Chores Division 2007 [618576]

ORIGINAL ARTICLE
Changing the Division of Household Labor:
A Negotiated Process Between Partners
Clelia Anna Mannino &Francine M. Deutsch
Published online: 28 February 2007
#Springer Science + Business Media, LLC 2007
Abstract This short-term longitudinal study expands on
previous theoretical approaches, as we examined howwomen ’s assertiveness and the strategies they use to elicit
more household labor from husbands help to explain the
division of labor and how it changes. Participants included 81married women with 3- and 4-year-old children who
completed two telephone interviews, approximately 2 months
apart. Results based on quantitative and qualitative analysesshow that (a) relative resource, structural, and gender ideology
variables predicted the division of housework, but not child-
care; (b) assertive women were closer to their ideal division ofchildcare than nonassertive women; (c) women who made a
larger proportion of family income were less assertive about
household labor than other women, but when they wereassertive, they had a more equal division of childcare; (d)
women who earned the majority of their household ’si n c o m e
showed the least change; and (e) the nature of women ’s
attempts to elicit change may be critical to their success.
Keywords Division of labor .Childcare .Household laborDespite the amelioration in women ’s political and economic
rights and their increased presence in the paid workforce sincethe 1960s, household labor and childcare remain divided
along traditionally gendered lines (e.g., Coltrane, 2000 ;
Thompson & Walker, 1989 ). Past researchers have examined
what factors influence and maintain this gendered division of
labor, and have noted the importance of everyday interactions
between husbands and wives in creating this inequitabledivision (e.g., Coltrane, 1990 ,2000 ;D e u t s c h , 1999 ;E n g l a n d
& Farkas, 1986 ; Ferree, 1990 ,1991 ; Hood, 1983 ; Risman &
Johnson-Sumerford, 1998 ; Szinovacz, 1987 ;T h o m p s o n&
Walker, 1989 ). Researchers have also begun to explore the
relationships between conflict, negotiation, and the division
of labor (Coltrane, 1989 ; Deutsch, 1999 ; Hood, 1983 ;
Risman & Johnson-Sumerford, 1998 ). However, further
examination is now needed of how specific strategies in
couples ’conflicts affect their division of labor and the
likelihood of their changing it. In the current study, we
explored these connections further.
Evidence for the Persistence of a Gendered Division
of Labor
Numerous studies document the persistence of a gendered
division of domestic labor (e.g., Coltrane, 2000 ;D e m p s e y ,
1997 ; England & Farkas, 1986 ; Ferree, 1990 ,1991 ;K l u w e r ,
Heesink, & Van de Vliert, 1997 ; Presser, 1994 ; Thompson &
Walker, 1989 ). In most couples, wives spend much more
time on household tasks than their husbands do (e.g.,
Coltrane, 2000 ;D e m p s e y , 1997 ,2002 ; Johnson & Huston,
1998 ; Kluwer, Heesink, & Van de Vliert, 1996 ,2000 ;
Presser, 1994 ;T h o m p s o n&W a l k e r , 1989 ). Even employed
wives assume the larger portion of household chores (e.g.,Dempsey, 2002 ) during what Hochschild ( 1989 )d u b b e dSex Roles (2007) 56:309 –324
DOI 10.1007/s11199-006-9181-1
This research is based on an undergraduate honors thesis at Mount
Holyoke College, conducted by Clelia Anna Mannino under the
direction of Francine M. Deutsch. Earlier versions of this article werepresented at the Eastern Psychological Association 2005 meeting in
Boston, and at the 2005 Association of Women in Psychology
conference in Tampa.
F. M. Deutsch ( *)
Department of Psychology and Education,Mount Holyoke College, 50 College Street,
South Hadley, MA 01075, USA
e-mail: fdeutsch@mtholyoke.edu
Present address:
C. A. Mannino
Department of Psychology, University of Minnesota,Minneapolis, MN, USA

their second shift. Not only do women retain responsibility
for completing the childcare and household duties (e.g.,
Coltrane, 2000 ; Dempsey, 2002 ; Thompson & Walker,
1989 ), but they also manage, plan, organize, and supervise
them. Women usually perform the most necessary and time-
consuming tasks on a daily basis. Men may help with thesetypical women ’s tasks but rarely assume equal daily
responsibility. Usually, men take on the responsibility only
for typical men ’s outdoor tasks that need to be completed
less frequently, such as mowing the lawn (Coltrane, 2000 ;
Ferree, 1990 ; Thompson & Walker, 1989 ).
Prominent theories
A number of theoretical perspectives have been proposed to
account for the gendered division of labor. The most
prominent among them are the resources, the structural
factors, the gender ideology, and the gender constructionapproaches (e.g., Berk, 1985 ; Coltrane, 2000 ;E n g l a n d&
Farkas, 1986 ;K l u w e re ta l . , 2000 ;P r e s s e r , 1994 ). Classic
resource theories propose that relative resources are a keydeterminant of how household labor is divided (e.g., Ferree,
1991 ). This approach suggests that a spouse ’se x t e r n a l
resources, such as income, education, and occupational
status, confer power. Thus, the spouse with the greater
resources also has the greater decision-making power to puthis/her wishes into practice (e.g., Coltrane, 2000 ; Johnson &
Huston, 1998 ; Presser, 1994 ), and that power is used to
reduce his/her own share of domestic labor. Thus, the greateras p o u s e ’s share of relative resources, the smaller his/her
share of housework and childcare (e.g., Coverman, 1985 ;
Deutsch, Lussier, & Servis, 1993 ; Presser, 1994 ).
The structural factors approach examines how personal and
family characteristics, such as the amount of time a wife works
outside the home as well as the number and ages of childrenpresent in the home, influence the division of household labor.
The demand-response model, for example, asserts that the
more structural demands placed on a spouse to participate inhousehold labor and the more time s/he has available to do so,
the greater the amount of household labor s/he performs (e.g.,
Coverman, 1985 ; Deutsch et al., 1993 ; Presser, 1994 ).
A third perspective, the gender ideology model, proposes
that attitudes toward gender roles are responsible for the
division of household labor (e.g., Coverman, 1985; Deutsch et
al.,1993;K l u w e re ta l . , 1997,2000) .M e na n dw o m e nw h o
have more egalitarian ideologies tend also to have more equal
divisions of labor (e.g., Shelton & John, 1996). The more
traditional husbands ’and wives ’gender ideologies, the less
husbands will participate in housework (e.g., Coverman, 1985).
The resources, structural factors, and gender ideology
perspectives typically ignore that the division of labor is
actively negotiated between spouses on a continual and daily
basis (Kluwer et al., 2000 ; Pittman, Solheim, & Blanchard,1996 ). Instead, these theories explain the gendered division
of labor as though it were static and changed little over time
(Pittman et al., 1996 ). In the last two decades, gender
construction theories have emerged to offer a more interac-
tional outlook. Gender construction theories propose that
men and women engage in different household tasks todemonstrate and reaffirm their gendered selves. The ongoing
and daily interactions concer ning the division of household
labor comprise a context in which gendered behavior asmasculine or feminine is created, maintained, and renegotiated
(Berk, 1985; Coltrane, 1989,2000; Fenstermaker, West, &
Zimmerman, 1991 ; Ferree, 1990 ; Kluwer et al., 2000 ;
Osmond & Thorne, 1993;R i s m a n , 2004; Thompson &
Walker, 1989).
In line with the gender construction theory, several
qualitative studies have been conducted to examine the
division of household labor through couples ’negotiations.
Hood ( 1983 ) examined couples ’negotiations over an 8-year
period to explore how and why families change after wives
take on paid employment. Coltrane ( 1989 ,1990 ) explored
how dual-earner couples discussed sharing childcare andhousework and how they negotiated these tasks. In his 1989
study, Coltrane found evidence for change in the meaning
of gender; when household labor was shared equally,
fathers developed maternal thinking. Likewise, Deutsch
(1999 ) explored couples ’retrospective accounts of the
interactions and discussions that enabled them to share
childcare equally. Risman and Johnson-Sumerford ( 1998 )
also interviewed “post-gender ”couples to determine how
they had developed the arrangement of sharing household
tasks equally. They concluded that the interactions of these
unique couples were characterized by egalitarian friend-ships and guided by fairness and sharing principles.
The gender construction approach has proved helpful in
addressing several limitations of previous perspectives(Ferree, 1990 ): specifically, why inconsistencies exist
between earlier theories and empirical findings, and why a
gendered division of labor persists. Even when womenmake as much money as men, work as many hours, and
believe in liberal gender ideology, they still do more
domestic labor than men do. In fact, more income cansometimes lead women to do more housework. Although a
woman ’s share of housework decreases when her share of
household income increases, as would be predicted byresource theory, once her income exceeds her husband ’s,
some studies have shown that her share of household labor
rises again (Bittman, England, Folbre, Sayer, & Matheson,2003 ; Brines, 1994 ; Ferree, 1990 ; Tichenor,
1999 ). Gender
construction theorists explain this curvilinear relationship as
couples ’attempts to reduce the threat to the husbands ’
masculinity and to reaffirm the wives ’femininity in the face
of their “masculine ”income-generating behavior (Bittman
et al., 2003 ; Brines, 1994 ; Ferree, 1990 ; Thompson &310 Sex Roles (2007) 56:309 –324

