o. .Io—“‘p“_.¢-—§.Co I-Q“‘“~”-wfi V

ANTECEDENTS AND CONSEQUENCES
OF JOB INVOLVEMENT

Thesis for the Degree of M‘A.‘
MICHEGAN STATE UNIVER-sm
ALAN LAWRENCE SIEGEL

1.971

293 10147

mmumnu mum ”111%!!!an

 

 

 

 

 

 

ABSTRACT

ANTECEDENTS AND CONSEQUENCES OF JOB INVOLVEMENT

By

Alan Lawrence Siegel

This study was designed to investigate relationships

between the construct Job involvement and both personal-

background variables and job design characteristics. The

most fruitful framework within which to focus such an

investigation consisted of examining possible antecedents

of involvement, in the form of background and situational

variables, and also consequences, such as performance,

absenteeism, and turnover. It was to this end that the

following hypotheses were formulated:

1.

2.

ha.

Job involvement will be positively related to the
degree of participation in decision making which
the Job provides the individual.

Job involvement will be positively related to
level of education.

The relationship between job involvement and
participation in decision making will be more
positive for more highly educated individuals
than for those individuals with less education.

The relationship between Job involvement and
participation in decision making will be more
positive for workers from a rural background
than for those of an urban background.

There will be mean differences in Job involvement
between individuals of urban and rural backgrounds,
with higher involvement for the rural workers.

Alan Lawrence Siegel
5. The relationship between Job involvement and
participation in decision making will be more
positive for those individuals with low religious
devotion than for those with high religious
devotion.

5a. Job involvement will be positively related to
intensity of religious devotion.

6. The relationship between Job involvement and
participation in decision making will be more
positive for those individuals with a high
intrinsic orientation than for those with a high
extrinsic orientation.

7. Job involvement will be positively related to
productivity.

8. Job involvement will be negatively related to
absenteeism.

9. Job involvement will be negatively related to
turnover.

The subject sample consisted of 2628 employees of
six medium sized manufacturing companies in the Midwest. Of
this sample various sub-samples were used in investigating
the predicted moderated relationships. Simple correlational
techniques were employed in testing the results.

Strong support was found for Hypothesis 1, but no
support was generated for Hypothesis 2. Several of the
variables did moderate the relationship between Job
involvement and participation in decision making. However,
only level of education and intrinsic orientation (Hypotheses
3 and 6) moderated the relationship in the predicted
direction. Urban-rural background (Hypothesis 4) moderated
the relationship but the results were not in the predicted
direction. Similarly, Hypothesis ha was not supported, in
that Job involvement was higher for the urban than for the

Alan Lawrence Siegel
rural group. Intensity of religious devotion was the only
one of the variables which did not moderate the relationship.
On the other hand, job involvement was positively related
to the intensity of religious devotion. In regard to the
consequences of job involvement, only turnover exhibited a

significant relationship with Job involvement.

Thesis Committee:

Dr. Frederic Wickert, Chairman Approved:
Date:

 

 

Dr. Frank Schmidt
Dr. Robert Buh

ANTECEDENTS AND CONSEQUENCES OF JOB INVOLVEMENT
By

Alan Lawrence Siegel

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology
1971

DEDICATION

To my wife Ronni: She has given me the confidence to

persevere and has endured for the past two years in one

of the most difficult roles, wife of a graduate student.

11

ACKNOWLEDGEMENT

To acknowledge everyone who has made a contribution to
this thesis and to my first two years in graduate school
would require a listing of almost all the faculty and
graduate students with whom I have had any contact. My
committee has formally consisted of three, but the inputs
of those around me have been frequent and worthwhile,

Dr. Fred Wickert, my committee chairman, has been an
invaluable source, not only throughout the many stages of
the thesis but for the last year which I have spent working
with him. Any stylistic polish exhibited throughout these
pages could not have been achieved without his outstanding
assistance.

Dr. Frank Schmidt was always available whenever questions
concerning design and measurement arose. In addition, his
many suggestions between rough and final drafts have
resulted in a tighter end product.

Dr. Robert Huh provided much of the impetus and general
direction from the initial planning stages to the completion
of this thesis. He was a constant source of encouragement
whenever ”theoretical confusion" set in. Also, his rare
ability to get right to the focal point of a problem was

appreciated.

111

 

AIIIIII‘III'D-

.lllllyl/[II‘AIAflI

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . .
LIST OF FIGURES . . . . . . . . . . . . . . . .
REVIEW OF THE THEORETICAL LITERATURE . . . . . .
REVIEW OF THE EMPIRICAL LITERATURE . . . . . . .
CONCEPTUAL FRAMEWORK AND HYPOTHESES . . . . . .
METHOD . . . . . . o . . . . . . . . . . . . . .
RESULTS AND DISCUSSION . . . . . . . . . . . . .
CONCLUSIONS AND IMPLICATIONS FOR FUTURE RESEARCH

BIBLIOGRAPHY o o o o e o o o e o o o o o o I o 0

iv

Page

vi

17
35
43
53
68

71

"lx‘l‘rsllll' all!" I :.

[fl-I'll» {Ill aIII-I

' [lull I‘ /. 1 I :III » l Illlly. Jill .I

LIST OF TABLES
Page

Table
1. Median and range of inter-item correlations,
item-cluster correlations, and item
reliabilities
o e 50

communalities. Scale
(alphas) are also included.

Summary of the relationship between Job

2.
involvement and FUN moderated by
level of education. . . . . . . . . . . . 56
3. Summary of the relationship between Job
involvement and PDM moderated by
O O O O O O O O 58

urban-rural background. .
A. Summary of the relationship between Job

involvement and PDM moderated by

intensity of religious devotion . . . . . 61

Summary of the relationship between Job

5.
involvement and PDM moderated by
intrinsic orientation . . . . . . . . . . 63
6. Summary of the relationship between Job
involvement and PDM moderated by 6
. . 3

extrinsic orientation

LIST OF FIGURES

Figure ' Page
1. Hypothesized antecedents and consequences

of job involvement. . . . . . . . . . 37

vi

Review of the Theoretical Literature

In an attempt to understand man and his motivation to
work, attitudinal and motivational theorists in industrial-
social psychology and sociology have generated an abundance
of terms indicative of their lack of consensus as to what
motivates man in the work setting. Terms such as commitment,
dedication, ego involvement, intrinsic motivation, identifi-
cation with work role, and Job involvement have frequently
appeared throughout the literature with only minimal attempts
to reconcile the similarities and differences inherent in
these various theoretical constructs. However, there do seem
to be common threads running throughout the various constructs,
as the theorists have tended to focus on the following:

1) the extent to which the Job is an integral part of the
individual's self-concept, the intensity of his dedication to
his work, and his interest in his work role: or 2) the extent
to which performance on the Job affects his feelings of
self-esteem, and the degree to which high performance is
valued by the worker. Although it is possible to distinguish
between the various conceptualizations, most theorists have
tended to include both of the above conceptualizations in
their theorizing while at the same time focusing on one

aspect of the complexity.

2

Vroom (1962), who is concerned with motivation and
its relationship to worker performance, defines ego
involvement with regard to the effect performance has on
an individual's self-esteem. A highly involved worker is
one whose self-evaluation is increased by high performance
and decreased by low performance. He seems to draw little
distinction between what others have termed internalized
motivation and ego involvement. An individual is internally
motivated to the extent that his performance is relevant
to the maintenance of his self-identity. That is, he
conceptualizes a situation in.which some people may derive
satisfaction from a high level of performance even though
external rewards are absent. It is also possible that they
may experience dissatisfaction from a low level of
performance in the absence of punishment. A person becomes
ego involved in his work performance to the extent that
performance is perceived to be relevant to certain aptitudes,
abilities, or other attributes which are central to his
self-conception. It therefore becomes necessary to take
into consideration not only the individual but also the
organizational constraints inhibiting this relevancy.
Factors which are a function of the characteristics of
both the individual and the work setting, such as autonomy
and degree of participation, can thus be assumed to play
an important role. It is with Vroom's theorizing about
individual differences that we see a glimpse of the first
definition of Job involvement in the preceding paragraph.

3

He states that for some individuals the major part of the
self-concept is composed of attributes which are relevant

to the work situation, while for others the central

components relate to other spheres of activity (Vroom,

1962). In Vroom's terminology, what is being measured are
"differences in the centrality of the work relevant attributes
in the self-conception of the worker.“

Others who have viewed the construct similarly include
Faunce (1959), Gurin et a1. (1960), and French and Kahn
(1962). Faunce uses the term occupational involvement in
referring to the extent to which success or failure in one‘s
occupational role affects the self-image. It is viewed as a
commitment to a task in which "successful role performance
is regarded as an end in itself and not a means to some
other end." Likewise, Gurin views involvement as an
indicator of the degree to which the individual validates
performance in terms of the degree to which performance
affects self-esteem.

Lodahl and KeJner (1965), who pioneered in the
construction of a Job involvement scale, seem to fall prey
to the complexity of this theoretical issue since they
verbalize two definitions of involvement without ever
discussing why they settled on these two or explicating
the differences between them. They initially discuss the
construct in terms similar to those used by Vroom, referring
to the effect performance has on self-esteem. It is only

in the abstract of their article that their second definition

4
appears. Here they define Job involvement as ”the degree

to which a person is identified psychologically with his
work, or the importance of work in his total self-image.”
Lodahl and Kejner do not refer to the esteem-performance
definition in the abstract at all, and the reader is left
wondering on which of the two definitions they were

actually focusing.

Despite this confusion, their theoretical labors do
have value. We see for the first time the introduction of
background factors as an attempt to explain the development
or causes of Job involvement. They hypothesize that the
main determinant of this attitude may be a value orientation
toward work which is learned by the individual during early
socialization. It may thus be the result of the intro-
Jection of certain work values into the self (Lodahl, 196A).
In line with this assumption is their speculation that Job
involvement, as a value orientation within the individual,
may be resistant to changes due to the nature of a Job.
Although they do not rule out the effect which.work condi-
tions may have on an individual's Job involvement, they do
seem to imply that it is a deep-seated value orientation
that is developed early in the life of the individual and
thus may be difficult to influence. The local organizational
conditions which they feel may possibly have an effect are
mainly social ones. This thought stems from their view
of the Job-involved individual as possibly having stronger
affiliative needs and needing to identify with the

5

organization. A good deal of social motivation seems to
run throughout their overall composite "picture” of the
Job-involved individual. Lodahl (196“) goes on to state
that the tendency to identify with the Job, company, or
product may be part of a general social obligation process
having its basis in the ”incorporation of the values and
attitudes of the generalized others in the self.” This
interpretation, although overlapping to an extent with
the theories of Vroom and others to be discussed, does
seem to approach an awareness of the complexity which may
have to be explored if there is to be progress in this
aspect of attitude-motivational research.

