"J:
’wlu \

...‘ “L“ ‘2
. ,wmm ' ;.
flimyfiv .

9!

. - J!

.J. "er
. '1‘

5%

_.._ u .
#3153

”4.

iii
3;? It
.:\t' 1-"‘ ‘z.'. 33'
-. I: ’31:. x;:,:‘ '
h. ‘ ‘3'," r,‘ 2

vi, ‘ 9"?“
, I. 5;

L ”any, ‘7":

-..»~w
n.
J u“: fig...»
IL...“-
any
.1 5'

fl

rv‘

,.p-e~ .

‘ 33v A
.o'ww-

w’.

'm.

~ . 2 .‘15'
i-'§‘."":552’5-
u '

I
,
., .
’ fir?
( l'

4o
.

.7
=2}:

mm a.

"w 52

w...”
v
an!”

 

_-.« o, >
,m.

'm--

-a

”m."-

n
.Ao
luv 1-.»
'nu
v... -.
.m-A

W
m
4

“#25;
“
fizuva
:x..

'1

 

E33
5""

THES‘H S

This is to certify that the

dissertation entitled

The Linkage between Social Class, Occupational
Conditions, and Father-Adolescent Son
Interactions: Examining the Work-Family Interface

presented by

Guenter H. Wasner

has been accepted towards fulfillment
of the requirements for

Ph.D degreein Psychology

‘ rofessor

W

Date June 22, 1993

 

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

Illllllllllllllll

 

 

 

 

LIBRARY
Mlchlgan State
Unlverslty
PLACE RETURNB unmov- uchhockouttom am I'm
TO AVOID FlNESrotum manor abdonddo om

    
  

DATE DUE DATE DUE DATE DUE

I LL}:
Sari
-m

Tflm

MWMMVEm-l olpponunlty lndtltutlon

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Wanna-9.1

THE LINKAGE BETWEEN SOCIAL CLASS, OCCUPATIONAL CONDITIONS,
AND FATHER-ADOLESCENT SON INTERACTIONS:

EXAMINING THE WORK-FAMILY INTERFACE

BY

Guenter H. Wasner

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1993

ABSTRACT
THE LINKAGE BETWEEN SOCIAL CLASS, OCCUPATIONAL CONDITIONS,

AND FATHER-ADOLESCENT SON INTERACTIONS:
EXAHINING THE WORK-FAMILY INTERFACE

BY

Guenter H. Wasner

There is a growing awareness that a fuller
understanding of family processes requires greater knowledge
of the interface between family life and ecological
variables such as social class and occupational conditions.
This study attempted to clarify the process by which
fathers' work experiences influenced their relationship with
their adolescent sons by testing a path model which
predicted a linkage between social class, work conditions,
fathers' values and personal efficacy, parenting style and
confidence, and sons' experience of autonomy and relatedness
with their fathers. Self-report questionnaires assessing the
above variables were completed by 204 undergraduates and
their fathers.

The study found that social class and the job condition
of skill utilization influenced the father-son relationship
through two primary pathways. One pathway was mediated
primarily by fathers' values (conformity versus self-
direction) and supported a "values shaping" hypothesis.

Fathers' valuing of conformity over self-direction was found

to be predictive of parental authoritarianism which in turn
predicted less encouragement of autonomy in sons. The second
pathway was mediated largely by fathers' sense of
powerlessness which was found to negatively influence
parental confidence which in turn was predictive of sons'
experience of support and relatedness from their fathers.
Job autonomy, job satisfaction and job-related stress were
found not to be significant links in the tested model.
Overall, the study demonstrated the value of examining
the mediating processes between work and family
interactions. Methodological limitations and future areas

for research were also discussed.

Copyright by
GUENTER HORST WASNER
1993

ACKNOWLEDGEMENTS

I would like to first of all thank the chairperson of
my dissertation committee, Susan Frank, for her
encouragement and guidance from the initial idea to the
completion of this project. Her willingness to take a risk
with this topic, her conceptual and practical counsel, and
her steady patience were essential to it's success. I also
wish to acknowledge and thank the other members of my
committee. Tom Luster, was particularly helpful in focusing
the direction of this study, introducing me to the
literature, and providing helpful feedback along the way.
Frank Floyd has been supportive throughout this entire
process and his intellectual rigor, especially in regard to
methodology and design, has been a valuable guide. Dan
Ilgen also added useful comments and instruction along the
way. I also wish to thank the chairperson of my doctoral
guidance committee, Gary Stollak, who while not actively
involved with this project, has been supportive and
encouraging throughout my graduate career.

I especially wish to thank all the friends and family
who have helped and sustained me along this journey. The
stress and strains of this work has no doubt influenced my

relationships with them in many ways. For their patience,

support, and love, I am grateful.

Finally, I am indebted to all the fathers and sons who
took the time and energy to participate in this study.
Their willingness to allow me a glimpse into their lives and
relationships has been the most crucial link to this

project's success.

vi

TABLE OF CONTENTS

LIST OF TABLES O O O O O O O O O O O O O v11 1

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

INTRODUCTION . . . . . . . . . . . . . 1
Social Class and Family Process . . . . . . 6
Modes of Parental Control . . 7
Emphasis on Independence and Achievement . 10
Parental Affection and Support . . . . 10
Critique of Social Class Differences . . . 12
The Relationship of Work Environment and
Family Functioning . . . . . 14
Value Shaping Models and Kohn' s Hypothesis . 14
Work, Personal Control and Orientation . . 19
Stress and Spill-over Models . . . . 23
Adolescent Development and Family Functioning . . 27
Father and Son Relationships . . . . . . . 30
Parental Mediators Between Work and
Adolescent Development . . . . . . 31
Proposed Conceptual Model and Hypotheses . . . 33
METHOD . . . . . . . . . . . . . . . 37
Subject Recruitment . . . . . . . . . . 37
Subject Demographics . . . . . . . . . . 39
Measures . . . . . . . . . . . . . 41
Social Class . . . . . . . . . . . 41
Measures of Job Characteristics . . . . . 42
Values . . . . . . . . . . . . . 46
Powerlessness . . . . . . . . . . 47
Family Process Measures . . . . . . . 48
Data Reduction Procedures . . . . . . . . 52
RESULTS . . . . . . . . . . . . . . . 55
Proposed Hypotheses of Path Model . . . . . 55
Goodness of Fit of Predicted Path model . . . 59
Model Modifications . . . . . . . . . . 62
Alternative Model Results . . . . . . . . 63

vii

DISCUSSION . . . . . . . . . . . . . . . 68
Methodological Issues . . . . . . . . 78
Considerations for Future Research . . . . . 80
Conclusion . . . . . . . . . . . . . 82

LIST OF REFERENCES . . . . . . . . . . . . 83

APPENDIX A: Student Demographic Questionnaire . . . 95
APPENDIX 8: Parent Relationship Inventory Subscales . 99
APPENDIX C: Father Demographic Questionnaire . . . 101

APPENDIX D: Skill Utilization and Job Autonomy Scales 103

APPENDIX E: Job Satisfaction Scale . . . . . . 105
APPENDIX F: Job Stress and Powerlessness Scales . . 106
APPENDIX G: Values Scale . . . . . . . . . 107

APPENDIX M: Parental Authority Questionnaire . . . 108

APPENDIX I: Family Experiences Questionnaire

(Parental Confidence Subscale) . . . 111
APPENDIX J: Parent Letter and Informed Consent . . 112
APPENDIX K: Student Informed Consent . . . . . 113
APPENDIX L: Student Debriefing Form . . . . . 116

Table
Table

Table

Table

Table

Table

LIST OF TABLES

Demographic Characteristics of Sample

Correlation Matrix of Latent Variables

i n Path Made 1 O I O O O 0

Summary of Residuals of Endogenous

Variables for Full and Predicted Models
and Chi-Square Goodness-of-Fit Test

Variance Explained by the Full
and Predicted Models . . . .

Revised Model: Summary of Residuals of
Endogenous Variables for Full and Predicted

Models and Chi-Square Goodness-of Fit Test .

Revised Model: Variance explained
by the Full and Predicted Models

ix

4O

56

6O

61

66

67

LIST OF FIGURES

Page

Figure 1 Theoretical Path Model . . . . . . 34
Figure 2 Predicted Model with Path Coefficients . . 57

Figure 3 Alternate Model with Path Coefficients . . 64

INTRODUCTION

Within the past two decades there has been increasing
interest in the study of the family system as a context for
human development. This approach has broadened the lens by
which researchers examine childhood development and
dysfunction, moving away from models of individual pathology
towards paradigms that look more closely at the
interactional patterns and inter-relatedness of behavior
within the family group. The family however, is also a part
of a larger social network, so that ecological systems
theorists have argued that understanding the relationship
between the larger environment and family functioning is
equally important to examining the more micro processes that
occur within individual family systems (Belsky, 1984:
Bronfenbrenner, 1986). The proposed research is embedded in
this broad ecological question of how intrafamilial
processes are affected by extrafamilial conditions.

One of the environmental conditions which has
consistently been shown to impact on family functioning and
child development is social class, or socioeconomic status.
While there has been considerable disagreement over the
accurate measurement of social class (e.g., Haug, 1977: Haug

& Susman, 1971: Hollingshead, 1971; Mueller & Parcel, 1981)

1

2
there is general agreement that its' important components
include occupational prestige or status, educational level,
and income (Haug, 1977). Research over the past several
decades (e.g., Bronfenbrenner, 1958: Gecas, 1979; Hess,
1970: Kohn, 1963: Luster, Rhoades & Haas, 1989) has
suggested consistent social class differences in child
socialization practices and parenting styles, including
differences in discipline techniques, attribution of
motivation to children's behavior, expression of parental
affection, and differences in communication style between
parent and child.

Social class, however, is a broad and complex variable,
and evidence that a family's placement on the social status
hierarchy can influence its interactional patterns tells us
little about how and why this influence occurs, and what the
salient mediating or moderating variables might be. Since
occupational type comprises one of the key indicators of
social class, one major area of exploration may be to look
at the influence of job characteristics on differences in
family functioning. Indeed, research in this area has
suggested that the specific content and nature of the work
activity may influence the worker's personal orientation and
sense of powerlessness (Kohn & Schooler, 1982: O'Brien,
1980), their psychological state or moods (Crouter, Perry-
Jenkins, Huston, & Crawford, 1989, Repetti, 1987), and
their parenting values and behaviors (e.g., Gecas & Nye,

1974: Kohn, 1963, 1980; McKinley, 1964: Miller & Swanson,

3
1958). More specifically, two conditions of work which have
shown particular salience in regards to personal orientation
and parenting values have been the extent of autonomy or
self-direction at the work place (Kohn, 1969,1980: Kohn &
Schooler, 1982: McKinley, 1964) and the opportunity to
utilize one's skills and perceive the work as challenging
and rewarding (O'Brien, 1980).

However, the path by which these differing occupational
conditions influence family processes is not clear and
little is known regarding the more proximal and direct
linkages between job experiences and specific family
interactions. Several different theoretical hypotheses have
been made regarding the processes which may mediate this
work-family interface. These include (a) frustration or
stress models which explore how stressful experiences at
work either "spill over" into negative affective experiences
in family relationships (Piotrkowski, 1979: Small & Riley,
1990) or are compensated for by a greater valuation of
family experiences at home (McKinley, 1964: Staines, 1980:
Hoffman, 1986) and (b) socialization or value-shaping models
(Kohn, 1969, 1980: Ogbu, 1981), which suggest that work
experiences affect parents' ideas about what qualities are
important in adulthood and thus influence attitudes and
behaviors regarding child rearing and marital interactions.

A final limitation of the research examining the
relationship between social class, occupational conditions

and family functioning has been that it looked primarily at

4
parenting practices among parents of school-aged and younger
children. Little work has been done on the influence of
these variables on parent/adolescent relationships and
subsequently on successful adolescent development. Yet,
outcome variables examined to date, such as modes of
parental control, emphasis on independence and achievement,
and the level of parental support, affection, and
involvement have relevance for adolescent development as
well as development during earlier childhood. In fact,
because the successful transition from adolescence to young
adulthood has often been characterized in terms of the
development of autonomy (Hill & Holmbeck, 1986) and
independence (Douvan and Adelson, 1966) within the context
of continued support and relatedness (Grotevant & Cooper,
1986: Murphy, Silber, Coelho, Hamburg, & Greenberg, 1963),
this life stage may prove to be particularly vulnerable to
the influences of occupational experiences on parents'
values and behaviors and the consequent level of both their
encouragement of autonomy and their affective involvement
with their children.

Finally, most of the recent literature examining the
impact of occupational conditions on parenting and family
relationships has looked at the influence of mothers'
employment. With the growing interest in fathers' increased
involvement with their children (e.g., Lamb, 1986, Bronstein
& Cowan, 1988), a greater understanding of the influence of

work on fathers' relationships with their children would

5
seem to be increasingly important.

This study proposes to more fully explicate the
relationship between social class, occupational conditions
and family functioning by looking more closely at variables
that potentially mediate the relationship between work
environment and family interactions. Specifically, it will
focus on the conditions of autonomy and skill utilization at
the father's work place and their linkage with late
adolescent sons' perceptions of their fathers' emotional
support, and encouragement of autonomy. It will also look
at fathers' perceptions of their own parental style and
parental confidence, and how this influences adolescent
perceptions. In addition, several potential mediating
variables between work conditions and family functioning
(i.e., parental values, job dissatisfaction and job stress,
and powerlessness) will be examined to test the utility of a
proposed model that integrates both value-shaping and
spill-over hypotheses. In the literature review that
follows, I will provide a background for the proposed
research model by looking first at the extensive research on
social class and parenting practices. Secondly, I will
examine research relating occupational conditions to both
child socialization and personal orientations such as
perceived control, as well as focusing on theoretical models
exploring the work/family interface. Finally, a section will
explore the relevant issues in adolescent development and

their relationship to family processes that may be

6
influenced by the work environment. The introduction will
conclude with a more extensive conceptual model and specific

research hypotheses.

WW

While the question of social class influences on family
interaction, and more specifically, child-rearing practices,
has produced a substantial literature over the past fifty
years, it has often been contradictory and difficult to
evaluate. This is due in part to the influence of various
disciplines and theoretical perspectives in this area as
well as significant differences in research design, sampling
and measurement techniques, and construct definition.
Despite these difficulties however, the research data does
suggest relatively consistent differences in child rearing
and parenting behavior between social classes.

One of the major difficulties in looking at research
results across studies has been the differences in defining
the variable of social class itself. There has often been a
lack of consensus on whether social class is a discrete or
continuous variable, whether it is unidimensional or
multidimensional, which of its component parts (occupation,
education, income) are most salient, and whether it is a
subjective or objective phenomenon to be measured by
self-report or observable criteria. (Hess, 1970: Otto,
1975). The most frequently used measure of social class in

psychological research has been the Hollingshead Index of

7
Social Position (Hollingshead, 1957), which is based on a
two-factor combination of father's education and occupation,
and places families along a five-category continuum. A more
recent four-factor version includes mother's education and
income and ranks families on a 9-point scale (Hollingshead,
1975). Other approaches range from simple dichotomies,
( i.e., middle class versus lower class), to continuous
scales based on large sample rankings of occupational
titles, such as the Duncan Socioeconomic Index (SEI)
(Duncan, 1961).

Despite these methodological difficulties, several
generalizations regarding the relationship of social class
and parent-child behaviors can be made. Gecas (1979) in his
review of this literature provides a useful framework by
dividing the relevant outcome variables into three major
categories: modes of parental control: parental emphasis on
independence and achievement: and parental support,
affection, and involvement. It should be noted that the
first two categories are most closely related to this
study's interest in the tolerance and encouragement of
adolescent autonomy versus control, while the final category
corresponds to parental support and relatedness vis-a-vis
the adolescent child.

M9§£§_9£_EQ£§2§21_QQDEIQI

Modes of parental control, referring to the manner in
which parents discipline, constrain or channel a child's

behavior, have been a frequent focus of research in parent-

8
child relationship. Gecas's meta-analysis of the research
in this area shows a weak but generally consistent negative
relationship between social class and parental use of
punishment (Gecas, 1979). There also appear to be some
differences not only in what parents do to discipline their
children, but in when they choose to discipline. Kohn
(1969), in interviews with 339 mothers of fifth-grade
children, found that middle-class mothers were quite
discriminate in their use of physical punishment and were
more likely to use it when the child's behavior was defined
as "loss of temper", as opposed to "wild play". Lower-class
mothers, on the other hand, reported the use of physical
punishment with equal frequency in both situations.

In a related vein, Gecas and Nye (1974) found middle-
class parents are more likely to respond to the child on the
basis of their interpretation of the child's motivations and
intentions while lower-class parents are more likely to
respond on the basis of overt acts. They found a greater
difference in the way middle-class parents acted toward a
child when he "accidentally breaks something" versus when he
”intentionally disobeys" than in the responses of
lower-class mothers.

Also several studies suggest that lower-class parents
rely more on the use of commands and imperatives, while
middle-class parents are more likely to explain the reasons
for a rule or request. M. L. Hoffman (1960) defined

parental imperatives as "unqualified power assertions",

9
which take the form of threats and direct commands, and
include the refusal to discuss alternatives, explain the
request, or reason with the child. Hoffman's research found
a negative relationship between social class and parental
use of "unqualified power assertions”. Elder and Bowerman
(1963) and Miller and Swanson (1960) found a similar
relationship to social class when looking at the extent to
which parents explained rules, as opposed to unquestionably
demanding obedience. Finally, studies by Waters and
Crandall (1964), Hess & Shipman (1965), and Erlanger (1974),
which looked at the coerciveness of suggestions, as well as
the use of imperatives, found both increased use of
imperatives and greater coerciveness among lower-class
parents.

Findings regarding the power relationship between
parent and child, or more specifically, the distinction
between an authoritarian or autocratic relationship style
and an equalitarian or democratic style are also relevant.
While the number of studies looking at this variable is
relatively small, there does appear to be a trend for middle
and upper—class parents to have a more equalitarian
relationship with their children and working class parents
to have a somewhat more autocratic parenting style (Gecas,
1979). Specific to our interest in adolescent development,
Elder and Bowerman (1963) found that lower-class adolescents
were more likely to perceive their parents as autocratic

than middle-class adolescents.

