Citation
The relationship between employment status and sex role attitudes

Material Information

Title:
The relationship between employment status and sex role attitudes a problem of selectivity
Creator:
McCarty, James Michael
Publication Date:
Language:
English
Physical Description:
113 leaves : ; 29 cm

Subjects

Subjects / Keywords:
Sex role in the work environment ( lcsh )
Sex role in the work environment ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 104-110).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Department of Sociology.
Statement of Responsibility:
by James Michael McCarty.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
20983869 ( OCLC )
ocm20983869
Classification:
LD1190.L66 1989m .M32 ( lcc )

Full Text
The Relationship between Employment
Status and Sex Role Attitudes:
A Problem of Selectivity
by
James Michael McCarty
B.A., University of Colorado at Denver, 1986
A thesis submitted to the
\
Faculty of the Graduate School of the
University of Colorado in fulfillment
of the requirements for the degree of
Master of Arts
Department of Sociology
1989


This thesis for the Master of Arts degree by
James Michael McCarty
has been approved for the
Department of
Sociology
by
Pat e*%/


iii
McCarty, James Michael (M.A., Sociology)
The Relationship between Employment Status and Sex
Role Attitudes: A Problem of Selectivity
Thesis directed by Associate Professor Richard H.
Anderson
This study addresses the problem of
selectivity occurring in the relationship of
employment status to sex role attitudes, which
previous research has neglected. Specifically,
this study proposes that white women with certain
backgrounds will be more likely to be employed and
also hold less traditional sex role attitudes.
Although the findings, which derive from a
secondary analysis of the General Social Survey,
1972-1987, are not conclusive, the problem of
selectivity cannot be discounted. Findings
indicate that there is a complex interaction
between background variables, employment status,
and sex role attitudes. Differences in sex role
attitudes between employed women and homemakers
suggest that educated women may form two groups,
those who have attained the twelfth grade or less
and those with college backgrounds. An unexpected
finding is that women with either high status


iv
mothers or fathers, in terms of education, tend to
be less likely to be employed and more traditional
in sex role attitudes than other women with
parents of lower educational backgrounds.
Although the possibility of selectivity is highly
likely, findings show that differences in sex role
attitudes between employed women and homemakers of
similar backgrounds remain. A plausible
interpretation is that women take into account
their own statuses and the need to support them
when answering items measuring sex role attitudes,
which is especially relevant in a society where
women's statuses are problematical. The
possibility of interdependence of the variables
cannot be dismissed. Future research should
address the problem of women's statuses and also
be more sensitive to the problem of selectivity.


CONTENTS
CHAPTER
I. INTRODUCTION............................. 1
THE PROBLEM.............................. 1
II. RATIONALE................................ 6
Hypotheses............................ 10
III. LITERATURE REVIEW....................... 11
Traditionality in Sex Role
Attitudes: Definition.............. 11
Parental Background................... 15
Education............................. 18
Religious Background.................. 21
Spouse's Background................... 24
Other Background Characteristics... 27
Age/Cohort.......................... 28
Race................................ 30
Summary............................... 31
Specific Conditions of the
Second Hypothesis................. 32
IV. METHODS.............................. 3 4
Measurement of the Dependent
Variable: Sex Role Attitudes....
34


vi
Measurement of Independent and
Control Variables................... 39
Employment Status................... 39
Parental Background................. 40
Education........................... 40
Religious Groups and
Denominations..................... 41
Spouse's Background
Characteristics................... 42
Husband' s EduCat i on............ 42
Husband's Occupation.............. 42
Husband's Occupational
Prestige........................ 43
Other Background
Characteristics................... 43
Age............................... 43
Cohort Analysis Over Time......... 44
Data Analysis......................... 44
V. RESULTS.................................. 45
Employment Status...................... 45
Parental Background.................... 47
Educational Background................. 53
Religious Background................... 56
Spouse's Background.................... 68
Age/ Cohort............................ 76
Summary................................ 83
VI. DISCUSSION
89


Vll
VII. CONCLUSIONS............................. 100
BIBLIOGRAPHY.................................. 104
APPENDIX
A. TABLES WHICH ARE CITED IN THE TEXT____ 111


TABLES
Table
1. Sex Role Attitudes by Employment Status
(Reported in Percentages)................. 4 6
2. Employment Status of White Women by
Father's Education (Reported in
Percentages)................................ 48
3. Employment Status of White Women by
Mother's Education (Reported in
Percentages)................................ 49
4. Percentage Holding Least Traditional
Sex Role Attitudes by Employment
Status, controlling for Father's
Education.................................. 50
5. Percentage Holding Least Traditional
Sex Role Attitudes by Employment
Status, controlling for
Mother's Education.......................... 51
6. Employment Status of White Women by
Respondent's Education (Reported in
Percentages)................................ 54
7. Percentage Holding Least Traditional
Sex Role Attitudes by Employment
Status, controlling for
Respondent's Education...................... 55
8. Employment Status of White Women by
Religious Group of Youth (Reported
in Percentages)............................. 57
9. Percentage Holding Least Traditional
Sex Role Attitudes by Employment
Status, controlling for Religious
Group of Youth.............................. 58


ix
10. Percentage Holding Least Traditional
Sex Role Attitudes by Religion at
age 16, controlling for Education
for Women who are Homemakers............... 59
11. Percentage Holding Least Traditional
Sex Role Attitudes by Religion at age 16,
controlling for Education for
Women who are Employed..................... 60
12. Difference between White Employed Women
and White Homemakers in Percentage
Holding Least Traditional Sex Role
Attitudes by Religion at age 16,
controlling for Respondent's Education.... 61
13. Employment Status of White Women by
Denomination of Youth (Reported
in Percentages)............................ 63
14. Percentage Holding Least Traditional Sex
Role Attitudes by Employment Status,
controlling for Denomination of Youth..... 64
15. Percentage Holding Least Traditional Sex
Role Attitudes by Denomination of Youth,
controlling for Respondent's Education
for White Women who are Employed........... 65
16. Percentage Holding Least Traditional Sex
Role Attitudes by Denomination of Youth,
controlling for Respondent's Education
for White Women who are Homemakers.......... 66
17. Difference between White Employed Women
and White Homemakers in Percentage
Holding Least Traditional Sex Role
Attitudes by Denomination, controlling
for Respondent's Education.................. 68
18. Employment Status of White Women by
Spouse's Education (Reported
in Percentages)............................. 69
19. Percentage Holding Least Traditional
Sex Role Attitudes by Employment Status,
controlling for Spouse's Education.......... 70


X
20. Employment Status of White Women by
Spouse's Occupation (Reported in
Percentages)................................. 71
21. Percentage Holding Least Traditional Sex
Role Attitudes by Spouse's Occupation,
controlling for Employment Status.......... 73
22. Employment Status of White Women by
Spouse's Occupational Prestige (Reported
in Percentages).............................. 74
23. Percentage Holding Least Traditional Sex
Role Attitudes by Employment Status
controlling for Spouse's Occupational
Prestige.................................. 75
24. Employment Status by Age (Reported
in Percentages)............................. 76
25. Percentage Holding Least Traditional Sex
Role Attitudes by Employment Status,
controlling for Age........................ 78
26. Difference in Percentage Holding Least
Traditional Sex Role Attitudes between
White Employed Women and White Homemakers
by Age, controlling for Respondent's
Education.................................. 79
27. Percentage of White Women who are
Homemakers by Cohort, controlling
for Year (Reported in Percentages).............. 81
28. Percentage of White Employed Women and
Homemakers Holding Least Traditional Sex
Role Attitudes by Cohort, controlling
for Year................................
82


CHAPTER I
INTRODUCTION
THE PROBLEM
A consistent finding of studies of American
women's sex role attitudes is that employed women are
less traditional than homemakers in sex role attitudes
(Andersen and Cook 1985; Mason, Czajka, and Arber
1976; Molm 1978; and Stokes and Peyton 1986).
Traditional sex role attitudes posit that men and
women should perform different roles, because they
have distinct natures. The finding of differences in
sex role attitudes between employed women and
homemakers apparently warrants Molm's (1978) argument
that being employed, which is viewed in terms of
performing a less traditional role, causes less
traditional sex role attitudes. While the studies of
Molm (1978) and Andersen and Cook (1985) have found
inconsistent results regarding the causal effect of
employment status on sex role attitudes, both of these
studies fail to ask the most important question
concerning the relationship of employment status to
sex role attitudes: Are employed women less


2
traditional in sex role attitudes due to their being
employed, or is it that women, who are more likely to
be employed, are also less traditional in sex role
attitudes? This is the focus of this study.
In the literature, the issue of this study is
known as a problem of selectivity (Hyman 1955). In
the case of employed women and their sex role
attitudes, it is possible that women who are likely to
engage in employment are also likely to hold less
traditional sex role attitudes. The other possibility
is that the experience of employment may lead to less
traditional sex role attitudes. Selectivity assumes
the voluntary participation of women in employment.
Employed women of the past may have been involuntarily
employed, as is suggested by the discussion of
Mullings (1986). More recently, since Ross, Mirowsky,
and Huber (1983) indicate that there are changing
norms regarding women's employment, it is highly
likely that many women may voluntarily seek
employment. However, as I have already indicated,
other studies of sex role attitudes have not addressed
the problem of selectivity.
While studies have controlled for the effects
of other variables when examining the relationship of
employment status to sex role attitudes, a rationale


