Available via license: CC BY-NC 4.0
Content may be subject to copyright.
https://doi.org/10.1177/2378023120924805
Socius: Sociological Research for
a Dynamic World
Volume 6: 1 –17
© The Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/2378023120924805
srd.sagepub.com
Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-
NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and
distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages
(https://us.sagepub.com/en-us/nam/open-access-at-sage).
Original Article
Introduction
For individuals in intimate relationships, relationship satis-
faction is an important determinant of relationship stability
and psychological well-being (Carlson and Kail 2018;
Frisco and Williams 2003; Hawkins and Booth 2005;
Williams 2003). Couples who are satisfied with their rela-
tionships are more likely to stay together and less likely to
exhibit symptoms of psychological distress. Although
numerous factors shape relationship satisfaction, the way
couples divide housework appears to matter a great deal.
The association appears strongest for employed women,
but research finds that sharing housework is optimal for
couples’ overall relationship quality (Carlson et al. 2016;
Carlson, Miller, and Sassler 2018; Schieman, Ruppanner,
and Milkie 2018).
Why egalitarian sharing of housework is associated with
relationship satisfaction remains unclear, nonetheless. One
factor that may play an important role linking the division of
labor to relationship satisfaction is partner communication.
Individuals who have partners with strong communica-
tion skills generally report greater relationship satisfaction
(Gottman 1994; Litzinger and Gordon 2005; Zamir et al.
2018), and strong communication skills may be an outcome
of sharing domestic work. According to equity theory
(Lively, Steelman, and Powell 2010; Walster, Walster, and
Berscheid 1978), feelings of fairness stemming from the
division of labor may shape partners’ positive and negative
communication, affecting not only their own relationship
satisfaction but also that of their partners. Egalitarian divi-
sions of housework are seen, on average, as more fair than
other arrangements and are associated with greater overall
relationship satisfaction (Amato et al. 2003; Carlson et al.
2016; Frisco and Williams 2003; Schieman et al. 2018;
Wilkie, Ferree, and Ratcliff 1998). The quality of one’s part-
ner’s communication may therefore constitute a mechanism
through which the division of labor and feelings of equity
affect the one’s relationship satisfaction.
Another explanation for the association between egalitari-
anism and relationship satisfaction is that some partners may
have the communication skills necessary to build close, inti-
mate bonds that are associated with both egalitarian divi-
sions and greater levels of relationship satisfaction (Giddens
1992). Some research suggests that the quality of partner
communication shapes the division of housework and feel-
ings of domestic equity (Komter 1989; Miller and Carlson
924805SRDXXX10.1177/2378023120924805Socius: Sociological Research for a Dynamic WorldCarlson et al.
research-article2020
1University of Utah, Salt Lake City, UT, USA
2University of Indianapolis, Indianapolis, IN, USA
Corresponding Author:
Daniel L. Carlson, Department of Family and Consumer Studies,
University of Utah, 225 E 1400 S. Alfred Emery Bldg Rm 228, Salt Lake
City UT 84112, USA.
Email: daniel.carlson@fcs.utah.edu
Division of Housework, Communication,
and Couples’ Relationship Satisfaction
Daniel L. Carlson1, Amanda J. Miller2, and Stephanie Rudd2
Abstract
The gendered division of housework is an important predictor of relationship satisfaction, but the mechanisms linking
these variables remain poorly understood. Using data on N = 487 couples from the 2006 Marital and Relationship
Survey, the authors examine the association of heterosexual partners’ communication quality with the division of
housework and the role of partners’ communication quality in the association between the division of housework
and relationship satisfaction. Results from instrumental variable models and Actor-Partner Interdependence Models
indicate that the quality of women’s communication with their male partners predicts how couples divide housework.
The quality of men’s communication with their female partners, however, appears to be an outcome of domestic
arrangements. Men’s communication quality mediates the association between the division of housework and women’
relationship satisfaction, while women’s communication quality confounds the association for men.
Keywords
housework, relationship satisfaction, communication quality, perceived equity, fairness
2 Socius: Sociological Research for a Dynamic World
2016). Direct, open communication is associated with a
greater likelihood of egalitarian labor arrangements at home,
while negative communication (i.e., ignoring, hostility) is
associated with conventional arrangements. The associations
between the division of housework and each partner’s feel-
ings of equity and relationship satisfaction may therefore be
spurious and confounded by how partners communicate with
one another.
In this article, we use the 2006 Marital and Relationship
Survey (MARS) to (1) assess causal directionality between
partners’ communication and the division of housework
using instrumental variable (IV) models and (2) examine
how partners’ communication quality accounts for the asso-
ciation of housework divisions with individuals’ relationship
satisfaction. Results from Actor-Partner Interdependence
Models (APIM) indicate that the quality of partner’s com-
munication largely explains the associations between the
division of housework and perceived equity with one’s rela-
tionship satisfaction. However, the causal pathway varies for
men and women.
Background
In and of itself, marriage is emotionally advantageous for
both men and women (Carlson 2012); however, being in a
low-quality or unhappy marriage is negatively associated
with psychological well-being (Hawkins and Booth 2005;
Williams 2003). In particular, being in an unhappy marriage
is linked to lower levels of self-esteem, health, happiness,
and overall life satisfaction (Frisco and Williams 2003;
Hawkins and Booth 2005; Williams 2003). In fact, being in a
low-quality marriage is worse for mental and physical health
than dissolving a low-quality relationship (Hawkins and
Booth 2005). Relationship satisfaction is also crucial to rela-
tionship stability. Relationship satisfaction and stability are
considered by Amato and colleagues (2007:41) to be “con-
ceptually distinct, but empirically correlated.” For both men
and women, higher levels of relationship satisfaction are
associated with greater relationship stability (Frisco and
Williams 2003; Ruffieux, Nussbeck, and Bodenmann 2014;
Williams 2003).
Relationship satisfaction is important for both individ-
ual- and couple-level well-being, and perhaps in no arena
of couples’ lives is it more scrutinized than the day-to-day
aspects of household living: shared routine housework.
Couples who share routine housework—cleaning, dishes,
cooking, laundry, and shopping—report higher levels of
relationship satisfaction and sexual satisfaction as well as
less relationship discord than couples who do not, though
the sharing of certain tasks may matter more than the shar-
ing of others (Carlson et al. 2016; Carlson et al. 2018;
Schieman et al. 2018). Despite the growth of egalitarian
arrangements over time, couples who share routine house-
work equally remain among the minority (Carlson et al.
2018; Gerson 2010).
Variations in the Division of Housework and
Relationship Satisfaction
Couples divide housework in different ways. These range
from conventional arrangements (in which the female partner
does most of the routine housework), to egalitarian (often
defined as within a 35–65 percent split) (cf. Risman 1998), to
counterconventional (male partners take on most of the rou-
tine housework). On average, egalitarian divisions of routine
housework are associated with higher relationship satisfaction
among women compared to conventional arrangements (e.g.,
Barstad 2014; Carlson et al. 2016; Carlson et al. 2018; Demaris
and Longmore 1996). For men, findings are more equivocal.
Some research shows that men are more satisfied in egalitarian
arrangements (Ruppanner, Brandén, and Turunen 2018;
Schieman et al. 2018), but other research shows no difference
in relationship satisfaction between conventional and egalitar-
ian housework arrangements (e.g., Barstad 2014; Carlson
et al. 2018) or indicates that men are more satisfied in conven-
tional arrangements than in egalitarian ones (Wilkie et al.
1998). This may be because, for men, specific tasks (shopping
especially) rather than overall divisions of housework are
related to relationship satisfaction (Carlson et al. 2018), and
measurement of housework isn’t consistent across studies.
Although research is scant, counterconventional divisions of
routine housework, in which the male partner does most of the
housework, appear to be associated with less relationship sat-
isfaction compared to egalitarian arrangements (Carlson et al.
2016; Carlson et al. 2018; Schieman et al. 2018).