 

The relationship between social class and parents'
emphasis on independence and achievement appears to be a
positive, consistent, and fairly strong relationship.
Bronfenbrenner's (1958) early review of these studies found
this to be one of the strongest class-related child-rearing
variables, with middle-class parents consistently having
higher academic aspirations for their child than working-
class parents. Particularly germane to variables examined
in this study was the research by Bayley and Schaefer (1960)
which found a strong correlation between SES and the degree
of autonomy granted to children. Little research however
has focused on the perceptions of allowance of autonomy or
independence for adolescents as opposed to younger children,
across social class.

2aIQnL§l_A££§QEiQn_§nQ_§BRRQI§

Research has also linked social class to the affective
dimension of the parent-child relationship, or the degree of
support, affection, and parental involvement. This variable
is often viewed as equally as crucial a dimension in family
process as parental control, and as such, a significant
variable in influencing child development and dysfunction
(Jacob, 1975). A review of the findings shows a consistent
positive relationship between social class and parental
affection and support. This was found in diverse samples
including a study by Sears, Maccoby, and Levin (1957) with

mothers of kindergarten children; Kohn's (1969) study of

11
mothers and fathers of fifth-grade children: Scheck and
Emerick's (1976) survey of ninth-grade males: and Thomas,
Gecas, Weigert, and Rooney's (1976) questionnaire survey of
high school adolescents and college freshmen. In addition,
studies by McKinley (1964) and Hess and Shipman (1968) found
a positive relationship between SES and parental involvement
with the child, (i.e., high SES parents appear to be more
attentive, helpful, talkative and playful with the child).
However, these findings are not uniform, with considerable
variance in strength. Also, sex of parent seems to be an
important feature for this relationship with most studies
finding greater class differences in fathers', as compared
to mothers,' support and involvement.

Luster, Rhoades and Haas (1989), in a more recent
study, confirmed several of the earlier findings regarding
the relationship of social class and parenting values and
behaviors. This study sampled 65 mother-infant dyads and
found that social class was related to global parenting
values, specific parenting beliefs regarding child-rearing,
and mothers' supportive and constraining behaviors towards
their babies. Specifically, lower social class was found to
be related to greater valuing of conformity, greater concern
over spoiling the child, greater valuing of discipline and
control, less valuing of giving children freedom to explore
the home environment, and less valuing of verbal
stimulation. These specific beliefs were in turn related to

mother's supportive and constraining behaviors towards their

12
babies with greater constriction and less overall supportive
behaviors related to lower social class rankings.

It should be noted that much of the classic research
linking social class to parenting behavior occurred in the
two decades between 1960 and 1980. Since that time there has
been very little substantive research in this area as more
recent studies have tended to focus on work-family linkages
and references to social class are less frequent (Crouter &
McHale, 1993). Research on the work-family linkage will be
reviewed in a subsequent section.

W

The above findings are fairly consistent: a lower
position on the social hierarchy appears to be associated
with more controlling parenting styles, less emphasis on
independence and achievement, and less perceived positive
nurturance and support. However, a few investigators have
challenged the notion that a consistent relationship between
social class and parent-child relationships actually exists.
Perhaps the strongest critique has been made by Henggeler
and his colleagues (Borduin, Henggeler, Sanders-Walls &
Harbin, 1986; Henggeler & Tavormina, 1980). Henggeler has
argued that observed social class and race differences in
family interactions are often the result of the failure to
consider confounding factors, specifically father absence,
age and sex of child, family size, and cultural differences
between researcher and subjects. In a study testing this

hypothesis, Henggeler & Tavormina (1980) looked at the

13

interaction patterns of 64 well-adjusted families that
included both mother, father and a 14 to 16-year-old son or
daughter, and matched subject and interviewer ethnicity.
The results revealed few social class or race effects in
regards to measures of family conflict, affect, and
dominance. Likewise, in another study that looked at social
class differences in maternal control, sensitivity, and
child compliance (Borduin et al., 1986) among 32 pairs of
mothers and their 4-year-old sons, no significant social
class differences were found in mother-child interactions.
Family size, father absence and child's intelligence were
controlled leading to the conclusion that previously
observed social class effects on family functioning may have
been due to the failure to control for such covariates in
defining social class. The investigators also speculate
that social class differences in maternal control that may
have existed 15 to 20 years ago are no longer evident.

While the critique of Henggeler and his colleagues are
important additions to this literature, it is unlikely that
the rather consistent findings regarding social class
differences in parenting practices through the last several
decades are due primarily to confounding variables or
methodological flaws. The Henggeler argument is further
weakened by studies such as that of Kohn (1969), which
statistically controlled for such potential confounds as
family size, and the several large studies which indirectly

controlled for father absence by focusing primarily on the

14

interactions between fathers and sons (Kohn, 1969: McKinley,
1964).
7,; ;; . .,., . . w. . .1 g.,u-, 1,. ,u g 1‘, ., ,.

In attempting to better understand why differences in
social class are related to family functioning, several
researchers have looked towards the content and climate of
the work environment. They propose that because
occupational characteristics change across class
hierarchies, work environment may mediate effects of SES on
family functioning. As Moos (1986) has pointed out, since
most workers spend from one third to one half of their
waking lives at their jobs it seems reasonable to
hypothesize that the quality and nature of one's work
environment has the potential to impact on one's sense of
self in regard to the broader social environment, and also
provide a model for social interactions with other systems
such as the family.

- ' ode s 's s's

Two of the major theories that have been proposed to
explain the mediation process between the work environment
and family functioning are the socialization or value-
shaping models, and the stress and frustration, or
spill-over models. One of the pioneering studies
investigating a value-shaping hypothesis was that of Miller
& Swanson (1958). They hypothesized that the child-rearing
patterns of parents would reflect the values and goals

consistent with traits that men experienced as being

15
valuable at work. They contrasted two different styles of
occupational organization, entrepreneurial and bureaucratic.
The entrepreneurial pattern was characterized by a simple
division of labor, mobility within the work setting through
competition and individual initiative, and risk-taking.
This organizational pattern was predicted as being
consistent with child-rearing practices that encouraged
achievement, responsibility and self-control. In contrast,
the bureaucratic occupational structure was characterized by
multiple levels of authority, greater job security, mobility
through interpersonal skills, and smooth relations within
the organization. This style was hypothesized to be
consistent with child rearing that emphasized accommodation
over self-control, and interpersonal interaction rather than
independence. On the whole, the findings of Miller and
Swanson (1958) supported these hypotheses. When a father
was employed in an entrepreneurial setting, the child-
rearing patterns in the family were more consistent with the
goals of independence, responsibility and self-control,
whereas when the father's occupation was of the bureaucratic
type, child-rearing practices were more permissive and
consistent with the goal of getting along with others.
While this was an important pioneering study in this area,
the research suffered from weak operational definitions of
the work setting and was unable to clearly suggest any
directionality of effect. Thus, while occupational type

could influence a father's child-rearing values it could be

16
equally likely that fathers with particular values and
orientations chose certain occupations.

The most comprehensive socialization or value-shaping
theory is that of Melvin Xohn (1959, 1963, 1969). Kohn's
model, which assumes a functional link between social class
structure and parenting practices, proposes that differences
in conditions of life associated with social class position
give rise to parents' values and orientations, which in turn
affect their patterns of child rearing. The causal sequence
proposed by Kohn can be diagrammed in the following manner:
social class --> conditions of life --> parental values -->
parental behavior.

The conditions of life which Kohn sees as most
important to family functioning are those present in the
parent's work environment. Specifically, he argues that the
occupational conditions characteristic of blue-collar and
white-collar work give rise to different adaptive values
which are transmitted by the parents in the socialization of
their children, both directly as conscious attempts to
inculcate values in their offspring, and indirectly through
different styles of parent-child interactions. He further
proposes that the most salient occupational conditions were
those that were conducive or restrictive of the expression
of self-direction. By occupational self-direction, Kohn
lneans "the use of initiative, thought and independent
judgment in work" (Kohn, 1969, p. 140). In addition, Kohn

theorizes that occupational self-direction is determined by

17
the substantive complexity of the actual work tasks (i.e.,
how much thought and independent action is required): the
closeness of the supervision; and the extent of
routinization of the work. Most middle-class or white-
collar occupations are viewed as promoting self-direction
because they involve manipulating ideas, symbols and
interpersonal relationships, and require more flexibility of
thought and judgment. Lower-class or blue-collar
professions are seen as constrictive to the development of
self-directive values as they involve less complexity, more
standardization, and closer supervision. Thus, while white
collar workers are more likely to enunciate values dealing
with self direction, such as freedom, individualism,
initiative, creativity and self—actualization, blue-collar
workers are more likely to stress values of conformity to
external standards, such as orderliness, neatness, and
obedience. The essential difference between these two sets
of values is that self-direction focuses on internal
standards of behavior, whereas conformity focuses on
externally imposed rules.

Kohn also links parental values and parental behavior,
maintaining that value orientations are reflected in the
style and circumstances of parental discipline. Because of
the greater emphasis white-collar parents place on
self-direction and internal standards of conduct, they are
hypothesized as more likely to discipline their children on

the basis of interpretation of the child's intent. Blue-

18
collar parents on the other hand, placing greater stress on
conformity and adaptation, are more likely to react on the
basis of the consequences of the child's behavior.

In regards to empirical evidence, results from several
major studies have tended to support Kohn's predictions
regarding the influence of social class and job content on
parenting values. In three major investigations including a
national study of 3,101 men across all social-class
categories, Kohn (1969) found a relatively strong (.38 to
.50) relationship in the expected direction between social
class and parental values of self-direction versus
conformity. In a replication study also using a large
national survey, Wright 5 Wright (1976) supported these
results, finding that social class accounted for 41 to 63%
of the total variance in parental values of self-direction
explained by the model. Moreover, their analysis confirmed
Kohn's finding that a substantial relationship between
values and social class remains when numerous other factors,
including national background, religion, urbanicity, race,
and region of country, were statistically controlled (Kohn,
1969: Wright & Wright, 1976).

When occupational content was focused on, Kohn and his
colleagues (Kohn, 1969: Pearlin & Kohn, 1966: Kohn &
Schooler, 1969) found a positive relationship between
increased autonomy and self-direction at work and parental
values stressing independence, achievement, and

self-reliance. Furthermore, when analyses were done that

19
controlled for social class, occupational variables
continued to predict parental values beyond the contribution
of class alone (Kohn & Schooler, 1969).

Although the evidence linking social class to parental
values is impressive, much less attention has been paid to
the relationship between parenting values and behaviors.
While the findings of Kohn (1969) and Gecas and Nye (1974)
were primarily confined to disciplinary practices, the study
by Luster, et a1. (1989), referred to earlier, did find a
relationship between global parental values, specific
beliefs about child-rearing, and home observations of
mothers interacting with toddlers. Their findings did
suggest that Kohn's value-shaping hypothesis could be
extended to predict parenting behaviors at least with
mother-infant dyads.

W

More recent research by Kohn has extended his model to
examine not only the relationship between work content and
parental values, but also the interplay between work and
adult personality and intellectual functioning. When Kohn
and his colleagues (Kohn & Schooler, 1969) looked at the
differences in worker values and attitudes across
occupations, they discovered that workers in jobs that had
more substantive complexity (i.e., required more autonomous
thought and action) valued self-direction over conformity,
demonstrated greater intellectual flexibility and were less

likely to attribute their problems to external causes.

20
Later studies by Kohn supported the notion that
self-directed occupations were significant predictors not
only of intellectual flexibility, but also, although to a
lesser degree, of self-esteem, job satisfaction and
commitment, and decreased feelings of powerlessness (Kohn,
1976).

Adding strength to the hypothesis that conditions of
work can directly influence personal orientation and style
have been longitudinal studies (Kohn 8 Schooler, 1981),
which have suggested that, to some extent, the job content
has more influence on personal variables such as
intellectual flexibility and self-esteem than these
variables have on job choice. In a path analysis model
derived from their longitudinal data, Kohn and Schooler
(1981) suggest that the structural aspects of a job are
mediated by the extent of allowed occupational
self-direction which impacts on variables such as
flexibility, sense of personal control, and to a lesser
extent, overall level of distress. They argue that their
findings hold regardless of social class, sex, type of
industry, or formal organizational structure of the work
place.

In a critique of Kohn and Schooler's findings, O'Brien
(1986) argues that while their data are substantial, their
interpretation is limited by idiosyncratic and unclear
constructs. He argues that there is little validity for the

notion of intellectual flexibility, as it is unclear whether

21
what is being measured is a fluid state or a more stable
trait, such as general intelligence. O'Brien makes a
similar criticism of the construct of self-directed
orientation, questioning whether it measures something
different from self-efficacy or locus of control. He argues
that locus of control, which refers to the generalized
expectancy about the extent to which reinforcements are
under external or internal control, is a more validated and
conceptually clear construct in measuring a worker's
adaptive style to his work environment.

In addition to these methodological and conceptual
critiques, O'Brien suggests that not only autonomy and
complexity, but also the extent of skill utilization or
challenge (i.e., the degree of match between the employee's
skills and the skills required for the job) is crucial to a
sense of power and control on the job. In a large-scale
study of Australian workers, focusing primarily on job
characteristics and their impact on worker satisfaction and
overall mental health, O'Brien (1983) found that employees
with a good job-skill match were more likely to develop an
internal orientation in regards to locus of control than
those who felt their skills were under utilized. In an
earlier series of studies, across a range of population
groups, O'Brien (1980, 1982,) also found that the
self-report of skill utilization was a strong predictor of
job satisfaction and was found to have an independent

contribution to job satisfaction beyond other work factors

22
such as skill variety, task significance, autonomy, and job
feedback. In addition, Kornhauser (1965), in a study of
Michigan automobile workers found that skill utilization was
the major factor accounting for variations in job
satisfaction and mental health, with employee mental health
being measured as a composite of manifest anxiety,
self-esteem, hostility, sociability, morale, and general
life satisfaction.

More generally, while O'Brien and Kohn have focused on
somewhat different constructs, work by both suggests that
the nature of the job and its specific activities can have
important influences on how workers perceive their efficacy
or potential for power or control in their broader social
context.

Although the above theoretical models and empirical
findings suggest relationships between the work environment
and the development of values and personal efficacy which
could impact on family functioning, they do not make any
clear linkage to direct parental or child behaviors.
Piotrkowski and Katz (1982) attempted to examine this
linkage by looking at the indirect socialization effects of
parents' occupational conditions on childrens' school
behaviors. Utilizing a sample of 60 inner-city,
predominantly minority families, and following Kohn's model
of indirect socialization, they hypothesized specific links
between parents' job characteristics and similar school

behaviors in adolescents. Specifically, they predicted that

23

mothers' job autonomy and consequent valuation of
self-directedness versus conformity would negatively relate
to school attendance (assuming that frequency of school
attendance indicated the extent of conformity to external
authority): that mothers' job demand (the effort required to
do the job) would relate positively to children's school
effort: and that job skill utilization would relate
positively to their children's skill acquisition or academic
achievement. Their results provided moderate support for two
of their predictions in that the less autonomy the mothers
exercised on the job, the more frequently their children
attended school, and the greater the extent of mothers'
skill utilization, the higher their academic achievement.
While their attempt to extend Kohn's hypothesis is to be
applauded, the attempt to create isomorphic linkages between
a child's school behaviors and parent's work context
excludes too many crucial mediating factors such as the
quality of the family relationships and particularly, the
nature of the parent-child interactions. In addition, such
a study does not help explicate the processes that lead a
parent's work life to influence family functioning.

W

The above discussion has described how value-shaping
and socialization models have attempted to explain how
social class, through its relationship with differing
occupational conditions, can influence both parenting

behaviors and personal orientations towards control and

24
powerlessness. A second set of hypotheses can be broadly
categorized as frustration or stress models. Stress models
suggest that the work environment impacts on the family by
influencing the worker's overall psychological state or mood
(Bolger, Delongis, Kessler 8 Wethington, 1989: Crouter,
Perry-Jenkins, Huston 8 Crawford, 1989, McKinley, 1964:
Piotrkowski, 1979). One of the earliest of such models, the
frustration aggression model, suggested by the work of
McKinley (1964) argues that social class is related to the
degree of frustration experienced in life, particularly in
the occupational sphere, which in turn affects the degree of
aggression and authoritarianism in parent-child
interactions. McKinley further argues that aggression
generated by frustration on the job is often displaced onto
the family, since there are more constraints on its
expression on the job and in other more socially proscribed
situations. The family, as such, is used as a compensatory
source of need satisfaction. In other words, a man may feel
powerless at work and be at the bottom of the power
hierarchy but at home he can be at the top.

McKinley (1964) finds some support for this hypothesis
in a series of studies looking at social class and father-
son relationships, which included questionnaire responses
from 260 11th and 12th-grade boys, and semi-structured
interviews with fathers with sons 13 to 19 years of age. His
results indicate a relationship between less autonomy at the

work place and more aggressive statements made by fathers

25
toward sons, as well as reports by sons that their fathers
utilized more severe discipline techniques. This
relationship was found across social-class rankings and was
found in both sons' and mothers' reports of fathers'
behavior. This notion of increased parental
authoritarianism over children as a compensation for a sense
of powerlessness at work is also described in the
theoretical work of Blau and Duncan (1967) and Hoffman 8
Manis (1979).

The idea of the family as fulfilling compensatory needs,
not aroused or satisfied at work does not however have to
imply a negative effect on the family. As Hoffman's work on
the value of children to parents has suggested (Hoffman 8
Hoffman, 1973: Hoffman 8 Manis, 1979), children can be seen
as satisfying those psychological needs that are not met by
other aspects of life, such as the work setting. While
these may be needs for power or dominance, they may also be
needs for companionship, acceptance, or simply fun and
stimulation. One moderating variable for this process may
well be the extent of overall parental investment in work
versus family activities (Greenberger 8 Goldberg, 1989).

Another related conceptualization suggests that moods
at work are carried over to the family. This "spill-over"
hypothesis proposes that job satisfaction and overall job
distress influence the general morale of the worker which
influences the affect and quality of the interactions with

family members. Piotrkowski (1979), in a small-scale

26
naturalistic study of blue collar and middle-class families,
identified three interface patterns between work and family:
positive carry-over, negative carry-over, and energy
deficit. According to Piotrkowski, in positive carry-over,
good feelings and satisfaction at work bring positive energy
to the family's life. In negative carry-over, the stressful
experiences at work are brought back to the family and the
family members are required to expend resources to manage
the stress. The third pattern, energy deficit, proposes
that the worker has little emotional energy left for the
family, and that this personal depletion often results in a
kind of depression and withdrawal from family interactions
and activities.