3
providing the criteria for selection of the control
variables is not apparent in these studies. The
findings of studies of sex role attitudes, which are
not particularly concerned with the relationship
between employment status and sex role attitudes,
indicate that in addition to employment status, many
key variables representing variations in background
characteristics are related to sex role attitudes
(Mason et al. 1976; Thornton, Alwin, and Camburn 1983;
Mason and Lu 1988). However, only some of these
variables are controlled for in studies specifically
examining the relationship of employment status and
sex role attitudes. This suggests that a rationale
has not been developed for the application of specific
control variables, when examining this particular
relationship.
Perhaps these studies assume that the control
variables are randomly related to employment status.
This is the impression given by Andersen and Cook
(1985) who match employed women and homemakers based
on an approximation of background characteristics.
This is incorrect, because it is likely that
employment status is related to many of the other
variables, which are used as controls. Studies of the
relationship of employment status to sex role


4
attitudes have not adequately examined if the control
variables are related to the independent variable,
employment status.
An examination of the relationship of
employment status to sex role attitudes and to
variables representing background characteristics
requires an analysis of women within at least the last
twenty years. This is necessary because of the
historical circumstances of women. It can be
inferred from Ferree (1974) that a systematic
examination of sex role attitudes is reasonable for
women of more recent times. Bernard (1976) suggests
that changes in women's norms and attitudes have
occurred in this time period.
It is reasonable to propose that a systematic
study of the relationship between employment status
and sex role attitudes be conducted for women of the
1970s and 1980s. I propose to carefully select
variables, which represent variations in background
characteristics when examining this relationship.
Since these background variables tend to be related to
both employment status and sex role attitudes, it is
reasonable to propose that selectivity is occurring
in that women who are more likely to be employed also
hold less traditional sex role attitudes. An


5
examination of the relationship of employment status
and sex role attitudes, when controlling for the
effects of background variables, will most likely
reveal that employment status is a spurious variable,
which suggests that background variables are the
critical variables in explaining sex role attitudes.


CHAPTER II
RATIONALE
In examining the relationship between
employment status and sex role attitudes, researchers
have not fully taken into account that employment
status does not occur randomly. Although the
increased labor force participation of women is due to
factors external to individuals, personal background
characteristics influence whether or not women will be
employed (Bergmann 1986). In addition, the
occupational development of females is related to
variations in socialization contexts, which refer to
variations in backgrounds (Ireson and Gill 1988;
Falkowski and Falk 1983).
The literature indicates the importance of
the following personal background characteristics in
regards to either actual employment status or
occupational development: Parental background,
educational background, spouse's background, and one's
age (Blau and Ferber 1985; Falkowski and Falk 1983;
Evans 1987). By age, I am referring not only to
actual age, but also to one's cohort. Religious


background is also likely to be important (Umansky
1985; Homola, Knudsen, and Marshall 1987).
7
These same background characteristics are
also related to sex role attitudes (Mason et al. 1976;
Thornton et al. 1983; Mason and Lu 1988).
This suggests that selectivity is occurring
in sex role attitudes. It is reasonable to propose
that women who are likely to be employed are also more
likely to hold less traditional sex role attitudes.
Consistent with this reasoning is that the
effect of employment status on sex role attitudes is
likely to be spurious. The rationale for the
importance of employment status is provided by Molm
(1978). Molm (1978) argues that the experience of
being employed, which places the employed woman in a
group which encourages the pursuit of careers for
women, leads to less traditional sex role attitudes.
On the other hand, homemakers will hold more
traditional attitudes for similar reasons (Molm 1978).
Molm (1978) assumes that there is consistency
in women's behavior and attitudes. This is counter to
Myrdal (1969), who notes that there is often a
discrepancy between people's behavior and their
attitudes. The argument that women's employment
status leads to less traditional sex role attitudes is


8
based on a rational model. Historical studies suggest
that this rational model was unlikely for women of the
past, who often were employed, but who did not
necessarily hold less traditional attitudes
(Oppenheimer 1970; Gutman 1976; Mullings 1986).
Of course, we should ask if employment is
representative of less traditional behavior. Kahn and
Crosby (1985) argue that employed women hold less
traditional attitudes, but they are employed in
organizations which attempt to structure behavior
along traditional lines. Similarly, Benenson (1985)
argues that women who belong to dual-career families,
which are thought to be the least traditional in both
behavior and attitudes, engage in traditional behavior
in order to support the careers of their high status
husbands. These arguments seem to suggest that less
traditional sex role attitudes do not derive from
engaging in less traditional behavior.
Bernard (1976) argues that less traditional
sex role attitudes represent the diffusion of norms
and ideas from an elite group of women to other women.
Those women who resist the adoption of less
traditional attitudes tend to be less educated and
older than other women (Bernard 1981). These


9
variables are also related to employment status (Blau
and Ferber 1985).
The finding that employed women hold less
traditional sex role attitudes gives the semblance
that the behavior and attitudes which women engage in
are consistent. This consistency may be due to women
having backgrounds which provide the basis for the
selection of future employment status and also the
adoption of less traditional sex role attitudes. I
assume that backgrounds provide the individual with
both the means and values for pursuing future
employment. If we know women's backgrounds, there is
a possibility that we will also know both their
behavioral and attitudinal tendencies.
I assume that the effects of variables
representing background characteristics are not likely
to be reversible. However, a rationale which posits
the importance of employment status to sex role
attitudes implies that employed women and homemakers
will differ in sex role attitudes, even though they
may have similar backgrounds. I assume the opposite.


10
Hypotheses
My rationale suggests the following
hypotheses:
1. Employed women and homemakers who are similar
in background characteristics will hold
similar sex role attitudes.
2. Women with certain background characteristics
are more likely to be employed (or less likely
to be homemakers) and also hold less
traditional sex role attitudes. (The
literature review will specify the conditions
of these hypotheses).


CHAPTER III
LITERATURE REVIEW
I will now present the existing evidence
regarding the relationship of sex role attitudes to
background variables. I will also note how these
variables are also related to employment status or
occupational development. However, I will first
define what is meant by the concept traditional sex
role attitudes before examining the relationship of
various background characteristics to sex role
attitudes.
Traditionalitv in Sex Role
Attitudes: Definition
Attitudes consist of "valuations" and
"beliefs" (Myrdal 1969, p. 15). An individual's
"valuations" refer to how reality should be, while
"beliefs" are an individual's knowledge of reality.
Sex role attitudes have both of these aspects. Peopl
who support traditional sex role attitudes have
specific attitudes concerning what should be the
appropriate behavior of men and women. In addition,


12
people with traditional sex role attitudes have
knowledge regarding the distinctive natures of both
sexes.
People with traditional sex role attitudes
express the attitude that women's proper place should
be the domain of the home. This is a valuation. In
addition, it is considered to be ideal for a married
woman not to be working in business or industry. Of
course, this requires a husband capable of supporting
her. The theme of the adequacy of the male provider
is a part of traditional sex role attitudes.
According to Gill, Stockard, and Johnson
(1984), females are expected to be dependent, non-
instrumental, and emotionally expressive. Parsons
(1942) presented a theory of the family, which
reflects these traditional expectations of females.
The man is the instrumental leader in his family,
while the woman is the expressive leader. As a
homemaker, the woman is able to provide emotional
leadership in her family, while the man is the
breadwinner.
Individuals with traditional sex role
attitudes hold the belief that men and women have
distinct natures that make them suitable for some
tasks, but not others (Rothman 1978). For example,


women's natures are thought to be best suited for
nurturing and motherhood (Rothman 1978).
13
While the ideal in traditional sex role
patterns for married women is not to work outside the
home, when women do work, certain types of work are
thought to be unsuitable for them. Oppenheimer (1970)
indicates that in 1900, an issue that concerned many
was the hardship of women employed in industrial
occupations. Americans believed that this labor was
too physically demanding for women. Occupations, such
as nursing and teaching, have been considered more
appropriate for women, because they are viewed as
extensions of women's traditional functions of
nurturing and childcare (Oppenheimer 1979). The
occupation of clerical work, which is
disproportionately performed by women, is viewed as
"clean" and "ladylike" (Bergmann 1986, p.' 3 6).
Traditional sex role attitudes reflect the
valuation that employment should be secondary to a
woman's primary duty to her family. People with
traditional sex role attitudes believe that a woman
should either forego employment or terminate outside
employment if it inconveniences her husband or the
welfare of her children (Scanzoni 1978).


14
Oppenheimer (1970) indicated that the issue
of women's employment is influenced by the historical
period. In 1900, the issue for Americans was the
employment of all women, while in the 1950s, the focus
was on the employment of married women (Oppenheimer
1970).
When women work, there are traditional sex
role attitudes that reflect traditional assumptions
about women even though they are working. For
example, when married women work, traditional sex
roles demand that it should not be threatening to the
status of husbands (Coser and Rogoff 1971). On the
other hand, equality between the spouses in status is
likely to be representative of less traditional sex
roles (Ross, Mirowsky, and Huber 1983).
The issue of competition between men and
women not only refers to employment, but other
activities as well, such as education and politics.
In education, Komarovsky (1946) presents this as a
"cultural contradiction." During the 1940s, women
college students would have to present themselves to
men as inferior. People with traditional sex role
attitudes express the attitude that women will not
compete with men.