Domestic Division of Labor, Communication, and
Relationship Satisfaction
Despite evidence that egalitarian sharing of housework between
partners is generally associated with greater relationship satis-
faction, the reason for this relationship remains unclear. One
potentially important factor that has received little consider-
ation to date is communication. It is well established that good
communication is vital to relationship satisfaction (Gottman
1994; Litzinger and Gordon 2005; Zamir et al. 2018). Research
also suggests that good communication is associated with an
egalitarian sharing of housework (Komter 1989; Ledbetter,
Stassen-Ferrara, and Dowd 2013; Miller and Carlson 2016).
How communication is causally associated with the division of
housework, however, is open for debate. On one hand, equity
theory (Lively et al. 2010; Walster et al. 1978) suggests that
good communication is an outcome of an equitable division of
labor. On the other hand, a gender power perspective (Komter
1989; Sassler and Miller 2017) indicates that communication
may predicate equal sharing.
Communication and Relationship Satisfaction
As couples move away from companionate relationships
guided by social norms and held together by specialization in
Carlson et al. 3
the division of labor (Cherlin 2004), Giddens (1992)
explains, couples increasingly seek out relationships that are
personally fulfilling and that involve mutual self-disclosure
and emotional intimacy. Communication is core to this.
Indeed, compared to the romantic love of companionship,
which is often rooted in gendered roles and female subjuga-
tion, modern relationships are built on confluent love—open
communication, trust, cooperation, and a presumed “equality
in emotional give and take” (Giddens 1992:62). These ele-
ments all contribute to an overall sense of intimacy, togeth-
erness, and ultimately relationship satisfaction among
partners.
Communication is a multifaceted concept that has been
examined most often through a combination of self-reported
questionnaires and behavioral observations (e.g., Gottman
1994; Zamir et al. 2018). Thematically, communication is
often divided into (1) negative traits, such as negative affect,
contempt, criticism, withdrawal, and conflict (Gottman
1994; Graber et al. 2011; Sullivan et al. 2010; Woodin
2011); (2) positive traits, including positive affect, intimacy,
validation, and spousal support (Graber et al. 2011); and (3)
effectiveness, like understanding intent, communicating
before making important decisions, and not avoiding par-
ticular topics (Navran 1967). Taken as a whole, high-quality
communication involves low levels of negative traits and
high levels of positive traits and effectiveness (Gottman
1994).
The quality of partner communication is positively
associated with relationship satisfaction since the ways in
which one’s partner communicates can result in feelings of
validation—or rejection—increasing or decreasing both
personal and relational satisfaction (e.g., Gottman 1994;
Johnson, Zabriskie, and Hill 2006; Litzinger and Gordon
2005; Zamir et al. 2018). A study by Lavner, Karney, and
Bradbury (2016), using cross-lagged panel modeling, sug-
gests a possible reciprocal association between communica-
tion and relationship satisfaction. That is, being happier may
improve one’s own and one’s partner’s communication.
Findings, nonetheless, were inconsistent and limited to a
community sample of Hispanic adults. Given the body of
evidence regarding the association between communication
and relationship satisfaction, we propose the following
hypothesis:
Hypothesis 1: One’s partner’s communication quality
is positively associated with one’s relationship
satisfaction.
If partner’s communication quality is important to under-
standing relationship satisfaction, understanding the predic-
tors of partner’s communication is equally important. One
factor that may be especially germane to shaping partner’s
communication is partner’s perceived equity regarding the
division of housework.
Egalitarian Sharing, Perceived Equity, and
Communication
A derivative of exchange theory, equity theory (Adams 1965;
Homans [1961] 1974; Walster et al. 1978) suggests that
individuals try to maximize rewards in interpersonal relation-
ships. When individuals perceive relationships as equitable—
when the amount of rewards received is perceived as fair,
deserved, and in balance with one’s contributions—they are
most satisfied. In contrast, individuals who perceive inequi-
ties become distressed (Mirowsky 1985), with greater inequi-
ties’ leading to greater distress. Those who feel they get less
out of the relationship than they feel is fair are underbene-
fited, while those who get more out of the relationship than
they think is fair are overbenefited. Both forms of inequity
produce distress and negative feelings (Canary and Stafford
1992; Sprecher 1986). Those who are underbenefited exhibit
feelings of anger, resentment, sadness, frustration, and depres-
sion, while those who are overbenefited also exhibit feelings
of anger and depression in addition to guilt (Lively et al.
2010; Mirowsky 1985; Sprecher 1986).
Individuals who perceive an equitable division of house-
work in their romantic relationships are more satisfied with
their relationships than those who find their relationships
inequitable (Amato et al. 2003; Chong and Mickelson 2016;
Frisco and Williams 2003; Greenstein 1996; Wilkie et al.
1998), and those with egalitarian divisions of housework are
most likely to feel their arrangements are equitable (Amato
et al. 2003; Carlson et al. 2016; Frisco and Williams 2003;
Schieman et al. 2018; Wilkie et al. 1998). Although feelings
of equity are most strongly linked to egalitarianism, espe-
cially for women (Carlson et al. 2016; Frisco and Williams
2003; Schieman et al. 2018) not all women, even those
employed full-time, find conventional housework arrange-
ments inequitable (Demaris and Longmore 1996; Greenstein
1996; Nordenmark and Nyman 2003). The reason is that
feelings of equity depend on what people think they deserve,
and thus perceived equity surrounding the gendered division
of labor is conditional to some degree (Greenstein 1996;
Lavee and Katz 2002; Qian and Sayer 2016).
Although research has focused almost exclusively on how
feelings of equity are associated with one’s own relationship
satisfaction, one’s feeling of equity likely also affects one’s
partner’s relationship satisfaction through the way one com-
municates. Figure 1 presents a causal model detailing this
association. As shown, the division of labor affects an actor’s
own relationship satisfaction via two pathways. The first is
through one’s own feelings of equity. The other is through
one’s partner’s feelings of equity and communication.
Research shows being both under- and overbenefited is asso-
ciated with lower-quality communication with one’s partner
(Canary and Stafford 1992; Ledbetter et al. 2013; Sprecher
1986). The association between perceived equity and com-
munication appears, however, to apply largely to women
4 Socius: Sociological Research for a Dynamic World
(Canary and Stafford 1992; Ledbetter et al. 2013; Stafford
and Canary 2006). Negative communication for female part-
ners who perceive inequities may help explain the fact that
although men often prefer gender-conventional divisions of
housework, and are highly satisfied with them, they are gen-
erally not more satisfied with their overall relationships than
men in egalitarian relationships (e.g., Carlson et al. 2016).
Following an equity theory framework, we hypothesize the
following:
Hypothesis 2: A couple’s division of housework is associ-
ated with one’s relationship satisfaction.
Hypothesis 3: Both an actor’s and his or her partner’s per-
ceived equity mediate the association between the divi-
sion of housework and one’s relationship satisfaction.
Hypothesis 4a: The association between the division of
housework and one’s relationship satisfaction is medi-
ated by one’s partner’s communication quality.
Hypothesis 4b: The association between one’s partner’s
perceived equity and one’s relationship satisfaction is
mediated by one’s partner’s communication quality.
This is the case especially for men.
Gender Power, Communication, and Egalitarian
Divisions of Labor
Most research on the division of housework implies that
housework arrangements are causally antecedent to relation-
ship satisfaction. Nevertheless, most of these studies are
cross-sectional in nature and cannot establish causal ordering.
This is important since two studies on relationship quality
using longitudinal data and methods (e.g., fixed-effects
regression, lagged dependent variables)—one by Carlson and
colleagues (2011) and another by Schober (2012)—suggest
that relationship quality may be antecedent to the division of
domestic labor. Carlson et al. found that both mothers’ and
fathers’ relationship quality predicted fathers’ frequency of
engagement with children. Schober found that mothers’ rela-
tionship quality predicted fathers’ frequency of childcare but
that fathers’ frequency of childcare predicted fathers’ rela-
tionship quality. Focusing on the division of childcare,
Schober found that fathers’ shares of childcare predicted
mothers’, but not fathers’, relationship quality.