Recent research in this area has found additional
support for "spill-over" effects between work and family
life. Stressors and conflicts at work have been found to
increase the likelihood of arguments with spouses (Bolger et
al., 1989), to be related to overall marital quality
(Hughes, Galinsky 8 Morris, 1992), and to predict greater
social withdrawal in parent-child interaction (Repetti,
1989). Other researchers in this area (e.g., Crouter et al,
1989: Small 8 Riley, 1990) have attempted to further specify
the nature of work spillover by examining several processes
by which work influences family interaction, such as time
spent away from home, psychological or mental preoccupation
with work, and physical and mental fatigue. Finally, recent

work has examined the variables that mediate or moderate the

27
influence of work strain on family interactions, such as the
quality of the marital relationship (Hughes et al. 1992:

Repetti, 1989) and the level of investment in work versus

family (Greenberger 8 Goldberg, 1989).

 

Despite the range of research that has explored the
relationship between social class, occupational conditions
and child socialization there has been little exploration of
ways in which extrafamilial factors might influence the
developmental processes of adolescents, as opposed to
younger children. Particularly missing has been any
research on how these variables might relate to adolescents'
struggle for autonomy and individuation from their families
of origin. This seems particularly important in the light
of recent research and theorizing on normative adolescent
development.

Hill and Holmbeck (1986) in a review and critique of
the recent literature on autonomy and attachment during
adolescence argue that successful functioning in adolescence
involves the development of autonomy, seen primarily as
self-governance and self-regulation, as well as the
maintenance of connected and close relationships with
parents. The ability of parents to be able to encourage
independence in their adolescent children while facilitating
warm and supportive interactions is also viewed as crucial
in many theoretical writings on adolescent development, such

as that by Bell 8 Bell (1983): Grotevant and Cooper (1986):

28
and Murphy, Silber, Coelho, Hamburg, and Greenberg (1963).

Grotevant and Cooper (1986), in their theoretical
review of individuation in family relationships, argue that
adolescents who perceive their parents as facilitating their
autonomy in an atmosphere of connectedness are more
successful in personal identity formation and role-taking,
two tasks generally conceded to be crucial in successful
adolescent development. Studies conducted by Cooper,
Grotevant, 8 Condon (1983) and that of others (e.g.,
Campbell, Adams, 8 Dobson, 1984: Ryan 8 Lynch, 1989:)
support the proposition that family interactions that
provide a balance of connectedness and encouragement of
autonomy facilitate positive adolescent identity formation,
social competence and enhanced self-esteem.

Other family systems theorists, although not focusing
specifically on adolescence, (e.g., Beavers, 1977: Minuchin,
1974: Olson, Sprenkle 8 Russell, 1979) have also argued that
optimal family functioning requires a level of cohesiveness
or connectedness between members, as well as the ability to
allow members to be autonomous and independent. Further
support for the salience of these constructs is also
provided by Jacob (1975) in his review of family
interactional research. His review argues that the extent
of dominance or control in family interactions, which can be
perceived as the converse to independence and autonomy, as
well as the level of positive affect, are two of the major

variables that distinguish disturbed from nondisturbed

29
families.

Finally, Baumrind (1971, 1978), in her model of the
relationship of parenting practices to the development of
competence in children, provides support for the notion that
a balance between firm direction and warmth provides the
optimal environment for child development. Baumrind has
theorized that most parenting styles fall into three
prototypes (permissive, authoritarian, and authoritative).
Based on Baumrind's definitions, permissive parents tend to
make few demands upon their children, allowing them to
exercise as much control as possible over their own
activities. These parents are also generally nonpunitive and
noncontrolling in their interactions with their children.
Authoritarian parents, on the other hand, are apt to be much
more directive with their children, shaping and controlling
their behaviors and attitudes whenever possible. They
generally value unquestioning obedience and discourage
verbal give-and-take with their children. Between these two
extremes are the authoritative parents. These parents
generally establish firm, clear, and relatively demanding
expectations for their children, but they tend to exercise
their authority in rational, issue-oriented, and flexible
ways that encourage communication with their children.
Baumrind (1978) and other researchers using this theoretical
construct (e.g., Buri, 1989a; Dornbusch, Ritter, Liederman,
Roberts, 8 Fraleigh, 1987; Steinberg, Elmen 8 Mounts, 1989:

Steinberg, Lamborn, Dornbusch, 8 Darling, 1992) have found

30

that authoritative styles of parenting are most conducive to
the development of social and intellectual competence in
children as well as higher levels of self-esteem,
psychosocial maturity, and academic success in adolescents.
W

A few additional comments need to be made regarding the
decision to focus this study on fathers and sons. While
some of the earliest research attempting to examine the
effects of occupation on family life looked at fathers' work
experience, most of the recent scholarship in this area has
focused on mothers and the influence of mothers' employment
on child development. This limitation has been particularly
true of the burgeoning literature on dual-earner families
and child development. Reviewers of the literature
(Bronfenbrenner 8 Crouter, 1982; Crouter 8 McHale, 1993:
Kanter, 1977) have urged more research looking at the
processes that link fathers' work experiences with their
interactions with their children and the resulting
developmental outcomes. This seems particularly salient in
light of the growing interest in the increased involvement
of fathers in child care or, as Michael Lamb (1986) has
described it, the movement from "breadwinner to nurturer".
While the extent of the changes in fathers' role functioning
is not yet clear (see Almeida 8 Galambos, 1991: Cowan 8
Bronstein, 1988), a further understanding of the potential
constraints to fathers' parental involvement, such as

occupational conditions, would seem to be increasingly

31
important.

The rationale for focusing on sons is a more pragmatic
one. The limited literature that has examined the parenting
relationship between fathers and their adolescent children
has generally found that fathers spend more time and are
more involved with sons than daughters (Lamb, 1986: Youniss
8 Ketterlinus, 1987). Other findings regarding the influence
of social class on parenting and child development have
suggested that social class differences are greater and more
consistently found in interactions with sons (Erickson 8
Gecas, 1991; Gecas, 1979). These findings suggest that
finding a relationship between ecological variables such as
social class and occupational conditions and father-
adolescent relationships would be more probable when
focusing primarily on the relationship with sons.

;, . ,-.q._.g: ;- --, .g, 1,. :2-,-- -, u; - ..u,,

Given the above theoretical frameworks, it is
reasonable to propose that extrafamilial influences from the
parent's work environment, (i.e., adaptive values, feelings
of dissatisfaction and distress, a sense of powerlessness)
influence parent-child interactions, which in turn influence
the adolescent individuation process. Specifically, it
would appear that parental values, in regard to
self-direction and conformity, would influence parental
control and authority style, with parents more prone to
valuing conformity over self-direction being more likely to

utilize an authoritarian versus a permissive or

32
authoritative parenting style. This parenting approach
presumably will influence adolescents' perceptions of how
much the parent allows for and encourages autonomy.

The spill-over from work of a sense of distress and
dissatisfaction would most likely influence the amount of
positive affect and nurturance parents have available in
interacting with their children. This, in turn, would be
expected to impact on perceptions of support and relatedness
between family members. Finally, the sense of powerlessness
which might arise from a lack of challenge and autonomy at
the work place could generalize into feelings of a lack of
mastery and confidence in parenting itself. Parents may
alter and compensate for this sense of powerlessness by
increased parental control and harshness of affect which
would be reflected in both parent and child perceptions
regarding autonomy and relatedness. It is important to
point out that the value-shaping, spill-over, and
compensatory hypotheses, which have been suggested as
mediating processes along the work and family interface, are
not seen as mutually exclusive, but rather as having
independent effects on family functioning. Better
understanding of how these processes might differentially
influence family interactions can shed considerable light on
the influence of ecological variables on family processes

and healthy individuation in adolescents.

33
WW

While considerable research has confirmed differences
in family functioning and parenting styles across social
class, less is known regarding the process by which
socioeconomic status influences family interaction and
consequently, child development. It has been argued that
one of the more salient aspects of social class which
impacts on family life is the content and climate of the
parent's work environment.

The proposed study will attempt to explicate the
relationship between social class and family functioning by
examining the interface between the father's work experience
and family processes salient to healthy adolescent
development. Particular focus will be on clarifying the
proximal variables that mediate the influence between work
and family functioning. A causal path is predicted (See
Figure 1) in which social class influences the extent of job
autonomy and skill utilization, which in turn influences a
worker's values and beliefs regarding conformity and
self-direction, their overall job satisfaction and distress
level, and their personal sense of powerlessness and mastery
over their environment. These latter variables are
predicted to influence family interactional patterns,
particularly as related to parent-adolescent relationships,
including parental authority style and overall confidence in
parenting abilities. It is further proposed that these

parental variables will influence adolescent sons'

34

 

 

 

<5chm n
GOING”: dm.

iv

 

 

 

menu)?
nr>mm

 

 

 

 

 

how
>GHOZOKK

mflmfir
CHHFHN>HHOZ

mmFWIUHwMOHHOZ

 

I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

how >de0flHH>wH>B
w>wMZHHzn
m>HHmw>nHHOZ MHKFM
MZOOGN>DNZM2H
Om
>6HOZOZK
how
wmrmeu «GMMOfiH
mawmmm nur>anuzmmm
w>wnzu>b
oozwmomznm

 

MOSMNhMmmzmmm

 

mHWCHo H

admononwomw won: zodow

 

 

35
perceptions of their relationship with their fathers,
including their perception of their fathers' encouragement
of autonomy and their perceptions of their fathers' support
and relatedness.
The study proposes to test the following hypotheses:

1. It is proposed that father's social class as
indicated by educational level, occupational prestige and
income will be positively related to job autonomy and
greater skill utilization. It is also proposed that higher
levels of father's social class will directly relate to a
decreased valuing of conformity over self-direction and
decreased sense of personal powerlessness.

2. It is proposed that job autonomy will be positively
related to a greater valuing of self-direction over
conformity, a decreased sense of personal powerlessness,
increased job satisfaction and decreased job-related stress.

3. It is proposed that increased skill utilization on
the job will relate to increased job satisfaction,
decreased job-related stress and a decreased sense of
powerlessness.

4. It is proposed that valuing of self-direction over
conformity will relate to parental report of less
authoritarian parenting style.

5. It is proposed that greater job satisfaction and less
job stress will relate to decreased sense of powerlessness
in fathers.

6. It is proposed that a decreased sense of powerlessness

36
by fathers will be related to greater confidence in their
parental abilities.

7. It is proposed that a less authoritarian parenting
style by fathers will be related to perceptions of greater
levels of encouragement of autonomy, and increased perceived
support and relatedness by their adolescent sons.

8. It is proposed that fathers' increased parental
confidence will be related to their adolescent sons'
perceptions of both increased support and relatedness, and

greater encouragement of autonomy.

METHOD

Witness.

Subjects for this study were 204 male undergraduates
and their fathers. The majority of students were recruited
from introductory psychology classes in a large mid-western
university (N=193). A smaller group of students were
recruited from social science classes in a local community
college (N=11). The students were asked to fill out
self-report measures regarding their perceptions of their
parents and family and a brief demographic questionnaire
which included a description of their fathers' occupation
and work activity. Students were further asked to take a
self-report questionnaire home to their fathers, or a
questionnaire was mailed out, which asked the fathers for
information regarding demographics, occupational status,
work conditions, and values and self-perceptions regarding
parenting. A letter was included in the questionnaire,
which briefly described the study and established consent to
participate. Fathers were instructed to complete the forms
individually and return them in a sealed envelope to their
sons or return them by mail in a prepaid mailing envelope.
A total of 240 questionnaires were distributed to fathers
and their sons who were undergraduates at Michigan State

University and 193 were returned and proved eligible for

37

38
participation (e.g., students in the correct age range,
fathers were biological parent or step-parents for at least
the last 10 years) for a return rate of 80%. Sixty
questionnaires were distributed to fathers and sons at
Lansing Community College and 11 were returned and eligible
for a return rate of 18%. All together, 300 questionnaires
were distributed and 204 were returned and eligible, for a
return rate of 68%. The significant difference in the
return rate between the two groups of students most likely
was due to the fact that for the Michigan State students
there was increased incentive for the return of the
completed questionnaires in that each student was required
to accumulate a certain amount of research credits as part
of their introductory psychology class. The community
college sample had no such research credit system and thus
less overall incentive to have the questionnaires completed
and returned. A comparison of the four year-college and
community-college groups found no significant differences on
demographic variables such as occupational status, level of
education, students' and fathers' age. Fathers of students
at the community college did however average slightly less
income, with a median income between $40,000 and 50,000
compared with between $50,000 and 60,000 for the four year
students. Duplicate information regarding occupational,
parenting, and relationship issues was also gathered for
mothers as part of a larger study. However, data regarding

mothers was not analyzed for this study. Consent and

39
participation forms were provided in accordance with the
ethical principles in the conduct of research with human
participants as set forth by the American Psychological
Association (1982) and the relevant human subjects research
committee standards of Michigan State University.
Wire

A demographic profile of this sample can be found in
Table 1. Eighty percent of the student subjects were
between 18 and 20 years old with the median age being 19
(range = 18 to 22). Thirty-three percent were freshman, 39%
were sophomores, 23% were juniors, 6% were seniors. Ninety-
two percent of the students were Caucasian, with 1.5%
African-American, 4% Asian, and 1% Hispanic. Ninety-four
percent of the fathers were the biological parents of the
student participants, and step-parents had lived with the
student for at least 10 years.

In terms of socioeconomic status, 57% of the fathers
were college graduates, with 11% having a high school degree
or less, and 21% having advanced degrees. The median income
for fathers was between fifty and sixty thousand annually
with 7% of the sample making less than $25,000, 43% earning
between $25,000 and $50,000 and 50% earning $50,000 or more.

In regard to occupational title and status, although the
fathers from this sample were dispersed across a wide range
of occupations, the sample is somewhat skewed towards
professional specialties (e.g., teachers, physicians,

lawyers,) and executive and managerial positions (e.g.,

40

Table 1
Dem9grannis_Qharasteristiss_9f_§amnle

 

 

Characteristics Sons Fathers
Ethnicity (%)
White 92.2 90.7
Asian 4.4 5.4
Afican-American 1.5 1.5
Hispanic 1.0 1.0
Other 1.0 1.5
Age
Mean 19.6 48.8
SD 1.1 5.8
Median 19.0 48.0
Education (%)
Less than 9th grade .5
Less than 12th grade 1.0
High School Graduate 9.3
2 years or less college 20.1
Less than 4 years college 11.8
4 year college degree 26.5
Masters Degree 18.1
Doctorate Degree 12.7
Gross Yearly Income (%)
Under 7,000 .5
$7,000 - 15,000 1.5
$15,000 - 20,000 1.0
$20,000 - 25,000 3.5
$25,000 - 30,000 4.0
$30,000 - 40,000 16.5
$40,000 - 50,000 22.5
$50,000 - 60,000 9.5
$60,000 - 80,000 21.0
$80,000 - 100,000 2.5
Over $100,000 17.5
Occupational Categories (%)
Executive, administrative, managerial 21.6
Professional specialty 31.3
Technical and related 4.9
Sales 15.7
Administrative and clerical support 3.0
Service 2.9
Farming, forestry, fishing .5
Precison production, crafts, repair 15.2
Operators, fabricators, laborers 4.9

41
administrators or mangers in the private or public sector)
with over half of the fathers falling in one of these
categories. The mean score on Duncan's Socioeconomic Index
(SEI) was 55.3 (SD = 20.7) (range 16.1 to 89.6). This score
is considerably higher than national average scores derived
from several large scale demographic surveys of the civilian
labor force (e.g., 42.6 in Featherman 8 Hauser, 1978: 39.0
in Grusky 8 Diprete, 1990), further suggesting that this

sample is skewed towards higher status professions.

Measures
Wags

Father's social class was measured by combining
measures of occupational status, educational level and
yearly income. Occupational status was measured utilizing a
revision of Duncan's Socioeconomic Index (SEI) (Duncan,
1961: Stevens 8 Cho, 1985). Father's job titles were
matched with 1980 Census Occupational Codes and given an SEI
score per Stevens and Cho's (1985) revision of the
Socioeconomic Index. Examples of titles and scores are:
Insurance Adjuster (55.8), Mechanical Engineer (76.7), and
Welder (20.7). The SEI is one of the most widely-utilized
measures of occupational status and is based on prestige
rankings of occupational titles. Numerous studies have
found prestige rankings to be quite stable and have high
reliability over time (e.g., Hodge, Seigel, and Rossi, 1964:
Featherman 8 Hauser, 1978). The SEI has proven effective in

describing socioeconomic distances between occupations

42
(Featherman, Jones, 8 Hauser, 1975) and has been widely used
as a summary of socioeconomic status in studies of
educational and occupational attainment (e.g. Sewell and
Hauser,1980: Blau 8 Duncan, 1967: Grusky 8 DiPrete, 1990)
and other social processes (Waite 8 Spitze, 1981).

Education level was measured using an eight-point
scale, ranging from less than ninth grade to Doctoral-level
degree. Income level was measured with an 11-point scale of
gross annual income, and ranged from below 7,000 to over

100,000.

W
W

Job autonomy was measured with a 10-item instrument
composed of four items from an autonomy scale developed by
Bacharach and Aiken (1976) and six items from the autonomy
scale developed by Sims, Szilagyi 8 Keller (1976). On the
Bacharach 8 Aiken scale, subjects are asked to respond to
the extent that statements are true about their jobs (e.g.,
"at my job a person can make his own decisions without
consulting anyone else"). This scale utilizes a four-point
response dimension ranging from "definitely true" to
"definitely false". In a study of city administrators and
workers, alpha coefficients of .66 and .65 were recorded for
this instrument, and autonomy was correlated .34 and .49
with reported influence in work decisions.

On the remaining six items, drawn from the autonomy

subscale of Sims, Szilagyi and Keller's Job Characteristics

43
Inventory, subjects are asked to rate how several statements
best fit their jobs (e.g., "How much are you left on your
own to do your work?"), and on these items a five-point
Likert Scale is used ranging from "very little” to "very
much”. Alpha coefficients of .84 and .85 were found in
studies of managers and supervisors in a manufacturing firm
(Sims, Szilagyi 8 Keller (1976) and a study of clerical
employees (Pierce 8 Dunham, 1978), respectively. Convergent
validity for this measure has been established with a strong
correlation (.68) with the corresponding autonomy subscale
in Hackman and Oldham's Job Diagnostic Survey. Alpha
coefficients for the current sample were .81 for the
Bacharach and Aiken measure, and .88 for the Sims and
Szilagyi measure.