15
Traditional sex role attitudes have political
implications (Cherlin and Walters 1981). Women are
thought to be unqualified when compared to men for the
holding of political office, because they are
emotionally unsuited for politics. Once again, the
implication is that women should not perform
instrumental tasks, because of their emotional
natures. Furthermore, men and women should not
compete with each other, because they have distinct
natures.
In summary, people with traditional sex role
attitudes advocate the distinct specialization of sex
roles for men and women. This is predicated on the
belief that men and women have different natures. I
will now discuss the effect of variations in
background characteristics on both sex role attitudes
and employment status. I will first begin with a
discussion of parental background.
Parental Background
The literature indicates that parent's level
of educational attainment affects an individual's sex
role attitudes more than other parental
characteristics (Thornton et al. 1983). In a panel
study of Detroit mothers initially conducted in 1977


16
and followed up in 1980, it was found that well
educated fathers and mothers were more likely to be
egalitarian in attitudes and also to have more
egalitarian children (Thornton et al. 1983).
Mother's work experience has been found to be
unrelated to children's sex role attitudes (Thornton
et al. 1983). This suggests that of all the parental
background variables, father's and mother's education
are the most relevant.
The literature does not indicate whether
women's employment status is related to parents'
educational level. An indirect indication comes from
a study of young women, which found that those who
expected to be homemakers at age 30 were more likely
to have fathers with lower occupational status
(Falkowski and Falk 1983). Since occupational status
is often associated with educational level, I expect
that the higher the level of education of women's
parents, the more likely they will be employed and not
homemakers.
By homemaker, studies refer to a woman who
is a full-time homemaker, in the sense that she is not
engaged in employment for wages. While there are
"househusbands," they are not the norm (Beer 1983).


17
This is why it is reasonable to examine only women for
this proposal.
The literature views the homemaker as an
individual with a status and performing a role (Molm
1978). The difference between status and role is that
status refers to an individual's position in the
social system, while role refers to the behavioral
expectations of an individual in a particular status
(Abercrombie, Hill, and Turner 1984). For this study,
I view homemakers and employed women in terms of
status.
It may seem reasonable to conceptualize
homemakers and employed women in terms of roles,
instead of status, especially if sex role attitudes
represent individuals' attitudes concerning
behavioral expectations of men and women. However,
attitudes concerning behavioral expectations of men
and women are different from a group's expectations of
an individual performing a particular role. While an
individual may express a less traditional sex role
attitude in response to an interview, the expectations
of a group may direct a person's behavior differently
from the expectations reflected in an attitude. Sex
role attitudes are not necessarily indicators of
expectations of the roles performed by the individual


18
answering the interview. On the other hand, it is
reasonable to view differences in attitudes between
individuals who vary in status.
Although Matthew's (1987) study of the
homemaker reveals that the homemaker's status has been
devalued in the twentieth century, I am not examining
variations in prestige between homemakers and women
who are employed. The homemaker's status is defined
as being non-employed, while the employed woman's
status is defined as being employed.
I will now discuss the effect of educational
background on sex role attitudes and employment
status.
Education
Educational level is a significant
explanatory variable of sex role attitudes (Mason et
al. 1976; Mason and Lu 1988; Funk and Willits 1987).
Better educated women hold less traditional attitudes
than women with less education (Mason and Lu 1988).
College educated women were the least
traditional of all women in sex role attitudes in 1976
and also 1985, even when controlling for age (Mason
and Lu 1988). It should be noted that studies suggest
that college educated women have changed over time.


19
Komarovsky (1985) has shown that female college
students of the 1940s were more traditional in their
sex role attitudes than students of the late 1970s.
This suggests that the relationship of education to
sex role attitudes has become systematic within the
last two decades.
Education is an important factor in the
socialization of individuals and the formation of
their sex role attitudes. Evidence from a panel
study, which was initially conducted using more than
11,000 high school students from Pennsylvania in 1970,
shows that higher education has an impact on changes
in sex role attitudes (Funk and Willits 1987). In the
follow-up study in 1981, which examined a random
sample of individuals from the initial survey, the
researchers found that high school students who later
graduated from college, had the greatest change toward
less traditional sex role attitudes of all the
students. The least change occurred for those who did
not extend their formal schooling past high school.
Those who went to college, but did not graduate, were
between the two extremes (Funk and Willits 1987).
The study of Funk and Willits (1987) provides
support to the argument that socialization is an on-
going process. This is despite the argument of Nock


20
(1987), who states that the acquisition of sex role
attitudes comes before individuals reach childbearing
ages.
Bernard (1981) argues that the better
educated are more likely to accept changes in sex role
attitudes, while the less educated resist changes.
Degler's (1980) discussion of the women's movement
appears to support Bernard's (1981) argument.
According to Degler (1980), the women's movement,
which was instrumental in promoting ideas that
supported less traditional sex roles, was primarily
aimed at educated and middle class women.
Education is also related to employment
status. Stokes and Peyton (1986) found that while
there is little difference in age between homemakers
and employed women, homemakers are less educated than
employed women. However, Stokes and Peyton (1986)
assume that employment status has an independent
effect. In addition, Stokes and Peyton (1986) only
examined married women under 50 years of age.
According to Lopata (1984), a homemaker does not
necessarily have to be married, for example, she can
be a widow.
The findings of the National Longitudinal
Study of the occupational plans of high school


21
students in 1972 shows that young women, who expect to
be homemakers at age 30, have lower scholastic
performance and tend to be on a nonacademic track
(Falkowski and Falk 1983) .
The higher the level of education, the more
likely a woman will be employed (Blau and Ferber
1985). Geerken and Gove's (1983) review of the
literature indicates that education is a resource for
those seeking to be employed. Campbell (1974) argues
that it has become the norm for college educated women
to be employed. I expect that education is the cause
of both employment status and sex role attitudes.
Religious Background
The literature indicates that the religious
preference of women is related to sex role attitudes.
The validation of traditional sex roles comes
through religion, which has beliefs which invoke a
deity's demand that the sexes perform distinct roles
(Umansky 1985). The findings of a qualitative study
of women's political activism support the importance
of religion in validating sex roles (Klatch 1987).
However, not all religious groups are equal in their
validation of traditional sex roles. Mason et al.


22
(1976) found that Catholic women are more traditional
than Jewish women.
A later study, which used data from two
national probability samples conducted in 1977 and
1985, confirmed that Jewish men and women are less
traditional in sex role attitudes than both Catholics
and Protestants (Mason and Lu 1988). Catholics and
Protestants were about the same (Mason and Lu 1988).
Umansky (1985) states that Jewish women have become
ideologically liberal in their attitudes toward the
status of women.
Studies have not examined the effect of
religious background of women's youth on sex role
attitudes. Consistent with the argument of this paper
is that the religion the individual is raised in is
more relevant than her present religion.
The variations in sex role attitudes of women
raised in different denominations should also be
examined. Since this has not been examined by other
studies of sex role attitudes, my proposal to examine
the effect of variations in denominations can be
considered exploratory in this regard.
When examining the effect of religous groups
and denominations on sex role attitudes, the control
variable of education should be introduced. Homola et


23
al. (1987) found that Jewish men are better educated
than other groups. Men who are Methodists are better
educated than Baptists. Episcopalians have a
tradition of seeking higher education (Homola et al.
1987). However, since this study examined only men,
it is not certain whether or not the findings can be
extended to women. This suggests that education
should be controlled for.
I speculate that Baptists are more
traditional in sex role attitudes than other
denominations, because they are less educated than
other denominations (Homola et al. 1987).
The literature also indicates that there is a
relationship between employment status and religious
background. Homemakers are more religious than
employed women (Luckmann 1967; Harrington 1983).
Harrington's (1983) interpretation is that homemakers
are more likely to resist secular ideas. However,
there is an absence of evidence as to whether or not
there are variations in employment status among women
from various religious groups.
In summary, religious preference of one's
youth is likely to be related to both employment
status and sex role attitudes. Education should be
controlled for when examining this, because people


24
with various preferences are likely to also vary in
level of educational attainment, which is also related
to sex role attitudes.
Spouse/s Background
For married women, researchers have examined
the effect of spouses' social characteristics on
women's sex role attitudes. Studies have found that
women married to husbands with higher educational
backgrounds are less traditional in their sex role
attitudes (Thornton and Freedman 1979; Thornton et al.
1983; Mason et al. 1976).
Goode (1963) argues that the well educated
man is more ideologically liberal in issues related to
women's status than working class men. Goode (1963)
uses class and education interchangeably.
Centers' (1961) study of class conducted
during the 1940s found that middle class men were more
likely to be less traditional on attitudes toward
women's employment than working class men.
In addition to examining the relationship of
husband's education to women's sex role attitudes,
Mason et al. (1976) also analyzed the effect of
husband's income level on women's sex role attitudes.
They found an inconsistent relationship. However, I


25
believe that it is more reasonable to focus on other
aspects related to the husband's occupation.
Since there are a variety of occupations, an
examination of women's sex role attitudes may take
into account the effect of variations of husbands'
occupations. I expect women married to professionals
and managers, which I consider to be middle class, to
be less traditional in sex role attitudes than women
who are married to men who are in other types of
occupations, such as, craftsmen, operatives, transport
equipment operatives and laborers, which I consider to
be working class.
Occupations also vary in status. I propose
to examine the effect of husbands' status on women's
sex role attitudes. The inclusion of this variable
for my proposal is justified, because historically it
has been argued that the status of a man was dependent
on having a non-employed wife (Veblen 1899). However,
in more recent times, I expect this variable to have
an opposite effect on women's sex role attitudes
because higher status individuals are thought to be
less traditional in general social issues (Harrington
1983). I extend this to attitudes toward sex roles.
It is reasonable to propose that a woman's
spouse may have some influence on her sex role


26
attitudes. I will now discuss the relationship of
employment status to spouse's background
characteristics.
The literature regarding the relationship of
class to employment is not fully consistent.
According to Evans (1987), the literature indicates
that after 1940 and up to the 1970s, middle class
women have been increasing their labor force
participation, while in the post-war period, women
married to working class husbands have decreased their
labor force participation, preferring the role of
full-time homemaker instead. However, during the
1970s, economic conditions worsened, which was a
factor in women's labor force participation (Bergmann
1986).
Ferree's (1976) review of the literature
indicates that many researchers have argued that women
married to working class husbands prefer to be full
time homemakers instead of being employed outside the
home. However, Ferree (1976) notes that it has not
been examined empirically. I expect women married to
working class husbands to be less likely to be
employed and also more traditional in sex role
attitudes.


