Citation
Division of labor and dual-earner families

Material Information

Title:
Division of labor and dual-earner families effects of equity and subjective perceptions of fairness on spousal well-being
Portion of title:
Effects of equity and subjective perceptions of fairness on spousal well-being
Creator:
Alerta, Norman Suarez
Publication Date:
Language:
English
Physical Description:
ix, 75 leaves : illustrations ; 29 cm

Thesis/Dissertation Information

Degree:
Master's ( Master of arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Sociology, CU Denver
Degree Disciplines:
Sociology

Subjects

Subjects / Keywords:
Dual-career families ( lcsh )
Home economics ( lcsh )
Dual-career families ( fast )
Home economics ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 72-75).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Sociology.
General Note:
Department of Sociology
Statement of Responsibility:
by Norman Suarez Alerta.

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:
39719155 ( OCLC )
ocm39719155
Classification:
LD1190.L66 1997m .A44 ( lcc )

Full Text
DIVISION OF LABOR AND DUAL-EARNER FAMILIES:
EFFECTS OF EQUITY AND SUBJECTIVE PERCEPTIONS OF FAIRNESS
ON SPOUSAL WELL-BEING
by
Norman Suarez Alerta
B.A., University of Colorado at Denver, 1995
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Sociology
1997


This thesis for the Master of Arts
degree by
Norman Suarez Alerta
has been approved
by
Date


Alerta, Norman Suarez (M.A., Sociology)
Division of Labor and Dual-Earner Families: Effects of Objective Equity and
Subjective Perceptions of Fairness on Spousal Well-Being
Thesis Directed by Assistant Professor Candan Duran-Aydintug

Abstract
In looking at the ever-changing dynamics and processes of American
households, this study expands previous research on equity in dual-earner
families. Using data from the 1987 National Survey of Families and
Households (NSFH), comparisons are made between models of objective
reality and subjective perceptions of fairness. This study looks solely at dual-
earner families and how much input either spouse contributes towards the
household and whether or not they perceive fairness of housework based on
the objective reality of equitable input. Also, this study proposes an effect of
either the objective reality of equity and the subjective perceptions of fairness
on the well-being of either spouse.
Input variables include indicators such as hours spent performing
housework, hours spent at work (full- or part-time), income from employment
in full-or part-time jobs and other factors. All these indicators of inputs for
the relationship are accounted for in a model depicting objective realities of
equity. From the NSFH data set, questions related to input values in and out
of the household are used. The data set also contains measures of
m


subjective perceptions of fairness as well as well-being, which is determined
by indicators of physical health, self-esteem, and happiness.
The results of the study indicate no effects between objective equity
and perceptions of fairness in housework for either spouse. Well-being is
affected to some degree for wives based on a positive perception of
housework fairness. The more wives perceive housework to be fair, the more
likely are they to report good health and overall happiness. Other factors
were found to contribute to perceptions of fairness however, particularly in
regard to beliefs in housework sharing held by both spouses. Husbands who
believed working spouses should share housework felt housework was fair to
both. Meanwhile, wives who felt the same way perceived housework as
unfair. Also significant was the wives amount of hours contributed to
housework. The less housework wives reported performing, the more they
saw housework as being fair. They saw housework as fair if they reported
their husbands as doing more housework.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
IV


DEDICATION
This research is dedicated to the memory of my parents,
Carl P. Alerta and Estela S. Alerta.
Their respect for education has instilled
within me a drive for higher goals.
I can only hope to make them proud.
To my big brother Carl and my sisters Beverly, Helene, and Jennifer;
though far away, they have remained close in my heart.


ACKNOWLEDGMENTS
I extend much gratitude to the people who have helped me along the
way. My grateful thanks to the Students of Sociology club and MJ Lane for
easing the burden of working with secondary data sets. Thanks also to the
Department of Scoiology for keeping me in their wings and preparing me for
higher challenges. Special thanks to the Little Shanghai Cafe and the Lei
family who kept me afloat throughout my college years by making me a part
of the family. Very special thanks to Jeannie whose push and shove was
tempered by love. Most especially, though, I acknowledge the energy and
spirit of Dr. Candan Duran-Aydintug whose belief in me sparked an academic
fire that completely changed my life.


CONTENTS
Tables........................................................ix
CHAPTER
1. INTRODUCTION............................................ 1
2. REVIEW OF THE LITERATURE.................................7
Theoretical Background..............................7
Previous Studies...................................12
Purpose of the Study...............................28
Hypotheses...................................29
3. METHODS.................................................31
Sample.............................................31
Sample Characteristics.............................34
Measurement........................................36
4. RESULTS.................................................48
5. DISCUSSION..............................................63
Limitations and Suggestions........................70
vii


REFERENCES


TABLES
Table
4.1. Coefficients for Full Logistic Regression Models of
Objective Equity and Perceptions of Fairness.....................49
4.2. Coefficients for Logistic Regression Model of Objective Equity and
Perceptions of Fairness for Husband
(Housework and Equity Variables Only).............................52
4.3 Coefficients for Logistic Regression Model of Objective Equity and
Perceptions of Fairness for Wife
(Housework and Equity Variables Only).............................54
4.4 Coefficients for Full Logistic Regression Model of Indicators of
Husbands Well-Being..............................................56
4.5 Coefficients for Full Logistic Regression Model of Indicators of
Wives Well-Being................................................ 58
4.6 Coefficients for Logistic Regression Model of Indicators of Husbands
Well-Being (Housework and Equity Variables Only)..................60
4.7 Coefficients for Logistic Regression Model of Indicators of Wives
Well-Being (Housework and Equity Variables Only)..................61
IX


CHAPTER 1
INTRODUCTION
For family researchers, household labor remains a principal focal point
of inquiry. The reasons for such may vary, but the fact that women regularly
do more housework than men (Presser, 1994; Rogers, 1996; South & Spitze,
1994; Thompson & Walker, 1989) suggests theoretical applications of
sociological concepts such as power, expectations, exchange and equity
(Blair & Johnson, 1992; Demaris & Longmore, 1996; Greenstein, 1996; Hiller
& Philliber, 1986; Kollock, Blumstein & Schwartz, 1994; Mirowsky, 1985;
South & Spitze, 1994; Stafford, Backman & DiBona, 1977). From these
concepts, past research has developed various assumptions about
household labor and examined its relationship with various family structures
and dynamics. Remarried families, dual-earner families, cohabiting couples,
mothers work-hours, mens contributions, and the second shift are just a
few examples of structures and dynamics that have been viewed through the
lenses of household labor (Benin & Agostinelli, 1988; Coltrane & Ishii-Kuntz,
1992; Ferree, 1991; Greenstein, 1996; Hochschild SMachung, 1989; Ishii-
Kuntz & Coltrane, 1992; Mederer, 1993; Rogers, 1996; Presser, 1994; South
l


& Spitze, 1994; Thompson, 1991). In most cases, the use of some
theoretical premise provides the framework from which logical hypotheses
are derived and tested.
One prevalent theoretical approach for household labor research is
equity (Blair & Johnson, 1992; Demaris & Longmore, 1996; Greenstein,
1996; Kollock, Blumstein & Schwartz, 1994; Lennon & Rosenfield, 1994;
Thompson, 1991). Equity theory suggests that, in an interpersonal
relationship, each individuals outcomes (what they get out of a relationship)
must be proportional to their inputs (what they put into the relationship) in
order for the relationship to be just or fair (DeMaris & Longmore, 1996;
Hatfield,et.al., 1984; Walster, Berschied & Walster, 1976). Equity theory
finds its basis in exchange theory which assumes that participants seek
satisfaction from an interpersonal relationship through the constant giving
and taking of tangible and intangible rewards (Homans, 1961; Thibault &
Kelly, 1959).
Proponents of equity theory suggest that a relationships equity can
be gauged through the use of a ratio formula model (Adams, 1965; Harris,
1983; Hatfield et al., 1984; Walster, Berschied & Walster; 1976). In most
models, each persons ratio of inputs and outcomes are compared with one
another. If the ratios are equal, the relationship reflects equity, and
2


conditions are presumed to be just or fair for the individuals involved. If not,
then the relationship is unbalanced and inequity exists for its participants
(Adams, 1965; Harris, 1983; Hatfield, Traupman, Sprecher, Utne & Hay.,
1984; Walster, Berschied & Walster, 1976). A notable dilemma for some
researchers involves the determination of what constitutes an input and an
outcome. For family researchers, quantifying household labor into a value,
such as hours spent per week on specific tasks, provides an input for use in
an objective equity formula (DeMaris & Longmore, 1996, Kollock, Blumstein
& Schwartz, 1994). Often this formula also includes individual income,
education, and/or physical health as additional input and outcome values
(DeMaris & Longmore, 1996). For intimate relationships like a marriage, it
has been proposed that one persons inputs are also his or her partners
outcomes (Cook & Hegvedt, 1983). Thus, with determined input values,
congruent outcomes can be established for both partners in the development
of a ratio formula of objective equity.
However, using objective measures does not necessarily provide a
true understanding of a relationships equity. While a formula may
objectively show inequity or unfairness, an individuals subjective perception
of the same might differ (DeMaris & Longmore, 1996; Lennon & Rosenfield,
1994; Greenstein, 1996). As mentioned above, women do most of the


household labor. However, some women still judge their relationships as
being fair regardless of this discrepancy (Greenstein, 1996). Since people
do not realistically calculate every minute input and outcome in their
relationship, individual perception is highly essential to the accurate
measurement of equity. Current research proposes that gender ideology,
power differentials, justifications, and appreciation by husbands of wives
household labor also contribute to judgments of fairness, although the
differences in household labor performed remains pertinent to perceptions of
fairness (DeMaris & Longmore, 1996; Blair & Johnson, 1994; Greenstein,
1996; Thompson, 1991).
Equitys importance as a theoretical approach to household research
is more significant when it is examined in relation to other variables. In this
research project, I will apply the concepts of objective equity and perceived
fairness to determine what effects, if any, either may have on the dynamics of
a marriage in terms of its well-being. Since mothers and childrens well-
being has been explored previously in terms of perceived fairness (Acock &
Demo, 1994), I will focus on what effects objectively measured and
subjectively perceived fairness have on the well-being of spouses in a
marriage. Well-being takes into account physical and psychological
conditions of individuals and offers a key reflection on the status of an
4


intimate relationship (Acock & Demo, 1994; Gove, Hughes & Style, 1983;
Gove, Style & Hughes, 1990; Mookherjee, 1997; Ross, Mirowsky &
Goldsteen, 1990; White, 1992; Williams, 1988).
In this research only married couples will be studied. Further, I will
concentrate on households in which both partners are employed and, if any,
children in the home are younger than the age of five years. It is my view that
children and the dual-earner stipulation keeps the notion of equity in
perspective. If children are present in the home and contributing little to
household labor because of their young age, it is assumed that spouses
would perceive housework as a responsibility of each other. If both are
working outside the home, then household labor cannot be easily distributed
based on who stays home and who works at a paid job. With both partners
working outside the home, and children assumed to be contributing only
minimal labor, the division of household labor and levels of equity
(subjective and objective) are crucial elements in understanding the balance
of a dual-earner relationship and the well-being of the individuals in it.
The data for my research come from a national sample. The National
Survey of Families and Households (Sweet, Bumpass & Call, 1988) is a
representative data set containing all of the variables needed. Questions in
the survey provide measures of equity, well-being, and the division of
5


household labor. Dual-earner couples included in the sample present
adequate cases for statistical observation and analyses.
6