5.1511141211251121:

Skill utilization or job challenge was measured by the
combination of two instruments, a self-report measure of
skill utilization developed by O'Brien (1980) and the Job
Challenge subscale from the Michigan Organizational
Assessment Questionnaire (Cammann, Fishman, Jenkins 8 Klesh,
1979). O'Brien's scale, which is part of a larger job
attributes questionnaire, includes four items and asks
respondents to indicate how much they used their abilities
or training on the job. Items ask about opportunities for
learning new jobs, working in the way they think best, using
abilities, and using training and experience. It utilizes a

five-point Likert-like response scale with categories

44
including: "not at all", "very little", "a reasonable
amount", and "a great deal". O'Brien reports high internal
reliability for this scale, and in terms of content-related
validity, he has found a moderately strong correlation of
.51 between employee perceptions of their skill utilization
and expert ratings. In addition, in a series of studies
across different occupational categories, O'Brien (1982)
also reports correlations between skill utilization and job
satisfaction averaging .49.

The MOAQ (Cammann et al., 1979) contains a number of
scales of work attitudes and perceptions. The three items
from the Job Challenge subscale were utilized. This scale
asks subjects to rate statements about their job (e.g., "My
job is very challenging": "On my job I seldom get a chance
to use my special skills and abilities") utilizing a seven-
point scale with end points of "strongly disagree" and
”strongly agree." A high internal reliability alpha of .81
has been reported by the authors for this scale. Alphas for
the current sample were .71 for both the Skill Utilization
and Job Challenge Scales.

Job Satisfagtion and Job Stress

Job satisfaction was measured utilizing the Type of
Work Satisfaction subscale from the Job Descriptive Index
(JDI) (Smith, Kendall 8 Hulin, 1969). The JDI is a widely
utilized instrument that was developed to measure the
principal features of job satisfaction. It utilizes a list

of adjectives and phrases (e.g., fascinating, boring,

45
satisfying, tiresome), and respondents indicate whether each
adjective describes the job aspect in question. The Type of
Work subscale, includes 18 such items. Responses are Yes,
Uncertain, or No. The authors suggest that Uncertain
responses are more indicative of dissatisfaction than
satisfaction, therefore, responses are scored three, one,
zero, respectively. The authors report a high level of
internal consistency with a Spearman-Brown coefficient of
.84 for this subscale. A number of other studies across a
range of occupational types (e.g., Dunham, Smith and
Blackburn, 1977; Schuler, 1979) have also found adequate
internal reliability coefficients ranging from .79 to .90.
The reliability coefficient for the current sample was .78.
Several researchers have also correlated this JDI sub-scale
with measures of overall job satisfaction. Wexley,
Alexander, Greenwalt 8 Couch (1980) in a study of 194
employees from varied occupations found a correlation of .66
with the overall satisfaction score of the Minnesota
Satisfaction Questionnaire. In a review of the validity of
this measure, Cook, Hepworth, Wall, 8 Warr (1981) found
correlations ranging from .53 to .74 with several other
commonly used measures of overall satisfaction.

A sub-scale of The Anxiety-Stress Questionnaire,
developed by House 8 Rizzo (1972), was utilized to measure
more directly job-related stress. The Job-Induced Tension
subscale of this instrument, which was developed to measure

the existence of tensions and pressures growing out of job

46

requirements, includes seven items. Subjects are asked to
respond true or false to statements about how their job
affects them (e.g., "I work under a great deal of tension":
"Problems associated with my job have kept me awake at
night"). In a study of 200 managerial, professional and
technical employees the authors report internal reliability
coefficients of .83 for this subscale. In this sample the
alpha was .68. In terms of validity, several studies (e.g.,
Brief 8 Aldag, 1976: Miles, 1975) have utilized this
instrument as an outcome measure when examining the impact
of working conditions, such as role ambiguity or role
conflict, and have found moderate correlations as predicted.
The authors (Rizzo, House 8 Lirtzman, 1970) have also
reported a moderately strong correlation between this
measure (.40) and a job-related tension index by Lyons
(1971).
Ysluee_aeale

In order to measure personal values that would be
relevant to parenting and family interactions, Schaefer and
Edgerton's (1985) revision of Kohn's Rank Order of Parenting
Values was utilized to assess father's values regarding
conformity versus self-directedness. This measure is
comprised of 15 value statements, six reflecting
self-direction values (e.g., to think for him/herself), six
indicative of conforming values (e.g., to obey parents and
teachers), and three designated as socially desirable values

(e.g., to be kind and considerate). The values are divided

47

into three groups of five items: each group includes two
self-direction items, two conformity items and one social
desirability item. Parents are asked to rank order each of
the three sets of characteristics from one (most important)
to five (least important). A conformity versus
self-direction score is computed by adding the rankings for
each category. Schaefer and Edgerton (1985) studying a
sample of 212 mothers and their attitudes towards their
children found adequate internal consistency and test-retest
reliability among the three factors. They also found
significant negative correlations between parents' valuing
of conformity and teacher ratings of children's curiosity
and creativity. Parents' valuing of self-direction on the
other hand was positively correlated with these behaviors.
In addition, Luster et al., (1989) utilized this measure in
testing the relationship between parental values and
behaviors. In a sample of mothers interacting with their
infants, they found strong positive correlations between
valuing of conformity with beliefs regarding discipline and
freedom as well as with increased constraining behavior.
The reliability coefficients for the current sample were .89
for the Conformity subscale and .93 for the Self-Direction
subscale.
W1;

A seven-item scale developed by Pearlin, Menaghan,
Lieberman 8 Mullan (1981) was used to measure the father's

sense of perceived control and mastery over his environment.

48
This mastery scale asks subjects to agree or disagree with a
series of statements about their life (e.g.,"I have little
control over the things that happen to me": "There is
really no way that I can solve some of the problems I have")
and utilizes a five-point Likert scale ranging from
"strongly agree" to "strongly disagree". This scale has
been utilized to measure the impact of stress on one's sense
of self (Pearlin et al., 1981) as well as the relationship
of psychological empowerment to community activism
(Zimmerman 8 Rappaport, 1988). It also correlates
moderately with other measures of perceived control, such as
a generalized measure of self efficacy and perceptions of
internal versus external control (Zimmerman 8 Rappaport,
1988). The alpha coefficient for the current sample was .75.
W
n a ' a
In order to measure parents' authority style, the
Authoritarianism subscale of the Parental Authority
Questionnaire (PAQ: Buri, Louiselle, Misukanis, 8 Mueller,
1988) was utilized. The PAQ is a 30-item questionnaire,
which asks the subjects to respond to statements about their
parenting style (e.g., "I do not allow my children to
question any decision that I make") and utilizes a five-
point Likert-like scale ranging from "strongly disagree to
strongly agree". This questionnaire assesses Baumrind's
(1971) three prototypes of parental authority:

permissiveness, authoritarianism, and authoritativeness.

49
The authoritarianism subscale was utilized because it
theoretically was more syntonic with previous research
looking at social class, occupational conditions, and
parenting style. Adequate reliability and internal
consistency measures have been established for this
instrument, with test re-test correlations ranging from .77
to .92 and coefficient alphas ranging from .74 to .87.
Regarding criterion validity, in a study of the relationship
between parental authority style and self-esteem among 230
college students, Buri et al. (1988), found that parental
authoritativeness positively predicted self-esteem and
authoritarianism was inversely related to self-esteem, as
would be predicted if the PAQ was a valid measure of
Baumrind's prototypes. In an additional study, Buri (1989b)
also established discriminate validity for the constructs of
this measure as authoritarianism was moderately negatively
correlated to authoritativeness and permissiveness as
predicted by Baumrind's theory, and permissiveness was not
related to authoritativeness as expected.

The PAQ (Buri et al., 1988) was slightly revised to
apply to parents as opposed to adolescents ( e.g., My father
believes that children...changed to: I believe that
children....), similar to a procedure utilized by Buri
(1989a) in a study of adolescent versus parent appraisals of
parental authority and its influence on adolescent
self-esteem. The coefficient alpha of the authoritarianism

subscale for the current sample was .82.

50

Earental.£9nfidense.§sale

Parental report of their own confidence was measured
utilizing the 15-item confidence scale from the Family
Experiences Questionnaire (FEQ) (Frank, Jacobson, 8 Houle
1988). The FEQ was developed to measure a variety of
parenting constructs including confidence in parenting
abilities. Subjects are asked to agree or disagree with
statements about their parenting (e.g., ”I worry that I am
not doing the right thing as a parent") and rate each item
on a four-point scale (1=strongly disagree: 2=disagree:
3=agree: 4=strongly agree). The scale includes seven
positively and eight negatively-keyed items. The authors
(Frank, et al. 1988) tested the instrument on a sample of
760 parents who had at least one child between the ages of
one and 19 living in their home. This scale has also been
utilized in assessing the impact of stressors from child
behavior problems with parents of mentally retarded children
as well as control families (Floyd 8 Zmich, 1989). The
reliability coefficient for the current sample was .81.

WW:

The Parent Relationship Inventory (Stutman 8 Lich,
1985) was utilized to measure perceived relatedness and
encouragement of autonomy in the family. The PRI is a 50-
item questionnaire which measures an adolescent's perception
of his or her relationships with his or her parents. It
utilizes a four-point response format (1 = "strongly

disagree" to 4 = "strongly agree"). The inventory includes

51

six subscales focusing on different aspects of adolescent
differentiation. In this study, four of the subscales were
utilized. To assess adolescent's perceptions of their
father's encouragement of autonomy the Respect for
Intergenerational Boundaries scale and the Mutuality scale
were utilized. The Respect for Intergenerational Boundaries
scale is composed of nine items which describe the extent to
which parents are seen as nonintrusive and as accepting of
their adolescent child's striving for independence (e.g.,
"My father respects my desire to be an independent person").
The Mutuality scale is made up of five items and indicates
the extent to which the relationship with the parent is
experienced as an adult-adult relationship (e.g., "My
father doesn't seem to recognize that I have grown up").
Perceived relatedness and support was assessed by the 11-
item Relatedness scale which describes the degree of
closeness and warmth in the relationship with the parents
and the extent to which the parents are seen as willing to
provide emotional and practical support (e.g., "When I'm
feeling bad I can count on my father to remind me of my
worth"). Finally, the four-item Supply Seeking scale
describes the adolescent's comfort in seeking emotional and
practical support from his or her parents (e.g., "If I need
practical help, I would prefer not to go to my father for
it”).

Stutman and Lich report adequate internal reliabilities

for each of the scales with alphas ranging from .72 to .95.

52
The authors also summarize several studies providing initial
evidence of the discriminant and construct validity of this
measure (Anter, 1985: Lich, 1985: and Stutman, 1984). The
current sample revealed reliability coefficients of .81 for
the created Encouragement of Autonomy scale and .89 for the
Support and Relatedness scale.
Di§§_B§§Q£§12n_RIQ££dn22§

To reduce the number of predictor variables in the path
analysis and verify the conceptual grouping of several
measures, a principal components factor analysis of all the
measurement variables utilizing a Varimax rotation was
completed. An analysis of the factor loadings as well as
the univariate correlations among variables indicated that
several measurement subscales could be combined.

(1) O'Brien's skill utilization subscale and the Job
Challenge subscale were conceptually very similar,
correlated strongly with one other (r=.63, p<.001) and
loaded on the same factor (.80:.90 ). Hence, these scales
were combined by averaging Z-scores for the two scales to
form the variable Skill Utilization.

(2) Similarly, the two measures of job autonomy were
moderately correlated (;=.39, p<.01) and loaded on the same
factor (.73: .79 ), and hence these scales were combined,
again by averaging z-scores to form a Job Autonomy variable.
It should be noted that while the correlation between these
two scales was only moderate, an examination of their

univariate correlations with the other variables indicated

53
little difference, further supporting combining the two.

(3) The two measures of the parental values of
conformity and self-direction were also found to strongly
and negatively correlate with one another (rs-.67, p<.001)
and loaded together (.87: .82) on the same factor. Thus
they were combined by reflecting the items for the self-
direction scale, converting the scales to z-scores and then
combining the two scales to form a ”Traditional" Values
variable.

(4) Level of education, level of income and the current
Socioeconomic Index score were also combined, after
converting to Z-scores, to form the Social Class variable.
It should be noted that some scholars (e.g., Otto, 1975:
Mueller 8 Parcell, 1981) have argued against using composite
measures of social class, suggesting that such measures may
mask rather than reveal relationships between variables.
However, it was decided to combine the variables due to the
theoretical considerations discussed earlier, and their
forming a distinct factor (.88:.84:.66) in the factor
analysis. In addition, there was a strong correlation
between the Socioeconomic Index score and education (13.74,
p:<.001) and a moderate correlation (;;.44, p<.001 ) between
income and the other two components of Social Class. An
examination of univariate correlations of the full
measurement model also found very few significant
differences in the relationship between the individual

social class components and other variables and that of the

54
combined measure, further supporting their being merged. It
should be noted that the relatively weaker relationship
income had with education and occupational status is
consistent with previous findings that income is often the
weakest and least stable indicator of social stratification
(Mueller 8 Parcell, 1981).

(5) Finally, job satisfaction and job stress which had
initially been theorized as a combined variable did not have
a significant correlation with each other nor did they load
on the same factor and thus they were entered as two

separate variables in the path analysis.

RESULTS

W

A path analysis utilizing a least-squares procedure and
employing simultaneous regression analysis was completed to
test the hypothesized relationships. Table 2 provides the
univariate correlations for the latent variables and Figure
2 gives the beta coefficients for the predicted model.

In regard to the proposed hypotheses, higher social
class, as expected, was related to higher levels of skill
utilization (B=.37, p<.001) and to a lesser extent with
greater job autonomy (B=.14, p<.05). However, job autonomy
was not significantly related to any of the downstream
variables, and as such, failed to mediate relationships
between social class and other occupational and parenting
variables.

When skill utilization became the predictive
occupational condition, a much stronger relationship was
found with downstream variables. Greater skill utilization
on the job was highly predictive of greater job satisfaction
(5;.66, p<.001) and of a lower reported sense of
powerlessness (B= -.38, p<.001). Also, a significant
relationship was found between greater skill utilization and
less valuing of conformity over self-direction (B= -.16,

p<.06). No relationship was found between skill utilization

55

56

 

 

 

55ch _ n w a u a q a o 3 Z
_. woos: 0—38 _.ooo
N. m5: Cam—muons .31.. rooo
u. now >582: .39. bus: rooo
a. hoe mongoose: boa: one: .uuu: rooo
u. moo manna. bum .3» Jo: 33o race
a. mots—8338 JEN: lawn: Jug: .33.»: Lou: rocc
a. 5.7.8 Jud: 3.6.33 Jam. 33?. .oam .81. reg
883338 msgv
m. 5525181»: mum—o 33¢... lac Jouu 33¢: ban .8— .woc: race
o. 3388— Oosmaosoo hue: hog... Low .923 Jae: mew: L: .21 rooo
8. moose—duo 2.88.5. .33 loan .58 38¢ .owa .o-u 3308 Rue—3 .2» r80
2. mcoooiwafiaasamm .2“ .5— .owu ban .8» Jim. Jim. 33° .203 Moo: rooo

 

En Em m3 :5 page...“ 09.35363. Bones 368 _m_ 8 was
.953. 39.2. 25 8:3 8.4.8.

 

.s

 

 

 

 

 

    

 

 

 

 

 

 

 

 

 

 

 

 

 

57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I a <>r=nm -
. u H 8:3”: 3.
35-8388: .9016
u
sou \ \
>583 I. cm
I
I .88
fl. / ’ 9.: $3202 /
J. / /
. /
H O V I /
«85. r» I J I
853 / I .= I
/ _ T
. x 1 0 I
_ com I
/ pagans llllllll
_ n \ .09\ A ‘ mama I
_ was. \ \ / I .
5.83.5.3: . 0
_ I. 9% o / 400% C
_ / I
_ I
z
/ . o
_ 4x
_.I. I . om 0 gnaw—8mm
.(il
3.956 m mowwa Show I mwmswmwomsn oonnowonHOSm
wnmdwonoe zodow awn: won: nommmwowosnm cannon VHSom I zos-mwmswmwomsn oonnowmnwosm

m I wAbu“ q I MACH" o I bAbo: n I MALo

58
and the level of reported job related stress.

Fathers' values, as expected, were found to be
predictive of parenting style with greater valuing of
conformity and less valuing of self-direction predictive of
greater authoritarian parenting style (£8.36, p<.001).

In regard to the intervening occupational variables of
job stress and job satisfaction, higher job stress was
predictive of greater sense of powerlessness (B=.20, p<.01),
but did not predict sons' perceived support and relatedness,
as hypothesized. Although not part of the predicted model,
there was a trend for greater job stress to be related to
lower parental confidence (B= -.12, p<.10). Further, an
examination of the univariate correlations found significant
and similar correlations between job stress and
powerlessness and parental confidence (;=.18, p<.01: IF-19.
p<.01), suggesting that powerlessness was a mediating
variable between job stress and parental confidence.

Greater job satisfaction however did not directly predict
less powerlessness nor did it predict higher levels of
perceived support and relatedness, as proposed.

As hypothesized, fathers' decreased sense of
powerlessness significantly predicted greater parental
confidence (B; -.38, p<.001). In regard to the relationship
of parenting style to the perceptions of sons of their
fathers' level of both support and relatedness and
encouragement of autonomy, parenting style was predictive of

encouragement of autonomy with greater levels of

59

authoritarianism predictive of less encouragement of
autonomy (B= -.25, p<.01). Greater authoritarianism was not
significantly predictive of less support and relatedness,
but a trend in the predicted direction occurred (3; -.13,
p<.10). Higher parental confidence, however, was predictive
of higher support and relatedness as anticipated (£3.18,
p<.05), but was not predictive of higher encouragement of
autonomy.
WWW

An overall goodness-of-fit test was completed, which
compared the predicted model (i.e., only those paths
hypothesized to have significant beta coefficients) with the
full recursive model (i.e., all possible causal paths
between the latent variables) by examining the residual path
coefficients in a procedure recommended by Pedhazur (1982)
for testing over-identified models (See Table 3). This test
found no significant difference between the two models,
(X 2= 8.874, df=30, NS) suggesting that the data did fit the
proposed theoretical model. In all, 10 of the 22 predicted
pathways were significant at the .05 level. In comparison,
only one out of a possible 30 nonpredicted pathways was
significant. The single nonpredicted pathway, which was
significant, was a moderate direct relationship between
higher social class and increased job satisfaction (33.16,
p<.01) beyond that mediated by skill utilization and job
autonomy. In addition, Table 4 shows the amount of variance

explained in each of the endogenous variables by preceding

Table 3

-.'.'. '1'. Q

o- ;-‘

6O

l'.’x!'