Although Katz and Kahn (1966) spend more time on
organizational involvement, or identification, than on
Job involvement as we have defined it, they do touch on
the subject using dissimilar terms. In discussing the
character of commitment of people to the system. they
distinguish between a purely instrumental versus a purely
expressive cycle. The latter concept is somewhat related
to what Vroom (1962) was discussing under the heading of
intrinsic motivation. Katz and Kahn state that when an
activity is intrinsically rewarding it is ”directly
expressive of the needs and values of the individual.“
They also discuss motivation to work using the terms
degree of inclusion, potency of involvement, and priority
of commitment. What they are referring to by inclusion is

the extent to which organizations are based on the

—.l.l I. \{ili [ill '3

y .1. ll. J'I ll . o.‘ In! 1 l: lull. Ill l ‘l

6
sequential involvement of individuals. Only a small part
of a person's personality or life space is occupied by his
organizational role.

Like Lodahl and Kejner (1965), Katz and Kahn state that
this commitment may have been built in during the sociali-
zation process. They also discuss the effect which society
in general may have had in fostering partial inclusion or
segmental involvement of individuals. As a result of the
depersonalization and fragmentation created by bureaucratic
systems, the individual is forced to play many roles.
Because these roles do not demand all of an individual's
personality, the individual is freer than in the past to
pattern his life as he chooses. His Job no longer has to
be his main source of self-identification. Work no longer
looms as the sole expression of one's personality. It
would thus appear that Katz and Kahn are focusing more on
Job involvement as an expression of the extent to which
work is a part of the individual‘s self-concept, or his
interest in his work role as an expression of himself.

Lawler and Hall (1970) also focus more on the psycho-
lOgical identification with work. They have been quite
explicit in drawing distinctions between the two definitions
of Job involvement. In an attempt to distinguish between
various Job attitudes, they defined Job involvement using
Lodahl and KeJner's (1965) definition of the degree of
psychological identification with one's work. The other

frequently used conceptualization of involvement, centering

7

on the effects of performance on self-esteem, was considered
to be a measure of intrinsic motivation. More explicitly,
intrinsic motivation was referred to as the ”degree to which
a job holder is motivated to perform well because of some
subjective rewards or feelings that he expects to receive

or experience as a result of performing well.” This is
essentially a restatement of Vroom's earlier definition.
Lawler and Hall decided on this interpretation due to the
fact that intrinsic motivation beliefs fit nicely into an
expectancy theory framework. They stated that intrinsic
motivation should be related to performance. However, no
such assumptions regarding job involvement were readily
apparent. As others had suggested, Lawler and Hall
theorized that job involvement might be an individual
difference factor, not readily influenced by situational
factors, and questioned the impact it might have on worker
behavior.

Some of the most recent theorizing on this topic evolves
from the work of Patchen (1970). Using what he terms job
motivation indices, he is concerned with the general devotion
of energy to job tasks. One of these indices, which
supposedly taps general interest in the job, is very similar
to what we have been calling job involvement in the psycho-
logical identification, or dedication, aspects of the term.
Patchen admittedly views the concept of job involvement as
merely a convenient label summarizing several characteristics

which make the job more important and potentially satisfying.

8
The lack of knowledge regarding this construct is thus
embodied in his conceptualization. In a very general sense
he is conerned with personal goals (motivations and feelings
on the job). The job-involved individual is highly
motivated and feels a sense of pride in his work. This
viewpoint is closely related to the performance-esteem
aspects to which we have previously referred. Related to
this is his interest in the internal motivation to perform
a job well which was labeled intrinsic motivation by
Lawler and Hall (1970). Patchen also uses a measure,
termed “identification with one's occupation,” which is
purportedly measuring how important one's occupational
role is to one's self-image. This too can be seen to be
very closely allied to the first definition of involvement.
In addition, however, he hypothesizes that one who is
highly ”identified” will be more likely to feel pride in
job achievement than someone whose occupation plays a
minor role in his self-identification. It can thus be
seen that Patchen's theorizing is of a very general
nature and includes aspects of both popular conceptions
of job involvement.

In addition to the work of these theorists who have
directly focused their attention on job involvement, or
some similar term, the work of the participative management
theorists is also very applicable to this review. The
concepts of involvement, commitment, and dedication enter

into their discussions as a consequence of the fulfillment

9
of ego needs by the job. As such, most of their theorizing

does not center on job involvement, but rather on expli-
cating the need for ego need gratification and ways of
providing such gratification. However, since much of the
work on job involvement has concerned itself with man's
internal motivation and his identification with work, it
is beneficial to briefly review these theorists' contri-
butions.

Argyris (1964) states that one necessary condition for
the existence of ”psychological success" is internal
commitment to work that is meaningful. “Psychological
success," which is viewed as equivalent to self-esteem, is
dependent upon the individual experiencing: 1) self-
responsibility (self-control), 2) internal commitment to
meaningful work, and 3) utilization of the worker's
important abilities. More commonly experienced in industry,
however, is psychological failure which results from
organizational demands requiring individuals to be dependent
and submissive. Non-involvement is viewed as an adaptive
defense mechanism to the constant frustration. conflict,
and failure which he experiences on the job. In line with
what Lodahl and Kejner (1965). Lawler and Hall (1970), and
Katz and Kahn (1966) have stated, Argyris also feels that
workers have possibly learned through work experience or
early socialization not to expect much self-expression in
the work environment. One way to possibly respond to this
situation is to reduce the psychological importance of the

work. This theorizing touches upon the first definition of

10
job involvement. As Argyris states, the workers may
unconsciously say, "I'll block my need for self-actual-
ization until I get out of work. Then I will live." He
feels that a psychological set may be created in which the
worker feels that causes of dissatisfaction are built into
the job and that if he is to experience any satisfaction
it will only be achieved outside of the work setting.
According to either of the definitions though, such an
individual can hardly be said to be involved in his job.
This cycle from conflict between organizational demands
and "healthy” individual needs to feelings of apathy and
non-involvement leads to a decrease in the psychological
importance of work and an emphasis on the material aspects.
Argyris starts from the assumption that most individuals
desire these higher order needs which lead to involvement
in the job. He feels that the central factors causing
non-involvement are related to organizational structure and
technology. Those individuals who do not desire psycho-
logical success are in a small minority and have actually
repressed their true feelings due to past and present
deprivations in the work environment. Thus, although he
does take past experience into account, he is much less
concerned with individual differences in value orientations
and tends to focus more on the work environment itself. Any
individual differences would be due more to work experience
deprivation than to socio-cultural values. Every individual

wants to be involved, wants his job to be an expression of

11
his self, and wants to be satisfied with his job. However.
barriers and constraints presented by the organization
inhibit the realization of these needs. Thus, this viewpoint
differs from other theorists discussed who have not gone as.
far as proclaiming that the need to be involved is present
in all individuals, either consciously or unconsciously.

Another proponent of participative management viewpoints
is McGregor (1960) who states that the typical rewards
provided the worker for satisfying his needs can be used
only when he leaves the job. Wages, fringe benefits, etc.
yield needed satisfaction only away from the work situation,
creating an environment in which work is not an interest
central to the individual. For McGregor the major
deficiency of management practices is the over-emphasis
placed upon hierarchical authority. Within the framework
heplaces most of the emphasis not on an individual
difference factor but on the limitations imposed by the
structure of the organization. The most likely consequence
of the implementation of traditional management philosophy
(Theory X) is passive acceptance, indifference, or
resistance. These are terms which can be seen to have a
relation to job involvement.

Some of the most elaborate criticisms of the work of the
participative management theorists (Argyris and McGregor)
have come from Hulin and Blood (1968) whose theorizing has
much relevance to the concept of job involvement. While

participative management theorists have tended to disregard

12
the whole notion of individual differences, Hulin and Blood
have made this the focal point of their conceptualization.
They feel that as a result of extra-work socialization
processes, many blue-collar workers in urban industrial
environments have no desire for ego need gratification
while on the job. In contrast to the job—involved indi-
vidual, these workers mainly view their jobs as a means
to an end. Work enables them to satisfy their main needs
off the job. They are thus refuting the participative
management assumptions regarding workers' needs. The
focus of Hulin and Blood‘s work is ”alienation from middle-
class norms.” Closely approximating the Protestant Ethic,
middle-class norms are beliefs concerning the goodness of
work and man's spiritual necessity to indulge in hard labor
for the glory of God. It is the authors' contention that
living in an urban, industrialized, blue-collar environment
leads to alienation from these norms, while life in a rural
non-industrial community or other settings does not. As an
outgrowth of their thesis we find that for rural workers
the more satisfying jobs would demand greater personal
involvement, while city workers are more likely to be
satisfied when their jobs are less personally involving.
As Blood and Hulin (1967) state, ”At the integrated end of
the continuum which ranges from integration with middle-
class norms to alienation from middle-class norms are
individuals who desire personal involvement with their
jobs.” The job could be expected to play a more important

central role to the individual at this end of the continuum.

13
Workers at the opposite end have only an instrumental
involvement. Increased responsibility, autonomy, etc. are
not unfulfilled needs: they have no such needs.

Thus we find a very basic, decisive difference between
the theoretical assumptions of the participative management
theorists and Hulin and Blood, which in turn leads to
differences in their conceptualization regarding the concept
of job involvement. The former place minimal emphasis on
job involvement as a value orientation or motivational set
subject to individual differences. Instead, they view the
organization as blocking the gratification of ego needs, a
result which leads to the absence of individual involvement
on the job. Thus. they are viewing involvement as a
consequence of ego need gratification. Hulin and Blood.
however, view involvement not so much as a consequence.
but as an antecedent embodied in a general value orientation
fostered by processes of socialization. Their concept,
within an individual difference, socio-cultural framework,
seems to closely parallel other such conceptualizations
which we have discussed to a greater extent than does the
work of the participative management theorists.