Walker, 1989 ; Tichenor, 1999 ). These women may judge
their contributions and success as a wife and mother by
how much work they do inside the home (Ferree, 1990 ;
Thompson & Walker, 1989 ; Tichenor, 1999 ).
Moreover, by eschewing power, wives can avoid marital
conflict. Brines and Joyner ( 1999) argued that when a wife ’s
income becomes closer to her husband ’s, there is a greater
chance of disruption than of potential gain. The more a wife
earns relative to her husband, the greater the couple ’sr i s ko f
divorce (Brines & Joyner, 1999). Thus, when high-earning
women do not use their greater resources to claim more
power, they may be returning control to their husbands toa v o i dm a r i t a ld i s c o r da sw e l la st ol i v eu pt on o r m a t i v e
standards of femininity.
The effect of women ’s resources, and their influence in
encouraging or discouraging a more equitable division of
household labor, is also dependent on how provider roles are
defined in a marriage and on the meanings and value given towomen ’s paid employment (Hood, 1983 ; Potuchek, 1992 ;
Pyke, 1994 ). Income may only increase wives ’power under
some circumstances (Blumstein & Schwartz, 1991 ). For
instance, Perry-Jenkins, Seery, and Crouter ( 1992) found that
husbands of women who identified themselves as coproviders,
viewing breadwinning as a res ponsibility that husbands and
wives should share equally, spent more time completing
household tasks than did husbands of homemakers andwomen who viewed their income as helpful but not essential.
The gender ideology model also fails to account for why
women continue to do more domestic labor when both theyand their husbands have liberal gender ideologies. Such
women may judge their actions and worth as mothers by an
idealized standard of motherhood, even when they reject thatstandard in principle (Deutsch, 1999 ). Based on the gender
construction approach, living up to this ideal and accepting a
division of labor where they retain major responsibility forthe housework and childcare may be a way that women “do
gender. ”Women may feel discomfort when they move away
from their motherly roles (Major, 1993 ).
Differences between housework and childcare The resour-
ces, structural factors, and gender ideology perspectivesattempt to explain domestic labor as a whole; they fail to
distinguish between housework and childcare. However,
many researchers have examined housework and childcareas distinct areas, and found that their nature and predictors
differ (e.g., Bianchi & Raley, 2005 ; Coltrane & Adams, 2001 ;
Deutsch et al., 1993 ;H o o d , 1983 ; Ishii-Kuntz & Coltrane,
1992 ). For example, some empirical work indicates that
housework is more likely than childcare to be affected by
wives ’relative income (Deutsch et al., 1993 ; Ishii-Kuntz &
Coltrane,
1992 ). Other studies suggest that childcare may
involve more of a joint agreement between spouses and that
fathers spend more time in childcare than in housework tasks(Bianchi & Raley, 2005 ; Hood, 1983 ). Thus, to understand
changes in the division of domestic labor, it is important to
evaluate housework and childcare as distinct activities.
Negotiation, conflict, and the division of labor
Negotiation Couples who are married or living together make
decisions everyday about who will do what (Kirchler, 1993 ).
Those decisions determine not only their day-to-day routine
but also their future behaviors and the balance of power
between them (Coltrane, 2000 ; Zvonkovic, Schmiege, &
Hall, 1994 ). Although Hood ( 1983 ), Coltrane ( 1989 ,1990 ),
Deutsch ( 1999 ), and Risman and Johnson-Sumerford ( 1998 )
have explored connections between conflict, negotiations,and the division of labor, and their qualitative studies suggest
that links do exist, more detailed insight is needed to
understand fully how the three are related and what thismeans for change. Hood ( 1983 ) acknowledged the necessity
of further exploring, in specific detail, the relationships
among communication, conflict and the division of house-hold labor. Likewise, Szinovacz ( 1987 ) echoed a similar
sentiment; he stated that it is important to study family
interactions, such as in negotiation and conflict situations, tounderstand the processes of family power (Szinovacz, 1987 ).
A few researchers have explored the relationship between
communication, conflict and the division of labor in moredetail. For example, Pittman et al. ( 1996 ) examined the
division of household labor with a microprocess approach
and conceptualized housework as an activity that couplesdecide to engage in on a daily basis. They found that stress, a
constantly changing variable, helped to predict a more
dynamic allocation of housework time. Non-home-basedstress, such as a job stress, decreased time spent on housework
by the stressed party on a particular day, regardless of gender.
In addition, non-home-based stress from one day carried overto the next day and influenced that day ’s division of domestic
labor as well. Thus, examinations of stress illustrate how the
division of labor is a product of ever-changing and activedecision-making processes that are influenced by the sur-
rounding social environment (Pittman et al., 1996 ).
The role of conflict Although many women do not think that
the inequitable division of labor in their homes is unfair
(Baxter, 2000 ;D e m p s e y , 1999 ;S a n c h e z , 1994 ), many do. In
fact, the division of household labor, including housework
and childcare, is one of the greatest areas of conflict (e.g.,
Kluwer et al., 1996 ,2000 ) and dissatisfaction for married
couples (Kluwer, 1998 ). Wives often wish that their husbands
would participate in household tasks and childcare more than
they actually do (e.g., Johnson & Huston, 1998 ;K l u w e re ta l . ,
2000 ), and many prefer their husbands to share the daily
responsibility for these tasks instead of simply helping withSex Roles (2007) 56:309 –324 311

them (Dempsey, 2000 ). Wives are more likely than their
husbands to be angry over the division of household labor, to
express their unhappiness with the distribution of tasks, toraise these issues with their husbands, and to instigate change
(Deutsch, 1999 ;K l u w e re ta l . , 1996 ,1997 ;R i s m a n&
Johnson-Sumerford, 1998 ).
Conflict is not necessarily negative. Perhaps women ’sl e a d
role in conflicts over household labor indicates their increas-
ing sense of entitlement to a more equitable division ofhousework and childcare (Ferree, 1990 ;K l u w e re ta l . , 2000 ).
Marital interactions and conflicts may provide a means for
women to move away from more traditional gender roles(Kluwer et al., 1996 ,1997 ). However, change may depend
on the way in which conflicts are conducted. Kluwer et al.
(2000 ) studied spouses ’attempts, through hypothetical
conflict interactions concerning paid work, childcare and
housework, to maintain or change a gendered division of
labor. In many instances a wife-demand, husband-withdrawscenario was activated that further fueled the argument and
hampered change. In this type of conflict pattern wives
generally attempt to begin a discussion with their husbandsby pressuring, demanding, blaming, or criticizing. Their
husbands, in response, distance themselves by withdrawing,
defending, or becoming silent (Kluwer et al., 2000 ).
Assertiveness (i.e., the willingness to state one ’s desires
in a clear, direct, straightforward way) may be a keycomponent of change. In a study of plans for an egalitarian
marriage among college seniors, Deutsch, Kokot, and
Binder ( 2006 ) found that an assertive conflict resolution
style was associated with women ’s plans to have a marriage
in which both spouses scaled back on paid work to care for
children. Thus, women with assertive conflict resolutionstyles may expect to gain a more egalitarian division of
labor by negotiating with their husbands and clearly stating
their desires (Deutsch et al., 2006 ).
Although assertiveness may be important, as yet, we do not
know exactly how it works. Assertiveness and income may be
related in different ways. It is possible that assertivenessmediates the effect of women ’s income on their share of
domestic labor. Bringing income into the family may entitle
women to argue assertively for a reduction in their relativeshare of housework and childcare, except when their incomes
exceed their husbands ’. Then, the pressure to behave in more
gendered ways could reduce their assertiveness about house-hold labor which would explain the increase in wives ’share of
domestic labor when they outearn their husbands (Bittman et
al.,2003 ; Brines, 1994 ;F e r r e e , 1990 ; Tichenor, 1999 ).
Alternatively, income may inhibit assertiveness by acting as
a source of power on its own. Accordingly, to the extent that
women bring in income, they have less need to be explicitlyassertive to get what they want. Moreover, assertiveness
coupled with too much income may threaten husbands and
foment serious conflict. Finally, as we predicted, based onthe Deutsch et al. ( 2006 ) finding, the influence of assertive-
ness may be completely independent of income. Regardless
of their income, more assertive women may simply get theirhusbands to do more household labor.
Change may also be facilitated by a woman ’s plans to
achieve a more equal division of h ousehold labor. According to
the theory of planned behavior, a person
’s intentions are a
critical predictor of behavior (Ajzen, 2001). In fact, several
studies have demonstrated that forming intentions (i.e. plans)is effective in influencing behavior, such as eating healthier,
exercising, attending cancer screenings, attempting to quit
smoking, and graduating high school (Conner, Norman, &Bell, 2002 ; Davis, Ajzen, Saunders, & Williams, 2002 ;
Norman, Conner, & Bell, 1999 ; Rise, Thompson, &
Verplanken, 2003 ;S h e e r a n&O r b e l l , 2000 ; Verplanken &
Faes, 1999 ). We hypothesized that women who plan to
change the division of labor will be more likely than other
women to attempt change and that those who attempt changewill show the greatest change.
In the current exploratory study, we sought to extend
previous research and to examine how specific conflictprocesses regarding housework and childcare affect wom-
en’s success in creating a more equal division of labor. If
the division of household labor remains gendered, to what
extent do women want to change it? If they do want
change, how do they attempt to enact it and in what waysdo they discuss the issue? Are they assertive, demanding
change in a clear and direct manner? In essence, by paying
attention to conflict-related variables, such as assertiveness,can we increase our understanding of how household labor
is divided and how it might be changed to reduce
inequality? If creating a division of labor is an ongoingprocess, then it should be possible to examine both fleeting
changes in response to outside events, as Pittman et al.
(1996 ) did, and changes toward a more equal division of
labor over a relatively short period of time. To do so, we
used a short-term longitudinal design and interviewed
women about the division of labor in their families twiceover a period of approximately 2 months. We expected that
women ’s desire to change, plans to advocate for a more
equal division of labor, attempts to do so, and theirassertiveness would predict greater change over the
2 months above and beyond changes predicted by their
percent income and/or work hours.
Method
Participants
We used town hall birth records of a New England town for
the years 1999 and 2000 to identify couples who had either312 Sex Roles (2007) 56:309 –324