CHAPTER 2
REVIEW OF LITERATURE
Theoretical Background
The concept of equity is derived from the theoretical underpinnings of
exchange theory (Adams, 1965; Walster, Berschied & Walster, 1976; Hatfield
et al., 1984). Exchange theory, in turn, is developed mostly from the early
works of George Homans (1961) and John Thibault & Harold Kelley (1959).
Homans ideas of exchange theory come mostly from the influence of
psychological studies of animal behavior wherein rewards and/or punishment
are the primary determinants of action (Homans, 1961). The basic
underlying theme asserts that humans are responsive to positive rewards
from others in an interaction and will act accordingly towards those others as
a means of securing the rewards. Homans (1961) explains that people make
investments in a relationship and incur a profit gain or loss in the exchange
with interacting individuals. Profitable investments thus are viewed as
rewards with minimal cost, while unprofitable investments result in
unsatisfactory rewards at significant cost. This idea provides one of the key
assumptions of equity theory in that individuals will attempt to maximize their
7


profits in a relationship by maintaining a level of equality between what is
gained and what is given up (Adams, 1965; Walster, Berschied & Walster,
1976).
Thibault and Kelleys (1959) work developed the same general
principles of rewards and costs. Rewards and costs are abstractly defined as
positive and negative components of a relationship. Focusing on the
interactions of two people, or a dyad, Thibault and Kelley suggest that the
conditions of the relationship will be most ideal if more rewards are gained
by each person at the least possible expense. As an example, a husband
may help his wife more around the house because she has a full-time job. In
return, she allows him to lounge around the home on Sundays watching
television. However, the husband realizes that he always lounged around
the home on Sundays before his wife had started working outside the home
and he also realizes that housework is a lot harder than he had imagined. In
this case, he may feel that the arrangement is imbalanced against him.
The idea of imbalance in the relationship as a result of rewards and
costs forms the main premise of equity theory. Equity, as utilized in social
psychology, concerns itself with the specific rules humans apply in
determining whether or not valued resources are distributed justly and fairly
among actors in specific settings (McClintock and Kiel, 1982). If not, then the
8


relationship between the actors is characterized as an inequity (Adams,
1965). In Adams notion of inequity, costs and rewards are reconceptualized
as inputs and outcomes; inputs being what a person contributes to an
exchange relationship and outcomes are what is received (Adams, 1965).
According to Adams (1965), a person may assess her level of inputs and
outcomes in a relationship and determine that the proportion of her outcomes
to inputs is less than her partners ratio of outcomes to inputs. In this case,
the person presumably will judge the relationship as inequitable. Adams
(1965) represents inequity as either:
0 a o a
> or -2- < ; (Equation 2.1)
1 Ib / Ih
a b a o
where y2- denotes the ratio of outcomes (O) and inputs (I) for Person A and
Q*
h
is the ratio for Person B. While it may seem obvious that a person on
the wrong end of an inequitable relationship will feel adversity towards the
situation, Adams (1965) suggests that people who are getting too much from
a relationship will feel guilt and thus view the relationship in negative terms.
In either case, the assumed behavior for either person is to obtain a level of
equity in the relationship. For equity to be achieved, the proportion of
9


outcomes to inputs for each person in the exchange relationship must be
equal. Equity is represented by the formula :
= (Adams, 1965). (Equation 2.2)
1 a h
This ratio formula of equity, however, has been evaluated as being
logically incorrect because it does not take into account instances in which
Person A or Person Bs inputs are negative (Walster, Berscheid, and
Walster, 1976). An example of this problem could be illustrated by
assuming that Person As inputs in a relationship are measured as -8 (/ = -8)
and her outcomes were 16 (0 = 16), giving her an equity value of -2. If
Person Bs ratio values were 1 = 6 and O = -12, his equity value would also
be -2. However, it is obvious that in this instance Person B is contributing
more and getting less, while Person A is contributing less and receiving
more, which is not as equitable as would be desired. What is proposed by
Walster, Berscheid, and Walster (1976) is a reformulation of the Adams
equation that considers the possibility of negative inputs. The revised ratio
formula is as follows:
Is-= (Walster, Berscheid &Walster, 1976); (Equation 2.3)
W (M
10


wherein (|/a|) and (|/A |) account for the possibility of negative inputs for A and
B, respectively. (Oa-1 a) and (Ob-Ib) symbolize either persons gains, or
what is left of the outcomes after inputs are subtracted out. The exponents
Ka and Kb are equal to +1 or -1 based on the sign of Person A or Person
Bs inputs and gains. Walster, Berscheid and Walster (1976) point out that
Ka = sign(/J x sign(0a -Ia) and Kb = sign(/4) x sign(04 -Ib). In this
formula negative inputs are accounted for, giving a more accurate
measurement of a relationships equity. Using the example of Person As
inputs and outcomes (7 = -8, O =16) with Person Bs (7=6, O =-12) in this
new formula, the computation would show that Person As value of
(|-8| x 24=) 192 is greater than Person Bs (|6|x-12 =-72). The values
achieved by this formula show that, rather than being equitable, the
relationship is obviously in favor of Person A.
The development and modifications of equity theory and the ratio
formulas contributes to equity theorys use in a variety of settings (Adams &
Freeman, 1976; Hatfield et al., 1984; McClintock & Keil, 1982). Equity
theory has been used in the analysis of business relationships (Adams,
1965; Lawler, 1968), helping relationships (Walster, Berscheid and Walster,
1976) and, most especially, intimate relationships (Demaris and Longmore,
li


1996; Hatfield etal., 1984;Walster, Berscheid, and Walster, 1976; Kollock,
Blumstein, and Schwartz, 1994; Sprecher, 1986). This study will further the
application of equity theory in intimate relationships by examining the
relationship of marriage and the equity determined and perceived about the
accomplishment of household tasks.
Previous Studies
How household labor is measured is relative to what is being
analyzed. The most common measurement of household labor is a
specification of tasks and an inquiry into how many hours a week individuals
perform each task. However, some researchers feel that this measurement
misses certain aspects of household labor. Rather than just measuring
tasks, it has been suggested that household labor be looked at in terms of
the creation and maintenance of a household where family members and
matters are cared for in a manner acceptable to all (Mederer, 1993;
Thompson, 1991). In this view, the division of tasks among members is
irrelevant without understanding who determines what the tasks are, when
the tasks should be accomplished and who takes the initiative in
accomplishing them. While this view of household labor delves deeper into


structural and ideological elements of family life, it is more appropriate for
studies aimed exclusively at measuring perceptions of equity. In this study,
perceptions of equity are not the primary focus. Rather, perceptions of equity
will be looked at in conjunction with objective equity as an evaluation of
individuals well-being. Still, the research dealing with household labor and
perceptions of equity rely on-quantifying specific types of tasks as a means of
evaluating who accomplishes what around the household.
There are differences in the types of tasks that men and women do.
Routine tasks are ones that are done on a regular basis. Cleaning house,
cooking, washing dishes, and doing laundry are routine jobs and often the
responsibility of women. Men are more likely to perform the sporadic tasks
of the household, which include auto maintenance, yard work or home repair
(DeMaris & Longmore, 1996). As such, household labor is sometimes
referred to as a gendered activity(DeMaris & Longmore, 1996; South &
Spitze, 1994).
The key difference between routine and sporadic housework is that
the former is considered essential to the day to day operations of the
household, while the latter is subject to discretions of necessity. The
distinction of types of tasks performed across gender lines is an imperative
consideration when examining equity. Since womens routine work is more
13


time intensive than mens sporadic work, a better assessment of equity would
take note of mens participation in routine tasks (DeMaris & Longmore,1996).
One of the overlying assertions of household labor is that women do
much more of it than men (South & Spitze, 1994; Hochschild &
Machung,1989; Presser, 1994; Coltrane & Ishii-Kuntz, 1992). Generally,
research has shown that the discrepancy of housework between men and
women is steady across various household structures (South & Spitze,
1994). In households consisting of the following: never married and living
with parents, never married and living independently, cohabiting, married,
divorced and widowed, South & Spitze (1994) determined that women spend
more time doing housework than men, with the disparity magnified in married
households and existent in dual-earner homes as well.
In dual-earner households, the distinction between routine and
sporadic tasks is a relevant issue. Assuming both partners work similar
hours outside the home, household labor cannot be divided along gender
activity lines without leaving the female in an inequitable situation. It would
be expected in these situations for each partner to contribute somewhat
equally to routine tasks. Past research, however, has shown that the majority
of routine tasks still fall into the hands of women (Hochschild & Machung,
1989). Hochschild and Machung (1989) contend that working women in
14


dual-earner households are subject to the second shift. The second shift
pertains to housework that must be accomplished at home, after the first shift
of paid employment is over. While it is arguable that both partners should
partake in the second shift, Hochschild and Machungs study (1989) revealed
the woman usually ends up doing the majority of chores around the home. In
one particular case, the dividing of the housework into upstairs and
downstairs left the wife with upstairs chores. Upstairs included the living
room, dining room, kitchen, bedrooms and bathrooms. The husband was left
to care for the garage and the dog. Again, the difference between the two is
large considering the types of tasks associated with each area of the home.
In other words, the wife is stuck with routine tasks and the husband is
relegated to sporadic household maintenance.
Who does what around the household can be attributed to several
factors. Presser (1994) suggests that employment schedules of dual-earner
couples play a significant role in the amount of female tasks completed by
men. According to Presser, when husbands work a non-daytime shift and
their wives work a daytime shift, it is more likely that the husbands share of
female tasks will increase. Pressers study also found that the occupational
prestige of both spouses affected to some degree the amount of female-
oriented housework performed by men. When spouses are both in
15


professional or managerial careers, the husbands participation in household
work is significantly increased.
The gender ideology of spouses is also believed to be a component
determining dynamics of household labor. Past research (Presser, 1994;
Hiller & Philliber, 1986) analyzed gender ideology and concluded that men
with egalitarian ideologies contributed more to household labor than men
who held more traditional beliefs about gender differences in the household.
Hiller & Philliber (1986) found that partners expectations had strong effects
on his or her behavior towards various aspects of household labor.
Greenstein(1996) took the idea a step further and analyzed the interactive
affects of gender ideology on household labor. What he determined was
that unless both husband and wife shared egalitarian views about gender
and marital roles, husbands contributed relatively less to household labor.
In a later study, Greenstein (1996) again applied gender ideology
measures to household labor, but tied it to perceptions of inequity for wives
and marital quality. According to Greenstein, the more inequalities in
household labor, the more inequity was perceived by egalitarian wives rather
than traditional wives. For egalitarian wives, too, the perception of inequity
in the marriage was highly related to the level of perceived marital quality.
16