27
On the other hand, Benenson (1985) argues and
presents evidence from 1977 that women married to high
status husbands, as measured by income, are less
likely to be employed than wives of lower status
husbands. I will not use income as a status variable,
since Mason et al. (1976) found an inconsistent
relationship between income and sex role attitudes.
Instead, I will use occupational prestige. I expect
that women married to higher status husbands will be
less traditional in their sex role attitudes.
However, I do not know if they are more likely to be
employed as well. I will have to explore this
possibility.
In summary, I expect women married to
husbands of the lower classes and of lesser education
to more likely be homemakers and also more traditional
in sex role attitudes than other women.
Other Background Characteristics
The literature indicates that the background
characteristics of age or cohort and race are
important in examining the relationship of employment
status to sex role attitudes.


28
Age/Cohort
Younger women are more likely to hold less
traditional sex role attitudes than older women
(Thornton and Freedman 1979). However, Mason et al.
(1976) found inconsistent results for age. They argue
that education may be the critical variable. Younger
women are better educated than older women (Blau and
Ferber 1985).
The literature indicates that older women are
less likely to be employed than younger women (Blau
and Ferber 1985).
In summary, I expect that the cohort of older
women are more likely to be homemakers than other
women and also more traditional in sex role attitudes.
It is reasonable to expect that homemakers
will become less traditional over time since younger
cohorts are likely to be less traditional in sex role
attitudes than older cohorts. Andersen and Cook
(1985) found that during 1972 and 1976 homemakers,
like employed women, were becoming less traditional in
their sex role attitudes, even though they were still
more traditional than employed women. This finding
needs to be confirmed, especially since it only covers
1972 and 1976.


29
Since society is becoming less traditional in
sex role attitudes, one also expects homemakers to be
a part of this trend, even though they may be more
traditional than other women. During the nineteenth
century, the ideal that women should be homemakers
arose among elite white women. However, this ideal
was also accepted by lower class immigrant women and
black women (Mullings 1986; Gutman 1976). In the
twentieth century, sex role attitudes became less
traditional. Initially, the change occurred with
better educated and younger women. It is likely that
these less traditional sex role attitudes will be
gradually accepted by homemakers as a result of the
times in which they live.
Evidence supports that society has changed in
the employment behavior of women (Blau and Ferber
1985). It is no longer the norm for women to be full-
time homemakers. Married women are no longer
interrupting their jobs for the rearing of small
children. About one-half of women in the age group
18-44, who had conceived a child in the preceding year
were participating in the labor force in 1985 (U.S.
Bureau of the Census 1986, cited in Rix 1987, p. 307).
Social scientists are arguing that the dual-earner
family has become the norm among married couples (Ross


30
et al. 1983). In 1900, only six percent of married
women worked (Oppenheimer 1970). As of April, 1986,
approximately seven out of ten women in the age group,
20-44, were in the labor force (U.S. Department of
Labor 1986, cited in Rix 1987, p. 302).
In summary, younger cohorts of women are more
likely to be employed and hold less traditional sex
role attitudes.
Race
Mason and Lu (1988) found that black men and
women in 1977 and 1985 were less traditional in sex
role attitudes than white men and women. Since the
historical conditions of blacks vary from whites,
researchers argue that what may be systematic for
whites, may not be systematic for blacks (Mason et al.
1976).
The literature indicates that there is a
relationship between race and occupational
development. White women are more likely than black
women to expect to be homemakers at age thirty
(Falkowski and Falk 1983).
I will omit black women from the analysis
since they are likely to be different from white women
and require a separate analysis.


31
Summary
The literature suggests that women with
various background characteristics hold less
traditional sex role attitudes than other women. In
addition, these background characteristics are likely
to be related to either occupational development or
employment status, which means that selectivity is
likely. It is reasonable to expect that the
relationship between employment status and sex role
attitudes will be reduced considerably if these
variables are controlled for.
I propose that variations in background
characteristics are related to sex role attitudes and
can account for the differences between employed women
and homemakers. I expect women with better educated
parents to be more likely than other women to be
employed and also less traditional in sex role
attitudes. I expect the same for better educated
women, women who are not Baptists, women with better
educated husbands, women with husbands who have higher
status, women with husbands in the professional or
managerial occupations, and women who are from younger
cohorts.


32
Specific Conditions of the Second Hypothesis
The following hypotheses are the specific
conditions of the second hypothesis of this study.
Employment Status
1. White homemakers are more traditional in sex role
attitudes than white employed women.
Parental Background
2. The greater the father's education, the less
likely white women will be homemakers, the more
likely white women will be less traditional in sex
role attitudes.
3. The greater the mother's education, the less
likely white women will be homemakers, the more
likely white women will be less traditional in sex
role attitudes.
Education
4. The greater the education, the less likely white
women will be homemakers, the more likely white
women will be less traditional in sex role
attitudes.
Religion of Youth (when controlling for education^
5. For white women who were raised in the Jewish
religion, the less likely white women will be
homemakers, the less traditional the sex role
attitudes.
6. For white women who were raised in the Baptist
denomination, the more likely white women will be
homemakers, the more traditional the sex role
attitudes.


33
Spouse's Background
7. The greater the husband's education, the less
likely white women will be homemakers, the more
likely white women will be less traditional in sex
role attitudes.
8. For white women married to middle class husbands,
the greater the likelihood of employment, the less
traditional the sex role attitudes (Specifically,
white women married to professionals and managers
will be less traditional than white women married
to men with other types of occupations).
9. For white women married to high status husbands,
the less traditional the sex role attitudes.
Other Background Characteristics
Age
10. The younger the age, the greater the education,
the more likely white women will be employed, and
will be less traditional in sex role attitudes.
Difference between Cohorts
11. When comparing white homemakers of the same age
group in 1975 with white homemakers in 1985, white
homemakers are becoming less traditional in sex
role attitudes over time.


CHAPTER IV
METHODS
In order to test the hypotheses, I will use
data gathered by the National Opinion Research Center
in conjunction with the General Social Survey (GSS),
which is an ongoing research undertaking. Successive
cross-sectional national probability samples of the
noninstutionalized population aged eighteen and older
in the continental United States are taken to survey
Americans on a variety of issues, which include sex
role attitudes.
Measurement of the Dependent Variable:
Sex Role Attitudes
The GSS has used four items which measure sex
role attitudes. These four items have been repeatedly
used across the years, 1974, 1975, 1977, 1978, 1982,
1985, and 1986. I will combine the data from these
years in order to test the relationship of sex role
attitudes to the variables: parents' education, age,
religious groups and denominations at age sixteen,


35
education, husband's education, husband's status,
husband's occupation, and employment status. I will
also combine the data from the years, 1972 to 1987, in
order to examine the relationship of employment status
to the background variables. I will only examine
white women for this study.
The sex role items present the respondent
with these questions:
1. Do you agree or disagree with this
statement: Women should take care of their
homes and leave running the country to men?
2. Do you approve or disapprove of a married woman
earning money in business or industry if she
has a husband capable of supporting her?
3. If your party nominated a woman for President,
would you vote for her if she were qualified?
4. Tell me if you agree or disagree with this
statement: Most men are better suited
emotionally for politics than are most women.
Those respondents who agree with the first
item, disapprove of the second item, give an answer of
no to the third item, or who agree with the last item
have given traditional answers. On the other hand,
those who have responded in an opposite manner have
given less traditional answers.
The four items are relevant to the dimensions
of sex roles. According to Cherlin and Walters
(1981), the first item and the last two items ask


36
about the political place of women. The first item
does have a political implication, but it also
measures the belief that men and women should have
separate spheres, with women's place being the home.
The last two items also examine this belief. The last
item is particularly relevant, because the belief that
men are better suited for their traditional domain of
politics, based on having a different emotional
nature, is measured. The second item measures the
attitude that a married woman should not make her own
income based on the condition of having an adequate
provider as a spouse. This of course also implies
separate domains for men and women. In addition, it
measures the traditional belief that married men and
women have specific obligations in marriage, namely,
the husband supports the wife, while the woman is not
employed earning income in business or industry.
The first two items and the last item of the
index appear to be valid on the face of it. The
second item could have been expanded to include
opinions about married women working under a variety
of conditions, especially when there is a presence of
young children, as Dowdall (1974) has done in her
study. However, this item, as well as the other


three, has the advantage of having been frequently
employed over a long period of time.
37
Despite the wide use of the third item, I
will have to discard it. This item on voting for a
woman President unfortunately adds the contingency
that the woman has to be qualified. It should have
asked respondents whether or not they believed men are
more qualified than women for politics, and only then
whether or not they would vote for a woman president.
A person who answers yes to this item, may not
necessarily think that women are qualified to be
President. It is thought that women often have to
prove that they are as qualified as men.
Unfortunately, the GSS has no indicators
asking respondents if they believe that a woman's
primary function should be as a mother. Also, there
are no indicators examining opinions regarding the
appropriateness of women in heavy industry versus
occupations, which are traditionally associated with
women, such as nursing and teaching. Also, there are
no measures of traditional sex role attitudes when
women do work. For instance, the GSS does not measure
people's opinions regarding status differences between
married men and women.