I. -

 

EQLL.BE§HB§I¥E_HQDEL EBEDIQIED_HQDEL
62° .930 62' = .930
d3e .990 63° = .990
d4° .704 64' = .721
65G .999 65' = .999
d6e .876 66' = .879
67‘ .938 d7° = .943
dse .920 d8° = .933
693 .894 69‘ = .926
d10° .940 d10° = .962
611° .966 611' = .972

 

Chi-Square Goodness of Fit (Pedhazur, 1982)

Q

W

.845 (Full Model)

.803 (Predicted Model)

(N-d) log e (0) =7¢ 2
8.87 < 43.77 (NS)

74" (30)

1.052

Table 4

 

1. Social Class

2. Skill Utilization

3. Job Autonomy

4. Job Satisfaction

5. Job Stress

6. Powerlessness

7. Values
(Conformity vs.

Self-Direction)

8. Authoritarian
Parenting Style

61

1-7

9. Parental Confidence 1-8

10. Encouragement of
Autonomy

11. Support and
Relatedness

1-9

1-9

.004

9*.
.233

if.
.120

iii
.153

**
.201

if
.117

.067

*

PBEDIQTED
2
EYIE 8
iii
1 = .135
t
2 = .020
it.
2-3 = .480
2-3 = .001
it.
1-5 = .227
it.
1-3 = .110
it!
7 = .130
*tt
6 = .147
ti
8-9 = .074
*
4-5,8-9 = .056

 

Note. PVIE = Predictor variables in the equation

t it it.
p<.05, p<.01, p<.001

62

variables in the predicted and full models. The BZfOr the
two models are similar with the exception that the full
model accounts for somewhat more variance in predicting
parental confidence. Most of this increased variance is
accounted for in the slight direct influence of social class
and job-related stress on parental confidence. The
difference in variance accounted for in encouragement of
autonomy is primarily related to the slight direct influence
of fathers' skill-utilization and values on this variable.
MedeLEedifigatiens

While the goodness-of-fit test suggests that the
predicted model did fit the data, a more parsimonious model
could be developed by eliminating the nonsignificant causal
paths. A closer look at the predicted model suggests two
problem areas. The first difficulty was the failure of job
autonomy to adequately mediate between social class and
other downstream occupational and parental variables. The
second weakness was the failure of job stress and job
satisfaction to sufficiently mediate between job conditions
and other parenting variables. Since skill utilization had
a moderate correlation with job autonomy (13.39 9,.01) and,
theoretically, may tap some similar dimensions, it was
decided to delete job autonomy and allow skill utilization
to be the primary intervening variable between social class
and the downstream variables. It was also decided to delete
job satisfaction and job stress from the model. The

rationale for the deletion of job satisfaction was two-fold.

63
First, there was a high correlation between job satisfaction
and skill utilization with a suggestion of multicollinearity
between the two variables. An exploratory factor analysis
also found it to load on the same factor as skill
utilization. Secondly, job satisfaction did not act as a
significant intervening variable between occupational
conditions and any of the downstream parenting or family
relational variables. The variable of job-related stress
proved to be an even weaker intervening variable between job
conditions and downstream variables with only a significant
linkage with perceived powerlessness and a moderate
correlation with parental confidence. Further, job-related
stress had not even a weak relationship to social class,
suggesting that it may be tapping a very different
phenomenon than the other job-related conditions examined.
Thus, it was also removed in the modified model. Finally,
the direct link between social class and fathers' perceived
sense of powerlessness was deleted as there was no direct
predictive relationship between these two variables.
'v o s s
In the alternative proposed model (See Figure 3), most
of the predicted pathways maintained their basic strength
and significance levels. Job skill utilization proved to be
a mediator between social class and fathers' values
regarding conformity and self-direction as well as
intervening between social class and fathers' sense of

powerlessness. Fathers' values continued to be predictive

64

 

mOOH>F
nr>mm

 

.uq

 

 

 

 

mwmcHo u

>Hnondmnm Koaow and: won: oommmwomenm

mHHrr
GHHrHN>HHOz

 

 

 

 

 

<>Hcmm I
condo”: 4m.
MNFNIUHNNOHHOZ

 

 

 

 

 

 

 

 

.\ «cwmown
r
\. wur>Hmvzmmm

 

 

 

 

 

u

wetmwhmwmzmmm

 

 

 

mowwn fiwsmm I memo»mwoosn noommwowoSnm
conned radon I zos-mwmsHmwom=n noommwooSnm
m I bfi.omu a I MA.OH“ o I mw.oown n I MA.Ho

65
of the level of their authoritarian parenting, and their
sense of powerlessness was again strongly predictive of
their level of confidence in parenting. Parental authority
style continued to be significantly linked with sons'
perceptions of their fathers' encouragement of autonomy, but
as in the data from the initial model, did not significantly
predict their perception of their fathers' support and
relatedness. Parental confidence also predicted perceptions
of fathers' support and relatedness, but did not quite
approach significance level in its prediction of
encouragement of autonomy.

A Goodness-of—Fit test again found no significant
difference between the fully-recursive model and this
alternative theoretical model (1(2= 12.40, df=20, NS),
suggesting that the data did fit the proposed theoretical
model (See Table 5). In this alternative model, 8 out of 10
predicted pathways were significant at least at the p:<.05
level. In comparison, only one out of 10 of the nonpredicted
pathways was significant. The nonpredicted significant
pathway was a direct link between social class and parental
confidence (5:.17, p<.05), which is not theoretically
dystonic with the proposed model. The amount of variance
explained by the full and predicted models are again very
similar, further suggesting that this revised predicted

model fits the data (See Table 6).

66

Table 5

Revised Model:

“Huge 0 4 : 9 9 = 0‘ Iaus‘oo : a a! ‘= ’0 i 3!!
O I

 

 

62° = .930 62, = .930
63° = .941 63' = .946
64° = .899 64° = .907
65° = .906 65° = .925
66° = .925 66° = .933
67° = .946 67° = .962
68° = .979 68‘ = .973

 

Chi-Square Goodness of Fit (Pedhazur, 1982)

Q = .627 (Full Model)
...... = 1.074
.584 (Predicted Model)

w = (N-d) loge (Q) =X2

752 (20) = 12.4 < 31.41 (NS)

Table 6

Revis

   

ed Model:

- ! ‘- .19

67

 

 

I it on
p<.05, p<.01, p<.001

FELL EBEDIQIED
EEDQQENQ!§.¥ABIAELE§
2 2
2213 B 2213 B
1. Social Class -------------------------
2. Skill Utilization 1 = .135"” 1 = .135""
N m
3. Powerlessness 1-3 = .191 1-2 = .178
4. Values 1-2,4 = .115"" 1-2 = .106""
(Conformity vs.
Self-Direction)
5. Authoritarian l-4,6 = .144"' 3 = .130"'
Parenting Style
. “t “t
6. Parental Confldence 1-5 = .180 6 = .145
7. Encouragement of 1-6 = .106" 5-6 = .074“*
Autonomy
8. Support and 1-6 = .062 5-6 = .053'
Relatedness
Note. PVIE = Predictor variables in the equation

DISCUSSION

This study was begun with the intent to explore the
role that extrafamilial processes might play in influencing
family functioning. In particular, it examined how social
class, occupational conditions, values and sense of
powerlessness influenced parenting processes and,
subsequently, sons' experience of autonomy and relatedness
in the father-adolescent son relationship. Previous
research and theorizing suggested several models to explain
the work-family interface, and this study also was designed
to test these formulations. Several observations and
tentative conclusions regarding the above questions can be
drawn from the data.

First of all, these findings tested and found partial
support for Kohn's value-shaping hypothesis which, suggests
a linkage between social class, occupational conditions, and
parental values regarding conformity and self-direction.
The data found that social class directly influenced
fathers' values, but also had a small indirect effect
through the intervening occupational condition of job skill
utilization. When this model was expanded to include
parenting style (i.e., the level of authoritarianism),
global values regarding conformity and self-direction values

significantly predicted fathers' authoritarianism,

68

69
indicating that fathers' values proved to be an effective
intervening variable between social class and authoritarian
parenting style. Further, when sons' perceptions of their
fathers' encouragement of autonomy was added to the model,
the level of fathers' authoritarianism predicted
encouragement of autonomy and acted as a mediator between
fathers' values and their relationship with their sons.
These findings thus add support to the value-shaping or
socialization models which link social class and
occupational conditions with parental values, attitudes, and
behavioral styles.

An unexpected finding was that job autonomy failed to
act as a mediator between social class and other downstream
variables such as a father's values, sense of powerlessness,
and other job-related variables, such as job satisfaction.
This was in sharp contrast to the stronger mediating
function of skill utilization. Several explanations could
be proposed for this finding. It may be that skill
utilization is simply a more powerful mediator between
occupational conditions and workers' personal attributes and
values. O'Brien (1980) has argued for the centrality of
skill utilization as a major factor in workers' job
satisfaction as well as in their sense of personal control.
This study tended to support O'Brien's highlighting of skill
utilization over autonomy as a central variable. The nature
of this sample may have also affected this finding. While

this sample had a range of working-class through upper-

70
middle-class families, it leaned toward white-collar and
professional occupations. It may well be that in white
collar and professional occupations, autonomy and
independence in the work place are more common place and
ingrained in the structure of the job. As such,
satisfaction and a sense of personal control on the job may
relate more to other factors such as time pressures,
quantity of tasks, external constraints of the market place
and the like. As Kline and Cowan (1988) have discussed, the
individual meaning of work may act as a filter through which
individuals interpret their work experiences as satisfying
or stressful. Work that is challenging and utilizes one's
potential may be increasingly more important than
independence and autonomy at the work place, particularly in
a growing professional work force.

The second major pathway by which social class and job
conditions were expected to influence parenting and father-
son relationships was through the variable of perceived
powerlessness. This was suggested by the work of O'Brien,
(1980, 1986) who proposed that workers' sense of personal
control was a crucial mediating variable between job
conditions and personal identity, which could influence non-
work interactions. Skill utilization was the major
occupational condition which O'Brien found influenced a
worker's sense of personal control. This study found that
increased skill utilization was, in fact, predictive of

decreased powerlessness and appeared to be a primary

71

mediator of social class influences on powerlessness.
When this model was extended to include parental confidence
and sons' perceptions of their fathers' supportiveness,
powerlessness did predict levels of parental confidence,
which, in turn, were predictive of sons' perceptions of
their fathers as being supportive. These findings thus
support O'Brien's initial hypotheses and findings, but
further suggest that fathers' sense of powerlessness may be
an important mediating variable between experiences on the
job and family variables, such as parental confidence.

Unexpected in this pathway, however, was the failure in
the path model to find a significant direct link between
social class and fathers' sense of powerlessness. Several
hypotheses can be suggested regarding this finding. Because
this sample was limited primarily to working-class and
upper-middle class families, with a leaning toward white-
collar and professional occupations, it may be that the
fathers' experience of respect and challenge on the job is
indeed the primary mediator of their overall sense of
mastery. Other conditions of life which one might expect to
influence one's sense of powerlessness (e.g., financial
failure, social discrimination, and poverty), might not
apply as strongly to this working population. It may also
be that powerlessness partially taps a dimension that is
more trait or personality based, such as a coping strategy
or overall locus of control. Such a coping or locus of

control variable might be less directly influenced by a

72
broader or more distal condition of life, such as social
class, than a more individualized variable, such as the
perception of how thoroughly one's skills were utilized and
maximized. This argument is further suggested by the
findings of Repetti (1987) that psychological well-being at
the work place was more closely tied to the proximal
individualized social environment than the more distal
common social environment. An alternate hypothesis would be
that workers who felt little mastery over their environment
would be less inclined to take jobs which were more
challenging or perhaps would be less inclined to be given
them.

A third theoretical model tested was the "spill-over
hypothesis" which suggested that dissatisfaction and
stressful experiences on the job could directly lead to less
supportive family relationships, specifically between
fathers and sons. This hypothesis was generally not
supported by the data. Neither job satisfaction nor job-
related stress directly predicted level of supportiveness,
as might be predicted from a stress spill-over model.
However, job stress was predictive of fathers' sense of
powerlessness, and powerlessness, in turn, predicted
parental confidence which in turn was related to the level
of support and relatedness perceived by sons. Thus, while
there is some suggestion that job stress might influence
father-son relationships through the intervening variables

of powerlessness and parental confidence, it is clearly not

73
an effective mediator between social class, other
occupational conditions, and downstream parenting and
relational variables.

What might explain the failure of job stress and job
satisfaction to be effective mediating variables between
occupational conditions and downstream parenting and
relationship variables? In regard to job stress, one
hypothesis might be that reported job stress is primarily a
transitory variable, subject to changes in job personnel and
specific time-bound conditions. As such, ongoing job stress
may be more difficult to tap with questions which ask a
subject to summarize their work experience than other
potentially more stable variables such as job autonomy or
skill utilization. Crouter et al., (1989) and Repetti
(1989) have emphasized the importance in distinguishing
between transient work stress and long term chronic work
strain in assessing the influence on family interactions.
Secondly, the responses on the "job stress" scale might be
tapping more personality or coping style characteristics
similar to the dimension of powerlessness. Thus, how one
perceived and coped with stress on the job might, to some
extent, be independent of the type of job one had. An
examination of the univariate correlations (See Table 2)
partially supports this notion in that the only variables
with which job stress significantly correlated were
powerlessness and parental confidence. In regard to the

failure of job stress to be related to sons' experience of

74
their father's supportiveness, there may be both substantive
and methodological explanations. Recent research on the
influence of work-induced stress on family life suggests
that while work strain may influence the marital
relationship, it has less direct influence on relationships
with children. Bolger et al. (1989) found that husbands
tended to have more arguments at home with their wives on
days that they had arguments at work, but that arguments at
work were not related to arguments with children. In a
related vein, other researchers (e.g., Barnett, Marshall 8
Pleck, 1992: Reppetti, 1989) have also found that the
quality of the marital relationship as well as the
satisfaction with the marital and parental role can buffer
the effects of job stress on family interactions. Thus,
while job stress may influence psychological outcomes such
as a sense of powerlessness and even personal confidence in
parenting there are a variety of moderators or filters that
can lessen its impact on parent-child relationships. A
final explanation for the failure of job stress to predict
sons' perception of father's support may relate to the
measure used to assess job stress. Several researchers
(Crouter et al., 1989: Small 8 Riley, 1990) have argued that
while stress scales, such as the Rizzo scale, which was used
in this study, are effective in measuring more immediate and
transient work strain, they are less effective in assessing
the more accumulative nature of chronic stress or of

establishing a more causal link between work stressors and

75
the resulting mood or psychological state of the worker.
Perhaps a specialized multidimensional measure such as the
Work Spillover Scale of Small and Riley (1990) would more
accurately assess the processes we were examining.

The failure of job satisfaction to be predictive of
parental variables such as a sense of powerlessness and
son's perceived support and relatedness is also puzzling. A
closer look at the data suggests several possible
explanations for this finding. One reason for the lack of
effect of job satisfaction with downstream variables such as
father's sense of powerlessness may be its relatively high
correlation with skill utilization (r=.69, p<.01). In the
path analysis, much of the relationship that job
satisfaction has to powerlessness may be subsumed by the
strong relationship that skill utilization has to
powerlessness. In fact, looking at the univariate
correlations, a significant relationship was found between
job satisfaction and powerlessness ( ;;-.31, p<.01) as well
as with other downstream variables such as parental
confidence ( 1:.24, p<.01), authoritarianism (rs-.18,
p<.01), and valuing conformity over self-direction (rs-.23,
p<.01). A comparison of the items making up the Type of
Work Satisfaction subscale of the Job Descriptive Index with
those of the Skill Utilization measure found several
overlapping items such as whether the work was
"challenging”, "routine", or "gave a sense of

accomplishment". This could account for the high

76
correlation between the two measures and the failure for job
satisfaction to provide unique variance beyond skill
utilization in the predicted path model. The use of a more
global job satisfaction measure with a more moderate
correlation with skill utilization might test the
relationship of job satisfaction with parenting variables
more accurately. The relatively high correlation between
job satisfaction and skill utilization does not fully
explain the lack of relationship between job satisfaction
and sons' perception of fathers' support and relatedness, as
might be predicted from a spill-over hypothesis. One
explanation for this lack of relationship might be similar
to that made in regard to job stress, in that job
satisfaction might be viewed as a more transitory state and,
as such, less influential to more continuous emotional
experiences, such as the sense of connectedness between
fathers and sons. Another explanation, however, might be
that the expected positive relationship between job
satisfaction and support and relatedness, based on the
hypothesis that the less negative spill-over from work, the
greater the likelihood of positive parent-child
interactions, might be counteracted by a compensatory
hypothesis which proposes that less job satisfaction at the
work place can lead to compensatory behavior, or putting
more time and emotional energy into family life (Hoffman,
1986). If some fathers compensated for their lack of

satisfaction at work with more positive interactions with

77
their son, while others had less positive interactions due
to being overly stressed or being in bad moods, they would
cancel each other out and the result would be a lack of
relationship between these variables.

A final theoretical question centered around whether a
father's attitudes and behaviors, which might be influenced
by work conditions, would directly impact on the father-son
relationship. It was theorized that sons' perceptions of
their fathers' encouragement of autonomy and supportiveness
would be influenced by two different pathways: The first,
largely mediated by parental authority style influenced by
fathers' values, and the second, largely mediated by
parental confidence influenced primarily through fathers'
sense of powerlessness. This division of influence emerged
even more strongly than originally predicted and tends to
support the theory that issues of autonomy and control,
while related to issues of connectedness and support, are
separate processes influenced differentially by fathers'
parenting style and confidence, as well as by their global
child-rearing values and sense of personal efficacy.