How equity is perceived in a relationship remains one of the central
themes in family research. Thompsons (1991) study of womens sense of
fairness about family work utilizes three factors for determining fairness:
outcome values, comparison referents, and gender-specific justifications.
For outcome values, Thompson suggests that what is pertinent to women is
not just the amount of tasks and how much time is spent on tasks, but also
the quality of those tasks, the quality of time away from tasks, and their
husbands appreciation and responsiveness to the work they do (Thompson,
1991).
Thompson also asserts that women make between and within gender
group comparisons in regards to the work at home. For women, the fairness
they perceive of housework is based on the comparison to the superwoman
ideal of the household; superwomen get little help from husbands and feel
everything is all right as long as nothing changes in the family dynamics.
Regarding husbands, women tend to compare them with other men in order
to assess their situations at home. If they believe that other men are doing
less work at home than their husbands, then their situation at home is
perceived as more fair.
Thompsons third factor for women perceiving fairness is justifications
about who does what around the household. Men may justify not doing
17


housework by claiming they are not experienced at it, their own jobs leave
them tired and out of time, or they do not enjoy it as much as women do.
These justifications are dependent on womens acceptance, however, if
women accept the reasons as justifying mens participation (or lack of) in
housework, then they may judge the division of household labor as fair. If not,
then the perception about the relationship leans towards unfairness
(Thompson, 1991).
Other explanations for perceptions of fairness rely on the concept of
power (DeMaris & Longmore, 1996). It is believed that women who have
fewer alternatives to their present relationship are apt to judge an inequitable
relationship as fair. The lack of resources for some women offer them little
leverage in bargaining for a more equitable relationship. The chosen
recourse is a perception of fairness as a means of justifying staying in the
relationship.
While perceptions of fairness may be correlated with many other
factors, this research will focus on levels of objective equity in the overall
household as they relate to perceptions of fairness. The other factors are
important, but the relationship between perceptions of equity and the actual
measurement of it regarding contributions to housework and other areas of a
marriage remain a relevant relationship for analysis.
18


As mentioned above, past research in household equity has relied on
a ratio formula or logged ratios to assess objective equity (DeMaris &
Longmore,1996; Presser, 1994; Ward, 1993). In order to achieve an
objective evaluation of a relationships equity, all possible forms of inputs and
outputs must be included. While household labor inputs are an important
part of the ratio formula, input values for other domains of the relationship
need to be considered (DeMaris & Longmore, 1996; Kollock, Blumstein &
Schwartz, 1994). Inputs may not always be equal in terms of household
labor, but the difference can be made up with contributions in other areas of
the relationship. This may include input in areas such as income, education,
hours in paid labor, child care or health status of the individual.
Kollock, Blumstein and Schwartz (1994) examined equity and
perceived fairness between married and cohabiting couples by using a
questionnaire to obtain information from respondents. The key dependent
variable used was in the form of a question asking persons to compare what
they put into a relationship with what they get out of it and what their
partners put into it and what the partners get out of it, then determine who is
benefiting more. This perception of fairness was measured on a nine-point
scale with I benefit the most being the high end of the scale and (s)he
benefits the most being on the low end of the scale. This variable was
19


analyzed with independent variables that included hours in housework,
income, employment status, and years of education (Kollock, Blumstein &
Schwartz ,1994). Also included were independent variables measuring
physical attributes, and emotional expressiveness.
Using regression analysis, for married couples Kollock, Blumstein and
Schwartz (1994) found the relationship between husbands housework and
perceptions of equity to be statistically significant. In other words, the more
work the husband did around the home in relation to what his wife did, the
more he was to rate the relationship as benefiting him more. Also significant
was the relationship between expressiveness and perceptions of fairness;
the more expressive the husband was towards his wife, the more he
perceived her as benefiting more.
For wives, the study determined that income and employment were
statistically significant in regards to the dependent variable. The more
income a wife made relative to her husband, the more she was to assume
that he benefited more than her. If the wives spent more hours in paid labor
relative to their husbands, the more they felt that they were benefiting from
the marriage (Kollock, Blumstein & Schwartz ,1994).
In one of the more recent studies using a formula for measuring
objective equity in marital relationships, DeMaris & Longmore (1996) relied


on similar input values and proposed that a husbands inputs are also a
wifes outcomes and vice versa. This follows most of the theoretical
assertions regarding intimate relationships that either partners inputs are the
others outcomes (Cook & Hegvedt, 1983). Also, the formula is more a
modification of Adams (1965) formula rather than Walster, Berscheid and
Walsters (1976) version. According to DeMaris and Longmore (1996), in an
intimate relationship such as marriage, it is unlikely that either partner would
make negative inputs toward the relationship. The need for absolute values
in the numerator is unnecessary. Thus, for this research the logged ratios of
inputs and outputs for partners in both married and cohabiting households is
utilized in an equation formula. In the formula, the product of each partners
outcome/input ratios in different areas of the relationship are compared. This
is represented as:
(q \(n Vr\ \
la
O,
2 a
^ Aa ^ A a y V A a '
o
3 a

J
o
V Aa J

O*
v Az> )
f n \
2 b
o.
ro ^
'-'3b
V 12b J ^ Aft y
'O'
V Ijb J
(Equation 2.4)
(Demaris & Longmore, 1996); where, la, 2a, 3a, and ja symbolize the different
areas of a marriage in which Person A makes inputs and receives outcomes.
For instance, la and lb might be the domain of contributions to housework.
Person A could be noted as doing 20 hours a week of housework and Person
B could be doing only 10 hours. If la and 2b were income values, Person A
21


might be recorded as contributing $10,000 annually, while Person B
contributes $30,000 annually. The determination of equity using the formula
would be:
(10V30,000^1 (20V 10,00(A .
V 20/V 10,000/ ~ V10/V30,000/
wherein Person As ratio is equal to 1.5, while Person Bs ratio is equal to
.6667. In this case, Person Bs ratio is less than Person As making the
relationship inequitable for Person B. However, if Person A were to increase
her income to $22,500, the resulting ratio would bring the relationship to
equity. Thus, the closer the ratios are to being equal, the more equitable is
the relationship (DeMaris & Longmore, 1996).
In Demaris & Longmores (1996) analyses, perception of fairness was
measured using a question from the 1987-88 NSFH (Sweet, Bumpass & Call,
1988) data set asking both husbands and wives to rate the fairness of
housework. The responses were rated on a 5-point Likert scale and ranged
from unfair to me as one extreme, fair to both as the middle value and
unfair to spouse as the other extreme. DeMaris and Longmore only sought
to determine the relationship of housework in regards to whether or not it was
unfair to the wife. Their findings reflect significance for the husbands
perception of unfairness of household tasks to the wife with the independent
22


variables of age, race, wifes years of schooling, and equity variables
including health, logged sporadic hours in housework ratio, and child
socialization. For the wives perception of unfairness of housework,
significance was determined for Hispanic couples in relation to white couples,
wifes health, both husbands and wives support for housework sharing as
well as wives perception of routine hours of housework, child socialization
activities, and ratio of hours in paid labor (Demaris & Longmore, 1996).
What is pertinent about DeMaris and Longmores (1996) study is the
provision of a ratio formula that purports to measure equity across the
various domains of the relationship. For this study, DeMaris and Longmores
formula will be used, although the analyses will take a different approach.
While DeMaris and Longmore measured only the perception of fairness of
housework as being unfair to the wife, this study will regard perceptions of
housework fairness as being equitable to both spouses.
While perceptions of fairness will be used as a dependent variable, it
will also be used as an independent variable to measure its effect on well-
being for both spouses. Well-being has been examined previously as a
factor in relationship to marital status, wifes employment, family social
status, parenthood, and family type (Acock & Demo, 1994; Gove, Style &
Hughes, 1990; Gove, Hughes & Style, 1983; Mookherjee, 1997; White, 1992;
23


Ross, Mirowsky & Goldsteen, 1990). With different studies the
measurement of well-being has varied as well, although most studies capture
the essence of the variable.
Acock and Demo (1994) examined well-being in relation to various
family types in the United States. For their study, they looked at four types of
American families: unwed mothers, divorced mothers with children, remarried
families, and two-parent families and studied the levels of well-being for
mothers and children against different factors. Within these family types,

variables such as income, race, education, household size, marriage length,
marital stability, marital satisfaction, parent-child interaction, and marital
equity were used as independent variables in the analysis of well-being
(Acock & Demo,1994).
Well-being was measured in four ways with different questions found
on the 1987-1988 NSFH (Sweet, Bumpass & Call, 1988) data set. For
psychological well-being, Acock & Demo (1994) used measures of self-
esteem, overall happiness, and depression. To measure physical well-being,
an overall health assessment question was used. Each indicator of well-
being was measured with a question or set of questions that asked
respondents to evaluate themselves in terms of their perceived health as
compared to others their own age, whether they were persons of worth or
24


ability, whether they were satisfied with themselves, how happy they were
and how often in the past week did they exhibit signs of depression (Acock &
Demo, 1994).
In their findings, Acock & Demo (1994) determined that first-married
mothers have higher well-being than continuously single mothers and are
happier than divorced mothers. First-married mothers also have slightly
higher self-esteem and better health than other types of mothers. Also, the
study showed that background variables such as education and income are
significant predictors of some mothers self-esteem and overall health (Acock
& Demo, 1994).
Other studies examining well-being tend to show that marital status
effects to some degree the physical and mental health of individuals (Glenn
& Weaver, 1988; Gove, Style & Hughes, 1990; Gove, Hughes & Style, 1983,
Mookherjee, 1997, White, 1992). Gove, Style and Hughes (1990) theoretical
analysis discusses some facts about the relationship between well-being and
marital status. The first point made by Gove, Style and Hughes (1990) is that
married people have lower incidences of treatment for mental conditions or
illness compared to unmarried people. The facts also point out unmarried
men being more likely to suffer mental illness than unmarried women. In
terms of overall happiness and satisfaction with life, Gove, Style and Hughes


(1990) report that married people tend to have higher levels of both. This is
consistent with other studies (Gove, Hughes & Style, 1983; Glenn & Weaver,
1988; Mookherjee, 1997). Mortality rates are also lower for married men
and women, suggesting that the overall physical health of married people is
better than others. Fewer married people are institutionalized for mental
health problems than are unmarried people. Being institutionalized is an
indicator of psychological and physical well-being, implying again the
advantage of being married over being single (Gove, Style & Hughes, 1990).
The reasons behind the advantages married people enjoy over
unmarried people rests in the fact that a marriage provides the means of
establishing and evaluating ones self through the consistent interactions and
role expectations of its participants (Gove, Style & Hughes, 1990).
However, this is assuming that the interactions within a marriage are positive.
If the interactions are negative and stability is shaken, then levels of well-
being tend to decrease (Gove, Styles & Hughes, 1990, Glenn & Weaver,
1988).
One of the studies that found no relationship between marital status
and well-being was Whites (1992) study in Canada. From a sample of
11,131 Canadians, White(1992) used life satisfaction, subjective health,
visits to physician, and absence of problems as indicators for well-being.
26