Blauner (1964), in discussing the industrial condition
termed alienation, overlaps to a certain extent both the
work of the participative management theorists and Hulin
and Blood. He refers to alienation as a general condition,
or composite, consisting of various different objective
conditions and subjective feeling states which emerge from

relationships between the workers and the work setting.

1n
Just as Katz and Kahn emphasize the depersonalization
caused by society and limitations imposed by the industrial
environment, Blauner discusses four types of alienation.
One of these, self-estrangement, is related conceptually
to job involvement. Self-estrangement is described as a
condition in which the individual experiences deperson-
alized detachment rather than immediate involvement or
engrossment in his job. As a result of this lack of
present time involvement, the work becomes mainly instru-
mental, a means toward future off-the-job considerations
rather than an end in itself. Lack of control over work
processes and inability to develop a sense of purpose to
the overall functioning of the organization are cited as
causes of alienation. On the other hand, when the work
is inherently fulfilling, when it satisfies needs for
control and meaning, or when it is highly integrated into
an individual's total social commitments, self-estrangement
is absent. Non-involving work is viewed as monotonous and
results in heightened time-consciousness. The similarities
between these assumptions and those of the participative
management theorists are readily apparent. Like Hulin and
Blood, however, Blauner does seem to take into account
individual differences, stating that self-estranged workers
are dissatisfied only when they have developed higher order
needs, such as for control and meaning in their work. He
makes the important point of noting that a worker can be
satisfied and self-estranged at the same time. It is also

possible for a worker to be integrated in the plant

. 1 5
community and loyal to the company but nevertheless fail
to have a sense of involvement or self-expression in the
work itself. Blauner makes no mention of the performance-
esteem aspects of involvement but does state that work is
not self-estranging when it is self-expressive and
contributes to the worker's personal identity and selfhood.
Thus he seems to emphasize the first definition of
involvement.

This distinction between the viewpoints of the partic-
ipative management theorists and their critics seems to have
been crystallized in what Schein (1965) terms “complex man”
assumptions about man. His theory points to the need to
take into account the full complexity of man and, as such,
refutes the participative management assumptions which
neglect individual differences and only serve to stereotype
man's needs. Schein states that organization and management
theory has adhered to and fostered these simplified and
generalized conceptions about man. They have failed to
recognize the variability of man's needs and motives, not
realizing that employee A is complex within himself and
different from employee B. Because man is variable and
capable of learning new motives through organizational
experience, his ”psychological contract“ with the organ-
ization is a function of both internal needs and organi-
zational experiences. The practical implications of these
assumptions are not that traditional principles of organ-

ization are misguided, but rather that there is no one

16
correct management strategy. These assumptions are more
in line with Hulin and Blood's and Lodahl and Kejner's
regarding the necessity of leoking at personal background
variables in determining individual differences in needs

and motives.

Review of the Empirical Literature

Having muddled through the literature and become aware
of the complexity and lack of general conceptual agreement
regarding this theoretical construct, the reader, however,
should be able to formulate a general outline of the nature
of the problem. What has been described by the previous
discussion of the construct is an orientation in which work
is a very important and meaningful part of life. It is not
viewed as a necessary evil or as an economic survival tool.
It is such an integral part of the individual's self-concept
that it also affects his self-esteem. It would be helpful
at this point to review the empirical work relevant to this
problem in order to gain insight into how this motivational
or attitudinal index fits into the broader perspective of
employee responses on the job.

As stated earlier, Lodahl and Kejner (1965) conducted
one of the first studies concerned with the definition and
measurement of job involvement. ’Their two definitions of
job involvement seem to embody all of the various definitions
and conceptualizations which have been reviewed. Starting
with 110 statements potentially related to their conception
of job involvement, they submitted 87 non-duplicated items

to "expert" judges. After calculating means, medians,

17

18
standard deviations, and Q-values, 40 of these items were
retained and administered to 137 nursing personnel. An
item analysis and factor analysis were performed. Five
factors emerged with loadings accounting for 92 per cent
of the variance in the total involvement scores. By
considering item total correlations, communalities. and
factorial clarity, this set of items was further reduced to
20 and was administered to a group of engineers. Factor
analysis was again performed on the data for both samples.
Loadings on the first principal axis indicated the presence
of a general involvement factor. However, only eight of the
items had their highest loading on this factor. Using both
samples and rotating to the varimax criterion, three factors
emerged for the nurses and four for the engineers. Factor 1
seemed to deal with the rejection of items expressing high
involvement and had the highest correlation with the total
job involvement score. Items composing Factor 2 indexed a
response of indifference to work. Factor 3 was termed a
duty-bound positive involvement, while the fourth factor
which emerged for the engineers' data seemed to be tapping
boredom and the general unimportance of work. Because of
the factorial similarity across the samples, they concluded
that job involvement is multidimensional with at least three
dimensions. Corrected split-half correlations in the samples
ranged from .72 to .89 for the 20 item scale. This led
Lodahl and Kejner to comment on the only moderate internal

consistency, which was probably due to the multidimensionality

19
of the scale. As evidence of validity they performed an
analysis of variance which indicated that three samples
(students, nurses, and engineers) differed in respect to
involvement scores.

Although this is but slight evidence of the validity
of their measure, further evidence comes from a study by
Goodman, Furcon, and Rose (1969). In an examination of the
convergent and discriminant validity of creative ability
measures, job involvement items developed by Lodahl and
Kejner (1965) were used as control traits in a multi-trait
multi-method matrix design. The job involvement items were
utilized since they were considered independent of the
creative ability traits. Using a sample consisting of 63
employees in a government research laboratory, reliability
of the involvement scale was .83. Of primary interest was
the finding that the involvement items exhibited both
substantial convergent and discriminant validity. Convergent
validity was indicated by the values in the diagonals of the
matrix and discriminant validity was indicated by the degree
to which a diagonal value exceeded its corresponding row and
column values.

Lodahl and Kejner (1965) also correlated the data with
other variables pertaining to the technological nature of the
work and job satisfaction as measured by the Job Description
Index and demographic data. The job involvement scale
correlated positively with only two of thirteen job

variables, number of people contacted per day on the job

20

(r = .30) and necessity of working closely with others
(r = .34). Four of five satisfaction variables were
associated with high job involvement; satisfaction with
the work itself (r = .29), promotion (r = .38), supervision
(r = .38), and people (r a .37). In terms of background-
demographic data, the only significant correlation was
between job involvement and age (r a .26), with older
workers being more involved. Years of college, years of
work experience, job status, and marital status did not
correlate with job involvement. Sample sizes in the inves-
tigation of these relationships ranged from 46 to 137.
Factor analysis also showed that the 20 item involvement
scale had its highest loading on the same factor as the
satisfaction variables. They concluded that, for the
sample of engineers used, job involvement had the same
factorial structure as job satisfaction. However, they
did point out that there was no justification in concluding
that the two constructs are identical although they might
have similar determinants. Combining their data to form
a composite profile of a job involved individual, they
state that such an individual: 1) is older, 2) is less
considerate as a leader, 3) has a more highly interdependent
job, and 4) is more satisfied with the work, coworkers, and
promotional opportunities.

It must be recognized, however, that their sample was
restricted to a narrow range of occupations. 0f further

interest is their lack of discussion regarding the two

21
conceptualized dimensions of involvement. Having performed
a factor analysis they never attempt to reconcile the
difference between the various factors that emerged and
their initial dichotomization of the construct. Also in
question is their use of orthogonal rotations while
performing the factor analysis since this implies, or
builds in, an assumption of independence between the items.
They never address themselves to the issue of whether or
not they expect the various items of the involvement scale
to be correlated or whether they feel that they are tapping
distinct job attitudes as their two definitions imply. This
seems to leave a major point unanswered since the goal of
this research is to gain as clear a conceptualization as
possible concerning what attitudes and feeling states the
construct involvement is actually measuring.

An earlier study by Lodahl (1964) made use of different
samples and technologies. Fifty male assembly line workers
in an automobile plant and 29 female assemblers in an
electronics company constituted the two samples utilized.
Lodahl was trying to get at ”deep emotional reactions to
work.” He was attempting to understand how attitude
variables measuring job satisfaction, motivation, and job
involvement are related to each other. Also of interest
was the question of the generality of these attitude
factors across various technologies. Data for the study
was obtained from a content analysis of interviews conducted

with each of the subjects. A principal components factor

22

analysis of the automobile assemblers' data was performed.
Eleven factors, accounting for 78 per cent of the variance,
emerged: however, only five were clearly interpretable. One
of the factors was termed involvement since all of the
involvement variables clustered together and were unrelated
to any other class of variables. Product involvement, job
involvement, company involvement, and number of men working
nearby loaded on this factor. The inclusion of this last
variable led to the speculation that involvement is part of
a more general social-affiliation process. The data also
indicated a tendency for men with more education to be less
involved in the assembly line job (r = -.38).

Analysis of the data for the electronics workers
yielded similar results. Eight factors, accounting for
85 per cent of the variance, emerged, but only six were
interpretable. Again, job involvement broke out as a
separate factor unrelated to satisfaction, motivation, or
technological variables. The items which clustered together
were product knowledge, job involvement, and team involve-
ment. From these results Lodahl theorized that job
involvement may be looked upon as a general value orien-
tation which deals with why people prefer work over idleness.

In a somewhat similar study to the above, Lawler and
Hall (1970) investigated the empirical and theoretical
interrelationships among job involvement, intrinsic
motivation, and satisfaction, and the relationship of these

attitudes to job design characteristics. As was stated

23

elsewhere, they were interested in investigating the
distinctness of these three job attitudes. To facilitate
their work they used Lodahl and Kejner's first definition
as their conceptualization of involvement and Lawler's
definition of intrinsic motivation. Using a sample of 291
scientists in research and development laboratories,
questionnaires were administered consisting of the following:
six items designed to measure the scientists' satisfaction
of autonomy and self-actualization needs: six items (from
Lodahl and Kejner's scale) measuring job involvement: four
items tapping intrinsic motivation: five items measuring
job perceptions dealing with the degree of control over
job, degree to which the job was perceived to be a relevant
test of one's abilities, and degree of influence and
control in the department: and two items measuring self-
rated performance and effort. The measure of satisfaction
was obtained by first asking how much of the factor was
associated with the job and comparing this response with
the rating of how much they felt the factor should be
associated with the job. In addition to these questionnaire
items, structured group interviews were conducted in which
the investigator was able to rate the amount of challenge
in the job and the amount of contact the subject had with
project clients.