A few comments should also be made in regards to how
these findings relate to the broad question of how social
class influences parenting and, specifically, father-son
relationships. While this study was more interested in the
mechanisms or processes that mediated social class and
family interactions, rather than direct effects, a few

cautious conclusions can be drawn. First of all, this study

78

found a significant negative relationship between social
class and fathers' emphasis on conformity over self-
direction with their sons. This validated previous studies,
as reviewed by Gecas (1979), in which higher social class
was correlated with emphasis on independence. Secondly, in
regards to parental control, a small but significant
negative correlation was found between social class and
authoritarian parenting style. This confirms earlier
findings of Kohn (1963) and Elder and Bowerman (1963) with
adolescent boys and their parents. In the path model,
however, most of the impact of social class was mediated
through parental values. Finally, in looking at the
relationship of social class and parental affection and
support, this study found no direct relationship between
social class and adolescent sons' perceptions of their
fathers' support and relatedness. However, there may be
some indirect effects through powerlessness and parental
confidence, especially when mediated by skill utilization.
This study clearly suggests that social class alone
contributes only a very small portion of the variance in
influencing parenting and father-son interaction. It
appears to be more useful to examine its influence on
specific intervening variables, such as parental values, and
specific occupational conditions.
WW

There are several methodological issues which limited

the generalizability of these results and which should be

79
considered for any future replication of this research.
First of all, a greater range of social class among the
sample, particularly the inclusion of more unskilled
laborers, would have been desirable. While a moderate range
of working-class to upper-middle-class professionals was
achieved, the median occupation was skewed towards
professional, or management, professions. The weakness of
some of the direct relationships found between social class
and individual and family functioning variables may have
been due to the constrained range of our sample. A broader
sample may have allowed deeper analysis of the differential
effects of social class variables versus conditions of work.

Secondly, a greater range of occupational conditions
might have been useful in assessing intervening variables
between social class and family functioning. Perhaps the
use of a multiscale measure, such as Hackman and Oldham's
Job Diagnostic Survey (Hackman 8 Oldham, 1975) or Sims,
Szilagyi 8 Keller's Job Characteristics Inventory (Sims et
al., 1976), might have tapped a broader range of work roles,
tasks, and climate, and allowed a more complex evaluation of
the aspects of work which influenced parental perceptions
and family relationships.

Finally, although path analysis procedures can suggest
the likelihood of particular causal models fitting a set of
correlational relationships among variables, a cross-
sectional study such as this project cannot clearly answer

the question of causality. To more clearly ascertain

80
whether there is a causal linear relationship between social
class, occupational conditions, parental values and
behaviors, and subsequent family relationships, longitudinal
research is needed. Ideally, data would be collected before,
during, and at the latter parts of a career and family life
cycle. Such longitudinal studies could also better test
alternative models of the work family interface which
suggest that personality disposition and character traits
can concurrently influence job choices and parental values
and behaviors. This seems particularly salient in light of
several of our findings regarding the relative independence
of job stress and powerlessness from social class.
s e se

In addition to the above methodological suggestions,
there are several other areas of future exploration that
could prove useful. First of all, research in adolescent
development and family processes has suggested that the
marital and parental relationships between fathers and
mothers are important mediators between parent-child
relationships. Understanding how the work environment
affects marital cohesion and satisfaction might help
explicate further the work-family interface. In fact,
several recent studies (Bolger et al., 1989: Hughes et al.,
1992: Repetti, 1989) have suggested that the marital
relationship, and particularly the wife's response, may be a
powerful moderator or buffer to the impact of work on the

overall family relationship.

81

Secondly, this study only looked at fathers'
perceptions of their jobs and relationships with their sons'
at a particular life stage, that of having young college-age
sons. It is unlikely that connections between work
conditions, parenting and family variables remain stable as
fathers move through their careers. As Kline and Cowan
(1988) have suggested, paid work may well change in both
salience and meaning at each stage of the family life cycle.
Research in fathers' involvement in parenting has also
suggested that fathers' active involvement in work and
family often changes throughout the life cycle (e.g. Almeida
8 Galambos, 1991: Cowan 8 Bronstein, 1988: Lamb, 1986).
Research which examines working fathers and their family
relationships across their work lives would help explain
this process.

Finally, this research only looks at the interface
between fathers' work experience and their relationship with
their sons. Whether these linkages hold also for fathers
and daughters, or for working mothers and their adolescent
children would be fruitful areas to explore. Moreover, as
Bronfenbrenner and Crouter (1982) have pointed out,
researchers studying the impact of work on family life have
treated the job situations of mothers and fathers as
separate worlds. What is needed is a better understanding,
not only how of work influences both fathers and mothers,

but also how these processes interact and influence both the

82

parenting process and child, adolescent, and young adult
development.
921121115126

Researchers and theoreticians studying family processes
have tended to emphasize interactions within the family and
have largely ignored the influence of external environmental
influences such as the work place. This research has been
an attempt to better understand the linkages between larger
system variables, such as social class and occupational
conditions, and father-son relationships. The study found
that social class and the job condition of skill utilization
influenced the father-son relationship through two primary
pathways. One pathway was mediated primarily by fathers'
values and supported a "values-shaping" hypothesis.
Fathers' valuing of conformity over self-direction was found
to be predictive of parental authoritarianism, which in turn
predicted less encouragement of autonomy in sons. The
second pathway was mediated largely by fathers'
powerlessness which was found to influence parental
confidence, which in turn was predictive of sons' perception
of support and relatedness from their fathers. Unexpected
were the findings that job autonomy, job satisfaction, and
job-related stress were relatively weak or ineffective links
in the model. While this model requires further validation
and empirical support, it hopefully has helped in further
explicating the complex relationship between the worlds of

work and family life.

LIST OF REFERENCES

LIST OF REFERENCES

Almeida, D. A., 8 Galambos, N. L. (1991). Examining father
involvement and the quality of father-adolescent

relations. Ionrnal_of_Eesearsh_on_8dolessenss. 1(2).
155-172.

American Psychological Association. (1982). Ethiggi
participants. Washington, DC: APA.

Anter, W. J. (1985). Children of Holocaust survivors:
Differentiation from family of origin and its
relationship to family dynamics. (Doctoral
dissertation, California School of Professional

Psychology. Los Angeles. 1985)- Dissertation_8bstraots
International

Bacharach, S. B., 8 Aiken, M. (1976). Structural and
process constraints on influence in organizations: A

level-Specific analysis. Administratixe_§siense
Quarterly 21. 623- 642.

Barnett, R. C., Marshall, N. L., 8 Pleck, J. H. (1992).
Men's multiple roles and their relationship to men's
psychological distress.

Enmilx. 54. 358-367-
Baumrind, D. (1971). Current patterns of parental

authority-_Dexe1onnent81_£sxsnolosx_uonogranns. A
(2, Pt. 2).

Baumrind, D. (1978). Parental disciplinary patterns and
social competence in children. 293th_§ng_§ggigty, 2,
239-276.

Bayley, N., 8 Schaefer, S. (1960). Relationships between
socioeconomic variables and the behavior of mothers

toward young children. Ionrnsl_of_§enetis_zexsnologx.
gg, 61- 77.

Beavers. W- R- (1977) £81282;neranx_and_grontn_1a_fanilx
gygtems_p§;_pggtiyg. New York: Bruner-Mazel.

83

84

Bell, D. C., 8 Bell, L. G. (1983). Parental validation and
support in the development of adolescent daughters. In
H. D. Gotevant 8 C. R. Cooper (Eds.), Adgigsggnt

ggygigpmgnt. San Francisco: Jossey Bass.

Belsky, J. (1984). The determinants of parenting: A process
model. gnild_nexelonnent. 55(1). 83-96.

Blau, P. M., 8 Duncan, 0. D. (1967). Ih§_Amgtigan
ossnnationnl_strnstnre- New York: WileY-

Bolger, N., DeLongis, A., Kesssler, R. C., 8 Wethington, E.
(1989). The contagion of stress across multiple roles.
, 51, 175- 183.

Borduin, C. M., Henggeler, S. W., Sanders-Walls, M., 8
Harbin, F. (1986). An evaluation of social class
differences in verbal and nonverbal maternal controls,
maternal sensitivity and child compliance. Child_§tngy

Journal. 15. 95-112-

Brief, A. P., 8 Aldag, R. J. (1976). Correlates of role
indioes. Jonrnnl_of_Annlied_2§xsnolosY. 59. 182-186-

Bronfenbrenner, U. (1958). Socialization and social class
through time and space. In E. E. Maccoby, T. M.

Newmcomb 8 E. L. Hartley (Eds.), ngding§_in_§ggiai
pgyghgiggy, New York: Wiley.

Bronfenbrenner, U. (1986). Ecology of the family as a
context for human development: Research

perspectives. Dexelonnental_zsxsnologx. 22. (6). 723-
42.

Bronfenbrenner, U., 8 Crouter, A. C. (1982). Work and family
through time and space. In S. Kamerman 8 C. Hayes

(Eds. ), s a o h’
wgzid (pp. 39-83). Washington, DC: National Academy
Press.

Bronstein, P., 8 Cowan, C. P. (1988L

Eatherhood.todaxi
Een_s_2nansing_role_1n_tne_fanilx New York: Wiley 8

Sons.

Buri, J. R. (1989a). Self-esteem and appraisals of parental
behavior. J2nrnal_of_Adolessent_Benaxior. A. (1).
33-49.

85
Buri. J- R- (1989b)- An_1nstrnment_for_tne_measnrement_of___

o . Paper presented at the
61st Annual Meeting of the Midwestern Psychological
Association, Chicago.

Buri, J. R., Louiselle, P. A., Misukanis, T. M., 8 Mueller,

R. A- (1988). Persona1112.1n6_§ooial_zsxsnolosx
Bulletin. 11. 271-282.

Cambell, E., Adams, G. R., 8 Dobson, W. R. (1984). Familial
correlates of identity formation in late adolescence.

I2nrnal_of_Yontn_snd_Adole§sense. 13. 509-525-

Cammann, C., Fichman, M., Jenkins, D., 8 Klesh, J. (1979).
The Michigan organizational assessment questionnaire.
Unpublished Manuscript, Ann Arbor: University of
Michigan.

Cook, J. D., Hepworth, S. J., Wall, T. D., 8 Warr, P. B.
(1981). -
' . London: Academic Press.

Cooper, C. R., Grotevant, H. D., 8 Condon, S. M. (1983).
Individuality and connectedness in the family as a
context for adolescent identity formation and role-
taking skill. In H. D. Grotevant 8 C. R. Cooper
(Eds. ), s v o

Ado1e_2en1_de_e1_onent_1n_tne_familx1_89n
direst1ons_for_shild_dexslonment (pp- 43- 60)- San

Franc1sco. ° Jossey-Bass .

Cowan, C. P., 8 Bronstein, P. (1988). Fathers' roles in the
family: Implications for research, intervention, and
change. In P. Bronstein 8 C. P. Cowan (Eds. ),

9‘ 1"! '1' ' y‘l'f‘ 1° 1° ,' ‘ I I 1!!

(pp. 341-348). New York: Wiley 8 Sons.

Crouter, A. C., 8 McHale, S. M. (1993). The long arm of the
job: Influences of parental work on childrearing. In

T- Luster & L- Okagaki (Eds-I. Paren11n91_sn_eoolosioal
pgtgpggtiyg (179-202). Hilsdale, N.J.: Erlbaum

Crouter, A. C., Perry-Jenkins, M., Huston, T.L., 8
Crawford, D. (1989). The influence of work induced
psychological states in behavior at home. B§§i9_ang

Annlied_§22111_2§xsnolosx. 19. 273-292-

Dornbusch, S., Ritter, P., Leiderman, P., Roberts, D., 8
Fraleigh, M. (1987). The relation of parenting style to
adolescent school performance. ghiig_pgygigpm§nt, 5g,
1244-1257.

86

Douvan, E., 8 Adelson, J. (1966). Ih§_gdgig§ggnt
gxpetigngg. New York: Wiley.

Duncan, 0. D. (1961). A socioeconomic index for all
occupations. In A. J. Reiss, Jr. et al. (Eds.),

Qs2nnation§_snd_sosial_siatus- New York: Free Prese-

Dunham, R. B., Smith, F. J., 8 Blackburn, R. S. (1977).
Validation of the index of organizational reactions
with the JDI, the MSQ and Faces scales. Agademy_gt

Managementnlournal. 29. 209-223-

Elder, G. E., Jr. 8 Bowerman, C. E. (1963). Family
structure and child rearing patterns: The effect of
family size and sex composition. Amgtiggn_§ggigiggiggi
892191. 25. 891-905-

Erickson, R. J., 8 Gecas, V. (1991). Social class and
fatherhood. In F. W. Bozett 8 S. M. H. Hanson (Eds. ),

Fa1nernood_and_fanilie§_in_snltnral_sontext New York:

Springer.

Erlanger, H. S. (1974). Social class and corporal punishment
in child rearing: A reassessment.

882111. 22. 68-85.

Featherman, D. L., 8 Hauser, R. M. (1978). Qppg;tnnity_gng
ghgngg. New York: Academic Press.

Featherman, D. L., Jones, F. L., 8 Hauser, R. M. (1975).
Assumptions of social mobility research in the U.S.:
The case of occupational status. aggigi_§gigngg
EEEEQIQD. A. 329-360-

Floyd. F-. 8 Zmioh. 0- (1989)- Parentins_narinersnin_of
motners_and_fatners_of_nentally_retarded_snildren

Paper presented at the annual mmeting of the American
Psychological Association, New Orleans.

Frank, S., Jacobson, S., 8 Houle, C. B. (1988). In;

oo 1' 1 0.0 o: 1 7.! 1!: 1°;

gng_p§zgnting. Unpublished manuscript.

Gecas, V. (1979). The influence of social class on
socialization. In W. Burr, R. Hill, F. I. Nye, 8 I.
Reiss (Eds.),

(pp. 365-404). New York: The Free Press.

Gecas, V., 8 Nye, F. I. (1974). Sex and class differences
in parent-child interaction: A test of Kohn' s

hypothesis. 1onrna1_of_uarriage_snd_tne_Eamily.25.
742-749.

87

Greenberger, E., 8 Goldberg, W. A. (1989). Work, parenting,
and the socialization of the children. ngygigpmgntai

WI 251(1). 22-35'

Grotevant, H. D., 8 Cooper, C. R. (1986). Individuation in
family relationships: A perspective on individual
differences in the development of identity and role-

taking skill in adolescence. Hnmgn_pgxglgnment, 22,
82-100.

Grusky, D. B., 8 DiPrete, (1990). Recent trends in the
process of stratification. ngggzgphy, 21(4), 617-637.

Hackman, J. R., 8 Oldham, G. R. (1975). Development of the
Job Diagnostic Survey. WWW.
fig, 159-170.

Haug, M. R. (1977). Measurement in social stratification.
WM. 2. 51-77-

Haug, M. R. 8 Sussman, P. M. (1971). The indiscriminate
state of social class measurement. Sggiai_£gtgg§, 12,
568-563.

Henggeler, S. W., 8 Tavormina, J. B. (1980). Social class
and race differences in family interaction:
Pathological, normative, or confounding

methodological factors? Wis
W: 1.31. 211-222.

Hess, R. D. (1970). Social class and ethnic influences upon
socialization. In P. H. Mussen (Ed. ), Cazmighagi_§

Wigwam. (Vol- 2) Chicago: Markham.

Hess, R. D., 8 Shipman, V. (1968). Maternal attitudes
toward the school and the role of the pupil: Some
social class comparisons. In A. H. Pasow (Ed.),

ms
gigggygntgggg. New York: Columbia University Press.

Hill, J. P., 8 Holmbeck, G. (1986). Attachment and autonomy
during adolescence. In G. Whitehurst (Ed.), Anggi§_gt

ghiid_figygigpmgnt (Vol. 3, pp. 145-189). Greenwich, CT:
JAI.

Hodge, R. M., Siegel, P. M., 8 Rossi, P. H. (1964).
Occupational prestige in the United States, 1925- -63.
, 1Q, 286- 302.

88

Hoffman, L. W. (1986). Work, family, and the socialization
of the child. In R. D. Parke (Ed.),
' (pp. 223-282).

Chicago: University of Chicago Press.

Hoffman, L. W., 8 Hoffman, M. L. (1973). The value of
children to parents. In J. T. Fawcett (Ed.),
. New York:
Basic.

Hoffman, L. W., 8 Mannis, J. D. (1979). The value of
children in the United States: A new approach to the

study of fertility. Harrisge_and_tns_fanily. 12.
583-596.

Hoffman, M. L. (1960). Power assertion by the parent and
its impact on the child. child_Deyelonnent. 21.

129-143.

Hollingshead. A- B. (1957) Iwo:£nctor.index_of_sosial
pggitign. Unpublished manuscript, Yale University, New
Haven.

Hollingshead, A. B. (1971). Commentary on "indiscriminant
state of social class measurement." Sggiai_figzgg§, 12,
563-567.

Hollingshead. A B. (1975). E9nrzzastor_index_of_social
stgtnfii Unpublished manuscript, Yale University, Dept.
of Sociology, New Haven.

House, R. J., 8 Rizzo, J. R. (1972). Toward the measurement
of organizational practices: Scale development and

validation. 2onrnal_of_Annlied_P§yshology. 11. 388-396.

Hughes, D., Galinsky, E., 8 Morris, A. (1992). The effects
of job characteristics on marital quality: Specifying
linking mechanisms. 2onrna1_of_narriage_and_tne_£anily.
51, 31-42

Jacob, T. (1975). Family interactions in disturbed and
normal families: A methodological and substantive

review. Psychological_nnlletin. 22. 33-65.
Kanter. R. M. (1977) Eork_and_fan1ly_in_tne_united_fitates1

!! -°’!°.' 0 1:7. , .go so .

New York: Russel Sage Foundation.

Kline, M., 8 Cowan, P. (1988L Re-thinking the connections
among "work" and "family" and well-being. Jgntngi_gt

29cia1_Behayior_and_zersonality. 2. (4). 61-90.

89

Kohn, M. L. (1959). Social class and parental values.
, g5, 471-480.

Kohn, M. L. (1963). Social class and parent-child
relationships: An interpretation. Amgtigan_iggtnai_gt
292121991. 61. 337-351.