3- or 4-year-old children. Given that we sought to examine
the division of labor between husbands and wives, only
couples who were married and living together were eligiblefor the present study.
1In addition, we only interviewed
wives because wives tend to be more discontented with the
division of household labor than their husbands and, thus,more likely to desire a change (Deutsch, 1999 ; Kluwer
et al., 1996 ,1997 ; Risman & Johnson-Sumerford, 1998 ).
All participants were recruited through an initial letter ofintent and then a follow-up telephone call. Recruitment for
the second interview was based upon permission given at
the end of the first interview. Upon completion of both thefirst and second interviews, participants were entered into
two separate $50 raffles. Of the 127 participants initially
contacted for the first interview, five letters were returned
with incorrect addresses, 17 telephone numbers had either
been disconnected or were no longer in service, threeparticipants were not eligible because they lived in a single-parent household, seven declined participation, and 14 were
not able to be reached. Of the 102 eligible women, 81
participated yielding a response rate of 79.4%. Of the 81participants from the first interview, three declined to
participate in the second interview and three could not be
reached, thus yielding a response rate of 92.6%.
Participants ranged in age from 26 to 45 years ( M=34.9,
SD=4.8), and were all married and living with their
husbands. The majority was European American (95%). Awide range of education levels was represented. In fact, these
women were more educated than their husbands, χ
2(12,N=
81)=33.18, p<.01. Also, most women in this sample had
between one and three children. The majority (57%) had two
children, and only 5% had more than four children (see
Table 1for additional demographic information).
ProcedureApproximately 1 week after a letter to explain the study
was sent, participants were telephoned, the purpose of thestudy was reiterated, and each participant ’si n f o r m e d
consent was obtained orally. If women were available, the
15-min telephone interview was completed at that time. Ifnot, an appointment was made.
The first interview consisted of 73 items that explored:
current division of housework and childcare, ideal division ofhousework and childcare, desire for change, gender ideology,
use of power strategies in negotiation and conflict resolution,1Married and cohabiting couples differ in the way they divide
household labor (Clarkberg, Stolzenberg, & Waite, 1995 ; Shelton &
John, 1993 ), the factors that affect their union stability (Brines &
Joyner, 1999 ), and the way in which gendered power relations operate
(Cunningham, 2005 ). For example, married women spend more time
on housework than cohabiting women do (Shelton & John, 1993 ).
None of the participants in this sample identified themselves as
cohabitors.Participants Husbands
Mean age 34.94 37.25
Education level, n(%)
Less than high school 0 1 (1.2%)
High school graduate 9 (11.1%) 22 (27.2%)
Some college/trade school 28 (34.6%) 20 (24.7%)College degree 29 (35.8%) 22 (27.2%)Graduate degree 15 (18.5%) 16 (19.8%)
Race, n(%)
European American 77 (95.1%) –
Hispanic 1 (1.2%) –
Mixed 3 (3.7%) –
Employment and income at Time 1, n(%)
Employed 59 (74.7%) 79 (100.0%)Evening or night shift 16 (20.3%) 21 (26.6%)Hours worked per week 0 –50 30 –90
(M=21.71, SD=16.88) ( M=47.03, SD=10.29)
Mean percentage of household income 21.9% 78.1%
Employment and income at Time 2, n(%)
Employed 53 (70.7%) 74 (98.7%)Evening or night shift 14 (18.7%) 17 (22.7%)
Hours worked per week 0 –50 0 –70
(M=21.40, SD=17.38) ( M=46.73, SD=10.66)
Mean percentage of household income 22.9% 77.4%Table 1 Participant ’s and hus-
band ’s demographics.
Dashes indicate that data were
not obtained.Sex Roles (2007) 56:309 –324 313

and background information (e.g., “Are you currently
employed? ”). In addition, three free-response questions were
included to assess what household task or aspect of childcarewives wanted their husbands to do more of, what statement
wives might make to their husbands to receive assistance, and
what household task or aspect of childcare the couple alreadydivided successfully.
At the end of the first interview, each woman was asked for
permission to include her name in a list of potentialparticipants for a second interview. This request was made at
the end of the telephone interview to ensure that a participant ’s
responses were not compromised by the knowledge that herresponses could influence participation in the next stage of
research. At the end of each telephone interview, participants
were debriefed and thanked. Approximately 2 months afterthe first interview, a second 15-min telephone interview was
conducted. The second interview consisted of 34 items that
explored: current division of housework and childcare, desirefor change, and background information. In addition, the
second questionnaire included four free-response items to
examine changes over time and recent discussions aboutchildcare and housework. In these questions, each participant
was given the opportunity to elaborate on the last time she and
her spouse discussed how some aspect of housework or
childcare would be divided, as well as any changes between
her first and second interviews.
Measures
Housework and childcare Four scales measured actual
housework, ideal housework, actual childcare, and ideal
childcare. The questions included in these four measuresasked how specific housework and childcare tasks were
divided (see Table 2). Each of these items was scored on a 7-
point response scale that ranged from 1( Id oi ta l l ) to 7( my
spouse does it all ), and a mean was calculated for each
measure. Previous researchers (e.g., Coltrane, 1989 ;D e u t s c h
et al., 1993 ; Grote & Clark, 1998 ;M e d e r e r , 1993 ;P r e s l a n d
& Antill, 1987 )h a v eu t i l i z e dc o m p a r a b l el i k e r ts c a l e sw i t h
similar items (e.g., laundry, cooking, meal preparation,
feeding children, bathing children). Although some articlesdo not detail the reliability or validity of these scales, Deutsch
et al. ( 1993 ) reported a Cronbach ’s alpha of .85 for their
paternal participation in childcare measure and Grote andClark ( 1998 ) supplied alpha coefficients of .77 for household
tasks and .86 for childcare tasks.
Although factor analyses did not reveal an underlying
dimension for either housework or childcare, the items in each
of our categories were combined into scales. This procedure is
justified because the scales were not measuring psychologicaldimensions but rather the cumulative behavioral actions that
reflect our apriori definitions of household labor. In addition,
previous research supports our decision to distinguishhousework from childcare tasks (Deutsch et al., 1993 ;
Ishii-Kuntz & Coltrane, 1992 ).
Gender ideology Four items measured the gender ideology of
women who participated in the present study. These items
include: “It is better for the family if the husband is the
principal breadwinner, and the wife has primary responsibility
for the home and family ”(reversed); “Ideally, there should be
as many women as men in important positions in thegovernment and business ”;“When both parents work full-
time, fathers should be responsible for the same amount of
housework and childcare as mothers ”;“If a child gets sick, the
husband and wife should take turns staying home from work
to take care of the child. ”
Each item was measured on a 4-
point response scale that ranged from 1( strongly disagree )t o
4(strongly agree ). All four items were taken from the study
of Deutsch et al. ( 2006 ) about college seniors ’plans for an
egalitarian marriage. Althou gh these items produced a
reliability of .68 in the study of Deutsch et al. ( 2006 )a n d
formed a gender ideology scale, the items did not load on the
same dimension in the present sample. Therefore, they wereanalyzed separately.
Power strategies Fifteen items on the first questionnaire
measured participants ’power strategies. Twelve items were
adapted from the study of Falbo and Peplau ( 1980 )( a s k ,Table 2 Items for the housework and childcare scales.
ScaleaItem
Actual/ideal
houseworkbshopping for groceriescooking mealsdoing the evening dishes
cleaning the house
Actual/ideal
childcarehelping your children get
dressed in the morning
putting your children to bedplaying with your childrengiving your children a bath
making decisions about your
children ’s upbringing
reading to your childrenarranging for childcareresponding to your children ’s
requests and ongoing needs
aActual housework, ideal housework, actual childcare, and ideal
childcare are, in fact, four separate scales. The actual/ideal houseworkand actual/ideal childcare are listed together because they include the
same items.
bThe questions “How is taking out the garbage currently divided? ”
and “Ideally, how would you like taking out the garbage to be
divided? ”were removed from both the actual housework and the ideal
housework measures respectively due to their perception as a task
typically performed by men and because these two items did notcorrelate with any other housework tasks.314 Sex Roles (2007) 56:309 –324