38
Another limitation of using data from the GSS
is that it is cross-sectional data. Andersen and Cook
(1985) examined the causal relationship between
employment status and sex role attitudes using
longitudinal data.
Despite these limitations, I will still
examine data from the GSS. For those desiring to
examine long periods of time, secondary data analysis
is required. I am particularly interested in the
trends of the last two decades, because I want to know
whether or not homemakers' sex role attitudes are
changing. In addition, other large data sets do not
have improved measures of sex role attitudes. Cherlin
and Walters (1981) indicate that the second and third
items are the most widely used of all measures in
studies examining change in sex role attitudes.
Barnartt and Harris (1981) have made a "sex
role ideology index" using these items. Other
researchers, including myself, have also employed all
four items in studies examining various issues
(Granberg and Granberg 1980; McCarty 1987). The
advantage of employing these items in an index is that
the items are interrelated (Granberg and Granberg
1980). According to Barnartt and Harris (1981), the
index has a reliability coefficient of 0.64 for 1974


39
and a reliability coefficient of 0.68 for 1977 using
Cronbach's alpha, which means that the index is highly
consistent over time.
I will construct an additive index employing
the three items. For those who give a traditional
response to an item, a score of one will be given. On
the other hand, for a less traditional response, a
score of two will be given. In order to tally a total
score for a respondent, I will add the scores of all
three items. The higher the score, the less
traditional the respondent. I will compare
differences among respondents in the percentages who
are least traditional in sex role attitudes, which
means that the respondents have given less traditional
responses to all three items.
Measurement of Independent
and Control Variables
In this section, I will discuss how I have
operationalized the independent and control variables.
I will be employing measures of the variables used
by the GSS.
Employment Status
The GSS asks the respondent what she was
doing the prior week. Respondents are classified in


40
one of the following categories: Working full time;
working part time; with a job, but not at work because
of temporary illness, vacation, or strike; unemployed,
layed off, looking for work; retired; in school;
keeping house; or other. I will collapse the
categories, working full time, part time, or with a
job, but not at work, into one category, employed.
Women who respond that they are keeping house will be
classified as homemakers. I expect homemakers to have
a lower percentage holding least traditional sex role
attitudes than the employed.
Parental Background
Father's and mother's education is measured
by the number of years of formal schooling that the
respondent's father or mother has completed. In order
to examine cross-tabulations, I will categorize
education in the following manner: 0-9th grade; 10th-
12th grade; 1 year of college to 3 years of college;
and four years or more of college. I expect that the
higher the category of education, the higher the
percentage holding least traditional sex role
attitudes.


41
Education
Education is measured in the same way as
reported in the discussion of parental background.
Religious Groups and Denominations
The GSS categorizes religious groups using
the following demarcations: Protestant, Catholic,
Jewish, none, other, or no answer. I will eliminate
those responding no answer as they do not represent
too many people. The category of "none" is
problematical and will be omitted from the analysis
(see Hyman 1972).
Protestant denominations are categorized in
the following manner: Baptist, Methodist, Lutheran,
Presbyterian, and Episcopal. Other denominations will
be omitted from the analysis.
The GSS has measures of religious groups and
denominations for respondents at both the time of the
interview and when they were sixteen. I will use the
latter measure. I expect women who were raised in the
Jewish faith to have a higher percentage holding least
traditional sex role attitudes than women raised in
other groups. In addition, I expect women raised in
the Baptist denomination to have a lower percentage


42
holding least traditional sex role attitudes than
other denominations.
Spouse's Background Characteristics
Husband/s Education. The GSS measure is the
same for husband's education as it is for respondent's
education. For the purpose of cross-tabulation/ I
will use the same categories as I suggested for
education. I expect that women married to husbands in
the higher educational categories will have a greater
percentage holding least traditional sex role
attitudes than women married to husbands with less
education.
Husband's Occupation. The GSS classifies
occupations using the U.S. Bureau of the Census 3
digit occupation classification for 1970 occupations.
The following categories are used by the GSS:
Professional, technical; managers and administrators,
sales workers; clerical and kindred workers; craftsmen
and kindred workers; operatives, except transport;
transport equipment operatives, laborers; farmers,
farm laborers, etc.; and service workers. I expect
that women married to professionals and managers will
have greater percentages holding least traditional sex


43
role attitudes than women married to men in other
occupations.
Husband's Occupational Prestige. The GSS
uses the 2 digit Hodge, Siegel, Rossi prestige score,
which is a widely used measure of occupational
prestige. I will use Richardsons (1979)
categorization of this variable: 0-29 (low); 30-49
(medium); and 50-89 (high). I expect women married to
men with high prestige to have a greater percentage
holding least traditional sex role attitudes than
women married to men of lesser prestige.
Other Background Characteristics
Age. The GSS records the date of birth and
then recodes it into actual age. For the purpose of
cross-tabulation, I will categorize age into the
following groups: 18-30 years; 31-45; 46-64; 65 and
older. I expect older women to be more traditional in
sex role attitudes than those in younger categories,
which means that the older category will have a lower
percentage holding least traditional sex role
attitudes than women in younger categories.


44
Cohort Analysis Over Time. Cohort analysis
will be performed on both white homemakers and white
employed women for the time period, 1975 to 1985. I
will combine data from 1975 with 1977 and data from
1985 and 1986, in order to have sufficient numbers for
analysis. I will use the following age categories:
18-29; 30-39; 40-49; 50-59; and 60 and over.
I prefer cohort analysis over trend analysis,
because the rate of change in sex role attitudes among
different cohorts should be examined.
Data Analysis
In examining the hypotheses of this proposal,
computer aided data analysis will be performed. I
will examine the percentages in the cross tabulations
between three and sometimes four variables. I will
use chi-square to test for significance. I will test
for significance at the 0.05 level. This statistic is
appropriate for nominal level variables, which are the
focus of this study.


CHAPTER V
RESULTS
In this section, an examination of whether or
not the hypotheses have been supported will be
presented. I will first examine the relationship of
sex role attitudes to employment status. In
subsequent sections, I will examine the relationship
of the background variables to employment status. I
will then examine the relationship of employment
status to sex role attitudes, controlling for the
effect of background variables which are related to
both sex role attitudes and employment status.
Finally, I will also present whether or not the effect
of employment status on sex role attitudes is reduced
by the introduction of the background variables.
Employment Status
In Table 1, White employed women are higher
than white homemakers in the percentage of women who
hold least traditional sex role attitudes, while
homemakers have higher percentages who hold the most
traditional sex role attitudes. In addition,


46
homemakers are more likely to be in the categories
between the most and the least traditional. My
findings are consistent with the literature.
TABLE 1
Sex Role Attitudes by Employment Status
(Reported in Percentages)* *
Employment Status
Sex Role Attitudes Employed Home- makers Unem- ployed Retired Students
Most
Traditional 6% 19% 5% 21% 1%
Between 14% 23% 14% 26% 5%
extremes 24% 24% 22% 24% 23%
Least
Traditional 56% 34% 59% 29% 71%
N 2286 1897 63 345 107
Note:
*p<.05.
Homemakers are more similar to retired women
than to employed women in the percentage holding least
traditional sex role attitudes. Employed women are
more similar to female students in the percentage
holding least traditional sex role attitudes.


47
Students have the highest percentage holding least
traditional sex role attitudes, while retired women
are the lowest.
Parental Background
I expect that the higher the parent*s
education, the less likely white women will be
homemakers and also the higher the percentage holding
least traditional sex role attitudes. The data from
Tables 2 and 3 show that the relationship between
employment status and parent's educational background
for white women is more complex than expected. There
is a steady increase in the percentage of employed and
a decrease in percentage who are homemakers up to and
including women with parents (either father or mother)
with educational backgrounds of 1-3 years of college.
However, for white women with parents with 4 years or
more of college, there is a surprising decrease in
employment and an increase in the percentage who are
homemakers.
In Tables 4 and 5, the percentages of white
women who hold least traditional sex role attitudes
are given by employment status, controlling for the
educational level of parents. A similar pattern
arises as in Tables 2 and 3, even though the dependent


48
TABLE 2
Employment Status of White Women
by Father's Education
(Reported in Percentages)*
Father #s Education
Employment Status 0-9th 10th- 12th 1-3 yrs. of coll. 4 yrs. or more
Employed 41% 58% 61% 41%
Homemaker 48% 34% 29% 44%
Other3 11% 8% 10% 15%
N 3621 2460 573 3530
Notes:
Retired, unemployed, and students.
*p<.05.


49
TABLE 3
Employment Status of White Women
by Mother's Education
(Reported in Percentages)* *
Mother's Education
Employment Status 0-9th 10th-12th 1-3 years of coll. 4 years or more
Employed 38% 56% 57% 39%
Homemaker 49% 35% 31% 46%
Other3 12% 9% 11% 16%
N 3592 3659 721 2212
Notes:
aRetired, unemployed, and students.
*p<.05.


50
TABLE 4
Percentage Holding Least Traditional
Sex Role Attitudes by Employment Status,
controlling for Father's Education
Father's Education
Employment Status 0-9th 10th-12th 1-3 yrs. coll. 4 yrs. or more
Employed 46% 62% 68% 58%
(722) (685) (182) (697)
Homemakers 28% 48% 57% 32%
(759) (374) (76) (688)
Difference 18% 14% 11% 26%
Note: All values p<.05, unless otherwise noted


51
TABLE 5
Percentage Holding Least Traditional Sex Role
Attitudes by Employment Status, controlling for
Mother's Education
Mother's Education
Employment Status 0-9th 10th-12th l-3yrs. coll. 4 yrs. or more
Employed 42% 63% 71% 56%
(656) (1019) (193) (418)
Homemakers 27% 46% 51% 29%
(771) (591) (101) (434)
Difference 15% 17% 20% 27%
Note: All values p<.05, unless otherwise noted.