A principal components factor analysis with varimax
rotation of the sixteen items indicated that a three

factor solution was most interpretable and accounted for

24
48 per cent of the variance. The first factor was a
satisfaction factor with all six satisfaction items
loading on it. All six involvement items loaded on
Factor 2 with four of the six loading highly (.55 - .84).
All four of the intrinsic motivation items loaded on
Factor 3. Lawler and Hall thus concluded that these
measures of the three job attitudes were factorially
independent and relatively distinct variables. Their
conclusions do seem justified in light of the factors
and loadings obtained. However, the methodology may be
suspect due to the nature of the scales used. The satis-
faction scale had different anchor points (minimum-maximum)
than the other two scales (Likert type). This may have led
to the clarity and artificiality of the results. It is
difficult to imagine attitudinal measures being as clearly
independent as this analysis indicates.

With regard to the relationship of these attitude
measures to job design characteristics, it was found that
the latter are more strongly related to satisfaction than to
job involvement and intrinsic motivation. However, the
involvement items were significantly related to four of
the five job design characteristics and the intrinsic
motivation items were related to three job design charac-
teristics. Jobs which allow the individual control, give
him a chance to be creative, and are appropriate to his
abilities are more satisfying. This suggests that job

design features can influence satisfaction levels (r's

25
ranged from .24 to .45). Lawler and Hall state that the
weaker relationships between the involvement measures and
job characteristics might further suggest that involvement
is more a function of the person than the job, as other
theorists have also proposed. They further state that
"people probably do differ as a function of background and
personal situation in the degree to which they are likely
to become involved in their jobs.“ However, the attitude
is probably dependent on both individual and job charac-
teristics. The involvement items correlated with a chance
to be creative, influence in the department, chance to ”do
the thing I do best,“ and appropriateness of job to
abilities (r a .18 - .21). Worth noting is the fact that
involvement was not related to the self-rated performance
measure, leading Lawler and Hall to state that an individual
can be involved in his job for reasons not related to
performance. It is possible that lack of variance in the
performance ratings could have led to the low correlation,
but they never touch upon this possibility.

Lawler and Hall (1970) have shown that research on job
involvement as a distinct attitudinal measure is a worth-
while direction to explore. Their results indicate that
people do differ in the degree to which they are involved
and outline what relationships may exist between attitude
and job characteristics. However, they have given no
indication of the intercorrelations between the intrinsic

motivation items and the involvement items. The correlatidns

26
of these two attitude measures with job characteristics are
not strikingly different. This is an important point
because, as previously noted, their definition of intrinsic
motivation is what others have proposed as an aspect of job
involvement.

Weissenberg and Gruenfeld (1968) further attempted to
clarify the relationships among these job attitudes. Using
Herzberg's dichotomization of ”motivator“ and ”hygiene”
variables, they investigated the relationship between job
involvement and satisfaction. The Wernimont Job Satisfaction
Scale and Lodahl and Kejner's short, six-item scale were
used. Results confirmed their hypothesis that involvement
increases with increasing satisfaction with motivator
variables. Significant correlations were found with three
of the five motivator variables and the motivator total.

As predicted, however, involvement did not correlate signif-
icantly with hygiene variables. These findings are what
might be expected by the very nature of the concept of job
involvement. As an attitudinal measure indexing how the
person is identified with his job and how important it is
to his self-concept, perceived content aspects of the work
would be expected to be related to job involvement. Since
these are higher order need satisfactions, somewhat similar
to the autonomy and self-actualization satisfaction items
used by Lawler and Hall, their work partially seems to
answer questions raised concerning the intercorrelations of

items. It also serves to reinforce the viewpoint that job

27
involvement is related to job design characteristics.
However, a more fruitful approach would appear to lie in
the investigation of the relationship between the importance
of these higher order needs and job involvement. This
could lead to understanding in regard to the influence
personal background has on the attitude.

In regard to this relationship between higher order
needs and involvement, Vroom (1962) conducted a study in
which the following question was asked of supervisory and
blue-collar workers: ”If a problem comes up in your work
and won't be settled by the time you go home, how likely
is it that you will find yourself thinking about it after
work?“ Those who thought about it after work were more
likely to be satisfied if the job offered them opportunities
for self-expression. These opportunities were found to be
less important to those who forgot about their jobs as soon
as they left work.

Goodman, Rose, and Furcon (1970) conducted an investi-
gation designed to compare four approaches to assessing the
motivational antecedents of work performance. The motiva-
tional models used were direction of motivational orien-
tation, source of motivational stimulation, job dedication,
and an expectancy model. Job dedication was measured by
Lodahl and Kejner's job involvement scale. Sixty-six
scientists and engineers in a government research laboratory
participated in the study. The investigators hypothesized
that the expectancy model should exhibit the strongest

28

relationship with performance because as a model it is more
multidimensional than the other three. Performance was
measured by self-reported publication output, actual
publication output, self-reports of unpublished papers and
reports, and self-reports of presentations at formal
scientific meetings. None of the expected relationships
between the first three motivational models and performance
was strong. Of particular interest to the present inves-
tigation are the weak relationships between job dedication
and the four performance measures. None of the relationships
was significant. As predicted, the expectancy model did have
the strongest associations with performance measures.
Significant correlations (p<.05) ranging from .24 to .39
were obtained. However, criticisms can be leveled at the
dependent measures. There is no ”quality" dimension in
any of the four performance measures and a gross measure
of quantity of publications may not be at all represent-
ative of an individual's worth to the organization.

Probably the most thorough investigation of the
relationship between job involvement and other attitudinal
and job design variables was conducted by Patchen (1965,
1970) at Tennessee Valley Authority. Although he only
used a two item scale, it was found to have fairly high
test-retest reliability (r a .83) with intercorrelations
between the items of .38. This low correlation in itself
points out that the two questions may be tapping different

aspects of work motivation. It also points to the

29
difficulty in generating a clear and general conceptual-
ization of job involvement. The multidimensionality
referred to by Lodahl and Kejner must be kept in mind.
Patchen labeled his index “general job interest” but the
items do seem to be in line with the involvement construct,
psychological identification. Utilizing a sample of 90
work groups. investigation with the index showed that there
are both within group and between group differences.
Patchen explored a myriad of relationships involving
general job interest and job design characteristics, finding
many to be significant. His investigation did tend to focus
on job characteristics, but some relationships involving
personal characteristics were also investigated. Perceived
opportunity for achievement, control over work methods,
feedback on performance, difficulty of the job, and chance
to learn new things all correlated fairly strongly with
general job interest (r s .25 - .53). Joint effects were
also investigated through several analyses of variance using
four job features: control over means, job difficulty,
feedback, and frequency of time limits. It was found that
job interest tends to be increased by combinations of:
1) high job difficulty and high control over means, 2) high
control over means and high feedback, and 3) high job
feedback and high difficulty. When the other three features
are constant, control over means has a marked relation to
job interest. These relationships are very much in accord

with the ideas espoused by the participative management

30
theorists regarding man's desire for higher order need
gratification. 0f further interest was the tendency for
a combination of high personal need for achievement and
poor opportunity for achievement to bring lower job interest
than is due to the effect of either personal need or
opportunity alone. This interaction, however, was small.
Although it was found that need for achievement was
unrelated to general job interest, highest levels of job
interest were recorded for those highest in need for
achievement, identification with occupation, influence
over work goals, and chance to do what best at. Conversely,
those with the lowest job interest had a high need for
achievement but were low on the other three characteristics.
In other words, those with the high need for achievement
were thwarted by job features and may have sublimated
their efforts to out of work activities in order to fulfill
this need. This lends support to what Argyris and McGregor
had been saying and to the idea that job involvement is
influenced by job characteristics.

One of the other scales which Patchen used, ”identifi-
cation with one's occupation,“ also looks as if it might be
closely related to our conceptualization of involvement and
to his general job interest scale. Using this scale as a
measure of commitment and pride in work, they found that
it was related to job interest (r - .32) even when other
variables were partialed out. Results also suggested,

however, that ”although commitment to an occupation

31
increases interest in work, it does not contribute to a
stronger motive for achievement." Thus occupational
commitment may reflect satisfaction with membership in
the occupational group, rather than a drive for achievement
in the work. This appears to refute what Lodahl and Kejner
(1965) and Vroom (1962) had been saying regarding the
performance-esteem aspects of involvement. They also
found a correlation of .38 between job interest and pride
in work. However, commitment to an occupation may not be
the same as commitment or involvement to a job. The latter
may be more influenced by organizational conditions and the
former can be considered more of a general attitudinal
measure.

It is interesting to note that both Patchen (1970) and
Lawler and Hall (1970) obtained results which were in accord
with Vroom's theorizing about involvement and the importance
of the individual's perceptions of how relevant his abilities,
aptitudes, etc. are to his performance on the job. The lack
of constraints imposed by the job, as measured by questions
of control, influence, etc., have also been shown to be
related to increased job interest as Vroom and others have
theorized.

Patchen's work is thorough in examining relationships
involving job design characteristics, but it does suffer in
its lack of investigation of personal and background
variables which would enable an examination of possible

antecedent conditions which may lead to job involvement.

32
Although he has generated a complex theoretical model, the
testing of it has been less than supportive. 0f the ten
determinants of job motivation investigated, only three
showed clear support for this theoretical position.
Statistically significant results accounting for a small
proportion of the variance were frequent throughout the
study. Yet, Patchen seems to attach unwarranted theoretical
significance to many of these findings. However, his work
does stand out as an attempt to approach problems and
relationships with a more complex perspective than is
usually found throughout the literature.

Schwyhart and Smith's (1971) study is one of the most
recent investigations which has emphasized the neglect which
the construct of job involvement has received. Using a
sample of 149 white-collar, salaried, middle managers under
the age of 40, they investigated the relationship between
job involvement and company satisfaction. Company satis-
faction was defined as "the degree to which an employee
derives satisfaction from and identifies with the company
in which he is employed.” It was indexed by 20 items found
to load on factors such as company identification, pride in
company, and personal pride in the organization. The job
involvement scale consisted of the 20 items developed by
Lodahl and Kejner (1965). The correlation between involvement
and company satisfaction was .45 (p<.05). Job involvement
was also found to correlate significantly (r u .18, p<.05)

with age. However, company tenure and number of promotions

33

showed no relationship with job involvement and company
satisfaction. Factor analysis using a principal compo-
nents method resulted in the emergence of six factors
accounting for 59 per cent of the common variance in the
unrotated solution. Only three of these factors were
consistent upon replication in a second sample. These
three factors accounted for 41 and 36 per cent of the
common variance in groups A and B, respectively. The
factors were termed job ambition, job centrality, and

job conscientiousness. Only the first factor was similar
to factors obtained by Lodahl and Kejner's (1965) analysis.
This finding led the authors to speculate that job
involvement may have a different factor structure across
occupational groups. They also pointed out the need for
future research to examine antecedents and consequences
of job involvement.