persuade, compromise, remind, silent, to be affectionate,
pout, to have importance, hints, discuss, tell needs, reason).
Participants indicated how often they used each of thesestrategies on 4-point response scales that ranged from 1
(never )t o4 ( always ). For example, they indicated how often
they “dropped hints and made suggestions ”when they and
their spouses had conficts about household chores or
childcare. They also indicated the frequency of an asymmet-
rical wife-demand/husband-withdraw interaction, an itemtaken from the study of Kluwer et al. ( 1997 ). Likewise, in
an item developed from the research of Deutsch ( 1999 ),
participants were asked how often they “explode ”when
having conflicts over childcare and household labor to assess
women ’s expression of being fed up. Lastly, participants
rated how often their response to a conflict over childcare orhousehold labor was to “avoid talking about the problem. ”
A factor analysis
2was subsequently conducted, and,
based on a varimax rotation, one interpretable factor
emerged from the 15 power strategies. The seven items
with loading coefficients above .4 on this dimension reflectan assertive conflict resolution style. Five items loaded
positively; two items loaded negatively. Those items that
loaded negatively were subsequently reverse-scaled, and theseven items comprised an assertive conflict resolution style
scale (see Table 3), which had an alpha reliability of .74.
Content analysis of data Free-response items in both the first
and second questionnaires were analyzed using a modified
grounded theory approach (Strauss & Corbin, 1990 ). Analysis
began with an examination of the text line-by-line; we
labeled each meaning unit with a code that was inductive andcame directly from the data itself. After the initial analysis,
codes were renamed and redefined if necessary. Similar
codes were then combined and placed into larger categories(Strauss & Corbin, 1990 ). Subsequently, a thematic analysis
was conducted, and themes were developed for each
response that illustrated the prominent emerging patterns(Taylor & Bogdan, 1998 ). After we developed themes for
each response, overarching themes were identified, and the
responses were placed into more specific categories.
Results
Initial division of labor
In line with past studies (e.g., Bianchi & Raley, 2005 ;
Coltrane, 2000 ; Dempsey, 2002 ), we predicted and found a
gendered division of labor. The tasks in both the housework
and childcare scales were rated on 7-point response scales
that ranged from 1( I do it all )t o7 ( my spouse does it all )
with a midpoint of 4, which represents an equal division of
tasks. One-sample t-tests confirmed the presence of
inequality at Time 1 by determining that the divisionof housework, t(80)=−13.64, p<.01 ( M=2.54, SD=.96),
and childcare, t(80)=−14.21, p<.01 ( M=2.92, SD=.68),
significantly deviated from the midpoint. Women did morehousework and childcare than their husbands.
At Time 1, the division of housework was positively
correlated with the division of childcare, r(79)=.33, p<.01.
A husband ’s greater contribution to housework tasks
corresponded to his greater contribution in childcare tasks.
However, childcare tasks ( M=2.92) were more evenly
divided than housework tasks ( M=2.54), t(80)=3.45, p<.01.
Role of resources, structural factors, and gender ideology.
Hypotheses derived from the resource, the structural
factors, and the gender ideology theories, respectively,
2In the factor analysis, first, the principal component analysis method
was employed. Next, two factors were chosen to extract based on the
component eigenvalues and corresponding scree plot. Finally, the
maximum likelihood extraction method was used to extract the factors,and the varimax rotation with Kaiser normalization was used to rotate
them. A loading coefficient of at least .4 was used as a cut-off.Table 3 Items and reliability for the assertive conflict resolution style scale.
Item Scale
reliability
1. When my spouse and I experience a conflict about household chores or childcare, we compromise and negotiate something
agreeable to both of us..74
2. When my spouse and I experience a conflict about household chores or childcare, we both avoid talking about the problem.
(Reversed)
3. When my spouse and I experience a conflict about household chores or childcare, I clam up and become silent. (Reversed)4. When my spouse and I experience a conflict about household chores or childcare, I tell him how important my request is to me.5. When my spouse and I experience a conflict about household chores or childcare, we talk about it and discuss our differences
and needs.
6. When my spouse and I experience a conflict about household chores or childcare, I tell him my needs and what I want.
7. When my spouse and I experience a conflict about household chores or childcare, I try to reason with him and argue my point
logically.Sex Roles (2007) 56:309 –324 315

predict that wives who contribute a greater percentage of
family income, work relatively more hours, and believe in
liberal gender ideology will do a smaller percentage ofhousehold labor than wives who contribute relatively less in
income, work relatively fewer hours, and believe in more
traditional gender ideology. We conducted correlations toexamine the relation between resources (wife ’s percent
income), structural factors (wife ’s hours at work, husband ’s
hours at work), and gender ideology on the division of bothhousework and childcare. Although none of these variables
were significantly related to the division of childcare
responsibilities, we found some support for resources,structural factors and gender ideology as predictors of the
division of housework. The percentage of a wife ’s total
household income was positively correlated with thedivision of housework, r(73)=.26, p<.05. Consistent with
the resource theory, the more income a woman contributed
to the family, the smaller her share of housework. Althoughhusbands ’hours at work were not significantly related to
housework allocation, r(77)= −.14, p>.05, wives ’work
hours were, r(77)=.23, p<.05.
3As predicted by the structural
factors approach, the more hours a woman worked, the
smaller her share of housework. Finally, only the gender
ideology item “It is better for the family if the husband is the
principal breadwinner, and the wife has primary responsibil-
ity for the home and family ”significantly correlated with the
division of housework, r(79)= −.33,p<.01.4Thus, husbands
performed a larger share of the housework at Time 1 if their
wives disagreed with this item and were, thus, less traditionalin their gender ideologies.
Role of assertiveness In addition to the more standard
theoretical predictors of the division of labor, and based on
the findings of Deutsch et al. ( 2006 ), we predicted that an
assertive conflict resolution style would be related to amore equal division of household labor. Correlations were
conducted between the assertive conflict resolution style
scale and the division of both housework and childcare atTime 1, as well as with the other predictors of domestic
labor. An assertive conflict resolution style did not
significantly correlate with the division of either house-work, r(79)= −.01,p>.05, or childcare, r(79)=.10, p>.05.
However, the assertive conflict resolution style scale
negatively correlated with the wife ’s relative percentage of
total household income, r(73)= −.42,p<.01; the higher the
wife ’s percent income, the less assertive she was inconflicts regarding housework and childcare. Clearly,
although women who earned a high proportion of family
income did less housework, that cannot be explained by theexplicit use of power, because they were less, not more,
assertive than other wome n. When assertiveness was
controlled in a partial correlation, the wife ’s percentage of
household income significantly correlated with the division
of childcare, r(72)=.30, p<.05.
5Thus it appears that higher
income typically restrains women from behaving asassertively as other women, but when they do their income
gets them a more equitable division of childcare.
Evidence that the current division of labor is unsatisfying
Ideal division of labor Given the presence of a gendered
division of labor, to what extent do women want change?
When the women were asked if they had ever thought
about changing the division of labor in their homes, 44(54%) said yes, and 37 (46%) said no. Thus, in line with
recent research (e.g., Johnson & Huston, 1998 ; Kluwer et
al.,2000 ), a majority of women were not satisfied with their
current division of labor. To examine the relation between
women ’s actual and ideal divisions of housework and
childcare, a two-way repeated-measures ANOVA was
conducted between the type of tasks (housework or child-
care) and division of labor (actual vs. ideal). The resultsindicated a significant main effect between actual and ideal
division of labor, F(1, 80)=202.44, p<.01. Participants ’
ideal division of labor ( M=3.74, SD=.07) was significantly
higher (i.e., closer to equal) than their actual division of
labor ( M=2.73, SD=.08). In addition, a significant interac-
tion was found between the type of household labor,namely housework and childcare, and the actual and ideal
division of labor, F(1, 80)=22.94, p<.01. Although
women ’s actual and ideal divisions for both housework
and childcare differed, women were further from their ideal
for housework (ideal M=3.79 vs. actual M=2.54) than for
childcare (ideal M=3.68 vs. actual M=2.92).
Women ’s aspirations were also closer to an ideal of equality
for housework than for childcare. Although women ’si d e a l
division for housework tasks ( M=3.79, SD=.99) and their
ideal for childcare tasks ( M=3.68, SD=.46) did not
significantly differ, t(80)=1.15, p>.05, when we compared
their ideals for each with an equal division, denoted by themidpoint of 4, their ideal housework division did not deviate
significantly from an equal division, t(80)= −1.90, p>.05, but
the division of childcare did, t(80)= −6.35, p<.05. Women
wanted to do more than one-half of the childcare.
3The wife ’s percentage of total household income and her number of
hours at work were highly correlated, r(72)=.80, p<.01.
4In a multiple regression this gender ideology item had the only
significant relationship ( p<.01) to the division of housework in a
model, which included the gender ideology item and the wife ’s
percentage of total household income, adjusted R2=.14, F(2,72)=6.88,
p<.01.5The same did not hold true, however, for a woman ’s hours at work,
r(74)=.12, p>.05.316 Sex Roles (2007) 56:309 –324