52
variables are different. There is a steady increase
in the percentage holding least traditional sex role
attitudes up to and including women with parents who
have educational backgrounds of one to three years of
college. However, for women with parents with four
years or more of college, there is a decrease. This
occurs for both employed women and homemakers.
The difference between employed women and
homemakers in sex role attitudes remains for each
educational level. However, it clearly is affected by
parent's educational level. The pattern of
differences between employed women and homemakers
varies for both father's and mother's education.
For father's education, the difference
decreases consistently up to and including the
educational level, 1-3 years of college. At this
level, the difference has been reduced to its lowest
point. However, the pattern does not continue after
this level. For four years or more of college, there
is a large difference in percentage points.
Homemakers whose fathers have backgrounds of
four years or more of college are more similar in
percentage holding least traditional attitudes to
homemakers whose fathers have backgrounds of nine


53
years or less of school. This pattern also occurs for
mother's education.
For mother's education, the pattern of
differences in percentage holding least traditional
sex role attitudes between employed women and
homemakers is more consistent. As the mother's
educational level increases, the difference between
employed women and homemakers in sex role attitudes
increases. In addition, the difference between
employed women and homemakers is greater for women
whose mothers have educational backgrounds of 1-3
years of college than for women whose fathers have
comparable backgrounds.
Educational Background
Table 6 shows that the lower the educational
level, the lower the percentage employed and the
greater the percentage of homemakers. The
relationship between employment status and the
respondent's own education is as the literature leads
us to expect.


54
TABLE 6
Employment Status of White Women
by Respondent's Education
(Reported in Percentages)*
Respondent's Education
Employment Status 0-9th 10th- 12th 1-3 yrs. of coll. 4 yrs. or more of coll
Employed 20% 46% 56% 67%
Homemaker 63% 45% 30% 22%
Othera 17% 10% 14% 11%
N 2653 2803 2757 1971
Note:
aStudents, retirees , and unemployed.
*p<.05.


55
In Table 7, the data for the percentage
holding least traditional attitudes are presented. A
similar pattern emerges for both homemakers and
employed women, the higher the educational level, the
higher the percentage holding least traditional
attitudes. However, the differences between employed
women and homemakers are inconsistent.
TABLE 7
Percentage Holding Least Traditional Sex Role
Attitudes by Employment Status, controlling for
Respondent's Education
Respondent's Education
Employment 0-9th 10th-12th 1-3 yrs. 4 yrs.
Status coll. or more
Employed 32% 49% 64% 75%
(158) (1176) (474) (478)
Homemakers 15% 38% 41% 61%
(465) (1028) (262) (142)
Difference 17% 11% 23% 14%
Note: All values p<.05, unless otherwise noted.
I expect that the relationship between sex
role attitudes and employment status will be reduced
when controlling for education. As Table 7 shows, the
differences between employed white women and


56
homemakers in sex role attitudes varies by educational
level. However, it does not vary as I expect.
Different patterns emerge for women who have attained
the twelfth grade or less, when compared to women with
college backgrounds. The difference between employed
women and homemakers decreases as education increases,
only when women with college backgrounds are
differentiated from those without college. The
greatest differences occur for women with the
educational backgrounds, 0-9th and 1-3 years of
college, while considerably less differences occur for
women with the educational backgrounds, 10th-12th and
4 years or more of college.
Religious Background
I expect that women who were raised in the
Jewish religion to have a lesser likelihood of being
homemakers than women raised in either the Catholic or
Protestant religions. As Table 8 shows, this is only
partially supported. Although women who were raised
in the Jewish religion are less likely to be
homemakers than women raised in the Protestant
religion, women raised in the Catholic religion are
about the same as women raised in the Jewish religion.


57
TABLE 8
Employment Status of White Women
by Religious Group of Youth
(Reported in Percentages)*
Religious Group
Employment Status Protestant Catholic Jew
Employed 45% 51% 52%
Homemaker 43% 38% 37%
Other3 12% 11% 11%
N 6085 2798 242
Note:
£ Students, retirees, and unemployed.
*p<.05.
The findings presented in Table 9 support
that women raised in the Jewish religion hold the
least traditional sex role attitudes of the three
religious backgrounds. In addition, although the
difference between employed women and homemakers
persists, it is reduced considerably. The differences
for both Catholic and Protestant groups are about the
same, approximately double the difference for the
Jewish group. However, Catholic homemakers have a
higher percentage than Protestant homemakers who hold


58
least traditional sex role attitudes. This also
occurs for employed women.
TABLE 9
Percentage Holding Least Traditional Sex Role
Attitudes by Employment Status, controlling for
Religious Group of Youth
Religious Group
Employment Status Protestant Catholic Jew
Employed 53% 59% 79%a
(1404) (730) (58)
Homemakers 31% 39% 69%a
(1263) (512) (49)
Difference 22% 20% 10%
Note: All values p<.05, unless otherwise noted.
p>.05insignificant.
In Tables 10 and 11, education is
controlled for when examining the relationship of
religion to sex role attitudes. There are not enough
individuals raised in the Jewish religion to conduct
an analysis. The data in Table 10 show that
homemakers who were raised as Catholics have higher
percentages holding least traditional sex role
attitudes than Protestants at each educational level.


59
For both Protestants and Catholics, as education
increases, the percentage of homemakers who are least
traditional in sex role attitudes also increases.
However, for Protestants, there is no difference in
percentage between homemakers in the educational
groups, 10th-12th grades and 1-3 years of college.
TABLE 10
Percentage Holding Least Traditional Sex Role
Attitudes by Religion at age 16, controlling for
Education for Women who are Homemakers
Respondent's Education
Religion 0-9th 10th-12th 1-3 yrs. of coll. 4 yrs. or more
Protestant 15% 34% 34% 52%
(319) (682) (176) (86)
Catholic 17%a 43% 49%a 64%a
(125) (290) (61) (36)
Jew 25%b 67%a 57%a ioo%a
(2) (18) (14) (13)
Note: In comparing employed to homemakers, all values
p<.05, unless otherwise noted.
ap>.05.
Statistics can't be calculated.
In Table 11, the same relationship discussed
in Table 10 is presented for employed women. Except


60
for the lowest educational group, 0-9th, Catholics
have higher percentages holding least traditional
attitudes for every educational level. The
differences between Catholics and Protestants are not
as pronounced for employed women as for homemakers.
TABLE 11
Percentage Holding Least Traditional Sex Role
Attitudes by Religion at age 16, controlling for
Education for Women who are Employed
Respondent's Education
Religion 0-9th 10th-12th 1-3 yrs. 4 yrs.
of coll. or more
Protestant 34% 45% 63% 71%
(106) (720) (273) (305)
Catholic 26% 54% 64% 80%
(46) (394) (167) (123)
Jew 25% 71% 71% 87%
(4) (14) (14) (30)
In Table 12, the findings of the percentage
differences occurring when comparing Table 11 to
Table 10 are presented. I have to omit women with
Jewish backgrounds from the analysis, because of
insufficient numbers. Table 12 shows that the


61
Protestant group follows the pattern found for women
in general when controlling for education. The
TABLE 12
Difference between White Employed Women
and White Homemakers in Percentage Holding Least
Traditional Sex Role Attitudes by Religion at
Age 16, controlling for Respondent's Education
Respondent's Education
Religion 0-9th 10th-12th 1-3 yrs. of coll. 4 yrs. or more
Protestant 19 11 29 19
Catholic 9 11 15 16
Jew ( -25) (4) (14) (-13)
Note: Parentheses refer to insufficient numbers.
educational backgrounds, 0-9th and 1-3 years of
college have the highest percentage differences. The
groups that follow them, 10th-12th and 4 years or
more, decrease in percentage differences. However,
the 0-9th group has the same percentage difference as
the group, 4 years or more. For Catholics, the
differences increase as education increases. For
every educational level, except 10th-12th, the
difference between homemakers and employed women are
less for Catholics than Protestants.


62
In comparing Table 12 with Table 9, it
appears that the introduction of education has reduced
the relationship between employment status and sex
role attitudes. The differences between employed
women and homemakers raised in the Catholic religion
have been reduced considerably. For Protestants, the
difference is reduced mainly for women with the
educational background, 10th-12th.
The findings of Table 13 do not support my
hypothesis that women raised in the Baptist
denomination are more likely to be homemakers than
women raised in other denominations. Table 13 shows
that women raised in either the Presbyterian
denomination or the Episcopalian denomination are less
likely to be homemakers than women raised in other
denominations. However, Baptists are similar to
Methodists and Lutherans in the likelihood of being
homemakers.
My expectation that women who were raised in
the Baptist denomination are more traditional in sex
role attitudes than women raised in other
denominations is supported in Table 14. The
percentage who hold least traditional attitudes is
consistently lower than the percentage for other
denominations for both homemakers and employed women.


63
TABLE 13
Employment Status of White Women by
Denomination of Youth
(Reported in Percentages)* *
Denomination
Employment Baptist Metho- Lutheran Pres- Episco
Status dist byterian palian
Employed 46% 42% 48% 46% 49%
Homemaker 43% 44% 42% 37% 35%
Othera 11% 14% 10% 17% 16%
N 1723 1350 870 494 242
Notes:
aStudents, retirees, and unemployed.
*p<.05.


64
The differences between homemakers and employed women
are consistently high for each denomination.
TABLE 14
Percentage Holding Least Traditional Sex Role
Attitudes by Employment Status, controlling for
Denomination of Youth
Denomination
Employment Baptist Lutheran Metho- Pres- Episco
Status dist byterian palian
Employed 47% 52% 56% 60% 72%
(410) (224) (297) (109) (57)
Homemakers 26% 28% 36% 28% 47%
(370) (190) (282) (87) (38)
Difference 21% 24% 20% 32% 25%
Note: All values p<.05, unless otherwise noted.
Tables 15, 16, and 17 present the findings
of the relationship between denomination of one's
youth and sex role attitudes, when controlling for
education. It is very difficult to conduct this
analysis, because the number of cases are reduced by
the addition of the control variable. The data for
employed women in Table 15 show that generally as
education increases, so does the percentage who hold
least traditonal attitudes. In Table 16, the data


65
TABLE 15
Percentage Holding Least Traditional
Sex Role Attitudes by Denomination of Youth,
controlling for Respondent's Education
for White Women who are Employed
Respondent's Education
Denomi- 0-9th 10th-12th 1-3 yrs. 4 yrs.
nation of coll. or more
Baptist 35% 41% 64% 64%a
(48) (239) (64) (59)
Methodist 21%a 49% 66% 66%a
(14) (145) (68) (70)
Lutheran 43%a 41%a 54%a 81%
(14) (117) (46) (47)
Presby- 50%a 49%a 55% 76%
terian (2) (49) (20) (38)
Epsico- 50%b 57% 67%a 80%a
palian (3) (14) (15) (25)
Note: In comparing employed to homemakers, all
values pc. 05, unless otherwise noted.
ap>.05.
DChi-square can't be calculated.