In addition to study design and a focus on childcare, an
important difference between these two studies and others is
the assessment of relationship quality rather than relation-
ship satisfaction. Both Carlson et al. (2011) and Schober
(2012) use multi-item scales for relationship quality that
measure one’s perception of his or her partner’s behavior.
Relationship quality scales in both studies include multiple
items that can aptly be labeled “partner communication qual-
ity” (e.g., partner listens to you, partner encourages you,
partner understands your needs, partner insults or criticizes
you). It is difficult, therefore, to compare these results to
other studies focused solely on relationship satisfaction (e.g.,
how satisfied are you with your relationship?) or to disen-
tangle the causal associations between partners’ communica-
tion, relationship satisfaction, and the division of labor. Yet a
data anomaly in the Millenium Cohort Study used by Schober
may shed light on this issue. The multiwave Millenium
Cohort Study data contained a relationship quality scale in
all waves except for the last, where instead only a simple
question about “relationship happiness” was asked. In this
final wave, fathers’ shares of childcare did not predict moth-
ers’ relationship happiness, although in all previous waves it
had predicted mothers’ relationship quality. Because the rela-
tionship quality scale included measures of women’s part-
ners’ communication, the findings suggest, consistent with
Figure 1. Communication as mediator of domestic labor.
Carlson et al. 5
equity theory, that male partners’ communication may be an
outcome of the division of household labor.
Taken together, nonetheless, the Carlson et al. (2011) and
Schober (2012) studies suggest that female partners’ com-
munication, especially, may be a predicate of the division of
domestic labor. If this is so, then the associations between
relationship satisfaction, perceived equity, and the division
of housework may be spurious rather than causal. This may
be especially so for men’s relationship satisfaction as it may
depend on their female partners’ communication quality.
Alternatively, the division of housework and feelings of
equity may mediate the association between partner’s com-
munication and relationship satisfaction.
Good communication skills may help foster egalitarian
divisions of housework in couples. In fact, it should be of
little surprise that shifts in relationship ideals from romance
to confluence are associated with significant shifts toward
gender equality during the latter half of the twentieth and
beginning of the twenty-first centuries. Couples do not spon-
taneously arrive at egalitarian arrangements. Rather, the
default division of housework is to follow hegemonic con-
ventional norms (Blaisure and Allen 1995). To overcome
hegemonic conventions, couples, women especially, must
utilize strong communication and negotiation skills to craft
an egalitarian arrangement (Sassler and Miller 2017).
Whether egalitarian beliefs are predictive of good com-
munication is an open question, but good communication
and sharing housework appear to be correlated. Sharing tasks
is correlated with greater positivity, openness, and assur-
ances between partners (Ledbetter et al. 2013). In her classic
qualitative work on gender and marital power, Komter
(1989) found that only when wives exercised manifest power
via direct, open communication were husbands willing to
contribute equal shares of housework. Recent work by Miller
and Carlson (2016), examining gender power among cohabi-
tors, found something similar, revealing through in-depth
interviews that egalitarian couples were more likely than
conventional couples to have explicit and detailed conversa-
tions about the division of housework. Couples who enacted
conventional divisions of labor in both studies used more
negative or ineffective communication tactics. In contrast to
women who used direct, open communication, women who
were hostile or passive-aggressive were less likely to achieve
an egalitarian division of labor. In efforts to rebuff female
partners’ attempts to achieve gender-egalitarian arrange-
ments, men also exercised negative or ineffective communi-
cation tactics like avoidance, ignoring/stonewalling,
criticizing, and hostility. Due to small sample sizes, neither
qualitative study was able to investigate counterconventional
arrangements. It also should be noted that even the most
egalitarian couples did not always seamlessly arrive at their
divisions, nor were the most equal among them always with-
out conflict (Miller and Carlson 2016). As ideology and
practice come into conflict, these divisions must be fre-
quently negotiated using what Hochschild and Machung
(1989) termed “gender strategies.” Frustration may further
bubble up as female partners in egalitarian relationships
often have the added responsibility of ensuring that their
divisions of labor remain equal (Risman 1998).
Figure 2 presents a causal path model for communication,
the division of housework, and relationship satisfaction
based on a gender power perspective. The model presents
two sets of pathways. The first pathway (marked by solid
lines) is one of spuriousness, where one’s partner’s commu-
nication quality is causally prior to the division of house-
work, an actor’s perceived equity, and an actor’s relationship
satisfaction, confounding these associations. The second
pathway (marked by dashed lines) demonstrates that the
division of labor, along with both partners’ feelings of equity,
mediates the association between communication and rela-
tionship satisfaction. From the model we hypothesize the
following:
Hypothesis 5: A couple’s division of housework is an out-
come of each partner’s communication quality, but
female partners’ communication quality is likely more
important than male partners’ communication quality.
Hypothesis 6a: For men especially, the association
between the division of housework and relationship
satisfaction is spurious, confounded by partner’s com-
munication quality.
Hypothesis 6b: For men especially, the association
between one’s perceived equity and relationship satis-
faction is spurious, confounded by partner’s communi-
cation quality.
Hypothesis 7: The association between one’s communica-
tion quality and relationship satisfaction is mediated
by the division of housework and both partners’ per-
ceived equity.
The Current Study. The gendered division of housework is
associated with feelings of equity and relationship satisfac-
tion for both men and women. The quality of communication
between partners also is associated with relationship satis-
faction, feelings of equity, and the division of domestic tasks.
The role of communication in this association is nevertheless
unclear. On one hand, partner communication may be a
mediator. A partner in an unequal arrangement, who feels his
or her relationship is inequitable, may communicate poorly
with the significant other and thus reduce the significant oth-
er’s relationship satisfaction. This may be especially the case
for male partners’ communication and women’s relationship
satisfaction. On the other hand, how partners, female part-
ners especially, communicate may determine whether cou-
ples divide domestic tasks equally, which may subsequently
affect significant others’ feelings of fairness and satisfaction.
It could also mean the relationships between the division of
labor, perceived equity, and relationship satisfaction, for men
in particular, are spurious. Using IV modeling and APIM,
this study aims to better specify these relationships and
6 Socius: Sociological Research for a Dynamic World
provide theoretical clarity about the causes of the division of
housework and its consequences.
Method
Data
For this study we use data from MARS. MARS is a cross-
sectional probability sample of heterosexual U.S. couples
with children conducted in March and April of 2006 by
Knowledge Networks. The MARS sample was restricted to
married and cohabiting couples with coresident minor chil-
dren and female partners younger than 45. MARS also overs-
ampled low- to moderate-income couples. The median
family income of MARS couples was $40,000. As of 2006,
the median family income for married couples in the United
States was $69,716 (DeNavas-Walt, Proctor, and Smith
2007). Information was collected independently from both
partners for a total of 1,095 individuals in 605 couples. We
limit our sample to the N = 487 couples among whom both
partners completed the survey. Missing values (n = 92) were
imputed using the set of “mi” procedures in Stata 14. We
produced and combined m = 10 iterations of the data.
Measures
Actor’s Relationship Satisfaction. Relationship satisfaction is a
continuous measure derived from a single item asking
respondents to note their level of satisfaction with their
romantic relationship. The item ranges from 0 to 10, with
higher scores indicating more satisfaction with one’s roman-
tic relationship. The distribution of scores for relationship
satisfaction is negatively skewed. We therefore conducted
supplemental analyses using the log(10) transformation of
relationship satisfaction as the dependent variable to correct
for skew. Results did not differ from those using the original
relationship satisfaction scale. We standardize (z-score) the
scale in our models for comparison with other studies.
Actor’s Reports of Division of Housework. To construct mea-
sures for the division of housework we use actors’ own esti-
mates of the division of housework given discrepancies in
actors’ and partners’ reports (partner’s reports are only
moderately correlated; r = ~.6). Using actors’ reports is
consistent with a large tradition of research on perceived
equity and relationship quality that measures the division of
labor from a respondent’s own point of view (e.g., Bianchi
et al. 2000).