Discrepancies in housework and childcare Discrepancy
scores were created between an individual woman ’s ideal
division and her actual division of housework and childcarelabor to explore further the difference between them. The
discrepancy of housework was positively correlated with
the discrepancy of childcare, r(79)=.33, p<.01. Women
who were dissatisfied in one aspect of domestic labor
tended to be dissatisfied in the other.
Correlations were conducted to evaluate the role of
resources, structural factors, and gender ideology in predict-
ing the discrepancy between a woman ’s actual and ideal
division of labor. These correlations showed that neither thewife ’s hours at work, husband ’s hours at work, wife ’s percent
income, gender ideology, nor the added assertive conflict
resolution style scale correlated with the discrepancy ofhousework at Time 1. The results for the discrepancy of
childcare were somewhat different. Although none of the
traditional items correlated with the discrepancy of childcare,an assertive conflict resolution style did prove to have a
significant negative correlation, r(79)= −.27,p<.05. The less
assertive a woman was in her conflict resolution style, thefurther she was from her ideal division of childcare. The same
held true when a partial correlation was conducted, after
controlling for women ’s percent income, r(70)= −.30,p<.05.
Intent to change At Time 1, when women were asked if
they actively planned to change the division of household
labor in the future, 34 (43%) of them said yes, and 46
(58%) of them said no. Thus, a sizeable minority of womenin fact planned for change. In addition, many women, when
asked whether they actively planned to change the division
of labor in their homes, referred to the active and ongoingnature of this process. For example, one woman explained:
It’s kind of a weekly thing for us here. … Well, it ’sa n
ongoing thing with us. I ’ve told him sometimes that I ’m
overwhelmed with how much I have to do and he doespitch in after that but it ’s only a short-term response. It
doesn ’t change in the long-run. There ’s nothing that he
does on a regular basis like daily chores or anything likethat. It depends if he has time or if I tell him that I ’dl i k e
him to do it because I have to work or something. Rarely
does he take the initiative to do it himself.
Others offered similar expressions about the continuing
battle they associated with the division of housework and
childcare. One woman even went on strike to achieve change
but reported that it had not succeeded as she had hoped. Yetanother woman expressed a feeling that may reflect the
sentiment of many women struggling to achieve a more
equitable division of labor; “I’m not asking for the world; I ’m
just asking for him to help out like 30%. ”Consistent with past
studies (e.g., Kluwer et al., 1996 ,2000 ), women in thepresent study were not content. The division of labor was an
ongoing struggle.
Division of labor at Time 2
At Time 2, a gendered division of labor persisted. Women
completed more housework, t(74)= −14.86, p<.01 ( M=
2.54, SD=.85), and childcare, t(74)= −12.18, p<.01 ( M=
3.14, SD=.61), than their husbands did. As at Time 1, thedivision of childcare at Time 2 ( M=3.14) was more equal
than the division of housework ( M=2.54), t(74)= −5.29, p<.01.
However, at Time 2 the division of housework and thedivision of childcare were no longer significantly correlated,
r(73)=.14, p>.05.
Role of resources, structural factors, and gender ideolo-
gySimilar to Time 1, the Time 2 results indicate support
for the relation between resources and structural factors,such as income and hours at work, and the existing division
of housework but not childcare. The division of housework
at Time 2 was positively correlated with both the wife ’s
work hours at Time 2, r(73)=.23, p<.05, and her percent-
age of total household income at Time 2, r(69)=.29, p<.05.
The more hours the wife worked and the greater her
percentage of total household income, the greater her
husband ’s share of housework. In addition, the husband ’s
work hours negatively correlated with his contribution,
r(73)= −.27,p<.05. As at Time 1, the gender ideology item
“It is better for the family if the husband is the principal
breadwinner, and the wife has primary responsibility for
the home and family ”negatively correlated with the
division of housework at Time 2, r(73)= −.24,p<.05.
6
Role of assertiveness The assertive conflict resolution style
scale did not significantly correlate with either the divisionof housework, r(73)=.04, p>.05, or childcare, r(73)=.18,
p>.05, at Time 2.
Change between Time 1 and Time 2
The division of housework and childcare at Time 2 was highly
correlated with the division of housework and childcare at
Time 1, r(73)=.72, p<.01, and r(73)=.74, p<.01, respec-
tively. To examine whether change did indeed occur, a two-way repeated-measures ANOV A was conducted between the
type of task, namely housework and childcare, and the time
at which these were evaluated. The results indicated a
6The results of multiple regression analyses indicated that, even
though the housework predictors of wife ’s hours at work, wife ’s
percentage of the total household income, husband ’s hours at work,
and the gender ideology item, all at Time 2, were not significant
individually, some aspect of their joint effects on housework made
them significant, R2=.14, adjusted R2=.09, F(4,66)=2.71, p<.05.Sex Roles (2007) 56:309 –324 317

significant main effect for the type of task, F(1, 74)=22.62,
p<.01, which was modified by a significant interaction
between the type of division of labor and the time at whichthese measures were evaluated, F(1, 74)=4.83, p<.05.
Although the change in housework ( M=−0.01, SD=.68)
was not statistically significant, t(74)=.11, p>.05, the change
in childcare was ( M=0.19, SD=.47), t(74)= −3.47, p<.01.
Men assumed proportionally more childcare at Time 2 than
at Time 1. On average, at Time 2, women were closer to theideal division of childcare.
Predictors of changeRole of resource, structural, and ideological variables
Partial correlations were conducted to examine the relationbetween variables derived from the resource, structural, and
gender ideology perspectives and change in the division of
household labor, and the relation between assertiveness andchange in housework and childcare. We predicted that
assertiveness would influence change, beyond changes
predicted by percent income and/or work hours. After wecontrolled for the division of housework at Time 1, neither
women ’s percent income, nor their work hours, nor their
spouses ’work hours, nor their responses to the gender
ideology questions at Time 1, nor assertiveness significant-
ly predicted housework at Time 2 (i.e., change). However,when childcare at Time 1 was controlled, the division at
Time 2 significantly negatively correlated with women ’s
work hours at Time 1, r(64)= −.25, p<.05, and women ’s
percent income at Time 1, r(64)= −.30,p<.01. Therefore, at
first glance, it appeared that the women who worked less
and made less income at Time 1 changed toward a moreegalitarian division of childcare at Time 2.
To understand this relationship, we created three cate-
gories based on the distribution of women ’s percent income
at Time 1: low income, 0 –15% ( n=33); medium income,
20–50% ( n=39); and high income, 60 –70% ( n=3). All
participants were included in a category. Although therewere only three high income women, it was important to
examine them in a distinct category because they were the
only women to earn a majority of their household ’s income.
A one-way analysis of covariance, with the division of
childcare at Time 1 controlled, was conducted to compare
women in the three income categories on the childcaredivision at Time 2. This analysis highlighted that the
relationship between income and change in the division of
childcare is curvilinear rather than linear. Although notstatistically significant, F(2, 72)=1.60, p>.05, women in
the medium income group had the most equal division of
childcare at Time 2 ( M
=3.26, SD=.62); women who
earned the majority of their household ’s income were
furthest from equality ( M=2.50, SD=.82), and women in
the low relative income group fell in between ( M=3.05,SD=.56). Thus, these data suggest a curvilinear relationship
between women ’s relative income and the division of
childcare that might have been confirmed if we had hadmore statistical power. Bittman et al. ( 2003 ) found a similar
curvilinearity for housework in their study of how spouses ’
contributions to family income affect its division. Amongthe 14% (approximately 314) of their sample where women
contributed 51 –100% of household income, couples divid-
ed housework more traditionally.
Role of plans and attempts Almost one-half (43%) of
women at Time 1 indicated active plans to change thedivision of household labor in the future. We hypothesized
that women who, at Time 1, planned to change the division
of labor would be more likely than other women to attemptto change and that those who attempted change would
show the greatest change. It was surprising that there was
no significant relationship between women ’s plan to change
and their actual attempt at change, χ
2(1,N=68)=.97, p>.05.
Only 12 women both indicated a plan to change and, in
fact, attempted to make a change. Thirty-eight women didsomething different than they had originally indicated.
Twenty-two women attempted change in the absence of a
stated intention to do so, and 16 women stated an intention
to try to change the division of labor, but did not do so. A
woman ’s plan to promote change may not be an accurate
measure of whether she actually will attempt to enact
change in the future.
In addition, women who made an attempt to change their
division of childcare were not more likely to achieve
change than those who did not actively try. One-way
analyses of covariance, with the division of either house-work or childcare at Time 1 controlled, were conducted to
compare women who attempted to change with those who
did not on their division of household labor at Time 2.Ironically, a woman ’s attempt to change was significantly
negatively related to the change in childcare, F(1, 67)=
9.64, p<.05, but not housework, F(1, 67)=2.20, p>.05.
Negotiating a division of labor
To gain more insight into how the division of labor is
actually negotiated and changed, responses to the open-
ended question “I’d like you to think back to the last time
you and your spouse discussed how you were going to
divide some aspect of housework or childcare, one that you
wanted your husband to do more of, and tell me the story ofwhat happened ”were analyzed qualitatively. Each response
was initially examined for the presence or absence of
change in the division of labor, then the responses wereevaluated for any underlying threads. Based upon these
evaluations, three categories were formed: No Change,
Limited Help, and Actual Change.318 Sex Roles (2007) 56:309 –324