66
TABLE 16
Percentage Holding Least Traditional Sex
Role Attitudes by Denomination of Youth,
controlling for Respondent's Education
for White Women who are Homemakers
Respondent's Education
Denomi- nation 0-9th 10th-12th 1-3 yrs of coll . 4 yrs. or more
Baptist 15% 32% 23% 50%
(123) (209) (30) (8)
Methodist 13% 43% 35% 44%
(61) (150) (46) (25)
Lutheran 12% 29% 41% 50%
(42) (113) (27) (8)
Presby- 27% 29% 22% 31%
terian (11) (45) (27) (13)
Episco 67% 15% 56% 83%
-palian (3) (13) (16) (6)


67
for homemakers show this pattern for only the
Lutherans (I omit Presbyterians and Episcopalians from
the analysis because of their low numbers). The
patterns of homemakers raised in the Baptist or
Methodist denominations are inconsistent. However,
this may be due to the lower number of cases in the
higher educational groups.
The differences between homemakers and
employed women are presented in Table 17. The
Baptists and Methodists appear to have greater
differences with the higher educational groups (1-3
years of college), while the Lutherans and
Presbyterians do not appear to have much variation
within each group, if one omits the cells with a low
number of cases. For specific educational groups, the
differences appear to be greatest for Baptists and
Methodists in the educational group, 1-3 years of
college, when compared to Lutherans and Presbyterians
of the same educational group. On the other hand, for
the educational group, 10th-12th, the greatest
difference occurs for Presbyterians. The other
denominations have much less differences for this
educational group.


68
TABLE 17
Difference between White Employed Women and White
Homemakers in Percentage Holding Least Traditional
Sex Role Attitudes by Denomination,
controlling for Respondent's Education
Respondent's Education
Denomination 0-9th 10th-12th 1-3 yrs. of coll. 4 yrs. or more
Baptist 20 9 41 (14)
Methodist (8) 6 31 (16)
Lutheran (31) 12 13 (31)
Presbyterian (23) 20 23 (45)
Episcopalian (-17) (42) (11) (-3)
Note: Parentheses refer to insufficient numbers.
Spouse's Background
The data of Table 18 show that as a spouse's
education increases, white women are less likely to be
homemakers. This supports the first part of my
hypothesis regarding the effect of spouse's education
on both employment status and sex role attitudes.


69
TABLE 18
Employment Status of White Women
by Spouse's Education
(Reported in Percentages)* *
Spouse's Education
Employment Status 0-9th 10th-12th 1-3 yrs. of coll. 4 yrs. or more of coll.
Employed 30% 44% 51% 50%
Homemakers 62% 52% 45% 32%
Othera 8% 4% 4% 18%
N 1164 2723 1077 5220
Note:
aStudents, retirees, and unemployed.
*p<.05.
As Table 19 suggests, the second part of my
hypothesis has not been supported. As the spouse's
education increases, the increase in those who are
least traditional in sex role attitudes occurs up to
and including women married to husbands with the
educational background of 1-3 years of college.
However, for women married to husbands with 4 years or
more of college, there is a decrease in percentage who
are least traditional.


70
Although this pattern holds for both employed
women and homemakers, the differences between employed
women and homemakers increase as spouse's education
increases, which is similar to the pattern occurring
for mother's education.
TABLE 19
Percentage Holding Least Traditional Sex Role
Attitudes by Employment Status, controlling
for Spouse's Education
Spouse's Education
Employment Status 0-9th 10th-12th l-3yrs. coll. 4 yrs. or more
Employed 32% 51% 62% 60%
(142) (584) (261) (1299)
Homemakers 20% 39% 41% 34%
(296) (592) (222) (787)
Difference 12% 12% 21% 26%
Note: All values pc.05, unless otherwise noted.
I expect that women married to professionals
and managers will less likely be homemakers than other
women. Table 20 indicates that although these women
generally have lower percentages of homemakers, there
are some exceptions. Women married to clerical
workers, service workers, and protective service


71
TABLE 20
Employment Status of White Women by
Spouse's Occupation
(Reported in Percentages)* *
Employment Status
Spouse's Employed Occupation Homemakers Othera N
Professional 51% 44% 5% 913
Manager/Sales 46% 49% 5% 1457
Clerical 51% 44% 6% 322
Craftsman 43% 53% 4% 1600
Operative (except transport) 40% 53% 4% 688
Transport operative 41% 55% 4% 341
Laborer (except farm) 38% 59% 3% 272
Farmer/ farm manager 25% 70% 6% 213
Farm laborer 25% 67% 8% 48
Service worker 46% 48% 6% 206
Protective service worker 45% 49% 6% 128
Private household worker 0% 100% 0% 1
Notes:
Students, retirees, and unemployed.
*p<.05.


72
workers have comparable percentages who are
homemakers. However, women married to professionals
and managers are less likely to be homemakers than for
many women married to men in blue collar occupations
or who are farmers.
I expect that women married to professionals
and managers have higher percentages holding least
traditional attitudes than other women. In addition,
I expect that the relationship between employment
status and sex role attitudes will be reduced when
controlling for spouse's occupation. The data
presented in Table 21 indicate that although employed
women and homemakers married to professionals and
managers are less traditional when compared to women
with the same employment status, the differences
between employed women and homemakers vary
considerably. When comparing Table 21 with Table 1,
the relationship between employment status and sex
role attitudes is affected by the introduction of this
variable.
I do not have a specific hypothesis regarding
the relationship between employment status and
spouse's prestige, because the literature is
inconsistent in this regard. Table 22 shows that the
relationship between employment status and spouse's


73
TABLE 21
Percentage Holding Least Traditional Sex Role
Attitudes by Spouse's Occupation, controlling
for Employment Status
Employment Status
Spouse's Occupation Employed Homemaker Difference
Professionals 63% 46% 17%
(213) (193)
Managers/Sales 60% 43% 17%
(327) (320)
Clerical 55% 31% 24%
(73) (52)
Craftsman 50% 38% 12%
(334) (359)
Operat ive (except 45% 26% 19%
transport) (1221 (171)
Transport 51%& 38% 13%
operatives (77) (79)
Laborer (except 54% 37% 17%
farm) (52) (68)
Farm/farm manager 30%a 23% 7%
(27) g (53)
Farm Laborer 40%a 63% -23%
(5) a (8)
Service Worker 36% 29% 7%
(33) (41)
Protective Service 58% 50% 8%
Worker (31) (18)
Private Household
Worker
Note: All values p<.05, unless otherwise noted.
p>.05insignificant.


74
occupational prestige is not fully consistent. Women
married to husbands with either low prestige or medium
prestige have similar patterns in percentages of
employed women or homemakers. The greatest contrast
is for women married to husbands with high prestige.
This group of women is less likely to be homemakers
and more likely to be employed.
TABLE 22
Employment Status of White Women by Spouse's
Occupational Prestige
(Reported in Percentages)*
Spouse's Prestige
Employment Status Low Medium High
Employed 41% 43% 48%
Homemakers 54% 55% 47%
Other3 5% 4% 6%
N 1055 3330 1804
Note:
aStudents, retirees, and unemployed.
*p<.05.
My expectation that as spouse's prestige
increases, women will have a higher percentage holding


75
least traditional sex role attitudes is supported for
both employed women and homemakers as Table 23 shows.
However, the differences between employed women and
homemakers remain. They also do not follow a
consistent pattern. The differences are less for
those married to husbands with either low or high
status. In comparing Table 23 with Table 1, which
examines the relationship of sex role attitudes to
employment status without the introduction of any
other variables, it appears that the introduction of
spouse's prestige has reduced this relationship.
TABLE 23
Percentage Holding Least Traditional Sex Role
Attitudes by Employment Status controlling for
Spouse's Occupational Prestige
Spouse's Prestige
Employment Status Low Medium High
Employed 44% 54% 59%
(194) (691) (409)
Homemaker 33% 36% 45%
(223) (740) (399)
Difference 11% 18% 14%
Note: All values p<.05, unless otherwise noted.