Looking at differences in gender, age, and marital status, White determined
that males are not as affected by being single as was reported by Gove,
Style & Hughes (1990). Further, Whites study found higher levels of
subjective health, fewer health problems and fewer doctor visits for unmarried
rather than married people. This would seem to contradict previous findings
and suggest marriage does not contribute to higher indicators for well-being.
27


Purpose of the Study
From the information above, this research will propose to show a
relation between variables of perceived and objective equity to spouses well-
being. Objective equity will be measured with a formulaic model of
comparison ratios for both partners in an intimate relationship. The model
will contain quantified indicators of inputs and outputs such as hours spent
on specific tasks, income, education, employed hours, etc., and follow the
assumption that a husbands inputs are a wifes outcomes and vice versa.
Subjective equity will be measured with questions previously used in
research that establish individual and partners perceptions of fairness in a
relationship. The point will be to compare subjective perceptions of fairness
with objective measures of equity and how each affect partners well-being.
Well-being will be measured following the precedent set by Acock & Demos
(1994) research. Self-esteem, overall happiness, depression, and physical
health will be used as indicators of well-being. Previous research has shown
well-being to be related to marital status, although well-being is conditional
on the marriage reflecting positive interactions (Gove, Style & Hughes,
1990). Considering that notions of equity are based on rules of determining^
what is fair and that inequity is usually related to emotional distress (Adams,
28


1965; Hatfield, et al., 1984; Walster, Berscheid & Walster, 1976), it seems
relevant to examine equity as it relates to well-being. One of the greater
concerns of family sociology is to understand the dynamics leading to
stability and/or quality of intimate relationships. The following research
intends to contribute to this understanding by offering this approach and
testing the following hypotheses.
Hypotheses:
1. The higher the level of objective equity in a relationship, the more likely
either partner will perceive the relationship as fair in regards to the division of
household labor.
2. The higher the level of objective equity in a relationship, the more likely
either partners level of overall happiness will be positively affected.
3. The higher the level of objective equity in a relationship, the more likely
either partners health will be positively affected.
4. The higher the level of objective equity in a relationship, the more likely
either partners self-esteem will be positively affected.
5. The higher the level of perceived fairness of housework in the relationship,
the more likely either partners level of overall happiness will be positively
affected.
29


6. The higher the perceived fairness of housework in a relationship, the more
likely either partners health will be positively affected.
7. The higher the perceived fairness of housework in a relationship, the
more likely either partners self-esteem will be positively affected.
30


CHAPTER 3
METHODS
Sample
The sample for this study comes from the 1987-88 National Survey of
Families and Households (NSFH) (Sweet, Bumpass & Call, 1988). The 1987-
88 NSFH is a large probability sample that is representative of American
households and families. A total of 13,017 households were administered a
survey which consisted of a main interview completed by a primary
respondent (R1), a set of self-administered questions distributed only to
specific primary respondents, and another self-administered questionnaire
completed by a spouse or cohabiting partner who was present in the
household (R2). For each of the 13,017 households selected for the study,
one individual was selected as a primary respondent. The primary
respondent was subjected to the face-to-face interview as means to collect
data on American households including information on marital conflict,
division of household labor, marital and nonmarital relationships,
relationships with parents, relationships with children, gender expectations of
sex roles, and physical and mental health status of family members.
31


Extensive demographic information was also collected, providing information
on race, income, education, and overall wealth for all respondents.
Additionally, a second, shorter questionnaire was given to spouses or
cohabiting partners who were present in the household. While shorter, the
second questionnaire provided information about the household from the
spouse/partner perspective, giving a more complete examination of family
dynamics and structure.
Although a probability sample was used, some groups were
oversampled. Oversampling was done to ensure that enough cases of
specific types of households were represented for adequate statistical
analyses. Single-parent families, families with stepchildren, Blacks and
Hispanics were included in the oversample.
For this particular study sample, certain households were excluded for
analysis based on specific criteria. Since this study focuses on equity in
married dual-earner households, the criteria for sample selection were as
follows:
1. Primary respondent was married, spouse was present, and
completed the secondary questionnaire.
2. Both spouses reported income from a job for the year of the
survey.
3. Primary respondent and secondary respondent were of the
opposite sex.
32


4. Both spouses worked at least 20 hrs. a week at a main job.
Furthermore, cases were eliminated in instances where either
respondent did not provide a valid answer to questions regarding hours spent
on specific household tasks. The reasoning for this criterion was the use of
each household labor response in the creation of an objective equity ratio
variable. While the option to recode missing responses was considered, I
decided to eliminate the cases with missing responses so as to keep intact
the original responses of the sample respondents. Finally, any case in which
the primary and secondary respondents were of the same sex was excluded
from this study. While same-sex relationships are akin to marriage for many
individuals, at the present time I will consider same sex relationships as
cohabiting relationships, thus excluding them from my study. Under all these
criteria, my final sample was reduced to 529 couples.
In the NSFH study, the survey questionnaire includes in excess of
4,000 variables covering a wide range of topics designed to describe fully the
dynamics and structures of various households. Questions were used to
gauge such topics as marital status, number of people living in the
household, views on gender roles, and child socialization. Of primary
interest for this study are the variables associated with housework
contributions, individual well-being, and perceptions of fairness regarding the
33


housework. Other variables of interest are: income from salary/wages for
each spouse, education levels, hours worked in employment, and
background variables including race, age, and sex of both primary
respondent and his/her spouse.
Sample characteristics
As mentioned above, specific cases were selected from the NSFH
study that were pertinent to my analysis. Setting conditions of marriage,
income and employment (at least twenty hours of work per week) for both
primary and secondary respondents as well as filtering out cases that did not
respond to key variables of the study, the final sample consisted of 529
married, dual-earner couples.
Since the data set includes both males and females as primary and
secondary respondents, steps were taken to convert primary and secondary
respondents variables into husband and wife variables. This required
selecting out cases in which the primary respondent was female and the
secondary respondent was male, then transforming the respective, pertinent
variables into wife and husband variables. This procedure was repeated
again in cases where the primary respondent was male and the secondary
respondent was female. Once all primary and secondary respondent
34


variables were converted into husband and wife variables, the cases were
merged together into one data set for which this studys analysis was done.
Because the reporting of age for most of the husbands was missing,
accurate descriptions for age were not available. For this study, then,
statistics for age will not be reported. Because of recoding and conversion of
primary and secondary respondents into husbands and wives, the education
variables were converted into categories depicting levels of education. 0-8
years of formal education are coded as 1, 9-12 years are coded as 2,13-16
years are coded as 3 and any years over 16 years of formal education are
coded as 4. Most of the respondents had at least a high school and some
college education. Of 529 wives, 259 had 13-16 years of education, 212 had
9-12 years of education. For 529 husbands, 245 had 13-16 years of formal
education and 195 had 9-12 years of education. 465 (87.9%) of the
husbands are White, 36 (6.8%) are Black, 19 (3.6%) are Hispanic and the 7
(1.3%) remaining husbands are categorized as other. For the wives, 466
(88.1%) are White, 29 (5.5%) are Black, 20 (3.8%) are Hispanic and 10
(1.9%) wives are identified as being members of other racial/ethnic groups.
Income was also recoded into categorical ranges of $5000 each.
Husbands or wives that made between $0-$5000 annual income were coded
as 5, those making between $5001 and $10,000 were coded as 10 and so
35


on. For husbands, the modal category for income is between $15,001 and
$20,000 with 91 husbands (17.2%) reporting their income within this range.
The wives modal income range is $10,001 to $15,000, with 124 wives
(23.4%) falling within this range. Husbands mean hours worked in paid labor
is 47.643 hours per week while wives mean hours worked in paid labor is
40.968 per week.
Measurement
The key dependent variable of this study is the psychological well-
being for respondents. The NSFH survey provides several questions
indicative of well-being, that have been used in previous studies (Acock &
Demo, 1994). Overall happiness, level of depression, and self-esteem are
three key indicators of psychological well-being. Physical well-being is
measured by a single question regarding the respondents health. Since
data are missing for several respondents regarding depression measures,
depression will be excluded from this study as an indicator of well-being.
For the first indicator of well-being, overall happiness, respondents
were asked to assess their lives using a 7-point Likert scale that ranged in
reply from 1 for very unhappy to 7 for very happy. Four was the midpoint
for the scale. This variable is referred to as HHAPPY for husbands and
WHAPPY for wives. HHAPPY and WHAPPY are recoded into dichotomous
36


variables for use in a logistic regression model. HHAPPYHI is equal to 1 if
the primary respondent has a score that is greater than or equal to 5 on the
HHAPPY variable, representing scores toward the high end of the scale. If
the score is anything less than 5, then the value is 0. WHAPPY is recoded in
the same fashion, dichotomously with values of 1 and 0; WHAPPYHI
represents the high scores.
Physical health is considered a crucial part of well-being. In this
survey, health is measured by asking the respondent to rate their health in
comparison to others same age. A 5-point Likert scale was used again with
the available responses ranging from 1 for very poor to 5 for excellent.
While this measurement seems very simple, it has been proven previously to
be an effective indicator of a person physical health (Acock & Demo, 1994).
HHEALTH and WHEALTH are the variables referring to the health of the
primary respondent and secondary respondent, respectively. For the logistic
regression models, HHEALTH and WHEALTH are recoded into dichotomous
variables. HHELTHHI is coded as 1 if the respondent scores 4 or higher on
the HHEALTH variable and 0 if the score is less than 4. WHELTHHI is the
dichotomous variable for the wives scores on the WHEALTH variable and is
coded in the same way as the HHELTHHI variable.
37


Finally, self-esteem is measured in the survey by three questions
commonly used in self-esteem scales (Acock & Demo,1994; Rosenberg,
1979) The questions were rated on a 5-point Likert scale with responses
ranging from 1 for strongly agree to 5 for strongly disagree and 3
representing the mid-point value of the scale. The questions asked of the
respondents centered around feelings of self-worth, how well the respondent
could do things compared to others and whether or not the respondent was
satisfied with him or herself. The average score on these three questions is
labeled as HSELF for the husbands and WSELF for the wives. Again, these
variables are further recoded into dichotomous variables. Namely, HSELFHI
is coded as 1 when the respondents average score for the self-esteem
questions is greater than or equal to 2 (or scored towards the high end or
agree end of the scale) and 0 represents all the other scores (HSELF > 4).
WSELFHI is coded in the same manner, representing the secondary
respondents high average scores on the self-esteem measures.
Also to be used as a dependent variable is the perceived fairness of
house work for either spouse. This variable is measured using a single item
on the NSFH survey. The item asks respondents to assess the fairness of
chores in their relationship by providing these responses: very unfair to me,
somewhat unfair to me, fair to both, somewhat unfair to him/her, very
38