MARS respondents reported their divisions on the follow-
ing core/routine items (Coltrane 2000): washing dishes, doing
laundry, house cleaning, cooking and preparing meals, and
grocery shopping. We focus our analysis on routine house-
work for several reasons. First, routine housework is at the
center of gender inequality at home and is the area most con-
tested within couples (Goldscheider, Bernhardt, and Lappegård
2015; Hochschild and Machung 1989). Second, routine house-
work constitutes the vast majority of time spent in house-
work. Third, routine housework is arguably more important
Figure 2. Communication as confounder of division of domestic labor.
Carlson et al. 7
to relationship satisfaction than nonroutine housework
(Carlson et al. 2018). Last, the measures of nonroutine house-
work in MARS (e.g., yard maintenance, car repair, bill paying)
are sparse and not comprehensive enough to warrant analysis
of this housework domain.
Each routine housework item had 5 possible responses: 0
(I do it all), 1 (I do most of it), 2 (we share it equally), 3 (my
partner does most of it), and 4 (my partner does it all). Each
measure was recoded to indicate the gendered division of the
task (e.g., 0 = she does it all, 2 = we share it equally, and
4 = he does it all). Each item was summed, then averaged to
create a mean scale—male partner’s share of routine house-
work—that ranged from 0 to 4. Higher scores indicate that
the male partner does greater amounts of routine housework.
From the mean scale we created three dummy variables to
indicate conventional, egalitarian, and counterconventional
divisions of labor. Each dummy stands for approximately
one third of the distribution of housework shares. She does
most of the housework equals 1 when respondents have a
scale score of less than 1.4 (approximately less than 35 per-
cent of housework). Housework shared equally equals 1
when scale score is between 1.4 and 2.6 (between approxi-
mately 35 and 65 percent), and he does most of the house-
work equals 1 when scale score is greater than 2.6
(approximately 65 percent or more of housework). In line
with past research showing that counterconventional arrange-
ments of housework may be problematic for relationship
quality, routine housework is measured as both a scale and a
series of dummies to test for the possibility of nonlinearity in
the association with both communication and relationship
satisfaction.
Actor’s Report of Partner’s Communication Quality. Each
respondent reported on his or her partner’s communication.
Quality of partner’s communication is a 5-item mean scale
(alpha = .78) comprising the following items: (1) My
[spouse/partner] listens to me when I need someone to talk
to, (2) I find it hard to tell my [spouse/partner] certain things
because I am not sure how [he/she)] will react, (3) My
[spouse/partner] and I discuss things together before making
an important decision, (4) It is hard for me to talk with my
[spouse/partner], (5) When we are having a problem, my
[spouse/partner] often gives me the silent treatment. Item 1
measures partner support, an element of positive communi-
cation; items 2 and 5 indicate negative affect and withdrawal,
which are aspects of negative communication; and items 3
and 4 indicate mutual decision making and avoidance,
aspects of (in)effective communication. Each item ranges
from 0 = strongly disagree to 3 = strongly agree. Items 2, 4,
and 5 are reverse coded so that higher scores indicate higher-
quality communication. Although different dimensions of
communication are assessed, principal component factor
analysis and confirmatory factor analysis both indicated that
all five items load on a single factor, supporting the use of a
singular scale. We standardize (z-score) the scale for ease of
interpretation and comparison with the results of other
studies.
Actor’s and Partner’s Perceived Equity. Each respondent reported
his or her perceived fairness of the division of housework on
a five-point scale: 1 = very unfair to me, 2 = somewhat
unfair to me, 3 = fair to both of us, 4 = somewhat unfair to
my partner/spouse, and 5 = very unfair to my partner/spouse.
Perceived equity in housework is a dummy variable; respon-
dents are given a value of 1 if the division is reported as fair to
both of them.
Control Variables. We control for several variables in our
models. Unless otherwise noted, we include measures for
these variables for both partners in models. Controls include
respondent’s age (in years), education, religious affiliation,
hours of paid work per week, self-rated health, and egalitar-
ian gender ideology. Egalitarian gender ideology is a seven-
item summary scale (alpha = .60, range = 0–21) assessing
respondents’ level of agreement with the following state-
ments: (1) It is ok for a woman to keep her maiden name,
(2) A woman should quit working if her husband can support
her, (3) A mother working outside the home is just as warm
as a stay-at-home mom, (4) Better for everyone if husband is
the breadwinner, (5) Men should share housework with
women, (6) Women should make work a priority, and (7)
Men and women should share child-rearing equally. Each
item ranges from 0 = strongly disagree to 3 = strongly
agree. Items 2 and 4 were reverse coded.
We combine information from both partners for the fol-
lowing controls at the couple level: couples’ total yearly
income, female partner’s relative share of income, and whether
both partners attend religious services weekly (1 = yes).
Couple-level controls for the number of children younger
than age 2 in household, number of children ages 2 to 5 in
household, number of children ages 6 to 12 in household,
and whether the couple is married are derived from a
respondent’s own reports. Descriptive statistics for all vari-
ables are shown in Table 1.
Analytic Strategy
Analyses for this study are conducted using APIM. APIM is
appropriate due to the dyadic nature of the MARS data and
the nonindependence of observations. Preliminary analyses
showed that partners’ relationship satisfaction (r = .56;
p < .001), feelings of equity in housework (r = .31; p < .001),
and quality of communication (r = .53; p < .001) were all
significantly correlated. Both actor effects (A)—how one’s
own attributes predict one’s outcomes—and partner effects
(P)—how one’s partner’s attributes predict one’s outcomes—
are estimated in our models. The inclusion of partner effects
controls for the possibility of interdependence and reduces
bias in estimating both actor and partner effects (Cook and
Kenney 2005).
8 Socius: Sociological Research for a Dynamic World
APIM can be conducted using ordinary regression tech-
niques, structural equation modeling, or multilevel modeling
(Cook and Kenney 2005). For the purposes of this study,
regression modeling is used. Although regression models do
not allow the researcher to estimate disturbance term correla-
tion or estimate between and within dyad variation, the
model is ideal for this study for a few reasons. First, we use
IV models to estimate causal directionality between the
division of housework and each partner’s communication
quality. Postestimation statistics assessing the quality of
instruments can be calculated only using a regression
approach. Second, measures for the division of housework
are taken only from the actor’s point of view, necessitating
separate equations. Third, regression models do not require
researchers to specify mediating and confounding pathways,
which avoids possible model misspecification. In addition,
although regression modeling requires separate models for
each dyad member, it does not require specifying different
path models in the case that causal directionality between the
division of housework and communication varies by sex.
Our analysis proceeds as follows. First, as presented in
equation 1, we estimate the associations of actors’ reports of
the division of housework (Adol), with actors’ reports of their
partners’ communication quality (Pcomm), controlling for
actors’ communication quality (Acomm), actor’s characteristics
(Acov), and partners’ (Pcov) characteristics.
AbbP bA APe
dol comm comm covcov
=+ ++
++
01 2ββ (1)
We examine both continuous (ordinary least squares
[OLS] regression) and categorical (multinomial logistic
regression) measures of the division of housework to evaluate
linearity. This is important as the second set of analyses esti-
mate causal directionality between the division of housework
and communication. This requires not only establishing cor-
relation between variables but also utilizing appropriate esti-
mation techniques based on the functional form of variables.
For linear associations, we assess whether the quality of part-
ner communication is endogenous to the division of house-
work using two-stage least square (2SLS) regression with IVs
Table 1. Descriptive Statistics (N = 487).