As would be intuitively expected, empirical results
have also indicated a negative correlation between involvement
indices and absenteeism. Patchen (1965) found fairly strong
negative correlations (r a .43 - .53) among work units.
Wickert (1951) found turnover to be related to job
involvement. Using a sample of 600 telephone operators and
service representatives, he found that those who tended to
remain on the job reported: 1) having a chance to make
decisions on the job, and 2) feeling that they were making
an important contribution to the company. His results may

have been biased, however, by the fact that part of the

34
sample was still employed and part was no longer working,
especially since the questionnaires were administered after
the latter had already quit. Those who thought about their
jobs after leaving work were more likely to be satisfied if
the job offered them opportunities for self-expression.
These opportunities may have been found to be less important

to those who forgot about their jobs as soon as they left.

Conceptual Framework and Hypotheses

The literature review was designed to familiarize
the reader with previous research. More importantly,
however, it should make one aware that in spite of all
the disagreement and attention that involvement has
received within a conceptual framework, very few empirical
investigations have followed. Job involvement has been
described as an attitudinal state greatly influenced by_
both job characteristics which the individual is presently
experiencing and by values inculcated early in life through
socio-cultural training. These positions are most cogently
represented in their extremes by the participative
management theorists (Argyris, Likert, McGregor) and their
critics (Hulin and Blood, Strauss, Turner and Lawrence),
respectively. Schein (1965) in his ”complex man" perspec-
tive, however, has argued for the necessity of an approach
which takes both of these types of factors into consider-
ation. It is neither the individual differences, in the
form of background and demographic factors, nor job design
characteristics, such as amount of control, influence, and
use of abilities allowed on the job, algae that determine
this attitudinal state. The interaction between individual
differences and organizational constraints is viewed as the

key framework for understanding relationships which occur.

35

36

While these theoretical distinctions have been
proposed, empirical testing has lagged behind. Lawler
and Hall's (1970) work has encouraged the belief that the
use of the construct job involvement can be a worthwhile
direction for future research. Patchen (1970) and Lawler
and Hall (1970) have also shown that as a distinct construct
it is related to several job characteristics. But despite
the theorizing regarding the possible importance of
background and socio-cultural variables by Patchen, Lodahl
and Kejner, and Lawler and Hall, this area of research has
been severely neglected. Past research has been limited to
a rather restricted range of occupations and levels. Also
worthy of study, but deficient in investigation, have been
the consequences of job involvement such as performance and
overall satisfaction with the job and the company.

The most fruitful framework within which to focus an
investigation of the construct of job involvement thus
consists of examining possible antecedent conditions of
involvement (background and situational variables) and
the consequences of involvement (performance, absenteeism,
and turnover). It is to this end that the present study
is directed.

The following hypotheses to be investigated are an
outgrowth of the conceptual framework which has been
discussed. More specifically, this investigation will
test the hypotheses which have been generated by
theoretical positions advocated by participative

37
management theorists and proponents of individual difference
considerations. In keeping within the proposed framework,
the investigation will also concern itself with the inves-
tigation of consequences of involvement, such as produc-
tivity, turnover, and absenteeism. These relationships are
conceptually presented in Figure 1.

Figure 1. Hypothesized antecedents and consequences of
job involvement.

Performance
Job Design
Characteristics / Absenteeism

> Job Involvement
Background-Valu =::::::::::Satisfaction

 

Variables
Turnover

 

Although there are differences in points of emphasis,
in general the participative management theorists (Argyris,
McGregor, and Likert) have failed to introduce the concept
of individual differences into their basic theoretical
framework. The ego needs, which are central to their propo-
sitions, are viewed as being important to all workers, with
fulfillment dependent upon job design characteristics
experienced by the individual worker. If the job is capable
of providing the gratification of ego needs, the individual
will become involved. This is stated more explicitly in the
form of Hypothesis 1: job involvement will be positively
related to the degree of participation in decision making
which the job design characteristics provide the individual.

38
Needs for influence and control in work are thus viewed as
some of the many possible ego needs which the individual may
need fulfilled. This would be an operational test of one
theoretical position concerning the developmental nature
of involvement.

A second theoretical position which must also be
considered reverses the approach of the participative
management theorists and, while not actually ignoring the
effects of job characteristics, dismisses their importance
and relegates them to a secondary position. In discussing
the importance of background-value variables, Lodahl and
Kejner (1965) state that involvement may be resistant to
changes due to the nature of the job. Hypothesis 2 stems
from this focus: job involvement will be positively related
to level of education. It is assumed that the higher the
level of education attained by the individual the more the
socialization process will have fostered a need for
involvement. No consideration is given to job design
characteristics or to the violation of an individual's
expectations concerning the job.

More realistically, and in line with the "complex man"
perspective, it is hypothesized that job involvement is a
function of an interaction between fulfillment of ego needs
by the job and individual difference factors in the form of
personal-background variables. To explore this more complex
position moderator variables will be used.

Hypothesis 3 states that the relationship between job

39
involvement and participation in decision making will be
more positive for more highly educated individuals than for
those individuals with less education. Schein (1965), in
arguing for his ”complex man” perspective, discussed the
use of education as a moderator in the relationship between
ego need gratification provided by the job and employee
responses to the job. Educational level was viewed as an
index of expected or desired ego need gratification, with
those individuals who are less educated expecting or
desiring less ego need gratification from their jobs than
more highly educated individuals. Ruh and Wakeley (unpub-
lished manuscript) also found general support for the
hypothesized interaction.

Hypothesis 4 states that the relationship between job
involvement and participation in decision making will be
more positive for workers from a rural background than for
those of an urban background. This is an outgrowth of Blood
and Hulin's (1967) refutation of the participative
management theorists' propositions. As a result of extra-
work socialization and ”alienation from middle-class norms,“
lower class city workers are more likely to be satisfied when
their jobs are less personally involving. It is only the
rural workers who demand personal involvement. Their
individual needs are different and will thus result in the
stated differences in the relationship between job
involvement and participation in decision making. These are
differences which the participative management advocates

would not consider exploring.

40

Hypothesis 43 states that there will be mean differences
in job involvement between individuals of urban and rural
backgrounds with higher job involvement for the rural
workers. This is also a direct outgrowth of Blood and
Hulin's theorizing regarding individual differences
fostered by extra-work socialization and ”alienation from
middle-class norms.”

Hypothesis 5 predicts that the relationship between job
involvement and participation in decision making will be more
positive for those individuals with low religious devotion
than for those with high religious devotion.! Use of this
moderator stems from the general concept of the Protestant
Ethic which has been operationally defined as the belief in
the inherent goodness of work and the consequential moral
and religious justification for the work process itself.

As Weber (1930) pointed out, the average Protestant had

to work hard to rid himself of the fear of damnation. He
could not relax or enjoy the fruits of his labors. His
profits and savings went back into the business which in
itself was ”a serious calling ordained by God" (McClelland,
1961). Viewing one's intensity of religious devotion as
an operational index of this general value orientation, the
above hypothesis can then be more easily understood. Job
design characteristics will not be as strongly related to
job involvement for the individual adhering to this value
system as they will be for the individual who is not as
influenced by the Protestant Ethic. The former individual

41
will need to be involved in his job, regardless of job
conditions, just to insure that his fate is not one of
damnation. The religious dogma would compel this and
leave him with a less flexible response than his less

religious counterpart.

' 1:! 3 _HS_W
u - KH|
. .‘. " I;
" Iiz'h 1*.
i

I .-
3-"3a-l} l! L

Hypothesis 5g states that job involvement will be
positively related to intensity of religious devotion.
This is simply an operational statement of what has been
). discussed in the previous paragraph concerning the

[g adherence to the Protestant Ethic as a value system or

 

orientation.

 

Hypothesis 6 states that the relationship between job
involvement and participation in decision making will be
more positive for those individuals with a high intrinsic
orientation than for those with a high extrinsic orientation.
This reflects the assumption that those individuals who are
highly intrinsically oriented will be more involved to the
extent that their ego needs are fulfilled by job design
characteristics. However, those individuals for whom the
importance of intrinsic needs is low on their need hierarchy
will view the job solely as a means to a financial end.

Therefore, their degree of involvement will not be as

 

strongly influenced by participation in decision making.

One aspect of the conceptual framework as previously
introduced is concerned with consequences of job involvement.
In an attempt to investigate these possible resultant

behaviors the following hypotheses are introduced.

42

Hypothesis 1 predicts that job involvement will be
positively related to performance as measured in terms of
productivity. However, the relationship is not expected to
be strong. As Lawler and Hall (1970) point out, an
individual can be involved in his job for reasons entirely
unrelated and of no consequence to performance.

Hypothesis 8 states that job involvement will be
negatively related to absenteeism.

Hypothesis 2 states that job involvement will be
negatively related to turnover. Both of these results
would be intuitively expected, but of more importance are
the empirical findings (Patchen, 1965: Wickert, 1951) which

lend some support to this hypothesis.

Method

Subjects

The source of data for the study was the 1968 Michigan
State University “You and Your Job” attitude survey. The
subjects were 2755 employees of six medium sized manufac-
turing companies in the Midwest. Of 4162 questionnaires
distributed, 2755 usable ones were returned, providing a
return rate of .66. Of these 2755 subjects, 2628 responded
to at least three-fourths of the relevant items. Various
subgroups of these 2628 respondents were utilized in the

present analysis.
Data Collection

The questionnaire mentioned above provided for the
measurement of personal-background data. Those demographic
variables of interest to this study were level of education,
intensity of religious devotion, and urban-rural background.
Each of the above variables, except level of education, was
indexed by summing responses to the specific items and
calculating means. The education measure, however,
consisted of a single item. A detailed description of the

specific items if given below.