unfair to him/her5. The response scores ranged from 1 (very unfair to me)
to 5 (very unfair to my spouse). A score of 3 represented perceived
fairness for housework. Both husbands and wives were asked this question
and the variable is labeled HFAIR and WFAIR, for husbands and wives,
respectively.
HFAIR and WFAIR are recoded into dichotomous variables with
values of 1 and 0 for use in a model of logistic regression. When the
husband perceives housework as fair to both (HFAIR = 3), the variable
HFAIRBTH is equal to 1. When the husband perceives housework as unfair
to himself (noted by values 1 and 2 on the scale), the variable HFAIR1 is
equal to 1. If the husband perceives housework as unfair to his wife (4 and 5
on the scale), HFAIR2 is valued at 1. WFAIRBTH, WFAIR1 and RWFAIR2
are the dichotomous variables representing the wives scores on the WFAIR
variable. The three variables are coded the same as the dichotomous
variables for the primary respondents.
This perceived fairness variables (HFAIR and WFAIR) will also be
used as independent variables in analyzing relationship to the well-being
variables. Objective equity will be the other key independent variable.
Objective equity will be measured using a formulaic ratio developed by
DeMaris and Longmore (1996). In this formula, ratios are obtained for
39


separate domains of the marriage in order to assess the equity within each
domain. The product of the ratios for each spouse are compared against
each other to determine the level of equity. If there is no difference in the
product of the ratios between spouses, then equity is achieved. Previous
models of equity suggest that the proportion of one individuals outcomes to
inputs for a relationship should equal the proportion of the others outcomes
to inputs(Adams, 1965; Jasso, 1978; Walster, Walster & Berscheid, 1978).
Further, it has been suggested that in intimate relationships like a marriage,
a husbands input is a wifes outcome and vice versa (Cook & Hegvedt, 1983;
DeMaris and Longmore, 1996).
Thus, assessing equity of a marriage requires taking ratios of outcome
to inputs in key domains of the relationship ( DeMaris & Longmore, 1996;
Lennon & Rosenfield, 1994). For this study, the domains I will focus on are
housework hours, job income, education and hours in paid employment. For
housework hours, the NSFH survey asked husbands and wives to indicate
how many hours a week they spend on nine types of household tasks and
how many hours a week they perceived their spouses spend on the same
tasks. The tasks include: cleaning house, preparing meals, cleaning up after
meals, washing and ironing, paying bills, auto maintenance, going shopping,
driving family members around, and doing outdoor tasks. These were
40


measured continuously by the survey and responses for each spouse on all
tasks were summed. For this study, the tasks were separated into routine
and sporadic tasks: cleaning house, preparing meals, cleaning up after
meals and washing and ironing are routine tasks. Sporadic tasks are paying
bills, auto maintenance, going shopping, driving family members around and
doing outdoor tasks. Routine tasks are designated as HROUTINE as the
husbands summed contributions to routine housework and WROUTINE as
the wives contributions. The variables HSPRUTIN and WSPRUTIN are the
spouses perceived contributions to routine housework as reported by
husbands and wives, respectively. Sporadic tasks were recoded in the same
fashion. HSPRADIC and WSPRADIC represent the summed total of hours
spent doing sporadic tasks as reported by husbands and wives. HSPSPRDC
and WSPSPRDC are the total hours husbands and wives report for their
spouses performing sporadic tasks.
The summed contributions to routine housework were converted into
ratios by placing HSPRUTIN (husbands perceptions of wifes contribution) in
proportion to HROUTINE (husbands contributions) or
HSPRUTIN/HROUTINE for husbands and WSPRUTIN (wifes perceptions of
husbands contributions) in proportion to WROUTINE (wifes contributions) or
WSPRUTINA/VROUTINE for wife's ratio for housework contributions.
41


Sporadic tasks are not used in a ratio formula, but will be considered as
independent variables in logistic regression equations measuring effects on
dependent variables.
For the other domains where both spouses are assumed to be making
contributions, ratios are obtained with the same concept of husbands inputs
being wifes outcomes and vice versa. In the domain of employment-related
finances contributed to the relationship, income is measured with a question
asking respondents to report annual income from wages and/or salary.
HWAGES and WWAGES represent the income variables for husbands and
wives To use them in the ratio formula, these variables were recoded into
smaller values based on income dollar intervals of $5,000. As an example, if
a husband reported an annual income between $0 and $5,000, the value for
WAGESR was recoded into a new variable, HINCOME, as 5. If a husband
reported an annual income as between $5,001 and $10,000, HINCOME was
coded as 10 and so on. For wives, WWAGES is converted into WINCOME
with the same value intervals. Thus, the ratio for income is measured
WINCOME/ HINCOME for husbands and HINCOME/WINCOME for wives.
Education is measured by asking the primary respondent to indicate
his/her level of education. The responses ranged from 0 (no formal
education) to 20 (doctorate/professional degree). Secondary respondents
42


were asked to state the highest grade they had completed. This variable was
measured continuously and represented by HEDUCAfor husbands and
WEDUCAfor wives. For the ratio formula, theses two variables are recoded
into HEDUCATE and WEDUCATE, with years of education separated into
four categories. Zero to 8 years of education is coded as 1,9-12 years is
coded as 2, 13-16 years is 3 and anything else is 4. The ratio for education
is thus designated as follows: WEDUCATE / HEDUCATE for husbands and
HEDUCATE / WEDUCATE for wives.
For education, when used as a independent variable in the logistic
regression models, HEDUCATE and WEDUCATE are recoded into the
dichotomous variables, HCOLLEGE and WCOLLEGE, respectively. If the
husband or wife had more than 12 years of education, the value for
HCOLLEGE and WCOLLEGE is 1, if less than 12 years of education, the
value is equal to 0.
Lastly, hours in paid employment was measured using items that
asked how many hours each respondent worked the previous week in a paid
job. Also, measures were taken to determine if either spouse worked hours
at a second job; if so, those hours were summed with the regular job hours.
The hours worked for husbands is denoted by the variable HHOURS and
wives by the variable WHOURS. The ratio for this domain of the marriage is
43


thus, WHOURS / HHOURS for husbands and HHOURS / WHOURS for
wives.
The final result is two sets of ratios for each particular domain of the
marriage. The four separate ratios for husbands were then multiplied and for
wives ratios the same procedure was done. In other words,
HSPRUTIN
HROUTINE.
WINCOME\f-WEDUCATE] ( WHOURS
HINCOMEJ V HEDUCA TEJ V HHOURS.
equals the husbands
equity ratio and is represented by the variable EQUITY! The wifes total
equity ratio is the variable EQUITY3 and is determined by the same equation
as EQUITY! although the numerators and denominators are flip-flopped.
Taking EQUITY1 and subtracting from EQUITY3, the difference in scores will
be used as an indicator of the relationships objective equity, noted as
DFCOUP10. While the difference in some cases will be a negative value,
this would still show the same level of equity for the marriage as the
difference in scores that produce the same positive value. For analytical
purposes, all negative scores in this variable were converted to positive
scores and renamed DIFCOUP9. Summary measures were then taken to
determine the mean, median and percentile distribution of this equity
variable. The mean was shown to be 5.872, while the median score for
DIFCOUP9 was 1.718. The value at the 25th percentile was .678. This
44


value will represent the cut-off point for a high level of equity. Thus,
DIFCOUP9 was recoded as EQUITYHI when its value was less than or equal
to the 25th percentile value of .678. Once this is done, logistic regression
equations can be applied to examine relationships between the dependent
variables and independent variables.
In the full logistic regression model using HFAIRBTH as the
dependent variable, EQUITYHI will be the primary independent variable.
Also included in the model will be the variables HSHARE and WSHARE.
These two variables measure husbands and wives expectations of equity
regarding household tasks by asking whether or not they agree to the
statement, Working spouses should share household tasks equally. Other
independent variables to be used are HEDUCHI2, HHOURSHI, HHSWRKHI
and HWAGEHI2 which represent whether the husband has more education
than his wife (HEDUCHI2), works more at paid labor HHOURSHI, works
more at routine housework (HHSWRKHI) or earns more than his wife
(HWAGEHI2). WEDUCHI2, WHOURSHI, WHSWRKHI and WWAGEHI2
reflect the same values, albeit with the wife working, earning or having more
education than her husband. The remaining independent variables in the full
model are HHOURS, WHOURS, HINCOME, WINCOME, HROUTINE,
WROUTINE, HSPRUTIN, WSPRUTIN, HCOLLEGE, and WCOLLEGE. Child
45


care will be represented by HCHILD and WCHILD, representing any hours a
husband or wife spends in the caring and socialization of a child. Race will
be measured using dichotomous variables for both husbands and wives
ethnic categories. HRACE1 is for when the husband is White (White is 1, all
others are 0), HRACE2 is when the husband is Black (Black is 1, all others
are 0), HRACE3 for Hispanic husbands (Hispanic is 1, all others are 0), and
HRACE4 for when the husband identifies as being a member of other ethnic
groups (Other is 1, anything else is 0). Wives are categorized in the same
fashion. For logistic regression equations, HRACE1 and WRACE1 will be
left out of the equation. Any effects of the remaining categories on the
dependent variables will be interpreted by comparing with the White
category.
In an auxiliary model for HFAIRBTH, the variables HROUTINE,
HSPRUTIN, HSPRADIC, HSPSPRDC, HCHILD, HSHARE, HEDUCHI2,
HHOURSHl, HHSWRKHI, HWAGEHI2 and EQUITYHI will be used. The
intent of this model is to see what husband-related variables affect the
likelihood that the husband will perceive housework as fair to him and his
wife.
The same set of variables will be utilized in the full logistic regression
model for WFAIRBTH. Again, auxiliary models will be used to see what wife-
46


related variables affect whether or not the wife perceives housework as fair to
both.
Analysis of the well-being variables will use logistic regression models
as well. With HHAPPYHI, WHAPPYHI, HSELFHI, WSELFHI, HHELTHHI
and WHELTHHI as the dependent variable in 6 separate models, the same
independent variables as used in the first two models, with EQUITYHI,
HFAIRBTH, and WFAIRBTH as key dependent variables, will be used in the
regression equation. Using EQUITYHI, HFAIRBTH and WFAIRBTH will
provide a means to analyze whether objective levels of equity and spouses
perceptions of housework fairness affect either spouses well-being in
regards to the respective dependent variables.
47


CHAPTER 4
RESULTS
Logistic regression analyses were run for model 1 and model 2, which
are full regression models measuring the effects of objective equity and
various background, equity, and housework variables on spouses
perceptions of household fairness. Table 4.1 shows the results.
48


Table 4.1 Coefficients for Full Logistic Regression Models of Objective Equity
and Perceptions of Fairness (H = Husband, W = Wife)
Housework is Housework is
fair to both fair to both
INDEPENDENT VARIABLES (Husband) (Wife)
Couples objective equity .5047 .4149
H working spouses should share housework .7427** .3602
W working spouses should share housework .0449 -.8053*
Husband Variables
H has more than a high school education .3151 -.2302
H hours paid job -.0155 -.0022
H annual income -.0135 -.0008
H is Black (compared to White) .8271 -1.541
H is Hispanic (compared to White) -.3047 .3177
H is Other (compared to White) .6437 -.5815
H hours routine housework .0245 -.0384
H hours sporadic housework .0208 .0339
H spouse hours routine housework -.0068 .0112
H spouse hours sporadic housework . -.0263 -.0340*
H hours childcare, socialization .0075 -.0023
H has more education than W -.2702 .0374
H works more at paid job than W .3766 .1106
H perceives doing more housework than W -1.489** -.6797
H earns more than W .5760 .5299
Wife Variables
W has more than a high school education -.9258* -.3110
W hours paid job .0050 -.0025
W annual income .0021 -.0011
W is Black (compared to White) -.5386 .9169
W is Hispanic (compared to White) -.5411 -.4780
W is Other (compared to White) -1.002 .2215
W hours routine housework - 0393*** . 0464***
W hours sporadic housework -.0074 .0185
W spouse hours routine housework .0443 .1110***
Wspouse hours sporadic housework .0441* .0453*
W hours childcare, socialization -.0021 .0079
W has more education than H 1.294** .0708
W works more at paid job than H -.4506 .0051
Wperceives doing more housework than H .3560 .9400*
W earns more than H .2687 .8108*
(*p < .05, **p < .01, ***p < .001)
49