Individual-level controls
Men Women
M SD M SD
Relationship satisfaction .02 .97 .01 .99
Partner’s communication quality 2.10 .56 2.07 .64
Male partner’s share of routine housework 1.27 .87 1.00 .82
She does majority of routine housework .56 .69
Routine housework shared equally .38 .26
He does majority of routine housework .06 .05
Perceived equity in housework .53 .50
Hours spent in paid labor per week 36.93 21.21 15.31 18.49
Protestant .37 .40
Catholic .18 .21
Other religion .22 .24
No religion .23 .16
Age 37.90 7.55 34.68 6.20
Self-reported health 3.52 .99 3.46 .95
Less than high school .10 .09
High school .35 .33
Some college .37 .40
Bachelor’s degree or more .19 .18
Egalitarian gender ideology 11.00 2.80 12.22 2.78
Couple-level controls M SD
Hours of housework per week 42.25 34.56
Couples’ total income (in $) 53,368.00 153,593.81
Her share of income .29
Number of children younger than age 2 in household .11 .31
Number of children age 2 to 5 in household .57 .73
Number of children age 6 to 12 in household .87 .95
Currently cohabiting .11
Both attend religious service weekly .30
Carlson et al. 9
for communication. For nonlinear associations we use probit
with IVs. Analyses are conducted in Stata using the ivreg2
and ivprobit commands. Equations 2 and 3 demonstrate the
two-stage instrumentation process.
PZ
v
comm
=+ +ππ
0 (2)
AbbP APe
dol comm covcov
=+ +++
01
ββ (3)
Equation 2 is the IV estimator for predicted values of
actors reports of partner’s communication ( Pcomm
) derived
from a matrix of IVs (Z). As noted by Bollen (2012), instru-
ments must be related to the variable for which they are
intended to proxy
()π≠
0 and unrelated to the endogenous
outcome variable of interest {(,)}CovZ Adol =0 and its dis-
turbance term {(,) }CovZe=0
Equation 3 is the second-stage equation predicting the
division of housework using the IV estimator derived in
equation 2. Using instruments to proxy for actor’s reports of
partner’s communication helps assess causal directionality
between communication and the division of housework. If
partner’s communication is exogenous to the division of
housework (i.e., equal division of housework is an outcome
of good communication), estimates of communication quality
from 2SLS regression and IV probit models will be similar to
OLS regression and probit estimates. If, however, partner’s
communication quality is endogenous to the division of
housework (i.e., good communication quality is an outcome
of equal divisions), then estimates for communication should
vary considerably and will likely be close to b1 = 0.
Based on theory and past empirical research, two vari-
ables are chosen as instruments of actor’s report of partner’s
communication: actor’s reports of family time spent at home
in leisure activities and couple’s time together in social
activities with relatives. For family time at home in leisure
activities, partners were asked, “In a typical month, how
often do you, your [spouse/partner], and the children do
things at home as a family together, such as eat dinner at the
table, play games, or watch videos?” For time with relatives,
they were asked, “During the past month, about how often
did you and your [spouse/partner] spend time together in
social activities with relatives?” For both questions,
responses included 0 = almost never, 1 = once or twice a
month, 2 = once a week or almost every week, 3 = more
than once a week, and 4 = almost every day.
Spending time together as a couple or family, in any
capacity, is associated with stronger communication and
conflict resolution skills (Caughlin 2003; Kumpfer and Alder
2003). Although it has not been a major topic of research,
studies show no association between housework and its divi-
sion with couples’ or families’ domestic leisure time (Thrane
2000; Voorpostel, van der Lippe, and Gershuny 2010).
Although research informs our choice of instruments, we
take numerous steps to verify them. To assess the validity of
the instruments we estimate whether the IVs are correlated
with actors’ reports of partners’ communication, but not the
division of housework—a necessary but insufficient condi-
tion of instrumentation (Bollen 2012). We also report several
post hoc diagnostics from our IV models. The Anderson
canonical correlation likelihood ratio test (Hall, Rudebusch,
and Wilcox 1996) and Cragg-Donald Wald F statistic (Cragg
and Donald 1993) assess instrument strength, while the
Hansen-Sargan Test (Sargan 1958) assesses whether they are
associated with the disturbance term.
Once it is established whether quality of partners’ com-
munication is exogenous/endogenous to the division of
housework, our final step is to examine how communication
quality confounds or mediates the associations of the divi-
sion of housework and perceived equity with relationship
satisfaction. Analyses are conducted using OLS regression,
as represented in equations 4 through 10.
AbbP bA APe
relsat comm comm covcov
=+ ++
++
01 2ββ (4)
AbbA APe
relsat dol covcov
=+ +++
03 ββ (5)
AbbA bA bP
APe
relsat dol equity equity
covcov
=+
++
+++
03
45
ββ
(6)
AbbP bA
AP
e
relsat comm dol covcov
=+ ++
++
01 3
ββ
(7)
AbbP bA
bA
AP
e
relsat comm comm
dol covcov
=+ +
++
++
01 2
3
ββ
(8)
AbbP bA bA
bP AP
relsat comm dol equity
equity covcov
=+ ++
++
++
01
34
5
ββ
ee
(9)
AbbP bA bA
bA bP A
relsat comm comm dol
equity equity co
=+ ++
+++
01 23
45
β
vvcov
Pe++
β
(10)
In the first model (equation 4), we examine the association
of partner’s communication quality with relationship satisfac-
tion (hypothesis 1). In the second model (equation 5), we
examine the association of the division of housework with
actor’s relationship satisfaction (hypothesis 2). In the third
model (equation 6), we assess the association between actor’s
and partner’s perceived equity and relationship satisfaction
and how perceived equity mediates the association between
the division of housework and relationship satisfaction
(hypothesis 3). Model 4 (equation 7) tests hypotheses 4a and
6a by examining how partner’s communication quality medi-
ates/confounds the association between the division of house-
work and actor’s relationship quality. In the fifth model
(equation 8), we assess whether the association between
10 Socius: Sociological Research for a Dynamic World
actor’s communication quality and relationship satisfaction is
mediated by the division of housework (hypothesis 7). Model
6 (equation 9) assesses how partner’s communication medi-
ates or confounds the association between perceived equity
and relationship satisfaction (hypotheses 4b and 6b). Finally,
in the last model (equation 10), we assess whether feelings of
equity mediate the association between actor’s communica-
tion quality and relationship satisfaction (hypothesis 7).
Results
Association between Division of Housework and
Partner’s Communication Quality
Table 2 shows the results of OLS and multinomial logistic
regressions for the association of quality of a partner’s com-
munication with the divisions of housework. All analyses
included statistical controls. Examining women’s reports, we
see that higher-quality communication from male partners is
associated with increases in men’s shares of housework.
When examining the division of housework categorically,
we see that better communication among men is associated
with significantly lower odds of having the female partner do
the majority of housework compared to sharing this work
equally. Although quality of male partner’s communication
is associated with greater odds of the male partner’s doing
most tasks compared to sharing equally, the difference is not
statistically significant. The pattern of evidence, nonetheless,
suggests a linear association between men’s communication
quality and men’s shares of unpaid housework among
women.
From men’s reports we see that female partner’s commu-
nication quality is linearly associated with a man’s share of
housework, but in the opposite direction. A man’s share of
housework appears to decrease as female partner’s commu-
nication quality increases. Multinomial logistic regression
results show that this pattern is driven by differences in the
odds of sharing versus male responsibility for housework.
Higher-quality communication among female partners is
associated with significantly lower odds of counterconven-
tional (i.e., he does most) housework arrangements com-
pared to sharing housework equally. No differences are
found in the odds of conventional (i.e., she does most) and
egalitarian arrangements by female partner’s communication
quality. Overall, the results suggest that higher-quality com-
munication from one’s partner is associated with a lower
likelihood that one will be responsible for housework.
IV Analysis
The results provide evidence that the division of housework
at home is associated with quality of partner communication.
Nevertheless, it is not clear whether communication affects
the division of labor or whether communication quality is an
outcome of housework divisions. Tables 3 and 4 show results
from IV analyses used to establish the causal direction
between the division of housework and partner’s communi-
cation quality. Bivariate correlations for partner’s communi-
cation quality, the division of housework, and instruments of
communication quality in Table 3 provide evidence that the
IVs for communication quality are significantly associated
with partner’s communication quality but not the division of
housework, satisfying the first criteria of IVs.