43

44
I. Intensity of religious devotion
1. Under usual conditions how often do you attend
religious services?
a. Twice or more a week
b. Every week
c. Once or twice a month
d. On special occasions (such as Easter)
e. Do not attend services
2. Religion in your home was considered as:
a. An essential part of home life
b. One of several factors which were important
c. A relatively unimportant factor
d. Something to be left out of our family life
e. One's own business
3. As a young man, were you any of the following?
a. A church member and active in church activities
b. A nominal church member
c. A religious rebel
d. A non-churchman without any great religious
conviction
e. None of the above
II. Level of education
1. How many years of school work have you completed?
6 or less 7 8 9 10 11 12

some college college graduate work

45

III. Urban-rural background

1. The place in which you spent the most time during

your early life was a:

9.

Farm

Town of less than 2,000

Town of 2,000 or more, but less than 10,000
City of 10,000 to 100,000

City larger than 100,000

2. In what type of community are you now living?

In the country

Town of less than 2,000

Town of 2,000 or more, but less than 10,000
City of 10,000 to 100,000

City of larger than 100,000

3. In what size city would you prefer to live?

Rural or country
5,000 or less
5,000 to 50,000
50,000 to 200,000
200,000 or more

In addition to these demographic items, attitudinal

variables concerned with conditions surrounding the job

and employee reactions to the job were measured. Each

variable was indexed by averaging the responses to five-

point Likert scale items. Four scales were utilized for

the study:

job involvement, participation in decision

making, intrinsic orientation, and extrinsic orientation.

46

A detailed description of the specific items follows.

I.

II.

III.

Job involvement

My job means a lot more to me than just money.

How much are you personally involved in your job?
I'm really interested in my work.

To me, my work is only a small part of who I am.

I would probably keep working if I didn't need

the money.

Most things in life are more important than work.
The major satisfactions in my life come from my job.
The most important things that happen to me

involve my work.

Participation in decision making

1.

5.

To what extent are you able to decide how you do
your job?

In general, how much say or influence do you have
on what goes on in your work group?

In general, how much influence do you have on
decisions which affect your job?

In general, how much say or influence do you have
on how you perform your job?

My superiors are receptive and listen to my

ideas and suggestions.

Intrinsic orientation--the importance of each of the

following

1.
2.

Having good chances for promotion

Being recognized and appreciated for doing good work

4?
III. Intrinsic orientation (cont.)
3. Having responsibility on your job
4. Being able to decide how to do your job
5. A feeling that the work you are doing is important
6 Being able to learn new skills and gain experience
on your job
7. A sense of achievement in the work you are doing
IV. Extrinsic orientation--the importance of each of the
following
1. Having a supervisor who really knows his job
2. Having relations of trust and confidence between
superiors and subordinates
. Being trusted by the people you work with
. Receiving fair pay

3
L:
5. Steady work and steady wages
6 Working under good (safe, clean, pleasant) conditions
7

. Being liked by the people you work with

Three post-questionnaire measures were also utilized:
productivity, absenteeism, and turnover. During the time
for which the productivity measures were recorded, the
companies were utilizing a measured day rate system. This
necessitated the keeping of weekly records of each individ-
ual's percentage of standard produced, a percentage which is
calculated for an individual's primary job and for his total
performance. Machine down time and subsequent work stoppages
not under the control of the individual were taken into

account. The measure of productivity used in this

48
investigation was the average of an individual's perform-
ance rating on his primary job over a 20 week period.
This measure was adjusted for vacations and other such
leaves of absence. The data were collected in two plants
of the same manufacturing company located in different
cities. Performance ratings for 238 subjects were used
with values ranging from 36.5 to 162 per cent.

Absenteeism data were collected for the same five
month period as the productivity measure. Absenteeism was
operationally defined as the proportion of hours absent to
total hours worked. Here also, the measure was adjusted
for vacations, layoffs, and plant shutdowns. Data of
398 subjects were utilized with values ranging from 0.0
to .147.

Turnover data were collected for 1662 subjects at
fifteen different plants of the same company from which
productivity and absenteeism data were obtained. It was
simply a measure of whether or not an individual who
returned a questionnaire from the 1968 survey was still
employed by the same company. Deaths, retirees, military
leaves, and other similar involuntary departures were
adjusted for. The data were coded in the following way:

1 a no longer employed, and 2 a presently employed.
Selection 23 items

All items were selected on the basis of a series of

cluster analyses which were performed on the Michigan

49
State University C. D. C. 3600 computer. The cluster
analysis program which was utilized performs an oblique
multiple groups analysis on the variables specified. It
is not "blind” analysis since the investigator designates
which variables are to be grouped together. Items were
initially grouped into eight clusters on the basis of face
validity, theoretical expectations, and results of prior
investigations (Ruh, 1970: Morrison, 1970). Decision rules
were then employed using item communalities. alpha estimates
of internal consistency reliability, inter-item correlations,
and item-cluster correlations to regroup and eliminate items.
Summary results of the cluster analysis are presented in
Table 1.

The scale measuring job involvement initially consisted
of ten items, seven of which were from Lodahl and Kejner's
scale (1965). Cluster analysis resulted in the experimenter
retaining eight items for use in this investigation. Median
inter-item and item-cluster correlations were .36 and .64,
respectively. The alpha estimate of reliability was .81.

The initial five items grouped to form the participation
in decision making scale (PDM) were retained with no
additional items being added. Median inter-item and item-
cluster correlations were .48 and .69, respectively. The
alpha estimate was .81.

The measure of intensity of religious devotion was
reduced from four to three items, with median inter-item
and item—cluster correlations of .45 and .71, respectively.

The alpha estimate was .72.

50

 

 

ma. 0:. a“. mm. mm. as. mm. «a. ma. mm. :oaaaeaoswo caucaupxm
mm. on. em. mm. mm. as. on. me. em. mm. :oapeueoaso cameaaueH
ea. so. mm. em. am. on. ma. am. 0:. ms. engagemean Hausuueaepp
mm. mm. 0:. an. ma. no. Ha. mm. es. ms. cospo>oe maoamaaam
8. mm. em. as. we. mm. mm. mm. mm. we. wages eoanaooe
as :oaeuaaoapsua
am. an. en. me. «a. we. as. me. am. on. pcososaosna nos
ems; sea assess amen .soa assess new: _=oa cameos
sBAH< medpwaeadssoo BopH m.n hovmsaoisovH m.n empdlaougH

 

 

.cepsaosd omae one Amenaaev modpaddnedaoa oHsom
.medpaaessssoo Sop“ use .m:0dveaeaaoo hopenaotaepd .e:0dueaoaaoo Sevdnaoosd Mo smash use define: .a canny

51

All three original items in the urban-rural background
scale were retained, with the scale having an alpha value
of .74. Median inter-item and item-cluster correlations
were .45 and .75, respectively.

The seven items which compose the scale measuring
intrinsic orientation were initially separated into two
distinct clusters of four and five items each on the basis
of a previous factor analysis (Morrison, 1970). However,
cluster analysis indicated that such a dichotomy was unjus-
tified, and one single cluster of seven items was retained.
The median inter-item and item-cluster correlations were
.32 and .58, respectively. An alpha estimate of .78 was
obtained.

The extrinsic orientation items were similarly grouped
into two clusters of four and three items based on a previous
factor analysis (Morrison, 1970). Cluster analysis also
indicated that this split was unjustified and the seven
items were retained and regrouped into one cluster having
an alpha estimate of .72. Median inter-item and item-cluster

correlations of .25 and .53. respectively, were obtained.

Data codipg

To investigate the hypothesized relationships, responses
to the five scales (intrinsic orientation, extrinsic orien-
tatian, urban-rural background, intensity of religious
devotion, and level of education) were trichotomized into
groups of high, medium, and low. These subgroups were

compared on each particular variable. Responses to the

52
level of education item were grouped into the three categories
according to the following code: less than 12th grade
education as low, 12th grade education as medium, and more
than 12th grade education as high. Due to the nature of the
design, sample sizes utilized in the investigation of each
of the moderator variables varied somewhat. Simple
correlational techniques were utilized in analysis of the
data which was performed with the Michigan State University
C. D. C. 3600 computer. As the description of the
moderator design implies, various subgroups were utilized
in investigating each of the moderator relationships. The
statistical significance of differences in correlation
between job involvement and participation in decision making
among the subgroups were investigated using Fisher's r to z
transformation. Simple t-tests were also employed in testing

mean differences for the relevant variables.

Results and Discussion

Hypothesis 1, which stated that job involvement would
be positively related to participation in decision making,
was tested using a sample of 2628 respondents. As predicted,
a correlation of .51 (p<.01) was obtained between job
involvement and participation in decision making. The first
hypothesis was thus confirmed, supporting the participative
management assumptions which have been discussed. Although
the study focused on antecedent and resultant factors, the
design was correlational in nature and causality becomes an

issue about which only hypotheses can be made. As Campbell

and Stanley (1963) state,

. . .that the causal interpretation of correla-
tional data is overdone rather than underdone,
that plausible rival hypotheses are often
overlooked, and that to establish the temporal
antecedence-consequence of a causal relation;
ship, observations extended in time if not
experimental intrusions of X are essential. (P. 63)
Lack of control eliminates any possibility of determining
the direction of causality in the involvement-participation
in decision making relationship. It is possible that those
individuals who entered the organization with relatively
high, stable job involvement also tended to perceive organi-
zational conditions as more participatively oriented and

contributing to the continuation of a feeling of involvement

53

54
brought to the job. Alternately, one can assume from the
participative management perspective that increases in the
gratification of ego needs, such as participation-related
needs, do indeed lead to increases in job involvement.
While this second explanation appears more viable to the
investigator, the last paragraph has emphasized the fact
that both assumptions are nevertheless tenable in a
correlational study.

Nevertheless, this result does contribute to a belief
in the necessity of considering job conditions as well as
value orientations in a discussion of this topic. One
cannot conclude that Lodahl and Kejner's (i965) assumption
regarding the resistance to change, or stability, of job
involvement has no merit. However, this result points to
the need to consider other alternative factors such as job
design characteristics in any theoretical discussion. The
design did not allow for experimental changes in job charac-
teristics which would have made possible a look at the
permanence of involvement as an attitude unaffected by
organizational change. But the results of the testing of
the first hypothesis do indicate a relationship between job
involvement and perceived job design characteristics.