HFAIRBTH is the dependent variable for model 1 and WFAIRBTH is
the dependent variable for model 2. Based on the results, the first
hypothesis is not supported. The results for the models show that the
husbands and wives perceptions of the fairness of household labor are not
significantly affected by the level of measured equity in the marriage.
For the first model, other independent variables show significant
effects upon the dependent variable HFAIRBTH. If husbands believe that
working spouses should share equally in housework, then they perceive
housework as being fair to both spouses (B = .7427, p < .01). This finding is
somewhat interesting considering that the model also shows that if a
husband perceives he is doing more housework than his wife, he is not
inclined to perceive housework as fair to both spouses (B = -.1489, p < .01).
Model 1 also shows an affect for some wives variables on husbands
perceptions of housework fairness. A wifes education is significant in
relation to HFAIRBTH. If a wife has more than a high school education, her
husband is less likely to see housework as fair (B = -.9258, p < .05).
Further, if the wife has more education than her husband, HFAIRBTH is
positively affected (B = 1.294, p < .01).
The amount of routine and sporadic housework reported by the wife
for herself and her husband also has significant effects on HFAIRBTH. The
50


more routine housework the wife reported herself as performing decreases
the likelihood that her husband will judge housework as being fair (G = -
.0393, p < .001). If a wife reported her husband as doing more sporadic
housework, the husband was apt to judge housework as fair for both
(WSPSPRDC, ft =.0441, p < .05).
In model 2, wives attitude toward working spouses and housework
shows a significant negative effect upon their perceptions of fairness. If
wives believe that working spouses should share equally in housework, they
are less likely to judge housework as being fair to both (ft = -.8053, p < .05).
Also, if wives perceive themselves as doing more housework than their
husbands, then they are more likely to see housework as fair to both
(ft = .9400, p < .05).
Husbands reports sporadic housework performed their spouses show
significant effect upon WFAIRBTH. If the husband reported the wife as doing
less sporadic housework, then she was more likely to view housework as
being fair (ft = -.0340, p < .05).
For wives reporting the amount of hours spent on routine and
sporadic tasks for themselves and their spouses, significance was pretty
strong in regards to routine tasks. The less routine tasks a wife reported
herself as doing and the more she reported her husband as doing had
51


positive effects on WFAIRBTH (WROUTINE, 13. = -.0464, p < .001;
WSPRUTIN, IJ = .1110, p < .001). Additionally, the more hours a husband
spent in sporadic tasks, as reported by his wife, the more likely she was to
judge housework as fair (B = .0453, p < .05).
Finally, income showed significant affects upon WFAIRBTH. The
results for model 2 show that if the wife earns more than the husband, she is
likely to perceive housework as being fair (B = .8108, p < .05).
Models 1A and 2A depict the equation for objective equity and the
husband and wife perceived housework fairness variables. The summary of
results for both equations is depicted in Tables 4.2 and 4.3.
Table 4.2 Coefficients for Logistic Regression Model of Objective Equity and
Perceptions of Housework Fairness for Husband
(Equity and Housework Variables Only)
INDEPENDENT VARIABLES
Housework is fair to both
(Husband)
Couples' objective equity
H working spouses should share housework
.3774
.6807*
H hours routine housework
H hours sporadic housework
H spouse hours routine housework
H spouse hours sporadic housework
H hours childcare, socialization
.0355*
.0326*
-.0136
-.0296
.0020
H has more education than W
H works more at paid job than W
H perceives doing more housework than W
H earns more than W
-.1144
.1648
-1.279**
.1179
(*p < .05, **p < .01, ***p < .001)
52


Models 1A and 2A include only equity variables such as whether
either spouse believes working spouses should share household tasks
equally, the differences in spouses education, income, paid work and
household tasks hours as well as variables measuring the total number of
hours in sporadic and routine housework as perceived by husbands and
wives. In Table 4.2, Model 1A shows no support for objective equity and
husbands perceptions of housework fairness. However, a husbands belief
that working spouses should share the housework equally has a positive
affect in his judgment of housework (IS = .6807, p < .05).
The amount of routine and sporadic housework reported by husbands
for themselves and their wives is positively related to HFAIRBTH. The more
routine and sporadic housework a husband reports himself as accomplishing,
the more he is to judge housework as fair (HROUTINE, IS = .0355, p < .05,
HSPRADIC, IS = .0326, p < .05). Model 1A also predicts that if husbands
perceive themselves as doing less housework than their wives, they will
determine housework to be fair to both spouses (IS = -1.279, p < .01).
Model 2A does not support Hypothesis 1, but shows significance once
again for wives attitude towards working spouses and sharing housework.
Interpreting the results in Table 4.3, it appears that if wives do not believe
53


working spouses should share equally in housework, then they will see
housework as fair to both (IS = -.9108, p < .01).
Table 4.3 Coefficients for Logistic Regression Models of Objective Equity and
Perceptions of Housework Fairness for Wife
(Equity and Housework Variables Only)
Housework is fair to both
INDEPENDENT VARIABLES (Wife)
Couples' objective equity .2889
W-worlcing spouses should share housework -.9108**
W hours routine housework -.0356***
W hours sporadic housework .0091
W spouse hours routine housework 0793***
W spouse hours sporadic housework .0551**
W hours childcare, socialization .0051
W has more education than H .0364
W works more at paid job than H -.1342
Wperceives doing more housework than H 1.119**
W earns more than H .2695
(*p < .05, **p < .01, ***p < .001)
Wives reports of routine and sporadic housework for both themselves
and their spouses had very significant effects upon how they judged
housework fairness. Any decrease in the amount of routine tasks a wife
reported herself as doing, the likelihood that she would view housework as
fair to both increases (IS = -.0356, p < .001). Moreover, if wives report their
husbands as doing more routine and sporadic household tasks, they are
54


prone to seeing housework as being fair (WSPRUTIN, B = .0793, p < .001;
WSPSPRDC, B = .0551, p < .01). However, if wives report themselves as
doing more routine housework than husbands, their judgment of housework
as being fair increases (ft = 1.119, p < .01).
In the analysis of objective equity on spouses well-being, Models 3, 4,
and 5 measure the effects of both high levels of couples objective equity and
spouses subjective perceptions of housework fairness on husbands levels
of well-being. Model 3 has the husbands overall happiness as the
dependent variable (HHAPPYHI), model 4 has HHELTHHI as the dependent
variable measuring the husbands self-evaluation of overall health and
model 5 has husbands high level of self-esteem (HSELFHI) as the
dependent variable. Models 6, 7 and 8 reflect the measurement of effects of
objective equity and spouses perceptions of fairness with wives overall well-
being using WHAPPYHI, WSELFHI and WHELTHHI as dependent variables.
The results for all models are summarized in Tables 4.4 and 4.5.
Looking at table 4.4, it appears that a high level of objective equity has
no effect whatsoever upon the husbands overall well-being. Therefore,
Hypotheses 2, 3, and 4 were not supported.
55


Table 4.4 Coefficients for Full Logistic Regression Model of Indicators of Husbands
Well-Being
Husband is Husband is Husband has
INDEPENDENT VARIABLES happy healthy high self-esteem
Couples' objective equity -.5128 .7002 .4550
Housework is fair to both (Husband) -.1160 .2639 -.2186
Housework is fair to both (Wife) .6362* .1745 .5035
H working spouses should share housework .8007* .5507 .3997
W working spouses should share housework -.4847 .7577 -.2443
Husband Variables
H has more than a high school education .8567 .4445 .2003
H hours paid job -.0229 -.0122 .0006
H annual income .0079 .0279 .0075
H is Black (compared to White) -1.292 -.3667 5.793
H is Hispanic (compared to White) .2949 -.7638 .5926
H is Other (compared to White) 6.221 6.530 -.5577
H hours routine housework .0328 .0153 -.0133
H hours sporadic housework .0021 .0098 .0099
H spouse hours routine housework -.0038 -.0001 .0008
H spouse hours sporadic housework -.0075 -.0417 .0050
H hours childcare, socialization .0110 .0054 -.0119
H has more education than W -.2446 .1119 -.1248
H works more at paid job than W .1351 .0341 .0204
H perceives doing more housework than W -.1082 -.9216 -.5240
H earns more than W .2793 -.5974 -.1686
Wife Variables
W has more than a high school education -.5063 .2912 .3023
W hours paid job .0286 .0191 -.0124
W annual income -.0153 -.0318* .0006
W is Black (compared to White) 1.372 1.267 -6.385
W is Hispanic (compared to White) -.3905 7.994 -.5673
W is Other (compared to White) -1.078 6.472 -.0910
W hours routine housework -.0019 .0123 -.0033
W hours sporadic housework .0076 .0243 -.0157
W spouse hours routine housework -.0355 -.0391 .0050
W spouse hours sporadic housework .0034 -.0005 .0198
W hours childcare, socialization -.0142 .0047 -.0065
W has more education than H .6826 -.0438 .0526
W works more at paid job than H -.1485 -.8097 .5066
Wperceives doing more housework than H -.8437 .1949 .1430
W earns more than H .4146 -.2066 -.2963
(*p < .05, **p < .01, ***p < .001)
56


Husbands happiness is positively affected by wives perceptions of
housework fairness. Interpreting the results, if a wife believes that
housework is fair to both, the husbands reported level of overall happiness
increases (B = .6362, p < .05,). Additionally, if husbands believe that working
spouses should share equally in household tasks, then husbands level of
overall happiness is increased (B = .8007, p < .05).
The only predictor of husbands self-evaluation of health is the wives
annual income. According to Table 4.4, any increase in a wifes annual
income has a negative effect upon a husbands self-evaluation of health
(B = -.0318, p < .05).
Table 4.5 shows the results of logistic regression analyses for models
6, 7, and 9 using couples objective equity and spouses perceptions of
housework fairness as the key predictor variables with variables of self-
esteem, overall happiness and health for wives. Wives perceptions of
housework fairness has significant effects upon evaluations of overall health
and happiness. If a wife perceives housework as fair to both, she is more
likely to report her health as better compared to others her own age
(B = .7069, p < .05). Also, her overall happiness is more likely to be positive
if she views housework as being fair to both (B = 1.344, p < .001).
57