No change There were 25 responses in the no change
category. Some women in this category had attempted change
but achieved nothing, and others had already given up. Forexample, one response in this category was: “I’dl i k et o
change it [the division of labor] but I don ’t really foresee it
actually changing so I ’ve just stopped trying. ”Other women
shared this same hopeless feeling. As one explained:
I’ve given up trying to change the division of labor in
my house. After years of trying I ’ve given up. It ’s just
less stress on me if I just do it all and not even count
on him to help me.
The no change category also encompassed women who
believed that another division of labor was not presently
possible due to structural reasons, such as their husbands ’
work schedules. One woman indicated that, although changehad not occurred, it might come in the future. “It [the
division of labor] will only change when I change careers. …
I tell him, when I get a real job you ’ll have to take her [their
daughter] everywhere. ”Several women expressed this same
hope that the division of labor in their homes would change
in the future, especially as a result of alterations in their workstatus. Still others did not report a change because they were
happy with the division that was already in place. For
example, one woman explained: “You know, it really just
goes well. We don ’t really have to discuss anything. We just
do what we need to do and everything gets done. ”
Limited help The limited help category included 18 women
who had achieved assistance by asking their husbands to
accomplish a specific task. One woman stated: “We don ’t
really talk about it too much. If I want him to do something
I just ask and he ’ll do it. ”Another woman reported:
Usually when I need him to do something I just say
I’m gonna need help with whatever it is, picking up,
etc. I just have to give him a heads up. I ’m usually
satisfied with the outcomes. I just have to tell himahead of time so that he can plan his work around it.
Still other women asked and gained assistance by giving
their husbands lists of what needed to be done. Forexample:
Childcare ’s really not an issue. It ’s pretty much a
given. Household tasks on the other hand are the
problem. It ’s just easier if I give him a list of things
that I want to get done. … And as long as I give him alist then he does it and it usually gets done. If I don ’t
put it on the list it won ’t get done. … You have to nag
or else he won ’t do it.
This same woman wanted more help from her husband
in making dinner, but there had not been a consistentincrease in his participation. In fact, she had thought about
changing her schedule to accommodate for his lack of help:
I’m actually thinking of changing to work two nights
because it ’s getting late for the kids to eat when I
come home from working days. And if I ’m not here
then the meal doesn ’t get cooked. So if I work nights
then right after I get dinner on the table then I canleave for work.
Thus, although these women might have achieved more
assistance with a specific task, the change was temporary.
One particular woman indicated that a long-term changewas highly unlikely: “If I asked him [her husband] to do
more of something on a regular basis he would probably
tell me that I ’m crazy and he really does it all the time. ”
The division of labor remained an ongoing and daily battle
for these women; they did not experience a stable change.
Actual change The final category of how the division of
labor is negotiated included 26 women who reported actual
change. This fairly broad classification encompassedseveral subgroups of women who reported different reasons
behind the changes. The four women in the first subgroup,
uncertain , could not specifically explain why a change
occurred. When asked to clarify how a more equal division
of childcare had been achieved in her home, one woman
answered: “He’s just been better. ”
The next subgroup, circumstances , included eight
women who indicated that the change in their division of
household labor was due to alterations in householdcircumstances. A common e xample within this group
resulted when the husband ’s work schedule changed and
subsequently increased his time spent at home. Oneparticular woman had previously attempted to receive help
in transporting the couple ’s daughter to and from school,
but had never achieved the assistance she requested untilher husband became unemployed. She explained:
We talked about trying to work out a way for it
[transporting their daughter to and from school] to be
more even, even though where she goes is close towhere I go. He ’s home right now because he ’s been
laid off and he ’s looking for a new job. I ’ve asked that
he take a bigger role in that since he ’s home more. I ’ve
told him that it ’s really to our daughter ’s benefit to
spend more time with him even if it ’s in the car going
to and from school. So he said that while he ’s home he
would try and do more of that and he is.
For another woman, her pregnancy sparked changes in
the division of household labor:
The circumstances have changed such that I ’m expect-
ing twins and he has sort of stepped in on his own.Sex Roles (2007) 56:309 –324 319

Innately he knows that he has to do more. It was more a
discussion of physical necessity. There are just certain
things that I can ’t do. It went well. He is doing more.
The changes described by women in this subgroup resulted
from different household circumstances that promoted and
even fostered a more equal division of labor. However, the
circumstances that brought these women change appear to be
temporary and may not result in lasting change.
Indirect , the third subgroup in the actual change classifi-
cation, encompassed three women who altered their division
of labor not through a direct appeal for their husbands ’
participation but by indirectly involving their children. Forexample, one woman asked her children to approach their
father for help with homework. Another had her older
daughter help with chores around the house. A third womanattributed her husband ’s increased help to his having
overheard her complaining to her children:
I have had the general discussion with my kids more
than him that I am a mom and not a maid and that I
shouldn ’t be the one doing everything around here…
that it ’s our home and not my home that they ’re a
guest at. So maybe by having that general conversa-
tion he has thought about it more.
The last subgroup, discussion , included 11 women who
had discussions with their husbands in which they
explained the reasons behind their requests. For example,
one woman explained how she and her husband changedtheir division of taking out the garbage:
Well, like taking out the trash. I asked him to do it
because it ’s a lot easier for me not to have to worry
about doing it. It ’s one less thing for me to worry about.
Since he puts the trash out on the street for the garbage
men then it ’s easier for him to just collect it too. So, he
has done it since. He knows it helps me out.
In another example, one couple negotiated a joint decision
about the husband ’s contribution. The wife explained:
We mutually agreed to put it [the television] away. I
might have had to give him a little nudge but now he ’s
doing more reading and childcare and housework
because he doesn ’t just come home and sit on the
couch and watch it. We also took our child out of schooland decided to home-school him. Before, my husband
said he felt like a divorced father, and we all live in the
same house, because he would only see the kids for anhour when he came home…and on the weekends. Now
because we home-school the kids and put the TV away
he’s spending more time with them.
The women in this subgroup either worked with their
husbands to change some aspect of their division of labor orthey explained to their husbands why they desired a certain
change. The requests and discussions present in this
category appear to reflect a more long-term tone andinclude women ’s explanations, feelings, and motivating
reasons. Perhaps these techniques assisted women in
achieving and enacting their desired changes.
One woman ’s description provides evidence for the
potential positive effects of a discussion on the division of
labor:
I asked him if he would help more with the laundry.
He said yes, as he always does because he ’s very
agreeable with things like that. He helped me once and
then that was the end of that. … He ’s a very agreeable
person so when I ask him to do things he generallydoes that. … I think we had an actual discussion about
that [childcare] and I feel that I ’m getting a little more
help. Still not a lot of help, because it didn ’t change a
whole lot, but a little bit more. At some point I don ’
t
intend to be home full-time so things will have to
change. There is a plan that we ’re going to split the
work more evenly but whether that actually happens
or not I don ’t know.
The discussion about childcare appeared to elicit more
change than her simple request for help with laundry. This
particular result, however, must be interpreted with caution.
Although it may provide evidence of the potential influenceof a discussion, it could also reflect the very different nature
of the requests. The change in childcare and not laundry is
consistent with previous literature that shows that menspend more time in childcare than in housework (Bianchi &
Raley, 2005 ).
Discussion
Presence of a gendered division of household labor
The division of both housework and childcare within the
households of this sample was split along decidedly
gendered lines. Women completed more housework and
childcare than their husbands did. In fact, the genderedinequality persisted throughout the course of the study. This
general finding supports the great majority of research that
illustrates the persistence of an unequal division of
household labor (e.g., Coltrane, 2000 ; Dempsey, 2002 ).
Role of resources, structural factors, gender ideology,
and gender construction
The present study shows some support for the conventional
theoretical perspectives of relative resources, structural320 Sex Roles (2007) 56:309 –324

factors, and gender ideology as predictors of the current
division of household labor. At both Time 1 and Time 2,
women who earned a greater percentage of family incomeand worked relatively more hours than other women
completed a smaller share of housework. Likewise, at Time
2, men who worked relatively fewer hours than other mencompleted a greater portion of housework. In addition, at
both Time 1 and Time 2 husbands completed more
housework if their wives disagreed that “[i]t is better for
the family if the husband is the principal breadwinner, and
the wife has primary responsibility for the home and family. ”
Moreover, although speculative because of the small
number of women who outearned their husbands, the
curvilinear relationship between income and change in the
division of childcare supports the notion that the division oflabor is shaped by how men and women do gender. Women
who made the majority of their household ’s income also
assumed most of the childcare responsibilities, whereaswomen who earned approximately one-quarter to one-half
of the household income enjoyed the most equal division of
childcare. Similarly, Tichenor ( 1999 ) found that women
whose income and/or occupational status was higher than
that of their husbands still completed most of the household
labor, and Brines ( 1994 ) discovered that husbands who are
dependent on their wives for economic support “do gender ”
by completing less housework. Bittman et al. ( 2003 ) found
a similar curvilinearity for housework in their study of how
spouses ’contributions to family income affect its division.
It appears, then, that a threshold level exists; when womenbegin to earn a majority of the household income, the threat
to conventional gender relations can be thwarted by a more
conventional division of household labor.
What does assertiveness add?
Assertiveness interacted with income in complicated ways
to affect the division of housework and childcare. First, we
found that women who earned relatively more familyincome were relatively less assertive than other women.
Perhaps when women ’s incomes increased, the pressure to
behave in more gendered ways may also have lead them tobe less assertive about household labor (e.g., Bittman et al.,
2003 ; Brines, 1994 ; Ferree, 1990 ; Tichenor, 1999 )o rt o
avoid the topic entirely because of its potential to damagethe relationship. The more a wife earns relative to her
husband, which threatens his breadwinning role, the greater
the chances of marital discord and divorce (Brines &Joyner, 1999 ). By avoiding assertiveness, the high earning
women may be subtly communicating that they are not
going to try to convert money into power.
Alternatively, women with more income may not have
needed to be assertive, at least with respect to housework.
Money seemed to give them power anyway. Women ’sincomes may not give them as much power as men ’s, but
they do give them more power than women with less income
have (Blumberg, 1991 ). Women with more income got a more
equal division, even though they were less assertive than other
women. With childcare, however, income alone did not
predict a more equal division of labor. In fact, none of theconventional theoretical perspectives shed light on the
division of childcare in our sample. Earning more money,
working more hours, and believing in liberal gender ideologywere not related to a more equal division of childcare either
time we interviewed our sample. Nor were they related to how
close women were to their ideal divisions of childcare.However, assertiveness was important, at least the first time
we interviewed women. The more assertive a woman was, the
closer she was to her ideal division of childcare. In addition,when we controlled for assertiveness, higher earning wives
did have a more equal division of childcare. Thus, to the extent
that women were assertive, more money meant more help.Why did the effects of income depend on assertiveness for
childcare, but not housework?
Differences between housework and childcare
Housework is more tedious and offers fewer rewards than
childcare. A clean house is simply not as rewarding for
most people as the love of a child. This difference in therelative rewards of these two domains is reflected in several
aspects of our findings. Women sought equality in
housework, but not childcare. Mothers may identify moreclosely with childcare than housework. In fact, with
women ’s increase in paid work since the 1960s, they have
decreased the amount of time spent on housework, butcontinue to spend the same or more time on childcare
(Bianchi, Milkie, Sayer, & Robinson,
2000 ; Bianchi &
Raley, 2005 ; Sayer, Bianchi, & Robinson, 2004 ). Men too
seem to prefer childcare. Over the past 40 years, men have
increased their involvement more in childcare than in
housework (Bianchi & Raley, 2005 ). In our study, the
division of childcare was more equal than that of
housework, and men increased their contribution to child-
care more over the 2 months of the study.
These differences may explain why income is differen-
tially related to housework and childcare. Housework more
clearly fits the relative resource model in which resourcessuch as income translate into the power to opt out of
responsibilities. Our finding that income independently
predicts housework is consistent with past researcherswho have found evidence for the effects of relative
resources in housework but not childcare (Deutsch et al.,
1993 ; Ishii-Kuntz & Coltrane, 1992 ). With respect to
childcare, it is ambiguous whether women want more help
unless they are clearly assertive about it. Men may think
they owe their high earning wives more housework withoutSex Roles (2007) 56:309 –324 321