76
Aae/Cohort
My expectation that as age increases, the
likelihood of being a homemaker increases is supported
by the findings of Table 24.
TABLE 24
Employment Status by Age
(Reported in Percentages)* *
Age
Employment Status 18-30 31-45 46-64 65 and over
Employed 56% 59% 48% 10%
Homemakers 34% 38% 45% 56%
Unemployed 2% 1% 1% 0%
Retired 0% 0% 3% 32%
Students 7% 1% 1% 0%
N 2653 2803 2757 1971
Note:
*p<.05.
Education qualifies the finding concerning
the relationship between employment status and age.
For every educational background, except 0-9th, the
likelihood of homemaking increases with age (see


77
Table 1 in Appendix A). For women with the
educational background, 0-9th, homemaking is highest
for those between eighteen and thirty, which may be
due to considerations related to childbearing and
childrearing. The likelihood of being a homemaker
appears to decrease with age for women with this
educational background, except for a slight increase
for those 65 and over.
Table 25 supports that as age increases,
those who hold least traditional sex role attitudes
decreases. This pattern is consistent for both
employed women and homemakers. The difference in
percentage holding least traditional sex role
attitudes between employed and homemakers decreases as
age increases.
Table 26 (see Appendix A for Tables 2
and 3 which show the percentages holding least
traditional sex role attitudes for both employed women
and homemakers) shows that the differences in
percentage holding least traditional attitudes between
employed women and homemakers vary by age group, when
controlling for the respondent's education. The
differences for those 65 and over are fairly even,
while those for the group, 46-64, follow an


78
TABLE 25
Percentage Holding Least Traditional Sex Role
Attitudes by Employment Status, controlling
for Age
Age
Employment 18-30 31-45 46-64 65
Status and
over
Employed 65% 61% 42% 29%
(755) (832) (625) (79)
Homemakers 50% 44% 30% 19%
(497) (466) (516) (521)
Difference 15% 17% 12% 10%
Note: All values p<.05, unless otherwise noted.


79
TABLE 26
Difference in Percentage Holding Least
Traditional Sex Role Attitudes
between White Employed Women and
White Homemakers by Age, controlling
for Respondent's Education
Respondent's Education
Age 0-9th 10th-12th 1-3 yrs. of coll. 4 yrs. or more
18-30 17% 7% 25% 12%
31-45 26% 10% 13% 7%
46-64 11% 6% 12% 17%
65 and over 4% 4% 8% 6%
Note: All values pc.05, unless otherwise noted.


80
inconsistent pattern. The pattern for the young is
consistent with my findings when controlling for
education. Women with the educational backgrounds, 0-
9th and 1-3 years of college have the highest
differences, while the groups, 10th-12th and 4 years
or more of college, have much smaller differences.
My expectation that more recent cohorts of
white women will less likely be homemakers is
supported by the findings in Table 27. This is
supported for every cohort group.
I expect that more recent cohorts of
homemakers will have a higher percentage than cohorts
from the mid 1970s who hold least traditional sex role
attitudes. As Table 28 shows this is supported for
every cohort group.
However, the differences between employed
women and homemakers remain for both sets of years.
When comparing the differences between employed women
and homemakers in 1975 and 1977 with the differences
for 1985 and 1986, the most significant change in
differences occurs for the cohorts, 50-59 and 60 and
over. The gap between homemakers and employed women
for these age groups in 1975 and 1977 was not very
significant. However, in 1985 and 1986, the


81
TABLE 27
Percentage of White Women who are Homemakers
by Cohort, controlling for Year
(Reported in Percentages)
Year
Cohort 1975,1977 1985.1986
18-29 38% 23%
(359) (304)
30-39 41% 27%
(262) (320)
40-49 45% 29%
(235) (216)
50-59 42%a 36%a
(245) (177)
60 and over 62% 51%
(349) (428)
Note: When comparing differences in employment
status by year, all values p<.05, unless otherwise
noted.
a p>.05.


82
TABLE 28
Percentage of White Employed Women and Homemakers
Holding Least Traditional Sex Role Attitudes
by Cohort, controlling for Year
Cohort Employed Homemakers
Year
1975,1977 1985,1986 1975,77 1985,86
18-29 60% 69%a 50% 58%a
(187) (185) (124) (64)
30-39 55% 66%a 38% 52%a
(140) (220) (104) (83)
40-49 49% 60% 38% 49%a
(117) (139) (93) (57)
50-59 25% 53% 28% 35%a
(132) (100) (94) (57)
60 and over 19% 39% 15% 19%
(57) (41) (203) (195)
Note: All values p<.05, unless otherwise noted,
difference between 1975, 1977 and 1985, 1986 for
women of the same employment status is statistically
insignificant, p>.05.


83
differences are greatest for these groups when
compared to other groups.
The differences between employed women and
homemakers in percentages holding least traditional
sex role attitudes when controlling for cohort are
considerably less for most cohorts during either set
of years than for the differences occurring in Table 1
between employed women and homemakers when this
control is not introduced.
Summary
I will now present a summary of my findings.
As the literature leads us to expect, employed women
have a higher percentage than homemakers who hold
least traditional sex role attitudes, which is
consistent with the literature.
For white women, the higher the father's
education, up to and including 1 to 3 years of
college, the higher the percentage employed and the
lower the percentage who are homemakers. However, for
women with high status fathers, who have 4 years or
more of college, there is a decrease in percent
employed and an increase in percent who are
homemakers.


84
For both employed women and homemakers, as
father's education increases, up to and including 1 to
3 years of college, the percentage holding least
traditional sex role attitudes increases. For women
with fathers who have 4 years or more of college,
there is a decrease in the percentage holding least
traditional sex role attitudes.
The same pattern occurring for father's
education also occurs for mother's education.
For women, as education increases, the
percentage employed increases and the percentage who
are homemakers decreases.
For both employed women and homemakers, the
higher the education, the higher the percentage who
hold least traditional sex role attitudes.
Jewish women have a higher percentage who are
employed (and a lower percentage who are homemakers)
than do Catholics and Protestants. Catholics are more
similar to Jewish women than Protestant women.
Jewish women have a higher percentage than
Catholics or Protestants who hold least traditional
sex role attitudes. For both employed women and
homemakers, Protestants have the lowest percentage
holding least traditional sex role attitudes.


85
For employed women of all three religious
backgrounds, as education increases, the percentage
holding least traditional sex role attitudes
increases. For homemakers, this pattern holds for
Catholics. For Protestant homemakers, there is also
an increase, but those with educational backgrounds of
either 10th-12th or 1 to 3 years of college have the
same percentage. Low numbers for Jewish women makes
analysis problematical for this group.
Women raised in either the Presbyterian or
the Episcopalian denominations have a lower percentage
who are homemakers. Episcopalians and Lutherans have
the highest percentage employed. Presbyterians and
Baptists follow. Methodists have the highest
percentage who are homemakers and the lowest
percentage employed. Baptists, Methodists, and
Lutherans are fairly similar in percentage who are
homemakers.
For both employed women and homemakers,
Episcopalians have the highest percentage who hold
least traditional sex role attitudes. Among employed
women, there is a steady increase in percentage
holding least traditional sex role attitudes in this
order: Baptist, Lutheran, Methodist, Presbyterian,


86
and Episcopalian. For homemakers, the pattern is
dissimilar.
For employed women of all denominations, as
education increases, percentage holding least
traditional sex role attitudes also increases. For
homemakers, this pattern occurs only for Lutherans.
As spouse's education increases, the
percentage who are homemakers decreases, while the
percentage employed increases. However, there is a
slight decrease for women married to men with 4 years
or more of college, but this group is higher in the
category which includes students.
For both employed women and homemakers, as
spouse's education increases, the percentage holding
least traditional sex role attitudes increases.
However, there is a slight decrease for employed women
married to men with 4 years or more of college and a
larger decrease for homemakers of this background.
Women married to professionals and managers
have a higher percentage who are employed than do
women married to craftsmen, operatives, transport
operatives, laborers, farmers, and farm laborers.
Women married to professionals and clerical workers
have the highest percentage employed and the lowest
percentage of homemakers. Women married to managers


are very similar to service workers and protective
service workers.
87
For both employed women and homemakers, women
married to professionals and managers have a higher
percentage holding least traditional sex role
attitudes than do women married to men in working
class occupations, with the exception of protective
service workers and the omission of farm laborers due
to insufficient numbers.
Women married to husbands with high prestige
have a higher percentage employed, and a lower
percentage who are homemakers than women married to
men with lower prestige.
As spouse's prestige increases, the
percentage holding least traditional attitudes
increases.
As age increases, percentage employed
decreases.
As age increases, percentage holding least
traditional sex role attitudes decreases for both
employed women and homemakers.
Women in 1985 and 1986, who are in the same
age group as women in 1975 and 1977, have a lower
percentage who are homemakers than women of 1975 and
1977.


88
For both employed women and homemakers, women
in 1985 and 1986, who are in the same age categories
as women in 1975 and 1977, have a higher percentage
holding least traditional attitudes.
Complex patterns of differences between
employed women and homemakers in percentages holding
least traditional sex role attitudes emerge when
control variables are introduced.


CHAPTER VI
DISCUSSION
Differences between employed women and
homemakers are not eliminated by the introduction of
the control variables. The idea that both employment
status and sex role attitudes are the result of
various backgrounds has been supported. However, even
when homemakers and employed women have similar
backgrounds, the differences between the two in the
percentages holding least traditional sex role
attitudes remain.
The introduction of women's age considerably
reduces the relationship between employment status and
sex role attitudes. The reduction in differences
between employed women and homemakers occurring for
older women is not an effect of age. Instead, it
represents women who have been socialized earlier,
during a period more traditional in sex role attitudes
than present times.
Religion also reduces the effect of
employment status on sex role attitudes, especially
for women with Jewish backgrounds. For women with


90
Catholic backgrounds, the introduction of education
also reduces this relationship. This effect appears
to be the result of the combined effect of education
with Catholic background, because this does not occur
for women with Protestant backgrounds.
Father's education also affects the
relationship between employment status and sex role
attitudes. The effect of father's education suggests
that the historical patterns of family structure
should be controlled for. The decrease in percentage
holding least traditional sex role attitudes for both
employed women and homemakers whose fathers have
attained four years or more of college suggests that
the influence of high status fathers, who may have
been more likely to have traditional family
structures, may be operating here.
The pattern of differences between employed
women and homemakers for spouse's education and
mother's education are different than for father's
education. When either spouse's education or mother's
education increases, the differences between employed
women and homemakers increase. On the other hand, as
father's education increases up to and including the
background level of one to three years of college, the
differences decrease.