Table 4.5 Coefficients for Full Logistic Regression Model of Indicators of Wifes
Well-Being
Wife is Wife is Wife has high
happy healthy self-esteem
INDEPENDENT VARIABLES
Couples objective equity .7202 .3334 .1940
Housework is fair to both (Husband) -.1656 -.0292 -.1299
Housework is fair to both (Wife) 1.344*** .7069* .2983
H working spouses should share housework -.5882 -1.159 .0764
W- working spouses should share housework -.4737 -.1060 -.3071
Husband Variables
H has more than a high school education .5360 .4141 .8550
H hours paid job -.0230 .0112 -.0139
H annual income .0097 .0272 .0141
H is Black (compared to White) -.9052 1.118 .1406
H is Hispanic (compared to White) -.0742 .8710 .3979
H is Other (compared to White) -7.885 -1.254 .1167
H hours routine housework .0159 -.0960*** .0227
H hours sporadic housework -.0073 .0350 -.0009
H spouse hours routine housework -.0185 .0058 .0013
H spouse hours sporadic housework .0081 .0114 -.0338
H hours childcare, socialization -.0182 -.0286 -.0174
H has more education than W -.6287 -.9864* -.9030*
H works more at paid job than W -.0274 -.9477 -.0329
Hperceives doing more housework than W -.7926 .8404 -.1005
H earns more than W -.4953 .5282 -.5523
Wife Variables
W has more than a high school education -.3361 .1407 -.0435
W hours paid job -.0046 .0193 -.0093
W annual income .0275 .0061 .0084
W is Black (compared to White) .3146 -1.124 -.0418
W is Hispanic (compared to White) .4750 -1.221 -.7718
W is Other (compared to White) 11.902 .0141 5.978
W hours routine housework .0081 -.0108 -.0183
W hours sporadic housework -.0005 -.0298 .0424
W spouse hours routine housework -.0169 .0659* -.0031
W spouse hours sporadic housework -.0342 -.0364 .0005
W hours childcare, socialization .0088 -.0013 -.0019
W has more education than H .3942 -.3869 .1020
W works more at paid job than H -.1278 -1.144 -.2432
Wperceives doing more housework than H .4536 1.876*** .7453
W earns more than H -.2873 .5196 .7697
(*p < .05, **p < .01, ***p < .001)
58


Other factors contribute to wives evaluating themselves as being
healthy compared to others their own age. If a wife reports her husband as
doing more routine housework, she is likely to claim herself to be healthy
(B = .0659, p < .05). The less housework a husband reports himself as
doing also contributes to a wifes positive assessment of her health
(B = -.0960, p < .001). Table 4.5 also shows that if a wife believes she is
doing more housework than her husband, she is likely to report her health as
being better than others her age (B = 1.876, p < .001). Wives health is also
affected positively when husbands report fewer years of formal education
than their wives (B = -.9864, p < .05).
The only predictor of wives self-esteem in the full equation of model 6
is the variable comparing a husbands education with that of his wife. If a
husband had fewer years of education than his wife, she was more likely to
report higher levels of self-esteem (B = -.9030, p < .05).
Models 3A, 4A, and 5A are auxiliary models of measuring the effects
of objective equity and perceptions of housework fairness on indicators of
husbands well-being. The models include only equity and housework
variables for husbands and the results are depicted in Table 4.6.
According to the table, none of the hypotheses are supported. For all
three models high objective equity and husbands perceptions of housework
59


fairness have no effect on husbands well-being. The only predictor of the
well-being indicators is husbands attitudes about working spouses and
housework sharing. If a husband believes working spouses should share
household tasks equally, he will report himself to be happier overall (ft =
.8552, p< .01).
Table 4.6 Coefficients for Logistic Regression Model of Indicators of Husbands
Well-Being (Housework and Equity Variables Only)
Husband is Husband is Husband has
happy healthy high
INDEPENDENT VARIABLES self-esteem
Couples' objective equity -.4157 .5448 5127
Housework is fair to both (Husband) .0463 .2075 -.1143
H working spouses should share housework .8552** .6082 .4576
H hours routine housework .0199 -.0020 -.0066
H hours sporadic housework .0041 .0188 .0186
H spouse hours routine housework -.0006 -.0048 -.0031
H spouse hours sporadic housework -.0121 -.0294 -.0075
H hours childcare, socialization .0008 .0081 -.0175
H has more education than W .2457 .3723 -.0129
H works more at paid job than W -.2327 .2604 -.0677
H perceives doing more housework than W ,0595 -.9247 -.6746
H earns more than W .2789 .2976 .0117
(*p < .05, **p < .01, ***p < .001)
Table 4.7 shows the results of models 6A, 7A and 8A which use only
wives housework and equity variables and measures the effects of objective
60


equity and wives perceptions of housework fairness on indicators of wives
well-being.
Table 4.7 Coefficients for Logistic Regression Model of Indicators of Wifes
Well-Being (Housework and Equity Variables Only)
Wife is Wife is Wife has high
happy healthy self-esteem
INDEPENDENT VARIABLES
Couples objective equity .6987 -.0774 .2106
Housework is fair to both (Wife) 1 195*** .6885* .2006
W working spouses should share housework -.4322 -.1948 -.0788
W hours routine housework .0017 -.0097 -.0224*
W hours sporadic housework -.0087 -.0182 .0140
W spouse hours routine housework -.0069 .0006 .0174
W spouse hours sporadic housework -.0415* -.0177 -.0038
W hours childcare, socialization -.0084 -.0168 -.0099
W has more education than H .1530 -.2816 -.0420
W works more at paid job than H .2368 -.4261 -.0915
Wperceives doing more housework than H .5193 1.550*** .7575
W earns more than H .2387 -.1233 1.154**
(*p < .05, **p < .01, ***p < .001)
Models 6A and 7A produced support for Hypotheses 5 and 6, showing
that perceptions of fairness in housework positively affect a spouses overall
happiness and health. As depicted in the table, if a wife determines that
housework is fair to both spouses, she will describe herself as happy overall
(B = 1.195, p < .001) and her health as better than others her same age
(B = .6885, p < .05). A wifes happiness is also predicted significantly by her
61


reports of her husbands hours spent doing sporadic tasks. The fewer
sporadic tasks performed by her husband, the more likely she is to report
herself as happy overall (IS = -.0415, p < .05).
A wifes overall health is predicted positively by the view that she does
more housework than her husband. If a wife reports doing more housework
than her husband, she will evaluate herself as being healthier than others her
own age (IS = 1.550, p < .001).
For the third indicator of well-being, model 7A provides two variables
that significantly predict a wifes level of self-esteem. The total amount of
hours spent in routine housework by wives has a negative affect upon her
self-esteem. The more hours wives spend performing routine household
tasks, the lower self-esteem will be recorded for them (IS = -.0224, p < .05).
Also, if wives are recorded as earning more than their husbands they will
reflect higher levels of self-esteem (IS = 1.154, p < .01).
62


CHAPTER 5
DISCUSSION
One of the key elements of this study is to offer new insights into the
determination of housework fairness by accounting for both spouses
assessments rather than previous studies that focused only on womens
perceptions of unfairness (Demaris & Longmore, 1996). When both spouses
are working, neither spouse can be said to be the sole breadwinner or
homemaker. This implores that judging fairness of housework
contributions should include assessing contributions in other aspects of the
marriage. Spousal income, education, and hours spent in paid labor are
areas in which spouses inputs are assumed to effect judgments of housework
fairness. These areas are included in the overall equity formula of this study
and are also measured separately along with variables such as childcare and
race for effects on the dependent variable of perceptions of housework
fairness.
The results of this study reveal that working spouses judgment of
fairness regarding household labor is not significantly related to the level of
measured equity in the marriage. In other words, the contributions of both
spouses in the areas of housework, income, education, or paid labor has no
bearing whatsoever on how they perceive the fairness of household labor.
63


The lack of significant effects of objective equity on perceptions of
fairness in housework could be attributed mainly to an inconsistency between
the independent and dependent variables. The formula for objective equity
accounts for contributions by both spouses in four domains of the household:
housework, income, education, and paid labor. Yet the dependent variable
refers only to perceptions of fairness in relation to housework. Perhaps
future studies would be more suited to measure perceptions of fairness in a
more complete manner, encompassing other domains
Identifying predictors of perceptions of fairness in regard to
housework, husbands perceive housework as fair if their wives have less
than a high school education, or if the wives have more education than the
husbands. Also, husbands who believe that working spouses should share
household tasks equally are more likely to judge housework as fair, but if
they feel that they are doing more than their wives, they are likely to see
housework as being unfair to both.
Interestingly, wives showed the same positive effect for the variable
concerning belief in housework sharing and judging fairness in housework.
However, when wives reported doing more housework than their husbands,
they judged housework as being fair to both as opposed to the men who felt
housework was not fair. This may suggest that traditional gender ideologies
are still in play for both men and women. For men, the belief in sharing
64


housework may be ideal, but once inequity is revealed, the judgment of
fairness is easily swayed. For women, performing more housework than men
may still be acceptable and have a lesser effect on the perception that
contributions in housework are unfair to both.
In judging fairness in housework, the most important variables the
amount of hours contributed by either spouse in household tasks. The
results of this study show that this factor is particularly relevant to wives.
When wives report that they are doing less routine housework or if they
report their husbands as doing more routine and sporadic housework, they
are likely to judge housework as fair. However, they will perceive
contributions in housework as fair if they report doing more routine
housework than their husbands.
For husbands, no significant coefficients were found to have effects on
their (and their wives) reported housework contributions and how they judged
housework. The results seem to reveal a conscious evaluation by women
when it comes to judging housework ; who does what and how much is done
is taken sincerely into consideration when determining what is fair.
Perceptions of fairness regarding housework on the husbands part may take
little more than stating that it is fair or unfair to both without considering
closely what gets done and who does it.
65