Since job design characteristics are such an integral
part of the focus of this investigation, it is essential
to emphasize the distinction between perceived and actual
job conditions. The data collected were not measuring

actual, "objective” conditions surrounding the employee

55

in his work environment. What has been measured is the
individual employee's perceptions of organizational condi-
tions. However, a look at the variance of the participation
in decision making scores across job levels and within
levels adds some interesting data to this issue. Job

level was broken down into nine categories. In all but

two of these levels the variance in participation in
decision making scores was low relative to the variance
when job level was not held constant. This tends to support
the contention that the scores reflect mostly job charac-
teristics and not individual differences.

Support for Hypothesis 2, which stated that job
involvement would be positively related to level of educa-
tion, was not found. Utilizing a sample of 2628 subjects,
the obtained correlation between job involvement and level
of education was -.04. This may not be particularly
surprising because education is such a global measure that
it is not very indicative of any specific socialization
process.

The importance of fulfillment of ego needs by job
design characteristics becomes even more obvious with the
results of the testing of Hypothesis 3. Hypothesis 3
stated that the relationship between job involvement and
participation in decision making would be more positive
for more highly educated workers than for those workers with
less education. As predicted, level of education did

moderate the relationship between job involvement and

56
participation in decision making. Using the 2530 subjects
grouped into high (488), medium (919), and low (1123).
categories, correlations of .62, .hZ, and .51 were obtained
with the high, medium, and low groups, respectively. The
difference between the high and low groups was significant
at the .01 level. Summary information is presented in

Table 2.

Table 2. Summary of the relationship between job involvement
and PDM moderated by level of education.

 

 

 

Mean Mean
Job inv. :93 g N
High education 3.39 3.39 .62* #88
Medium education 3.1h 2.97 .42 919
Low education 3.3# 2.92 .51* 1123

 

I difference between r's significant—Ip<.Oi)

The apparent curvilinearity of this relationship (F statistic
indicates a significant non-linear relationship at the .01
level.) points even more strongly to the need for considering
both background variables and Job design characteristics in
a discussion of Job involvement. Those individuals with the
highest levels of education are also those with the highest
Job levels (Johnson, 1971). Their needs for participation,
control, influence, etc. are being met and they are involved
in their Jobs. Those individuals in the low education group
do not have a significantly different Job involvement mean,
perhaps indicating that their expectations regarding the

57
possibility of having a job have been more than satisfied

and resulting in their ”total immersion in the work process.“
Those subjects in the medium group might have had a higher
level of expectancy regarding how high in the organization
they would go and what level job they would eventually
attain, For many of them this expectancy has been violated
and consequently their involvement score is the lowest of
the three levels of education. If it is assumed that the
participation in decision making measure is also a
reflection of individual expectancy, then the above assump-
tions are even more plausible. Those individuals with the
highest levels of education have the highest need for
participation in the organization. Since they also have the
highest job level, they value these opportunities to satisfy
needs for participation and thus become involved to the
extent to which these needs are fulfilled. However, the
medium level of education group expected more in the way of
fulfillment of ego needs than what they had experienced.
Also of interest is the finding that the high education
group had the largest variance in job involvement scores of
the three groups. One could hypothesize that individuals
with a college (or some college) education are more hetero-
geneous, have experienced a more varied socialization process,
than those workers with less education. Their expectations
regarding work and their futures are also more varied. This
results in greater heterogeneity in the involvement scores.

Although this is not a major finding of the present study,

58
it does underline the need to examine not only mean differ-
ences but variances as well in an attempt to explain as much
of the data as possible. Future research would benefit
greatly from such an approach.

Hypothesis b, which stated that the relationship between
job involvement and participation in decision making would be
more positive for workers from rural backgrounds than for
those of urban backgrounds, was investigated using a sample
of 2572 respondents. The correlation between job involvement
and participation in decision making was found to be signif-
icantly different for the high and low urban groups.. However,
this difference was not in the predicted direction. The
correlations were .55 and .#3 for the high and low urban

groups, respectively. Summary information is presented below

in Table 3.

Table 3. Summary of the relationship between job involvement
and PDM moderated by urban-rural background.

 

 

 

Mean Job Inv. Mean PDM r N
High urban 3.h6* 3.29* .55** 857
Medium urban 3.23 2.9% .h8 858
Low urban 3.13* 2.81* .43** 857

 

3_—mean differences significant—Ip<.01)
** difference between r's significant (p<.01)

Similarly, Hypothesis he was not supported. The results
were opposite to those predicted. Mean job involvement

scores were higher for the urban than for the rural group

59
(p<.01). Hulin and Blood's assumptions regarding extra-work

socialization and "alienation from middle-class norms" which
results from living in an urban environment received no
support from the results of this analysis. But before these
propositions are dismissed as being fully disconfirmed, the
samples used in the two investigations should be compared.
Those individuals classified as being from an urban
environment in the present study are probably not even
remotely similar to the deep urban, ghetto-type individuals
composing the Hulin and Blood urban sample. It is imper-
ative that the reader keep in mind the difference in the
subgroups being compared. It is also possible that job

level is confounding the results in the present investigation.
These individuals with an urban classification may also have
the highest job level and consequently the most opportunities
for participation in decision making, as the data in Table 3
indicate. These individuals have high ego needs, perceive
the job as fulfilling these needs, and are thus highly
involved in their work.

For those individuals in the high rural group the
relationship between job involvement and participation in
decision making is not as strong and the mean on both of
these variables is significantly lower as indicated in
Table 3. Contrary to what Hulin and Blood hypothesized,
these individuals have lower involvement scores. Again,
this could be due to differences between the ”degree of

ruralness” of their sample and the one utilized in the

60
present investigation. These individuals may be more rural
than the rural sample utilized by Hulin and Blood which may
have been more suburban than rural. It is thus tenable
that the rural individuals in the present study were heavily
involved in their home duties since many of them also work
farms. Regardless of the participation in decision making
that they experienced, they might not become as involved.
That these results differ so strikingly from Hulin and
Blood's highlights the need to have similar samples and
operational definitions in order to compare results.

No support was generated for Hypothesis 5, which stated
that the relationship between job involvement and partici-
pation in decision making would be more positive for those
individuals with low religious devotion than for those with
high religious devotion. For the sample of 2556 subjects,
the correlations between job involvement and participation
in decision making were found to be .51 and .#6 for the
high and low religious groups, respectively. These differ-
ences were neither in the predicted direction nor statis-
tically significant. As predicted in Hypothesis 5a,
however, involvement was positively related to “intensity
of religious devotion." Utilizing the same sample of 2556
mentioned above, a correlation of .15 (p<.01) was obtained.
Summarized results are presented for Hypotheses 5 and 5a
in Table b.

In regard to the lack of support for Hypothesis 5, it
is likely that the measure of ”intensity of religious

61
devotion” does not reflect one's adherence to the Protestant
Ethic. The idea of work for the sake of work, or work as
a means of avoiding damnation, may no longer be at the
forefront of even strict religious ideals. The concept of
the "leisure society" may have infiltrated even the most
religious members of society to such an extent that the
work ethic is no longer prevalent. Thus, even the
individuals classified as highly religious in this sample
are interested in gaining fulfillment of these ego needs
which both work and the working environment can provide.

Table h. Summary of the relationship between job involvement
and PDM moderated by intensity of religious devotion.

 

 

 

Mean Job lflg- MeanigpM ;. N
High religious 3.U2 3,16 .51 853
Medium religious 3.29 2.99 .52 352
Low religious 3.11 2.88 .M6 851

 

The fact that Hypothesis 5a was supported does not make
the above assumptions less acceptable. Those individuals
who are classified in the high religious group may have a
strong need to become involved in whatever activities they
pursue. They are deeply committed to the activity they are
engaged in at a specific time, without having to relegate
other activities to permanent positions of less involvement.
Disregarding these theoretical explanations for the moment,

it is also necessary to look at the practical significance

62
of the obtained correlation. Although the correlation of
.15 is statistically significant, it accounts for less than
two per cent of the variance and raises the question of
whether it has any psychological significance in relation
to involvement and personal-background variables.

As predicted in Hypothesis 6, the relationship between
job involvement and participation in decision making was
more positive for those individuals highly intrinsically
oriented than for those highly extrinsically oriented (p<.01).
The correlations for the high intrinsic group and the high
extrinsic group were .51 and .n5, respectively. Summarized
results are presented in Tables 5 and 6.

These findings lend support to the assumption that those
individuals who are highly intrinsically oriented and have
ego needs to be fulfilled will be more involved to the
extent that participation in decision making is provided by
the job. Those individuals who are extrinsically oriented
will be more interested in the financial rewards which the
job provides and the attainment of extra-work related goals.
For the latter individuals, involvement on the job is less
related to fulfillment of ego needs.

It was also found that involvement scores for the low
extrinsic group and the high extrinsic group were signif-
icantly different (p<.01). As expected, the higher
involvement score was obtained by the low extrinsic group.
Similarly, the high intrinsic group obtained a higher

involvement score than the low intrinsic group (p<.01).

63

Table 5. Summary of the relationship between job involvement
and PDM moderated by intrinsic orientation.

 

 

 

Mean Job gnv. Mean PDM _;;_ N
High intrinsic 3.37* 3.17 .51 87h
Medium intrinsic 3.28 3.01 .50 878
Low intrinsic 3.15* 2.8# .h8 876

 

* mean differences significantilp<.01)

Table 6. Summary of the relationship between job involvement
and PDM moderated by extrinsic orientation.

 

 

 

Mean Jobwlnv. Mean PDM r N
High extrinsic 3.19* 2.8? .45 877
Medium extrinsic 3.22 2.97 .h9 875
Low extrinsic 3.h0* 3.18 .56 876

 

* mean differences significant (p<.01)

6#

Both of these results lend support to the assumption that
an individual's basic value orientation is reflected by his
involvement in work. The less an individual is involved in
his work, the more he is apt to value highly what Herzberg
(1959) has termed "hygiene factors.” The work is not
viewed as an end but rather more as a means to an end.

Before trying to integrate what has been reported and
discussed up to this point, results dealing with conse-
quences of job involvement will be presented since they
are of both theoretical and practical significance.