any discussion, but may justify their lack of participation in
childcare to themselves by assuming that their wives want
to do a disproportionate share. By suppressing theirassertiveness, high earning wives may be giving their
husbands the wrong message about childcare.
Finally, assertiveness, regardless of level of income, pre-
dicted how close women were to their ideal division of
childcare, but not housework responsibilities. Again, this
suggests that more than material power is important in thenegotiations over the less desirable housework. Women ’s
direct and persistent pursuit of their husbands ’help with
childcare is more likely to be successful, even when wivesare not high earning, because it holds potential rewards for
those husbands.
Intent, attempts, and change
The path to change is not a simple one. Initially, we
hypothesized that, to achieve change, a woman must
recognize an unequal division of labor, form an intention to
change the division, and then attempt to change it. However,women often did not do what they intended. Moreover,
sometimes the division became more egalitarian even
without the wife ’s efforts because of a structural change in
the family (e.g., a change in the husband ’s work situation).
Finally, women ’s attempts at change did not automatically
result in an egalitarian shift in either housework or childcare.
Clearly, simply making an attempt is not enough. The
type of attempt and the effort involved matter. Ourdefinition of “attempt ”might have been too broad. For
example, the efforts of women in the limited help category
may have been too limited and short-lived. As a result, theirhusbands might have helped with one or two tasks without
making an overall difference in the way housework or
childcare was divided. In contrast, women in the discussion
subgroup of actual change may have succeeded because
they engaged in lengthy conversations to enact change.
Ah u s b a n d ’s resistance or receptivity can also effectively
foster or prevent any changes from taking place. Although
women may try to bargain successfully for a change in the
division of labor, their attempts may be thwarted by men whorefuse to relinquish power or to change their beliefs and
attitudes (Dempsey, 2002 ). The words of one participant
summarize a husband ’s potential to exercise this veto power.
She explained: “The last time we talked about it [the division
of labor] was actually when you [the interviewer] last called.
I said he should clean the bathroom more and he justlaughed. ”In laughing, this husband was essentially stating
his opposition to her request, an act that quite probably will
thwart any other desires to pursue a different division ofhousehold labor. Such sentiments may reflect the entitlement
that men feel and the obligation women feel in relation to
housework. Because family work and childcare may be morecentral to a woman ’s identity, and breadwinning to a man ’s
identity, women typically feel entitled to less help, and men
to more help, at home (e.g., Hochschild, 1989 ;M a j o r , 1993 ).
Although husbands were not interviewed in the present
study, they should be in future research.
Limitations
Measurement Women who rated high on the assertive
conflict resolution style scale were assumed to have had
stable assertive conflict resolution styles. However, effects
of assertiveness were only found the first time participantswere interviewed, at the same time that assertiveness was
measured. Perhaps when initially questioned about their
assertiveness, participants referenced recent conversations,which may or may not have reflected their usual manner. In
fact, women in the discussion subgroup, who reported on
their assertive pursuit of change in the division of laborlater in the study, should have scored relatively high on the
assertiveness scale, if it measured a stable trait. However,
women in this subgroup did not score significantly higherthan other women. Given that this measure of assertiveness
is new, more work is needed on its test –retest reliability and
validity. If assertiveness varies over time, we may have
failed to detect the effects of “assertiveness ”the second
time we interviewed women with the quantitative measuresof the division of labor. The qualitative findings that
women in the discussion-based group did report change
suggest as much.
A second measurement issue involves gender ideology.
The four gender ideology items we used were analyzed
separately because they did not load on the samedimension. Other researchers have also found that their
gender ideology items were only moderately correlated or
did not all factor together (e.g., Coverman, 1985 ; Sanchez,
1994 ), which suggests that gender ideology is not a unitary
construct but, rather, is composed of different components.
Moreover, Kroska ( 1997 ,2000 ) has argued that identity
may predict behavior related to the division of household
labor better than ideology does.
In our study, only the gender ideology item “It is better
for the family if the husband is the principal breadwinner,
and the wife has primary responsibility for the home and
family ”significantly correlated with the division of
household labor. Perhaps that item was more predictive of
the division of labor because it relates to attitudes about
domestic labor and may better capture identity than otheritems do. The item “Ideally, there should be as many
women as men in important positions in the government
and business ”does not directly address the division of
domestic labor. The other two items ( “When both parents
work full-time, fathers should be responsible for the same
amount of housework and childcare as mothers ”and “If a322 Sex Roles (2007) 56:309 –324

child gets sick, the husband and wife should take turns
staying home from work to take care of the child ”)d o
address housework and childcare, but both items may referto more specific circumstances than fundamental beliefs. If
circumstances require that both parents work full-time, one
could believe that they should share housework andchildcare equally, yet believe that ideally men and women
should have a traditional division of housework and
breadwinning. Their ideal beliefs may better represent theiridentities.
Time frame As much literature indicates, the division of
household labor is an ongoing process (e.g., Deutsch, 1999 ;
Kluwer et al., 1996 ,2000 ; Pittman et al., 1996 ) that is
constantly changing, based on a couple ’s daily interactions.
As such, we can gain important insight by examining
factors that influence the division of labor and movement
toward equality within a short time frame. That being said,the short-term longitudina l design certainly limits the
generalizability of our results. The links between various
predictors and the division of housework and childcare mayhave differing effects over 6 months or 1 year. The effects
of assertiveness, for example, may need more time to
develop and become prominent.
The path to change
In the end, the path to change is complex. In women ’s
struggles for change at home they must grapple with howassertive to be, how to “do gender, ”and how to contend
with power issues in their marriage. Change seems to
involve women ’s awareness of inequality and their willing-
ness to approach and actively promote long-term discus-
sions with their husbands. These findings illustrate that the
division of household labor is an active and ongoingprocess. We suggest that researchers further explore how
assertiveness and other conflict-related variables work to
influence the division of housework and childcare. This is apromising avenue by which researchers can better under-
stand and illuminate how a couple achieves equality.
Acknowledgement We thank the Harap Fund for their generous
donation for expenses associated with this project.
References
Ajzen, I. (2001). Nature and operation of attitudes. Annual Review of
Psychology, 52 ,2 7–58.
Baxter, J. (2000). The joys and justice of housework. Sociology, 34 ,
609–631.Berk, S. (1985). The gender factory: The apportionment of work in
American households. New York: Plenum.
Bianchi, S. M., Milkie, M. A., Sayer, L. C., & Robinson, J. P. (2000).
Is anyone doing the housework? Trends in the gender division ofhousehold labor. Social Forces, 79 , 191 –228.
Bianchi, S. M., & Raley, S. B. (2005). Time allocation in families. In
S. M. Bianchi, L. M. Casper, & B. R. King (Eds.), Work, family,
health, and well-being (pp. 21 –42). Mahwah, NJ: Erlbaum.
Bittman, M., England, P., Folbre, N., Sayer, L., & Matheson, G.
(2003). When does gender trump money? Bargaining and time inhousehold work. American Journal of Sociology, 109 , 186 –214.
Blumberg, R. L. (1991). The triple overl ap of gender stratification, economy,
a n dt h ef a m i l y .I nR .L .B l u m b e r g( E d . ) , Gender, family, and economy:
T h et r i p l eo v e r l a p (pp. 7 –32). Thousand Oaks, CA: Sage.
Blumstein, P., & Schwartz, P. (1991). Money and ideology: Their
impact on power and the division of household labor. In R. L.
Blumberg (Ed.), Gender, family, and economy: The triple overlap
(pp. 261 –288). Thousand Oaks, CA: Sage.
Brines, J. (1994). Economic dependency, gender, and the division of
labor at home. American Journal of Sociology, 100 (3), 652 –688.
Brines, J., & Joyner, K. (1999). The ties that bind: Principles of
cohesion in cohabitation and marriage. American Sociological
Review, 64 , 333 –355.
Clarkberg, M., Stolzenberg, R. M., & Waite, L. J. (1995). Attitudes,
values, and entrance into cohabitational versus marital unions.Social Forces, 74 , 609 –634.
Coltrane, S. (1989). Household labor and the routine production of
gender. Social Problems, 36 , 473 –490.
Coltrane, S. (1990). Birth timing and the division of labor in dual-
earner families: Exploratory findings and suggestions for futureresearch. Journal of Family Issues, 11 , 157 –181.
Coltrane, S. (2000). Research on household labor: Modeling and
measuring the social embeddedness of routine family work.Journal of Marriage and the Family, 62 , 1208 –1233.
Coltrane, S., & Adams, M. (2001). Men ’s family work: Child-centered
fathering and the sharing of domestic labor. In R. Hertz & N. L.Marshall (Eds.), Working families: The transformation of the
American home (pp. 72 –99). Berkeley. CA: University of
California Press.
Conner, M., Norman, P., & Bell, R. (2002). The theory of planned
behavior and healthy eating. Health Psychology, 21 , 194 –201.
Coverman, S. (1985). Explaining husbands ’participation in domestic
labor. Sociological Quarterly, 26 ,8 1–97.
Cunningham, M. (2005). Gender in cohabitation and marriage: The
influence of gender ideology on housework allocation over thelife course. Journal of Family Issues, 26 , 1037 –1061.
Davis, L. E., Ajzen, I., Saunders, J., & Williams, T. (2002). The
decision of African American students to complete high school:An application of the theory of planned behavior. Journal of
Educational Psychology, 94 , 810 –819.
Dempsey, K. (1997). Trying to get husbands to do more work at
home. Australian and New Zealand Journal of Sociology, 33 ,
216–225.
Dempsey, K. C. (1999). Attempting to explain women ’s perceptions
of the fairness of the division of housework. Journal of Family
Studies, 5 ,3–24.
Dempsey, K. C. (2000). Men and women ’s power relationships and
the persisting inequitable division of housework. Journal of
Family Studies, 6 ,7–24.
Dempsey, K. (2002). Who gets the best deal from marriage: Women
or men? Journal of Sociology, 38 ,9 1–110.
Deutsch, F. M. (1999). Halving it all: How equally shared parenting
works . Cambridge, MA: Harvard University Press.
Deutsch, F. M., Kokot, A. P., & Binder, K. S. (2006). College
women ’s plans for different types of egalitarian marriage.
Journal of Marriage and Family . (in press).Sex Roles (2007) 56:309 –324 323