Kohn. M. L- (1969). Q1aee_and_eenfernity1_8_etndy_in
yaines. Homewood, Ill.: Dorsey Press.

Kohn, M. L. (1976). Occupational structure and alienation.
, 32, 111-130.

Kohn, M.L. (1977). -
(2nd ed.). Chicago: The University of Chicago Press.

Kohn, M. L. (1980). Job complexity and adult personality. In

N. J. Smelser 8 E. H. Erickson (Eds.), Ih§m§§_gt_ygtk
and_1eye_in_adnltheed (pp. 193-210). Cambridge. Mass-z

Harvard Univ. Press.

Kohn, M. L., 8 Schooler, C. (1969). Class, occupation and
orientation. Aneriean_§eeielegieal_8eyies. 21. 659-678.

Kohn, M. L., 8 Schooler, C. (1981). Job conditions and
intellectual flexibility: A longitudinal assessment of
their reciprocal effects. In D. J. Jackson 8 E. F.
Borgatta (Eds.). Eaeter_analyeie_ana_neaenrenent_in
-° 0 0° - ‘8‘: 9' : u- “9 U‘!‘ 'l: 9‘ =9: :.

London: Sage.

Kohn, M. L., 8 Schooler, C. (1982). Job conditions and
personality: A longitudinal assessment of their

reciprocal effects. Aneriean_lenrna1_ef_§eeielesy.
g1, 1257-1285.

Kornhauser. A- (1965). Henta1_healtn_ef_tne_inanetria1
ggtkgt. New York: Wiley.

Lamb, M. E. (1986). The changing role of fathers. In M. E.
W (Ede), 's .
(pp. 3-23). New York: Wiley.

Lich, S. S. (1985). Couples' experiences of first-time
parenthood: Intergenerational, marital, and social
factors as predictors of adjustment. (Doctoral
dissertation, California School of Professional

Psychology. Los AngeleS. 1985). Dissertatien_8betraete
International

90

Luster, T., Rhoades, K., 8 Haas, B. (1989). The relation
between parental values and parenting behaviors: A test

of the Kohn hypothesis. 2enrnal_ef_uarriage_and_the
11218111. 51: 139-147 .

Lyons, T. F. (1971). Role clarity, need for clarity,
satisfaction, tension, and withdrawal.
, §, 99-110.

McKinleY. 0- G. (1964). 2ee1al_elaee_and_fanilx_life.

London: The Free Press of Glencoe.

Miles, R. H. (1975). An empirical test of causal inference
between role perceptions of conflict and ambiguity and

various personal outcomes. Jggtngi_gttfippiigd
Psychology. 29. 334-339.

Miller, D. R., 8 Swanson, G. E. (1958). Ih§_ghgnging
aneriean_parent. New York: Wiley-

Miller, D. R., 8 Swanson, G. E. (1960). 1nng;_ggntiigt_gn§
ngtgngg. New York: Holt.

Minuchin. 8- (1974). Ean111ee_and_fanil¥_tnerapx-
Cambridge: Harvard University Press.

Moos, R. H. (1986). Work as a human context. In M. S.
Pallak, 8 R. O. Perloff (Eds.), Egyghgiggy_and_ggzki
' . Washington, D.C.:
American Psychological Association.

Mueller, C. W., 8 Parcel, T. L. (1981). Measures of
socioeconomic status: Alternatives and recommendations.

snild_2exelepnent. 52. 13-30-

Murphy, E. B., Silber, E., Coelho, G. V., Hamburg, D. A.,8
Greenburg, I. (1963). Development of autonomy and
parent-child interaction in late adolescence. Amazigan

learnal_ef_ertnepeyeniatry. 22. 643-652.

O'Brien, G. E. (1980). The centrality of skill-utilization
for job design. In K. Duncan, M. Gruneberg, 8 D.

Wallis (Eds ). Qhangee_in_xarting_life. Chichester:
Wiley.

O'Brien, G. E. (1982). The relative contribution of
perceived skill-utilization and other perceived job
attributes to the prediction of job satisfaction: A

cross validation study. figm§n_3§igtign§, 15, 219-237.

 

91

O'Brien, G. E. (1983). Skill utilization, skill variety,

and the job characteristics model. Anfitraii§n_ggntnai
QI_E§¥QthQQ¥r 55. 451-463-

O'Brien. G. E. (1986). Eeyehelegy_ef_werk_and_nnempleyment.
Chichester: John Wiley 8 Sons.

Ogbu, J. U. (1981). Origins of human competence: A cultural-

ecological perspective. 98118.nexelepment. 22(2). 413-
429.

Oldham, G. R., Hackman, J. R., 8 Stepina, L. P. (1978).
Norms for the Job Diagnostic Survey. Yale University
School of Organization and Management, New Haven.

Olson, D. H., Sprenkle, D. H., 8 Russell, C. S. (1979). i
Circumplex model of marital and family systems: I.
Cohesion and adaptability dimensions, family types, and

clinical applications. Egmiiy_£zgg§§§, lg, 3-27.

Otto, L. (1975). Class and status in family research.
, 11, 315-332.

 

Pearlin, L. I., 8 Kohn, M. L. (1966). Social class,
occupation, and parental values: A cross-national

study. Aneriean_Seeielegieal_Beyie1. 21. 466-479.

Pearlin, L. I., Menaghan, E. G., Lieberman, M. A., 8 Mullen,

J. T. (1981). The stress process. Jggtngi_gt_fig§ith
and_2eeia1_2enaxier. 22. 337-356.

Pedhazur. 8- J. (1982). Enltip1e_regreeeien_in_tenayiera1_
reeearen1_Explanatien_and_predietien. New York: Holt
Rinehart and Winston.

Pierce, J. L. (1978). The measurement of perceived job
characteristics: The Job Diagnostic Survey versus the

Job Characteristics Inventory. Agademy_gt_nanaggmgnt_
Journal. 21. 123-128-

Piotrkowski. C- S- (1979). Nerk_ana_tne_fanilx_exeten1_a
natnra1iet1e_etndx_ef_nettingzelaee_and_leuer:
migdig;gi§§§_tgmiiig§. New York: The Free Press.

Piotrkowski, C. S., 8 Katz, M. H. (1982). Indirect
socialization of children: The effects of mothers' jobs

on academic behaviors. thild_nexelepnent. 52.
1520-1529.

92

Repetti, R. (1987). Individual and common components of the
social environment at work and psychological well-
being. 2enrna1_ef_2eraana1ity_ana_§eeia1_£eyenalegy.
51(4), 710-720.

Repetti, R. (1989). Effects of daily workload on subsequent
behavior during marital interaction: The roles of
social withdrawal and spouse support.

lenrnal.ef
Eereena1ity_and_aeeia1_Payehelesx. 51(4). 651-659.

Rizzo, J., House, R. J., 8 Lirtzman, S. I. (1970). Role
conflict and ambiguity in complex organizations.
, 15, 150-163.

Ryan, R. M., 8 Lynch, J. H. (1989). Emotional autonomy
versus detachment: Revisiting the vicissitudes of

adolescence and young adulthood. Chi1d_peyeigpment, 69.
340-356.

Scheck, D. C., 8 Emerick, R. (1976). The young adolescent's
perception of early childrearing behavior: The
differential effects of socioeconomic status and family

size. Seeienetry. 22. 39-52.

Schaefer, E. S., 8 Edgerton, M. (1985). Parent and child
correlates of parental modernity. In I. E. Sigel
(Ed-) Earenta1_pe11ef_eyetene1_Tne_peyenelegieal
eeneesneneee_fet_enildren. Hillsdale. N. J.: Erlbaum

Schuler, R. S. (1979). A role perception transactional
process model for organizational
communication-outcome relationships. thanizatigngi

Behayier_and_Hnman_ EEIIQIEQBQQ. lfi. 93-107-
Sewell, W. H., 8 Hauser, R. M. (1980). Ing_fli§ggn§in

u na st d f soci a d o ' l
in_aen1tatiene_ana_aene1yemente1_Beeearen_1n_eeeielegx
ef_eaaeatien_ana_eeeia112atien. Greenwich. Conn.: JAI

Press.

Sims, H. P., Szilagyi, A. D., 8 Keller, R. T. (1976). The
measurement of job characteristics. Agadgmy_gfi

Management_lenrna1. 12. 195‘212-

Small, S. A., 8 Riley, D. (1990). Toward a multidimensional
assessment of work spillover into family life.
, 52, 51-61.

Smith, P. C., Kendall, L. M., 8 Hulin, C. L.(1969). Ihg
e .
Chicago: Rand-McNally.

 

93

Sears, R. E., Maccoby, E. E., 8 Levin, H. (1957). Patterns
gf_ghilg_zg§;ing. Evanston, IL.: Row 8 Peterson.

Staines, G.L. (1980). Spillover versus compensation: A
review of the literature on the relationship between

work and non-work. Human_3glatign§, 31, 111-129.

Steinberg, L., Elmen J.D., 8 Mounts, N. S. (1989).
Authoritative parenting, psychological maturity, and
academic success among adolescents. Qh11d_ngyg1gpm§nt,
6Q, 1424-1436.

Steinberg, L., Lamborn, s. D., Dornbusch, S. M., 8 Darling,
N. (1992). Impact of parenting practices on adolescent
achievement: Authoritative parenting, school
involvement, and encouragement to succeed. thld

Dexelnnnent. fig. 1266-1281-

Stevens, G., 8 Cho, J. H. (1985). Socioeconomic indexes and
the new 1980 census occupational classification scheme.

Sonial_§21ence_3e§eargh. 11. 142- -168-

Stutman, S. S. (1984). Family life cycle development:
Examination of the pathway linking differentiation from
the family of origin, marital adjustment, child-focused
triangulation, and adolescent adjustment (Doctoral
dissertation, California School of Professional

Psychology. Los AngeleS. 1984). Diseertatinn_bb§trant§
International

Stutman, S. S., 8 Lich, S. S. (1985). Ihg_dgyglgpm§nt_ang

'! ' 'L ! t ‘ -. '17! 9 l‘! .‘o

Unpublished Manuscript.

Thomas, D. L., Gecas, V., Weigert, A., 8 Rooney, E.(1974).
Eamilx_s22ializntinn_and_tne_ndnlesgent- Lexington:
Heath.

Waite, L., 8 Spitze, G. D. (1981). Young women's transition
to marriage. Demography, 1g, 681-694.

Waters, E., 8 Crandall, V. (1964). Social class and
observed maternal behavior from 1940-1960. guild

DQEQLQPEEDE. ii: 1021-1031-

Wexley, K. N., Alexander, R. A., Greenawalt, J. P., 8 Couch,
M. A. (1980). Attitudinal congruence and similarity
as related to interpersonal evaluations in
manager-subordiante dyads. Aggdgmy_gj_u§ngggmgnt
ionrnnl. 22: 320-330-

 

94

Wright, J. D., 8 Wright, S. R. (1976). Social class and
parental values for children: A replication and

extension of Kohn's thesis. Angriggn_figgiglggig§1
Benign. 11.

Youniss, J., 8 Ketterlinus, R. D. (1987). Communication and
connectedness in mother-and father-adolescent

relationshiPS- l2nrnal_9f_X92tn_and_Adnle§2enne. 16.
265-280.

Zimmerman, M. A., 8 Rappaport, J. (1988). Citizen
participation, perceived control, and psychological
empowerment.
1g, 725-750.

 

APPEN DI CES

 

APPENDIX A

STUDENT DEMOGRAPHIC QUESTIONNAIRE

95

APPENDIX A

STUDENT DEMOGRAPHIC QUESTIONNAIRE

 

Age:

Class: __Freshman __Sophomore __Junior
_Senior __Grad.

Ethnic Status: _Black __Caucasian __Oriental
Hispanic Native American Other

Marital Status: __Single Married Sep/Div

Were you adopted? __yes _no

Are your biological (or adopted) parents:

a) Married to each other? yes no

 

b) Separated from each other?__yes _no
If yes, how old were you when they separated?

c) Divorced from each other? __yes __no
If yes, how old were you when they divorced?

Previous to coming to MSU who did you primarily live with for the last ten

years? W221):
__ Biological or adopted Mother
__ Biological or adopted Father
_ Step-mother

__ Step-father

Other adult male relative (i.e. Uncle, Grandfather)

Other adult female relative (i.e. Aunt, Grandmother)

 

96

8. limntsz
FATHER
a) Age

 

b) Occupation
(Give job title, formal or informal of occupation for at least the last
10 years. If more than one title list both)

 

c) Description of Occupation

 

 

 

(Describe the kinds of activities your father does on the job i.e.,talk
with people, sell cars, work on a punch press, supervise other
employees in a repair shop)

d) Highest level of education

Less than 9th grade Less than 4

years of college
Less than 12th grade 4 Year College Grad.
High School Grad. Masters Degree
2 years or less of Ph.D., MD. etc.
college

e) Father's gross yearly income (approx)

_below 7,000 _4o,ooo to 50,000
_7,000 to 15,000 _50,000 to 60,000
_1s,ooo to 20,000 _60,000 to 80,000
_2o,ooo to 25,000 _80,000 to 100,000
____25,000 to 30,000 __Over 100,000

30,000 to 40,000

97

 

b) Occupation
(Give job title, formal or informal of occupation for at least the last
10 years. If more than one title list both)

 

c) Description of Occupation

 

 

 

(Describe the kinds of activities your mother does on the job i.e.,talk
with people, sell cars, work on a punch press, supervise other
employees in a repair shop)

(1) Highest level of education

_ Less than 9th grade __ Less than 4
years of college

__ Less than 12th grade _ 4 Year College Grad.

_ High School Grad. _ Masters Degree

_ 2 years or less of _ Ph.D., MD. etc.

college
e) Mother's gross yearly income (approx)

_below 7,000 ___40,000 to 50,000

_7,000 to 15,000 _50,000 to 60,000

_15,000 to 20,000 _60,000 to 80,000

____20,000 to 25,000 _80,000 to 100,000

_25,000 to 30,000 __Over 100,000

30,000 to 40,000

98

9. Siblings (including step-siblings)

Total Number of siblings you grew up with
10.

11.

12.

13.

14.

Age Sex Grade Living at Home
(Yes or No)

 

 

 

 

 

 

 

 

Do you live at home during summers and vacations? yes _no

Do you live at home during the school year? yes _no
(If yes, skip to #14)

What is the average number of phone contacts you have with your
parents each month?

With Father:
With Mother:

What is the average number of visits you have with your parents each
month?

With Father:
With Mother:

How many miles away from MSU do your parents live?

 

APPENDIX B

PARENT RELATIONSHIP INVENTORY SUBSCALES

. a.» 1‘1. A..."

 

99

APPENDIX B

PARENT RELATIONSHIP INVENTORY SUBSCALES

INSTRUCTIONS: The following questions refer to your current relationship with your father. If
you were not reared by your natural father, please respond in terms of the person who primarily
raised you until you were 18 in the role of father. If your father is no longer living answer in terms
of how you remember the relationship to have been as well as how you imagine the relationship
would be today if your father was still living.

PLEASE CIRCLE THE NUMBER WHICH BEST DESCRIBES YOUR AGREEMENT OR
DISAGREEMENT WITH EACH STATEMENT.

Strongly Strongly
Agree Agree Disagree Disagree
1. My father respects my desire to 4 3 2 1
be an independent person.
2. My father accepts my need for privacy. 4 3 2 l
3. My father wants to know all my thoughts. 4 3 2 1
4. My father doesn't try to influence the
decisions I make. 4 3 2 l
5. It is alright with my father if I
disagree with her. 4 3 2 1
6. My father doesn't try to tell me how
to run my life. 4 3 2 l
7. When my father gives me things, there
are generally “strings attached”. 4 3 2 1
8. I feel that my father tries to
interfere in my personal business. 4 3 2 1
9. My father often insists on making
me see things his way. 4 3 2 1

10. My father doesn't seem to recognize
that I've grown up. 4 3 2 1

11. My father frequently lets me know
that his generation knows best. 4 3 2 1

12. It's hard not to feel like a child
when I'm with my father. 4 3 2 l

 

 

 

100

Strongly Strongly
Agree Agree Disagree Disagree
13. The relationship I have with my father

feels like a relationship between equals. 4 3 2 1
14. My father is able to talk to me as one

adult to another. 4 3 2 1
15. It is fun to be with my father. 4 3 2 1

16. When I'm trying to reach a goal, I
can't depend on my father for

encouragement. 4 3 2 1
17. In my relationship with my father,

I often feel like an ”orphan“ 4 3 2 1
18. When I am feeling had, my father

shows little interest in my feelings. 4 3 2 1
19. My father and I feel like strangers

to one another. 4 3 2 1
20. I feel happy when I am with my father. 4 3 2 1
21. When I am feeling bad, I can count on

my father to remind me of my worth. 4 3 2 1
22. I feel tense when I am around my father. 4 3 2 1
23. I can't rely on emotional support from

my mother. 4 3 2 1
24. My father and I don't seem to have

very much in common with each other. 4 3 2 1
25. I feel very warm towards my father. 4 3 2 1

26. I” were in financial trouble, I would
feel comfortable asking my father to
lend me money. 4 3 2 1

27. I avoid asking my father for his

emotional support. 4 3 2 1
28. When I'm ill, I avoid asking my

father for sympathy. 4 3 2 1
29. If I needed practical help, I would

prefer not to go to my father for it. 4 3 2 1
Do these answers refer to your: Biological (or adoptive father)

Step-Father Other Father Figure

APPENDIX C

FATHER DEMOGRAPHIC QUESTIONNAIRE

 

 

101

APPENDIX C

FATHER DEMOGRAPHIC QUESTIONNAIRE

 

E l , D . .

1. Age:

2. Ethnic Status: _Black ____Caucasian __Oriental
_Hispanic Native American Other

3. Marital Status: _Single Married Sep/Div

4. Occupation

 

 

(Give job title, formal or informal of occupation for at least the last 10
years. If more than one title list them. If retired give last major
occupation.)

How long at this job title?