Hypothesis 7 predicted a slight positive relationship
between job involvement and productivity. A sample of 238
respondents were utilized in investigating this relationship.
The obtained correlation coefficient of .03 was not
significant. The reason for the lack of a more definitive
result may lie with the productivity measure which was
employed. Although there was variance in the productivity
measure, use of a measure based on a percentage of standard
achieved may be of questionable value. Since the sample
was composed of rank and file workers, it is also necessary
to question the amount of influence which these workers
actually had over their own productivity. 0f even greater
import may be Lawler and Hall's (1970) statement that
individuals can be involved in work for non-performance
reasons which may thus have no essential effect on perform-
ance. Perhaps Lodahl and Kejner's (1965) ideas about the

social-affiliative nature of involvement have merit and

65
an individual's involvement is entirely unrelated to
performance. One must also remember that the productivity
measures were not collected for the same time interval as
were the other questionnaire items. There was a time lag
of one year between the questionnaire and the performance
data collection. What was thus being employed was a
predictive design which may be of little value in inves-
tigating job involvement. During the time lag between
questionnaire and postquestionnaire measures, the company
changed to a measured day rate system from a piece-rate
system. This type of change might have had an effect on
an individual's feelings of involvement, thus making the
stability of the involvement scores from questionnaire
measure to productivity measure suspect. Essentially,
what would have been more desirable was a concurrent
measure of productivity. This should certainly be a goal
for future investigations of this construct. Future
research might also benefit from looking at moderators of
the involvement-productivity relationship such as salaried
versus non-salaried personnel.

There was no support generated for Hypothesis 8. Job
involvement and absenteeism correlated .02. Since this
result was so strikingly non-significant, there is no
urgent need to hypothesize concerning its positive direc-
tionality, although a negative correlation would have given
more support to the construct validity of the involvement

measure used. The lack of results indicates that there is

66
probably not a simple relationship between involvement and
absenteeism. Future researchers will need to look further,
possibly with the use of moderators, to clarify the
relationship. However, it is worthwhile to note once again
what type of measure of absenteeism was employed. The
measure was a ratio of days absent over a five month period
to the total number of days worked. Thus, consecutive days
absent were considered as two absences, not one. This may
not be the most legitimate method of indexing absenteeism
since an individual could have been absent for three consec-
utive days as a result of sickness without the cause of
absenteeism ever having been recorded. In light of this
fact, a better approach would have been to use a frequency
measure counting consecutive absences as a single occurrence.
Ideally, the reasons for specific absences should have been
recorded so that the measure could have been adjusted for
illegitimate absenteeism. However, this is only an ideal
which can infrequently be achieved in field research of
this nature.

Hypothesis 9 was the only one of the three ”resultant”
hypotheses that was supported. Using a sample of 1662
respondents, a significant (p<.01) correlation of .17 was
obtained between job involvement and turnover. This
reflects the hypothesis that those individuals with high
involvement means were more likely to still be employed
by the same corporation. Since this correlation is so small

and accounts for only a fraction of the variance, it would

67
be unrealistic to attach much importance and theoretical
significance to it. Again, there was the problem of not
knowing the specific reasons for an individual's withdrawal
from the organization. But what may have contributed to
this larger correlation was the adjustments made in coding
the turnover data. As reported earlier, deaths, retirees,
military and pregnancy leaves were not included. This may
have contributed to this index being ”cleaner” than the

other two postquestionnaire measures.

Conclusions and Implications for Future Research

The reader of this discussion should be aware that the
investigator did not expect "practical” or definitive
results. In dealing with this theoretical issue the primary
purpose was to investigate, and thereby explicate, relation-
ships involving the construct of job involvement. However,
the data may have hindered more than helped the achievement
of this end. Results have indicated that there are individ-
ual differences in job involvement, thus making investigation
of the construct a worthwhile task. The fact that three of
the four moderator relationships were significant, although
not always in the predicted direction, is encouraging and
points to the need for further research and more sophisti-
cated designs and measures dealing with this complex
theoretical issue. While results of Hypothesis 1 give
support to the more simplistic, straightforward approach
of the participative management theorists, no unequivocal
conclusions can be drawn. The results indicate that
perceived job characteristics are an important variable to
be considered. However, the differences achieved with the
various moderators also point out that job design charac-
teristics are but one of many factors which have to be
considered. Individual differences in the form of

background-personal variables serve to complicate and to

68

69
lessen the simplistic nature of relationships involving
job involvement. Such relationships are probably both
situation and person specific with interactions between
the two variables being of major importance. While there
was no definitive support generated for either of these
theoretical perspectives, it would appear that future
researchers in this area should not ignore the interaction
between individual and situational factors.

Results of the investigation of consequences of job
involvement were generally ambiguous. Although the
turnover-involvement relationship was significant and in
the predicted direction, the absenteeism-involvement and
the productivity-involvement relationships were not
supported. The most parsimonious explanation may lie
with the measurements. Refined measures, concurrent with
attitudinal indices, should be used in future research in
this area.

While results obtained in this investigation have
indicated that individuals with high intrinsic orientations,
high religious devotion, high levels of education, and from
urban areas have the highest job involvement scores, it
has not answered the question regarding the developmental
nature of job involvement. Is it socio-cultural and
resistant to change or is it organization dependent?
Correlational techniques have only allowed relationships,
not causality, to be outlined. What has been indicated is
the potential usefulness of an approach which takes into

account person and situation variables.

70

0f initial interest for future research would be an
investigation of the relationship between job involvement
and job satisfaction. Lawler and Hall (1970) have pointed
to the validity of distinguishing between these two
attitudes, and research findings could prove to be of both
theoretical and practical value. The use of moderator
variables, such as extrinsic-intrinsic orientation, might
also serve a useful function in investigating this relation-
ship. Job involvement might be found to be a useful
moderator itself, such as in the relationship between
satisfaction and performance. This would put Lodahl and
Kejner's (1965) performance-esteem defintion of involvement
to an empirical test. The present investigation has tended
to focus on the psychological-identification aspects of
involvement. Involvement has been viewed as an operational
measure of the degree to which the job is central to the
individual's entire life and an integral part of his self-
concept. Future testing of these related interpretations
would be of immense value in defining the construct.

Since this investigation has focused on personal-
background variables with some successful results, a
logical next step would be the investigation of the
relationship between job involvement and values. Another
area of potential interest grows out of Lodahl and Kejner's
(1965) theories concerning the relationship between
involvement and a social-affiliative need. Investigation of
this area could also prove enlightening and be integrated

with the investigation of values.

BIBLIOGRAPHY

Bibliography

Argyris, C. Integrating the Individual and the Organization.
New York: Wiley, 1963.

Blauner, R. Alienation and Freedom. Chicago: University
of Chicago Press, 1960.

Blood, M., and Hulin, C. Alienation, environmental charac-
teristics, and worker responses. Journal of Applied
Psychology, 1967, 51, 284-290,

Campbell, D., and Stanley, J. Experimental and Quasi-
experimental Designs for Research. Chicago: Rand
McNally, 1963.

Faunce, W. Occupational involvement and the selective
testing of self-esteem. American Sociology Association

Meeting. Chicago, 1959.

French, J., and Kahn, R. A programmatic approach to
studying the industrial environment and mental health.
Journal of Social Issues, 1962, 18, 1-47.

Goodman, P., Furcon, J., and Rose, J. Examination of some
measures of creative ability by the multitrait-multi-

method matrix. Journal of Applied Psychology, 1969,
53, zuo-2uu.

Goodman, P., Rose, J., and Furcon, J. Comparison of motiva-
tional antecedents of the work performance of scien-
tists and engineers. Journal of Applied Psychology,
1970. 53. #91-495.

Gurin, G., Veroff, J., and Field, S. Americans View their
Mental Health. New York: Basic Books, 1960.

Herzberg, P., Mausner, B., and Snyderman, B. The Motivation
to Work. New York: Wiley, 1959.

Hulin, C., and Blood, M. Job enlargement, individual differ-

ences, and worker responses. Psychological Bulletip,
1968: 6 ! “1‘55:

71

72

Johnson, H. Interrelations among sex, job level, and job
orientation. Unpublished Master's Thesis. Michigan

State University, 1971.

Katz, D., and Kahn, R. The Social Psychology of Organ;-
zations. New York: Wiley, 19 .

Lawler, E., and Hall, D. The relationship of job charac-
teristics to job involvement, satisfaction and
intrinsic motivation. Journal of Applied Psychology,
1970. 9. 305-312.

Lodahl, T. Patterns of job attitudes in two assembly

technologies. Administrative Science Quarterly,
1961‘s .8... “82-519.

 

Lodahl, T., and Kejner, M. The definition and measurement

of job involvement. Journal of Applied Psychology,
19659 Q2! 2h-33,

McClelland, D. The Achieving Society. New York: Free
Press, 1961.

McGregor, D. The Human Side of Enterprise. New York:
McGraw-Hill, 19 0.

Morrisson, J. Organizational climate, individual background,
and values and personal job goals in a sample of
Scanlon plan plants. Unpublished Doctoral Dissertation.
Michigan State University, 1970.

Patchen, M. Some questionnaire measures of employee
motivation and morale. Institute for Social Research

Monogyaph, 1965, fii, 1-70.

Patchen, M. Participation, Achievement and Involvement on
the Job. Englewood Cliffs: Prentice-Hall, 1970.

Ruh, R. Ego need gratification, extra-work socialization,
and attitudes toward the job. Unpublished Doctoral
Dissertation. Michigan State University, 1970.

Huh, R., and Wakeley, J. Education, ego need gratification,
and attitudes toward the job. Unpublished manuscript.
Michigan State University, 1971.

Schein, E. Organizational PsycholOgy. Englewood Cliffs:
Prentice-Hall, 19 5.

Schwyhart, W. R., and Smith, P. C. Factors in the job

involvement of middle managers. Ex erimental
Publication System, June, 1971, 12, Ms. No. 569-1.

73

Vroom, V. Ego-involvement, job satisfaction, and job
performance. Personnel Psychology, 1962, 15, 159-177.

Weber, M. The Protestant Ethic and the Spirit of Capitalism.
New York: Scribner, 1930.

Weissenberg, P., and Gruenfeld, L. Relationship between
job satisfaction and job involvement. Journal of
Applied Psychology, 1968, 52, h69-h73.

Wickert, F. Turnover and employee's feelings of ego-
involvement in the day to da operations of a company.

Personnel Psychology, 1951, _J 185-197.

W

N

"I71111111171111“11131118