When looking at secondary models regarding perceptions of fairness
in housework, the results again show no effects whatsoever for the objective
equity variable for both husbands and wives. However, reports of housework
contributions remain significant predictors perceptions of fairness for
housework for both husbands and wives.
The model presenting only husband variables shows that belief in
sharing housework for working spouses remains a factor in determining
housework fairness for husbands. For this model, husbands routine and
sporadic housework contributions have positive effects on whether or not
housework is seen as fair by the husband. Unlike the results of the full
model, this provides a different image of husbands as contributing more in
terms of household tasks while at the same time feeling that housework is fair
to both. Yet, when husbands report that they are doing more routine tasks
than their wives, they tend to judge housework as unfair to both.
For the wives, the results appear to follow the ones of the husbands.
Unlike the husbands, though, wives belief in sharing housework for working
spouses has a negative effect on how they perceive the fairness of
housework. The wives results also reveal that the less routine housework
they do and the more they report for their husbands positively affects their
perceptions of housework fairness. Opposed to that, however, is the fact that
if they perceive their husbands to be doing less than themselves around the
66


house, they still judge housework as fair. The discrepancy in these factors
may attributable once again to traditional gender ideologies. For both
husbands and wives the accepted expectation seems to persist that it is okay
for the wives to do more housework and the husbands to do less. In either
case housework is deemed fair by both spouses.
Looking at the well-being variables for both husbands and wives,
some predictors can be found that have an effect to some degree for either of
these spouses health, happiness, and self-esteem. The objective equity of a
marriage has no effect upon husbands or wives levels of well-being. Thus,
Hypotheses 2, 3, and 4 are not supported which assumed causal effects of
objective equity in the marriage upon either spouses levels of happiness,
health, and self-esteem, respectively.
Wives perceptions of housework as fair to both spouses has positive
effects on wives happiness and health. At least in the wives cases, this
supports Hypotheses 5 and 6, which assume that perceptions of fairness in
housework have positive effects upon either spouses happiness and health,
respectively.
Other predictors of indicators of well-being for wives are revealed by
this study. Wives report better health when they feel their husbands are
doing more routine tasks around the household or if they feel they are doing
more routine housework relative to their husbands contributions. Wives
67


health is also conditional upon husbands education; if husbands have fewer
years of formal education than their wives, wives are likely to report better
overall health. And, if husbands report performing fewer hours of routine
housework, wives health is positively affected. Levels of self-esteem for
wives is also affected by education wherein if husbands have fewer years of
formal education than wives, wives report better overall health.
Husbands well-being does not seem to be affected by many factors.
None of the hypotheses pertaining to the effects of objective equity and
perceptions of housework fairness upon levels of happiness, health, and self-
esteem for husbands found support in this study. Husbands reports of
happiness is affected by husbands belief that working spouses should share
equally in household tasks and wives perceptions that housework is fair to
both.
Husbands reports of overall health is positively affected by wives
annual income. The more annual income reported by wives, husbands self-
reports of being health tend to decrease. The relationship between wives
income and husbands health is unclear, but one might speculate that
households with healthy husbands may lean more toward the traditional role
of husbands as breadwinner. Increased health for husbands implies
increased ability to work and earn more, supplanting the relevance of the
wives income to the household. Perhaps when husbands are not as healthy,
68


the need for wives to work and earn more becomes essential to the
household.
Running logistic regression on just the wives housework and equity
variables with indicators of well-being most of the predictors hold true. Once
again, wives perception that housework is fair to both has positive effects on
both overall happiness and health for wives. Fewer hours in sporadic
household labor for husbands as reported by the wives also contributes
positively to happiness. Self-esteem is supported by fewer hours for wives in
routine household tasks and along with reports that wives earn more at paid
labor than their husbands. However, wives perceptions that they are doing
more routine housework than their husbands positively affects wives self-
reports of better overall health. Going back to the notion of gender
ideologies, it would seem pertinent to think that adequate performance of
certain roles would contribute somewhat to feelings of self and health. This,
of course, is dependent upon a persons acceptance and expectations of the
specific roles.
The husbands secondary models of objective equity and perceptions
of fairness affecting well-being shows positive effects only upon self-reports
of husbands haappiness. If husbands hold the belief that working spouses
should share equally in household tasks, then their overall happiness tends
to increase. No support is shown for any of the hypotheses regarding
69


objective equity and perceptions of housework fairness with husbands
indicators of well-being.
Limitations and Suggestions
Perhaps the most glaring limitation of this study is the use of the
perceived fairness of housework variable in relation to a formula of a
marriage overall objective equity. The formula used in this study incorporated
contributions across differing domains of marriage, specifically, spousal
income, education, hours in paid labor, and hours spent performing routine
household tasks. While this gives the researcher an ample view of equity in
the marriage, the data set used for this study did not provide perceptions by
spouses in these specific areas separately. It would be interesting to see
what results would be obtained with the equity formula and perceptions of
fairness model if those perceptions were to include spousal perceptions of
fiarness regarding education, income, and paid labor along with perceptions
of household tasks.
One other limitation to this study is the decision to forego any
measurement of gender beliefs for husbans and wives in relation to
perceptions of fairness and objective equity. As the results show, many
interesting questions are posed without benefit of an expplanation by way of
how men and women feel about the traditional roles of a marriage. While the
70


group under study consisted of married, working couples, it is not so easy to
dismiss the traditional expectations of breadwinner and homemaker for
contemporary men and women. As shown in Hochschild and Machungs
(1989) study, gender expectations of married spouses tend to follow tradional
beliefs regardless of the non-traditional situation of a dual-earner family.
Bearing this in mind, it is advisable that future research in this area
broaden the notions of equity and perceived fiarness. The continued use of
an objective equity formula in gauging perceptions of fairness is futile if all
that is purported to be measured objectively is not fully measured
subjectively. Once the relationship between the two is consistent, the
assumptions to be made about either in terms of other variables would most
likely find stronger support under appropriate models of statistical analysis.
71


REFERENCES
Acock, Alan and David Demo. (1994). Family Diversity and Well-Being.
Sage:Thousand Oaks, CA.
Adams, J. Stacy. (1965). Inequity in Social Exchange. Advances in
Experimental Social Psychology 2:267-300.
Adams, J. Stacy and Sara Freeman. (1976). Equity Theory Revisited.
Advances in Experimental Social Psychology 9:43-90.
Blair, Samuel and Michael Johnson. (1992). Wives Perception of the
Fairness of the Division of Household Labor: the Intersection of
Housework and Ideology. Journal of Marriage and Family 54:570-581.
Benin, Mary and Joan Agostinelli. (1988). Husbandsand Wives
Satisfaction with the Division of Household Labor. Journal of Marriage
and the Family 50:349-361.
Coltrane, Scott and Masako Ishii-Kuntz. (1992). Mens Housework: A Life
Course Perspective. Journal of Marriage and the Family 54: 43-57.
Cook, Karen and Karen Hegvedt. (1983). Distributive Justice, Equity and
Equality. Annual Review of Sociology 9:217-241.
DeMaris, Alfred and Monica Longmore. (1996). Ideology, Power and
Equity: Testing Competing Explanations for the Perceptions of Fairness in
Household Labor. Social Forces 74:1043-1071.
Ferree, Myra Marx. (1991). The Gender Division of Labor in Dual-Earner
Marriages. Journal of Family Issues 12:158-180.
Glazer-Malbin, Nona. (1976). Housework. Signs 1:905-922.
Glenn, Norval & Charles Weaver. (1988). The Changing Relationship of
Marital Status to Reported Happiness. Journal of Marriage and the
Family 50: 317-324
72


Gove, Walter, Michael Hughes and Carolyn Briggs Style. (1983). Does
Marriage Have Positive Effects on the Psychological Well-Being of the
Individual? Journal of Health and Social Behavior 24:122-131.
Gove, Walter, Carolyn Briggs Style and Michael Hughes. (1990). The
Effect of Marriage on the Well-Being of Adults. Journal of Family Issues
11:4-35.
Greenstein, Theodore. (1996). Gender Ideology and Perceptions of the
Fairness of the Division of Household Labor: Effects on Marital Quality.
Social Forces 74:1029-1042.
_______(1996). Husbands Participation in Domestic Labor: Interactive
Effects of Wives and Husbands Gender Ideologies. Journal of Marriage
and the Family 58:585-595.
Harris, Richard (1983). Pinning Down the Equity Formula. Pp. 207-241
in Equity Theory. D. Messick & K. Cook (Eds.). New York: Prager.
Hatfield, Elaine, Jane Traupmann, Susan Sprecher, Mary Utne and Julia
Hay. (1984). Equity and Intimate Relations: Recent Research. Pp. 91-
118 in W. Ickes (ed.), Compatible and Incompatible Relationships. New
York: Springer-Verlag.
Hiller, Dana and William Philliber. (1986). The Division of Labor in
Contemporary Marriage: Expectations, Perceptions and Performance.
Social Problems 33:191-201.
Hochschild, Arlie with Anne Machung. (1989). The Second Shift:
Working Parents and the Revolution at Home. New York: Viking.
Homans, George. (1961). Social Behavior: Its Elementary Forms. New
York: Harcourt Brace.
Ishii-Kuntz, Masako and Scott Coltrane. (1992). Remarriage,
Stepparenting and Household Labor. Journal of Family Issues 13:215-
233.
73


Kollock, Peter, Philip Blumstein and Pepper Schwartz. (1994). The
Judgement of Equity in Intimate Relationships. Social Psychology
Quarterly 57:340-351.
Lawler, E.E., III. (1968). Equity Theory as a predictor of productivity and
work quality. Psychological Bulletin. 70:596-610.
Lennon, Mary C. & Sarah Rosenfield. (1994). Relative Fairness and the
Division of Housework: The Importance of Options. American Journal of
Sociology 100: 506-531.
Matthews, Lisa, Rand D. Conger and K.A.S. Wickrama. (1996). Work-
Family Conflict and Marital Quality: Mediating Processes. Social
Psychology Quarterly 59:62-79.
McCIintock, Charles G. And Linda Keil. (1982). Equity and Social
Exchange. Pp. 337-387 in J. Greenberg and R. Cohen (eds.), Eguitv and
Justice in Social Behavior. Academic Press: New York.
Mederer, Helen. (1993). Division of Labor in Two-Earner Homes: Task
Accomplishment Versus Household Management as Critical Variables in
Perceptions About Family Work. Journal of Marriage and the Family
55:133-145.
Mirowsky, John. (1985). Depression and Marital Power: An Equity
Model. American Journal of Sociology 91:557-592.
Mookherjee, Harsha. (1997). Marital Status, Gender, and Perception of
Well-Being. The Journal of Social Psychology 137:95-105.
Presser, Harriet. (1994). Employment Schedules among Dual-Earner
Spouses and the Division of Household Labor by Gender. American
Sociological Review
59:348-364.
Rogers, Stacy J. (1996). Mothers Work Hours and Marital Quality:
Variations by Family Strucutre and Family Size. Journal of Marriage and
the Family 58:606-617.
74


Rosenberg, Morris. (1979). Conceiving the Self. New York: Basic
Books.
Ross, Catherine, John Mirowsky and Karen Goldsteen. (1990). The
Impact of the Family on Health: The Decade in Review. Journal of
Marriage and the Family 52:1059-1078.
South, Scott and Glenna Spitze. (1994). Housework in Marital and
Nonmarital Households. American Sociological Review 59:327-347.
Sprecher, Susan. (1986). The Relation Between Inequity and Emotions
in Close Relationships. Social Psychology Quarterly 49:309-321.
Stafford, Rebecca, Elaine Backman and Pamela DiBona. (1977). The
Division of Labor Among Cohabiting and Married Couples. Journal of
Marriage and the Family 39:43-57.
Thibault, John and Harold Kelley. (1959). The Social Psychology of
Groups. New York: John Wiley & Sons, Inc.
Thompson, Linda. (1991). Family Work: Womens Sense of Fairness.
Journal of Family Issues 12:181-196.
Thompson, Linda & Alexis Walker. (1989). Gender in Families: Women
and Men in Marriage, Work, and Parenthood. Journal of Marriage and
the Family 51:845-871.
Walster, Elaine, Ellen Berschied and G. William Walster. (1976) New
Directions in Equity Research. Advances in Experimental Social
Psychology 9:1 -42.
Ward, Russell. (1993). Marital Happiness and Household Equity in Later
Life. Journal of Marriage and the Family 55:427-438.
White, James. (1992). Marital Status and Well-Being in Canada.
Journal of Family Issues 13:390-409.
Williams, Dorie Giles. (1988). Gender, Marriage and Psychosocial
Well-Being. Journal of Family Issues 9: 452-468.
75