Results from Table 2 provide evidence that the associa-
tion of shares of housework with communication quality for
both men and women is linear. This enables us to use 2SLS
regression to examine endogeneity for quality of partner’s
communication. The results of IV analysis (see Table 4) indi-
cate that male partner’s communication quality is endoge-
nous to the division of routine housework whereas female
partner’s communication quality is exogenous. This provides
support for hypothesis 5 but for female partner’s communi-
cation only.
Results from OLS regression analysis with statistical con-
trols are identical to results from Table 2. A one standard
deviation increase in partner’s communication quality for
women is associated with a .17 point increase (p < .001) in
male partner’s share of routine housework. Results from
Table 2. Ordinary Least Squares and Multinomial Logistic Regression of Division of Housework on Partner’s Communication Quality
(N = 487).
Women Men
Male Partner’s
Share
She Does
Most
He Does
Most
Male Partner’s
Share
She Does
Most
He Does
Most
(Reference = Shared Equal) (Reference = Shared Equal)
Partner’s communication
quality
B .17*** −.31* .27 −.13*** .17 −.68*
SE .04 .13 .29 .04 .12 .27
Note: All models include statistical controls.
*p < .05. ***p < .001.
Carlson et al. 11
2SLS regression, nonetheless, indicate that instrumented
communication quality for women’s male partners is associ-
ated with a nonsignificant, .03 point decrease in men’s shares
of housework. Postestimation diagnostics indicate that the
instruments satisfy all of the criteria for IVs. The significant
(p < .001) Anderson canonical correlations likelihood ratio
test indicates that the instruments are sufficiently correlated
with partner’s communication quality. The Cragg-Donald
Wald F statistic of 19.01 indicates that the instruments are
strong predictors of communication quality. Using the
Staiger-Stock (Staiger and Stock 1997) rule of thumb, a
Cragg-Donald Wald F statistic less than 10 would indicate a
weak instrument. Finally, the nonsignificant Sargan statistic
demonstrates that the instruments are not associated with the
equation disturbance. Diagnostics for 2SLS analyses indi-
cate excellent instrumentation.
Men’s reports show that coefficients for female partner’s
communication quality are changed little between OLS
(b = –.13) and 2SLS (b = –.17) models, indicating that
female partner’s communication quality is exogenous to a
male partner’s share of routine housework. Despite a slightly
larger coefficient, the association is not significant in the
2SLS model due to inflated standard errors—a known limi-
tation of IV models (DiPerete and Gangl 2004).
Division of Housework, Communication, and
Relationship Satisfaction
In Table 5 we examine the associations of the division of
routine housework with women’s relationship satisfaction
and each partner’s perceived equity as well as how male part-
ner’s communication quality may mediate these associa-
tions. All models include statistical controls. We use
categorical measures for the division of housework to dif-
ferentiate between conventional, egalitarian, and counter-
conventional divisions of labor. Results using a linear
measure of division of housework are similar and are found
in Tables A1 to A2 in the online supplemental Appendix A.
In partial support of hypothesis 1, male partner’s commu-
nication quality is positively associated with women’s rela-
tionship satisfaction (p < .001). A 1 standard deviation
increase in male communication quality is associated with a
Table 3. Bivariate Correlations for Partner’s Communication Quality, Division of Housework, and Instrumental Variables.
1 2 3 4
Women
1. Partner’s communication quality —
2. Male partner’s shares of routine housework .15*** —
3. Time in social activities with relatives .17*** .05 —
4. How often family together at home .33*** −.06 .15** —
Men
1. Partner’s communication quality —
2. Male partner’s shares of routine housework −.21*** —
3. Time in social activities with relatives .10* .01 —
4. How often family together at home .31*** −.07 .18*** —
*p < .05. **p < .01. ***p < .001.
Table 4. OLS and 2SLS Analyses Assessing Endogeneity of Partner Communication Quality.
Women Men
Male Partner’s Share of
Routine Housework
Male Partner’s Share of
Routine Housework
OLS partner communication
B .17*** −.13**
SE .04 .04
2SLS partner communication
B −.03 −.17
SE .13 .17
Diagnostics for instruments
Anderson canonical correlation likelihood ratio 37.31*** 24.32***
Cragg-Donald Wald F statistic 19.01 11.99
Hansen-Sargan test χ2(1) = 1.69 χ2(1) = 0.06
Note: All models include statistical controls. OLS = ordinary least squares; 2SLS = two-stage least square.
**p < .01. ***p < .001.
12 Socius: Sociological Research for a Dynamic World
0.64 standard deviation increase in women’s relationship
satisfaction. A woman’s own communication quality is unas-
sociated with her relationship satisfaction. Supporting
hypothesis 2, model 2 shows that shouldering the housework
is associated with less relationship satisfaction among
women (b = –.24; p < .05) compared to sharing it equally
with their partners. Consistent with hypothesis 3, we find
that the association between the division of housework and
women’s relationship quality is mediated by both men’s
and women’s perceived equity. A woman’s own sense of
fairness (b = .47; p < .001) is more strongly associated
with her relationship satisfaction than is her partner’s sense
of fairness (b = .22; p < .05). When perceived fairness is
included in the model, differences in relationship satisfac-
tion between women who share housework and those who
do the majority of it are reduced by 71 percent and to statis-
tical nonsignificance.
Given evidence from 2SLS analysis that male partner’s
communication quality is endogenous to the division of
housework, the results from models 4 through 6 indicate that
male partner’s communication mediates the associations
between the division of housework, perceived equity, and
women’s relationship quality. These findings provide strong
support for hypotheses 4a and 4b. The inclusion of male part-
ner’s communication quality in model 4 reduces the coeffi-
cients for the division of housework substantially (71
percent) compared to model 2. As shown in model 6, male
partner’s communication quality also accounts for much of
the association between women’s perceived equity and their
relationship satisfaction, reducing the coefficient for
perceived equity from b = .47 in model 3 to b = .10 in model
6 and to statistical nonsignificance. Male partner communi-
cation quality also accounts for some of the association (~18
percent) between his perceived equity and women’s relation-
ship satisfaction, suggesting that although their perceived
equity is associated with men’s communication quality, com-
munication quality is not the primary pathway through which
men’s perceived equity is associated with the female part-
ner’s relationship satisfaction.
Table 6 shows the results of analyses assessing the rela-
tionships of the division of housework, female partner’s
communication quality, and perceived equity with men’s
relationship satisfaction. As with women, results support
hypotheses 1 and 2. Communication quality of both men and
women is positively associated with men’s relationship satis-
faction, although the effect size of female partner’s commu-
nication quality is twice as large. For both sexes then, results
show that partner’s communication quality matters for one’s
satisfaction more than one’s own communication quality.
Consistent with hypothesis 2, being largely responsible for
routine housework is associated with lower relationship sat-
isfaction for men compared to sharing responsibilities
equally with their partners.
Given that female partner’s communication quality is
exogenous to the division of housework, results from models
3 through 6 provide support for hypotheses 6a and 6b. Both
associations between the division of housework and perceived
equity with men’s relationship satisfaction is explained by
female partner’s communication quality. Differences in rela-
tionship satisfaction between men sharing housework and
Table 5. Relationship Satisfaction among Women.
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Partner’s communication quality
B .64*** .66*** .64*** .64*** .62***
SE .04 .04 .04 .04 .04
Actor’s communication quality
B .05 .05 .05
SE .04 .04 .04
Male partner’s share of housework (reference = shared equally)
She does most housework
B −.24* −.07 −.07 −.09 −.03 −.04
SE .11 .11 .09 .09 .09 .09
He does most housework
B .01 .24 −.10 −.09 −.01 −.00
SE .22 .22 .17 .17 .17 .17
Partner’s perceived equity
B .22* .18* .18*
SE .09 .07 .07
Actor’s perceived equity
B .47*** .10 .09
SE .09 .08 .08
Note: All models include statistical controls.