 

5. Description of Occupation

 

 

 

 

(—Describe the kinds of activities you do on the job i.e. talk with people, sell
cars, work on a punch press, supervise other employees in a repair shop)

6. Highest level of education:

_ Less than 9th grade _ Less than 4
years of college
_ Less than 12th grade __ 4 Year College Grad.
__ High School Grad. __ Masters Degree
_ 2 years or less of _ Ph.D., MD. etc.

college

102

7. Father's gross income (approx)
_below 7,000 _40,000 to 50,000
_7,000 to 15,000 _50,000 to 60,000
_15,000 to 20,000 _60,000 to 80,000
_20,000 to 25,000 _80,000 to 100,000
__25,000 to 30,000 _Over 100,000

30,000 to 40,000

 

APPENDIX D

SKILL UTILIZATION AND JOB AUTONOMY SCALES

 

 

 

103

APPENDIX D

SKILL UTILIZATION AND JOB AUTONOMY SCALES

Skill [1111113. Q'Qn

INSTRUCTIONS:Using the scale described below find: the number which best fits the following
statements about your work.

Not at all Very little Some A reasonable amount A Great Deal
1 1

 

 

z 4 s
1.1amyworkIhavethechancetolearnnewjobs. 1 2 3 4 5
2. In my work I have the opportunity to use my 1 2 3 4 5

own work methods.
3. In my work I am able to use my abilities. I 2 3 4 5
4. In my work I am able to use my training and 1 2 3 4 5
experience.
IQhAntanmx

PLEASE CIRCLE THE NUMBER FOR THE FOLLOWING STATEMENTS ABOUT YOUR
JOB WHICH BEST DESCRIBES HOW MUCH YOU AGREE OR DISAGREE WITH EACH
STATEMENT

Strongly Disagree Slightly Neither Agree Slightly Agree Strongly

 

Disagree Disagree nor disagree Agree Agree
1 2 ‘2 4 S 6 7

1. To be successful on my job requires 1 2 3 4 5 6 7
all my skill and ability.

2. On my job, I seldom get a chance 1 2 3 4 5 6 7
to use my special skills and abilities.

3. My job is very challenging 1 2 3 4 5 6 7

 

 

104

CIRCLE THE NUMBER WHICH BEST DESCRIBES WHETHER EACH STATEMENT IS
TRUE OR NOT ABOUT YOUR JOB.

Definitely More true More false Definitely

 

true than false than true false
1 9 1 4
1. People at my job are allowed to do almost 1 2 3 4

as they please.

2. How things are done at my job are left pretty 1 2 3 4
much up to the person doing the work.

3. At my job a person can make his own decision 1 2 3 4
without consulting anyone else.

4. Most people at my job make up their own rules. 1 2 3 4

USING THE SCALES DESCRIBED BELOW CIRCLE THE NUMBER WHICH BEST FITS
YOUR JOB

 

 

Very Little A moderate amount Very Much
1 7 1 4 s

1. How much are you left on your own to do 1 2 3 4 5
your work

2. To what extent are you able to work 1 2 3 4 5
independently of your supervisor in
performing your job function?

3. To what extent are you able to do your 1 2 3 4 5
job independently of others?

A minimum A moderate Very

amount amount Much
I 2 3 4 S

4. The freedom to do pretty much what I 1 2 3 4 5
want on my job.

5. The opportunity for independent 1 2 3 4 5
thought and action.

6. The Control I have over the pace of my work. 1 2 3 4 5

 

APPENDIX E

JOB SATISFACTION SCALE

105

APPENDIX E

JOB SATISFACTION SCALE

INSTRUCTIONS: For each of the following adjectives or statements, indicate whether or not it
describes your current job (or the last significant job you held).

Y = Yesitdescribesmyjob

U Uncertainifitdescribesmyjob

N No it doesn't describe my job

PLEASE MARK EITHER Y, U, OR N FOR EACH OF THE FOLLOWING WORDS THAT
DESCRIBE YOUR JOB.

l. Fascinating
2. Routine
3. Satisfying

4. Boring
5. Good

6. Creative

7. Respected
8. Hot

9. Pleasant

10. Useful

11. Tiresome
12. Healthful
13. Challenging
14. On your feet
15. Frustrating
16. Simple

17. Endless

18. Gives sense of accomplishment

APPENDIX E

JOB STRESS AND POWERLESSNESS SCALES

106

APPENDIX F

JOB STRESS AND POWERLESSNESS SCALES

INSTRUCTIONS: Answer true or false to the following question.
Circle either T = True or F = False

1. My job tends to directly affect my health.

2. I work under a great deal of tension.

. I have felt fidgety or nervous as a result of my job.

4. If I had a different job, my health would probably improve.

"fl'fl'l'l'rl'u
w

5. Problems associated with my job have kept me awake at night.

'1]
O\

. I have felt nervous before attending meetings at my job.

'-l '-l -l D-l -l -i -l

F 7. I often ”take my job home with me” in the sense that I think about
it when doing other things.

INSTRUCTIONS: Using the scale described below m the letter(s) that
indicate how much you agree with each of the following statements.

SD=Strongly Disagree D=Disagree A=Agree SA=Strongly Agree

1. There is really no way I can solve SD D A SA
some of the problems I have.

2. Sometimes I feel that I am being SD D A SA
pushed around in life.

3. I have little control over the things SD D A SA
that happen to me.

4. I can do just about anything I really SD D A SA
set my mind to.

5. I often feel helpless in dealing with SD D A SA
the problems of life.

6. What happens to me in the future mostly SD D A SA
depends on me.

7. There is little I can do to change many SD D A SA
of the important things in my life.

APPENDIX G

VALUES SCALE

107

APPENDIX G

VALUES SCALE

INSTRUCTIONS: For the following three sets of statements please mat the five statements from 1
to 5 with the most important thing you want your child (student) to learn ranked 1 the second most
importantthingrankedzandsoonwiththeleastimportantrankedi Giveeachfivestatementsa

different number. 1 8 Most important ............ Silent Important
M:
1. It is important for my child: RANK

a. to think for himself
b. to keep himself and his clothes clean
c. to be curious about many thing
(1. to be polite to adults
e. to be kind to other children
mm:

2. It is important for my child: RANK
a. to obey parents and teachers
b. to be responsible for his own work
c. to be kind and considerate
d. to keep thing neat and in order
e. to use imagination
Ihirslfiet:

3. It is important for my child: RANK

a. to be interested in how and why

thine happen
b. to have the ability to get along with people
c. to be a good student
d. to have the ability to look after himself

e. to have good manners

APPENDIX H

PARENTAL AUTHORITY QUESTIONNAIRE

 

108

APPENDIX H

PARENTAL AUTHORITY QUESTIONNAIRE

INSTRUCTIONS: For each of the following statements, circle the number on the 5~point scale
(1=strongly disagree, 5=strongly agree) that best indicates how that statement applies to you and
your father. Try to read and think about each statement as it applies to you and your father during
the last 5 years. There are no right or wrong answers, so don't spend a lot of time on any one item.
We are looking for your overall impression regarding each statement. Be sure not to omit any
items.

 

 

Strongly Strongly

Disagree Disagree Agree Agree
1 2 1 4 S

1. I feel that in a well—run home the children 1 2 3 4 5
should have their way in the family as often
as the parents do.

2. Even if my children don't agree with me, I 1 2 3 4 5

feel that it is for their own good if they
are forced to conform to what I think is right.

3. Whenever I tell my children to do something, 1 2 3 4 5
I expect them to do it immediately without

asking any questions.

4. Once family policy has been established, 1 2 3 4 5
I discuss the reasoning behind the
policy with my children.

5. I always encourage verbal give and take 1 2 3 4 5
whenever the children feel that family rules
and restrictions are unreasonable.

6. I feel that children need to be free to make 1 2 3 4 5
up their own minds and to do what they want
to do even if this does not agree with

what I might want.
7. I do not allow my children to question any 1 2 3 4 5
decision that I make.
8. I direct the activities and decisions of the 1 2 3 4 5

children in my family through reasoning
and discipline.

9. I feel that more force should be used by
parents in order to get their children to
behave the way they are supposed to.

10. I do 119$ feel that children need to obey

109

11.

rules and regulations of behavior simply
because someone in authority has
established them.

My children know what I expect of them in
my family, but they also feel free to discuss
these expectations when they feel they are
unreasonable.

12. I feel that wise parents should teach their

13.

14.

16.

17.

18.

19.

21.

children early just who is boss in the family.

I seldom give my children expectations and
guidelines for their behavior.

Most of the time I do what the children
in the family want when making family
decisions.

. I consistently give the children in my family

direction and guidance in rational and
objective ways.

I get very upset if my children try to disagree
with me.

I feel that most problems in society would be
solved if parents would _ngt restrict their
children's activities, decisions, and desires as
they are growing up.

I let my children know what behaviors are
expected them, and if they don't meet those
expectations, I punish them.

I let my children decide most thing for
themself without a lot of direction from me.

. I take the children's opinions into

consideration when making family decisions,
but I would not decide for something simply
because the children want it.

I do 391 view myself as responsible for
directing and guiding my children's behavior
as they grow up.

 

110

22. I have clear standards of behavior for my 1

children, but I am willing to adjust those
standards to the needs of each of the
individual children in the family.

. I give direction for my children's behavior 1
and activities and I expect them to follow my
direction, but I am always willing to listen to
their concerns and to discuss that direction
with them.

. I allow my children to form their own point 1
of view on family matters and I generally
allow my children to decide for themselves
what they are going to do.

. I feel that most problems in society would be I
solved if we could get parents to strictly and
forcibly deal with their children when they
don't do what they are supposed to do.

. I tell my children exactly what I want them to 1
do and how I expect then to do it.

. I give my children clear direction for their 1
behaviors and activities, but I am also
understanding when they disagree with me.

. I do m direct the behaviors, activities, 1
and desires of my children.

. My children know what I expect of them and 1
I insist that they conform to those expectations
simply out of respect for my authority.

. IfI make a decision that hurts my children, 1
I am willing to discuss that decision with my
children and admit it if I had made a mistake.

 

APPENDIX I

FAMILY EXPERIENCES QUESTIONNAIRE
(PARENTAL CONFIDENCE SUBSCALE)

 

111

APPENDIX I

FAMILY EXPERIENCES QUESTIONNAIRE
(PARENTAL CONFIDENCE SUBSCALE)

INSTRUCTIONS: Using the scale described below fish the letter(s) that indicate how much you
agree with each of the following statements.

SD= StronglyDlsagree D: Disagree A: Agree SA= StronglyAgr-ee

1. I know that I am doing a good job as a parent. SD D A SA
2. Being a parent turned out not to be as difficult

as I thought it would be. SD D A SA
3. Being a parent makes me feel drained and depleted. SD D A SA

4. I have the knowledge I need to be a good parent. SD D A SA

5. I should have read more books on parenting because
I often feel like I don't know what I am doing. SD D A SA

6. HI could do it over again I would raise my
children the same way I am raising them now. SD D A SA

7. I often worry that I am letting my children down. SD D A SA

8. Whenever I start feeling comfortable as a parent
something goes wrong and the doubts start all

over again. SD D A SA
9. Iworrythatlamnotdoingtherightthingas

a parent. SD D A SA
10. No matter how hard I try, I never seem to be

a good enough parent. SD D A SA
11. I often worry that I don't know enough to be

a good parent. SD D A SA
12. I often feel guilty about neglecting my children. SD D A SA

13. Juggling all the responsibilities of being a
parent is one of my talents. SD D A SA

14. Parenting means a lot of responsibilities and
problems, but I always feel that I can cope with
the difficulties that come along. SD D A SA

15. When there is a crisis with the children, I know
that I will do what needs to be done. SD D A SA

APPENDIX J

PARENT LETTER AND INFORMED CONSENT

112
APPENDIX J

MICHIGAN STATE UNIVERSITY

DEPARTMENT OF PSYCHOLOGY
EAST LANSING. MICHIGAN 48824

 

PARENT LETTER AND INFORMED CONSENT
WORK AND FAMILY INTERACTIONS STUDY
Dear Parent:

We are contacting you to request your participation in a
research project at MSU in which your undergraduate son or
daughter has voluntarily agreed to participate. The purpose of
this study is to examine the relationship between experiences
at the work place and interactions between parents and their
children. The study is being conducted by Jim Wasner, a
graduate student in clinical psychology under the supervision
of Susan J. Frank, Ph.D., Associate Professor of Psychology.

Participation in the study involves completing a questionnaire
asking your perceptions about your job conditions, your
beliefs about parenting, your relationship with your son or
daughter, and several questions about your family and life
situation in general. It will take approximately 1 hour to
complete these questionnaires. Your son or daughter who is
currently enrolled in an introductory level psychology class
at MSU also will be participating. Each student will be
receive extra credit in the course for completing a
questionnaire and arranging for his parents to participate.
If you or your son or daughter decide not to participate,
there are other options for earning extra credit.

Your participation is completely voluntary. You may choose
not to participate at all or not to answer particular items on
the questionnaires without any repercussions for you or your
child. However, all of your answers are valuable to this
study and your decision to omit various items may make it
difficult or impossible to use the information you do provide.
Although you may feel that some of the questions are personal,
several steps have been taken to protect your confidentiality.
First, we do not want you to put your name or address on the
questionnaire. Second, in order to protect the privacy and
confidentiality of your responses we have included an
addressed envelope for returning the questionnaire. When you
have completed the questionnaires you should put them in the
envelope provided, seal it, and return it by mail to the
address indicated. (If you write a separate letter or
postcard we will be happy to reimburse you for the cost of the
postage).

113

We have identified your and your child's questionnaires by a
common code number. This code number has been linked in a
list to your child's student identification number so that
they can be given assured of getting credit when the
questionnaires are returned. The primary researcher who is
not connected to the course has the only copy of this list and
it will be destroyed at at the end of the term.

We are aware that some of the parents may find it inconvenient
to return the questionnaires by'mail. You do have the option
of putting the completed questionnaire in the envelope
provided, sealing it, writing your first name over the seal,
and returning it with.your son.or daughter, who can turn it in
to a departmental mail box. You should be aware, however,
that if you chose to return it in this manner we are less able
to protect the full confidentiality of your responses.

To protect the validity of the study, please do not discuss
your answers with your spouse, friends, or children, until the
questionnaires are returned.

If you would like a summary of the results of this study you
may obtain them by writing: Jim Wasner, M.A., Department of
Psychology, MSU, East Lansing, Michigan 48824. If you have
any additional questions regarding this study’you may contact
Mr. Wasner at the above address and at (517) 355-9564 or at
487-6749.

Thank you for supporting the research and educational efforts
of Michigan State University and the Department of Psychology.

Jim Wasner, M.A.
Department of Psychology
Michigan State University

 

APPENDIX K

STUDENT INFORMED CONSENT

114
APPENDIX K

MICHIGAN STATE UNIVERSITY

DEPARTMENT OF PSYCHOLOGY
EAST LANSING, MICHIGAN 48824

 

STUDENT INFORMED CONSENT
WORK AND FAMILY INTERACTIONS STUDY

The purpose of this study is to examine the relationship
between experiences at the work place and interactions between
parents and their college aged children. The study is being
conducted by Jim Wasner, a graduate student in clinical
psychology under the supervision of Susan J. Frank, Ph.D.,
Associate Professor of Psychology.

Participation in the study involves completing a questionnaire
asking about your perceptions of your relationship with your
parents, as well as some general background information about
you and your parents. It will take approximately 30 minutes
to complete the questionnaire. In addition, I will ask you to
request that your parents participate in the study. Their
participation will also involve filling out several
questionnaires which can.be mailed to them or you can take to
them.directlyu .A letter will be enclosed in their'packet.that
will similarly explain this study and indicate the nature of
their participation.

You will receive extra credit in the course for completing
this questionnaire and arranging for your parents to
participate. Your own and your parents participation is
completely voluntary. You may choose not to participate at
all or not to answer particular items on the questionnaire
that you do not wish to answer. However, all of your answers
are valuable to this study and your decision to omit various
items may make it difficult or impossible to use the
information you do provide. Although you may feel that some
of the questions are personal, several steps have been taken
to protect you anonymity; We do not want you to put your name
or address on the questionnaire, and the data will only be
identified with a code number.

Your parents have been instructed to mail their questionnaires
directly to the Psychology Department in the envelopes that
have been provided for that purpose. A few parents may find
it more convenient to have you return their completed

questionnaires. You may return them to Jim Wasner's mailbox
in 135 Snyder or to Sharon the subjects pool coordinator in
137 Snyder; If your parents choose this option, the envelope
with the questionnaire must be sealed. You will notice that
the code numbers on your own and your parent's questionnaires

115

are the same. This will tell us that you are all members of
the same family. On the sign up sheet you should indicate:
a) your student identification number and b) the code number
on your envelope. The primary investigator will have the only
copy of the list linking the student numbers and code numbers,
and this will be utilized to determine for which students we
have received both student and parent questionnaires. At the
end of the term this list will be destroyed.

If you would like to know the results of this study , you can
contact Jim Wasner, Department of Psychology, 135 Snyder, MSU,
East Lansing, MI 48824 or at 355-9564 or 487-6749.

I, the undersigned, have been fully informed as to the nature
of this study and voluntarily agree to participate.

Signed

 

Date

 

 

APPENDIX L

STUDENT DEBRIEFING FORM

 

116

APPENDIX L

STUDENT DEBRIEFING FORM

Work and Family Relationships Study

Thank you for participating in this research study. I hope you and your parents
found the experience interesting and enjoyable.

This study is designed to explore the relationship between parent's experiences at
their jobs and the relationship they have with their children. The study focuses
particularly on how the stresses and demands of the workplace and the values
learned there might influence the interactions between parents and their children.

Your participation in this study required that you answer a variety of questions about
your relationship with your parents, as well as some description of your family and
your parent's jobs. Your parents were also asked to fill out several questionnaires
about their work, their attitudes and values about certain things, and their
relationship with their children. By analyzing this data we will be able to look at the
relationship between occupational conditions and family relationships and better
understand the mediation process between work and family life.

If you are interested in finding out the results of this study, please contact me at the
telephone number or address below. I will take your name and address and mail you
a brief summary of the results when the study is completed.

If you are interested in reading further on this subject, the books and articles below
will provide an introduction to the issues covered in the research. Thanks again for
your time and interest.

Jim Wasner, MA Home: 487-6749
Department of Psychology Office: 355-9564
Michigan State University

References:

Hoffman, LW. (1986). Work, family, and the socialization of the child. In R. D.

Parke (Ed) WWW
(223-282). Chicago: University of Chicago Press.

Mortimer, J. T., Lorence J., & Kumka, D. S. (1986).
W Norwood, N.J.: Ablex Publishing.

l’iotrkowski C S (1979) W
W New York: The Free Press

 

"llllllllllll'll'lll“