perseverance levels of the individual and of the work environment.
(pawls & Lofquist, 1984, p. 67-68).