*p < .05. ***p < .001.
Carlson et al. 13
those doing the majority of housework are reduced by 50
percent and to nonsignificance when female partner’s com-
munication quality is entered in model 4. The association
between female partner’s feelings of equity and men’s rela-
tionship satisfaction is reduced by more than 80 percent in
model 6. Taken together, these findings suggest that the asso-
ciation of the division of routine housework and perceived
equity with men’s relationship satisfaction is spurious.
Although men’s communication quality is associated with
men’s relationship satisfaction, results show that neither the
division of housework nor perceived equity explain this
association. We therefore find no evidence to support hypoth-
esis 7.
Discussion
How couples arrange their housework and how they feel
about those arrangements are related to feelings of satisfac-
tion. Partners find divisions of unpaid housework to be most
fair and are mutually most satisfied with their relationships
when they are equal (Amato et al. 2003; Frisco and Williams
2003; Wilkie et al. 1998). Why this is the case has not been
well explained. One integral factor linking the division of
housework to perceived equity and relationship satisfaction
that had not been explored fully, prior to this study, was part-
ner communication.
Although communication matters for housework and
relationship satisfaction, how it matters was less than clear.
This study aimed to explicate the role of partners’ communi-
cation quality to improve our theoretical understanding of
the role of housework in shaping couples’ well-being and the
processes through which couples shape their housework
arrangements. We tested two theories—equity theory and
gender power—and found evidence to support both. Partners’
communication is a primary factor linking the division of
housework to relationship satisfaction, but the way partners’
communication matters depends on the partner’s gender.
Given competing hypotheses regarding causal ordering,
and cross-sectional data, IV analyses were used to establish
the direction of association between communication and the
division of housework. Consistent with a gender power per-
spective, our results indicated that female partner’s commu-
nication quality is exogenous to the division of housework.
In contrast, male partner’s communication quality was
endogenous, consistent with equity theory and previous
findings on the division of childcare by Schober (2012). For
women, this means that male partner’s communication
mediated the association between the division of labor, feel-
ings of equity, and women’s relationship satisfaction. For
men, the results suggest that the association between the
division of housework and men’s relationship satisfaction is
spurious and owed to the quality of the female partner’s
communication.
One of the primary findings of this study is that female
but not male communication quality shapes couples’ house-
work arrangements. Given that a conventional division of
housework is the hegemonic norm (Blaisure and Allen 1995),
this makes sense since it is women, not men, who bear a
greater burden of initiating and crafting nontraditional divi-
sions of labor at home. Women’s communication, however,
Table 6. Relationship Satisfaction among Men.
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Partner’s communication quality
B .41*** .50*** .41*** .49*** .40***
SE .05 .04 .05 .04 .05
Actor’s communication quality
B .19*** .19*** .19***
SE .05 .05 .05
Male partner’s share of housework (reference = shared equally)
She does most housework
B .06 .12 −.01 .01 .02 .02
SE .10 .10 .09 .09 .09 .09
He does most housework
B −.38* −.29 −.19 −.23 −.14 −.19
SE .19 .19 .17 .17 .17 .17
Partner’s perceived equity
B .26** .05 −.03
SE .09 .08 .08
Actor’s perceived equity
B .14 .13 .12
SE .09 .08 .08
Note: All models include statistical controls.
*p < .05. **p < .01. ***p < .001.
14 Socius: Sociological Research for a Dynamic World
seems to matter largely in distinguishing whether men do
most of the housework. Women’s communication quality
didn’t vary between conventional and egalitarian arrange-
ments. It is interesting to note that men appear to do most of
the housework only when their female partners communicate
poorly (less positive and effective, more negative). Although
communication tactics like avoidance, withdrawal, and neg-
ative affect may be necessary for getting men to take respon-
sibility for housework, something they may be reluctant to
do, it appears to carry costs as men in counterconventional
arrangements express significantly lower relationship satis-
faction compared to men in conventional or egalitarian
arrangements. Although counterconventional arrangements
remain rare, research shows that the number of stay-at-home
dads and male homemakers has increased substantially in
recent years (Carlson et al. 2018; Livingston 2014).
Reversing gendered family roles may be a sign of progress,
but research shows that men in these situations are often at
home due to unemployment or health concerns and not out of
ideology (Livingston 2014). As such, men may need a great
deal of reinforcement to take on the home responsibilities
that their earnings and time availability suggest should be
theirs. That these situations appear thrust upon couples rather
than specifically chosen helps explain why couples’ relation-
ship quality seems to suffer in these arrangements (Carlson
et al. 2016; Carlson et al. 2018; Schieman et al. 2018).
The findings regarding women’s communication stand in
contrast to qualitative work on gender power, communica-
tion, and the division of housework, especially those that
suggest that good communication is associated with more
egalitarian divisions of labor. Yet although Komter (1989)
indicated that direct communication from women was neces-
sary to make changes in the division of labor toward more
equality, she also noted that women were often unsuccessful
even when they exhibited high-quality communication. On
average, our results indicate that high-quality communica-
tion from female partners is not sufficient to achieve egalitar-
ian divisions of labor.
Qualitative work also suggests that good communication
from men is necessary for egalitarianism. Our results support
the observation that high-quality communication among men
is associated with a greater likelihood of egalitarian arrange-
ments compared to conventional divisions of housework.
However, in line with equity theory, we find that the quality
of men’s communication appears to stem from housework
arrangements rather than determine them. Moreover, our
findings indicate that men’s communication quality is a pri-
mary mediating mechanism linking the division of house-
work to women’s relationship satisfaction. This is important
as the vast majority of research on the division of labor has
focused on women’s own characteristics and perspectives
and the way they alone shape women’s reaction to the divi-
sion of labor.
Although equity theory implies that men’s communica-
tion would be an outcome of men’s sense of fairness, we find
that communication accounts for only a small proportion of
the association between men’s sense of fairness and wom-
en’s relationship satisfaction. Instead, men’s communication
quality appears to explain nearly all of the association
between women’s perceived equity and relationship satisfac-
tion. Why this is the case isn’t entirely clear. On one hand,
men’s communication quality may affect how women per-
ceive the division of labor, making the association between
women’s perceived equity and relationship satisfaction spu-
rious. Male partners who tout their less-than-equal shares of
housework as being more than “most men” could lead their
female partners to judge their relationships as more fair (cf.
Tichenor 1999). On the other hand, women’s perceived
equity could influence men’s communication. Men may be
attuned to indicators of women’s sense of fairness, which
may shape men’s communication and thus subsequently
affect women’s relationship satisfaction.
Whatever the causal pathway, this study demonstrates that
men’s communication is implicated in women’s sense of
equity. Again, these findings depart somewhat from those of
other studies that suggest that the association between one’s
communication and one’s equity is limited largely to women
(Canary and Stafford 1992; Ledbetter et al. 2013; Stafford
and Canary 2006). One reason for these discrepancies is that
past studies examined this association separately for men and
women, whereas this study accounted for the interdepen-
dence of observations between partners using APIM and thus
the high degree of correlation among partner’s feelings of
communication and perceived equity. The strong correlations
observed between partner’s satisfaction, feelings of equity,
and communication demonstrate that future research looking
at the consequences of the division of labor should consider
the interdependence of both partners’ characteristics.
Because perceived equity is, by definition, an outcome of
the division of labor, we are confident in our assessment of
causal ordering among men for whom female partners’ com-
munication quality is exogenous to the division of labor.
Among women, however, we cannot say definitively whether
the male partner’s communication predicts the woman’s
sense of fairness or vice versa. Disentangling the causal
associations between communication, perceived equity, and
relationship satisfaction is unfortunately beyond the scope of
this article. Although theory and research suggest that rela-
tionship satisfaction is an outcome of communication and
the division of labor, some studies suggest these associations
may be reciprocal or reversely causal (e.g., Johnson et al.