Deutsch, F. M., Lussier, J. B., & Servis, L. J. (1993). Husbands at
home: Predictors of paternal participation in childcare and
housework. Journal of Personality and Social Psychology, 65 ,
1154 –1166.
England, P., & Farkas, G. (1986). Households, employment, and gender: A
social, economic, and demographic view. Hawthorne, NY: Aldine.
Falbo, T., & Peplau, L. A. (1980). Power strategies in intimate
relationships. Journal of Personality and Social Psychology, 38 ,
618–628.
Fenstermaker, S., West, C., & Zimmerman, D. H. (1991). Gender
inequality: New conceptual terrain. In R. L. Blumberg (Ed.),Gender, family, and economy: The triple overlap (pp. 289 –307).
Thousand Oaks, CA: Sage.
Ferree, M. M. (1990). Beyond separate spheres: Feminism and family
research. Journal of Marriage and the Family, 52 , 866 –884.
Ferree, M. M. (1991). The gender division of labor in two-earner
marriages: Dimensions of variability and change. Journal of
Family Issues, 12 , 158 –180.
Grote, N. K., & Clark, M. S. (1998). Distributive justice norms and
family work: What is perceived as ideal, what is applied, and
what predicts perceived fairness? Social Justice Research, 11 ,
243–269.
Hochschild, A. (1989). The second shift: Working parents and the
revolution at home. New York: Viking/Penguin.
Hood, J. C. (1983). Becoming a two-job family. New York: Praeger.
Ishii-Kuntz, M., & Coltrane, S. (1992). Predicting the sharing of
household labor: Are parenting and housework distinct? Socio-
logical Perspectives, 35 , 629 –647.
Johnson, E. M., & Huston, T. L. (1998). The perils of love, or why
wives adapt to husbands during the transition to parenthood.Journal of Marriage and the Family, 60 , 195 –204.
Kirchler, E. (1993). Spouses ’joint purchase decisions: Determinants
of influence tactics for muddling through the process. Journal of
Economic Psychology, 14 , 405 –438.
Kluwer, E. S. (1998). Responses to gender inequality in the division
of family work: The status quo effect. Social Justice Research,
11, 337 –357.
Kluwer, E. S., Heesink, J. A. M., & Van de Vliert, E. (1996). Marital
conflict about the division of household labor and paid work.
Journal of Marriage and the Family, 58 , 958 –969.
Kluwer, E. S., Heesink, J. A. M., & Van de Vliert, E. (1997). The
marital dynamics of conflict over the division of labor. Journal of
Marriage and the Family, 59 , 635 –653.
Kluwer, E. S., Heesink, J. A. M., & Van de Vliert, E. (2000). The
division of labor in close relationships: An asymmetrical conflictissue. Personal Relationships, 7 , 263 –282.
Kroska, A. (1997). The division of labor in the home: A review
and reconceptualization. Social Psychological Quarterly, 60 ,
304–322.
Kroska, A. (2000). Conceptualizing and measuring gender ideology as
an identity. Gender & Society, 14 , 368 –394.
Major, B. (1993). Gender, entitlement, and the distribution of family
labor. Journal of Social Issues, 49 , 141 –160.
Mederer, H. J. (1993). Division of labor in two-earner homes: Task
accomplishment versus household management as critical varia-bles in perceptions about family work. Journal of Marriage and
the Family, 55 , 133 –145.
Norman, P., Conner, M., & Bell, R. (1999). The theory of planned
behavior and smoking cessation. Health Psychology, 18 ,8 9–94.
Osmond, M. W., & Thorne, B. (1993). Feminist theories: The social
construction of gender in families and society. In P. G. Boss, W.
J. Doherty, R. LaRossa, W. R. Schumm, & S. K. Steinmetz(Eds.), Sourcebook of family theories and methods: A contextual
approach (pp. 591 –625). New York: Plenum.
Perry-Jenkins, M., Seery, B., & Crouter, A. C. (1992). Linkages
between women ’s provider-role attitudes, psychological well-
being, and family relationships. Psychology of Women Quarterly,
16,3 1 1 –329.
Pittman, J. F., Solheim, C. A., & Blanchard, D. (1996). Stress as a
driver of the allocation of housework. Journal of Marriage and
the Family, 58 , 456 –468.
Potuchek, J. L. (1992). Employed wives ’orientations to breadwin-
ning: A gender theory analysis. Journal of Marriage and the
Family, 54 , 548 –558.
Presland, P., & Antill, J. K. (1987). Household division of labour: The
impact of hours worked in paid employment. Australian Journal
of Psychology, 39 , 273 –291.
Presser, H. B. (1994). Employment schedules among dual-earner
spouses and the division of household labor by gender. American
Sociological Review, 59 , 348 –364.
Pyke, K. D. (1994). Women ’s employment as a gift or burden?:
Marital power across marriage, divorce, and remarriage. Gender
& Society, 8 ,7 3–91.
Rise, J., Thompson, M., & Verplanken, B. (2003). Measuring
implementation intentions in the context of the theory of plannedbehavior. Scandinavian Journal of Psychology, 44 ,8 7–95.
Risman, B. J. (2004). Gender as a social structure: Theory wrestling
with activism. Gender & Society, 18 , 429 –450.
Risman, B. J., & Johnson-Sumerford, D. (1998). Doing it fairly: A
study of postgender marriages. Journal of Marriage and the
Family, 60 ,2 3–40.
Sanchez, L. (1994). Gender, labor allocations, and the psychology of
entitlement within the home. Social Forces, 73 , 533 –553.
Sayer, L. C., Bianchi, S. M., & Robinson, J. P. (2004). Are parents
investing less in children? Trends in mothers ’and fathers ’time
with children. American Journal of Sociology, 110 ,1–43.
Sheeran, P., & Orbell, S. (2000). Using implementation intentions to
increase attendance for cervical cancer screening. Health Psy-
chology, 19 , 283 –289.
Shelton, B. A., & John, D. (1993). Does marital status make a
difference? Housework among married and cohabiting men and
women. Journal of Family Issues, 14 , 401 –420.
Shelton, B. A., & John, D. (1996). The division of household labor.
Annual Review of Sociology, 22 , 299 –322.
Strauss, A. L., & Corbin, J. (1990). Basics of qualitative research.
Thousand Oaks, CA: Sage.
Szinovacz, M. E. (1987). Family power. In M. B. Sussman & S. K.
Steinmetz (Eds.), Handbook of marriage and the family (pp.
651–693). New York: Plenum.
Taylor, S. J., & Bogdan, R. (1998). Introduction to qualitative
methods: A guidebook and resource. New York: Wiley.
Thompson, L., & Walker, A. J. (1989). Gender in families: Women
and men in marriage, work, and parenthood. Journal of Marriage
and the Family, 51 , 845 –871.
Tichenor, V. J. (1999). Status and income as gendered resources: The
case of marital power. Journal of Marriage and the Family, 61 ,
638–650.
Verplanken, B., & Faes, S. (1999). Good intentions, bad habits, and
effects of forming implementation intentions on healthy eating.
European Journal of Social Psychology, 29 , 591 –604.
Zvonkovic, A. M., Schmiege, C. J., & Hall, L. D. (1994). Influence
strategies used when couples make work-family decisions andtheir importance for marital satisfaction. Family Relations, 43 ,
182–188.324 Sex Roles (2007) 56:309 –324