2006). We were unfortunately limited in our ability to test
these causal relationships by the cross-sectional nature of
MARS and a lack of suitable IVs. It is entirely plausible that
an equal division of housework leads to more satisfaction
with one’s relationship, which leads to better communication
and subsequently to more perceived equity. This is a question
we hope future research will address.
Although this study makes numerous contributions to
our understanding of the links between the division of
Carlson et al. 15
housework, communication, and relationship satisfaction in
couples, there are a few additional limitations. First, the
MARS sample consists of middle- to low-income heterosex-
ual parents, so the findings may not be generalizable to
wealthier, same-sex, and childless couples. Nevertheless,
limited resources to outsource housework make communica-
tion between lower-income partners more salient than for
upper-income partners who make up only a small proportion
of couples. Second, MARS contains many indicators of
communication quality, and while the scale demonstrates
high reliability, the list is not entirely comprehensive. How
the addition of other indicators may affect results is not clear.
Third, the study is limited by the individual assessments of
the division of labor and partners’ communication in MARS.
It is unfortunate that we were not able to compare personal
assessments of one’s own communication skills with part-
ner’s assessments. These limitations may have some effect
on the validity of these measures. Finally, because the data
are cross-sectional, we are unable to capture partners’ (dis)
satisfaction and communication skills as they negotiate and
renegotiate their divisions of labor.
Egalitarian divisions of housework are best for couples’
well-being. When one person is primarily responsible for
housework—women in conventional arrangements or men
in counterconventional arrangements—that person’s rela-
tionship satisfaction most often suffers. Only in egalitarian
arrangements are partners mutually benefited. We find
domestic arrangements are associated with partners’ com-
munication quality, and this communication explains why
egalitarianism is associated with more relationship satis-
faction compared to one partner’s having responsibility for
housework. For contemporary couples who value self-dis-
closure and trust (Giddens 1992), communication skills
appear to be an essential part of crafting a fulfilling rela-
tionship. Strong communication skills help couples
improve the long-term, global health of the relationship in
a positive and constructive fashion that leads to mutual sat-
isfaction. For women, good, high-quality communication
appears to shape not only the division of labor in her part-
nership but also her partner’s relationship satisfaction. For
men, more equal sharing of labor appears to lead to better
communication with partners, enhancing their partners’
well-being.
Acknowledgments
The authors would like to thank Dan Lichter for providing access to
the data.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: The
Marital and Relationship Survey (MARS) was supported by the
Initiative in Population Research at The Ohio State University
(Daniel T. Lichter, principal investigator).
ORCID iD
Daniel L. Carlson https://orcid.org/0000-0001-9415-0656
Supplemental Material
Supplemental material for this article is available online.
References
Adams, J. Stacy. 1965. “Inequity in Social Exchange.” Pp. 267–99
in Advances in Experimental Social Psychology, edited by L.
Berkowitz. New York: Academic Press.
Amato, Paul R., Alan Booth, David R. Johnson, David R. Johnson, and
Stacy J. Rogers. 2007. Alone Together: How Marriage in America
Is Changing. Cambridge, MA: Harvard University Press.
Amato, Paul R., David R. Johnson, Alan Booth, and Stacy J. Rogers.
2003. “Continuity and Change in Marital Quality between
1980 and 2000.” Journal of Marriage and Family 65(1):1–22.
Barstad, Anders. 2014. “Equality Is Bliss? Relationship Quality and
the Gender Division of Housework.” Journal of Family Issues
35(7):972–92.
Bianchi, Suzanne M., Melissa A. Milkie, Liana C. Sayer, and John P.
Robinson. 2000. “Is Anyone Doing the Housework? Trends in the
Gender Division of Housework.” Social Forces 79(1):191–228.
Blaisure, Karen R., and Katherine R. Allen. 1995. “Feminists and
the Ideology and Practice of Marital Equality.” Journal of
Marriage and Family 57:5–19.
Bollen, Ken A. 2012. “Instrumental Variables in Sociology and the
Social Sciences.” Annual Review of Sociology 38:37–72.
Canary, Daniel J., and Laura Stafford. 1992. “Relational Maintenance
Strategies and Equity in Marriage.” Communication
Monographs 59:243–67. doi:10.1080/03637759209376268
Carlson, Daniel L. 2012. “Deviations from Desired Age at Marriage:
Mental Health Differences across Marital Status.” Journal of
Marriage and Family 74(4):743–58.
Carlson, Daniel L., and Ben Lennox Kail. 2018. “Socioeconomic
Variation in the Association of Marriage with Depressive
Symptoms.” Social Science Research 71:85–97.
Carlson, Daniel L., Amanda J. Miller, and Sharon Sassler. 2018.
“Stalled for Whom? Change in the Division of Particular
Housework Tasks and Their Consequences for Middle- to
Low-Income Couples.” Socius 4:2378023118765867.
Carlson, Daniel L., Amanda J. Miller, Sharon Sassler, and Sarah
Hanson. 2016. “The Gendered Division of Housework and
Couples’ Sexual Relationships: A Reexamination.” Journal of
Marriage and Family 78(4):975–95.
Carlson, Marcia J., Natasha V. Pilkauskas, Sara S. McLanahan, and
Jeanne Brooks-Gunn. 2011. “Couples as Partners and Parents over
the Early Years.” Journal of Marriage and Family 73(2):317–33.
Caughlin, John P. 2003. “Family Communication Standards: What
Counts as Excellent Family Communication and How Are
Such Standards Associated with Family Satisfaction?” Human
Communication Research 29(1):5–40.
Cherlin, Andrew J. 2004. “The Deinstitutionalization of American
Marriage.” Journal of Marriage and Family 66(4):848–61.
Chong, Alexandra, and Kristin D. Mickelson. 2016. “Perceived
Fairness and Relationship Satisfaction during the Transition to
Parenthood: The Mediating Role of Spousal Support.” Journal
of Family Issues 37(1):3–28.
16 Socius: Sociological Research for a Dynamic World
Coltrane, Scott. 2000. “Research on Housework: Modeling and
Measuring the Social Embeddedness of Routine Family
Work.” Journal of Marriage and Family 62(4):1208–33.
Cook, William L., and David A. Kenny. 2005. “The Actor-Partner
Interdependence Model: A Model of Bidirectional Effects in
Developmental Studies.” International Journal of Behavioral
Development 29(2):101–109.
Cragg, John G., and Stephen G. Donald. 1993. “Testing Identifi-
ability and Specification in Instrumental Variable Models.”
Econometric Theory 9(2):222–40.
DeMaris, Alfred, and Monica A. Longmore. 1996. “Ideology, Power,
and Equity: Testing Competing Explanations for the Perception
of Fairness in Housework.” Social Forces 74(3):1043–71.
DeNavas-Walt, Carmen, Bernadette D. Proctor, and Jessica C.
Smith. 2007. US Census Bureau. Income, Poverty, and Health
Insurance Coverage in the United States: 2007. Washington,
DC: Government Printing Office.
DiPrete, Thomas A., and Markus Gangl. 2004. “Assessing Bias
in the Estimation of Causal Effects: Rosenbaum Bounds on
Matching Estimators and Instrumental Variables Estimation
with Imperfect Instruments.” Sociological Methodology 34(1):
271–310.
Frisco, Michelle L., and Kristi Williams. 2003. “Perceived House-
work Equity, Marital Happiness, and Divorce in Dual-Earner
Households.” Journal of Family Issues 24(1):51–73.
Gerson, Kathleen. 2010. The Unfinished Revolution: How a
New Generation Is Reshaping Family, Work, and Gender in
America. New York: Oxford University Press.
Giddens, Anthony. 1992. The Transformations of Intimacy.
Cambridge, UK: Polity.
Goldscheider, Frances, Eva Bernhardt, and Trude Lappegård. 2015.
“The Gender Revolution: A Framework for Understanding
Changing Family and Demographic Behavior.” Population
and Development Review 41(2):207–39.
Gottman, John Mordechai. 1994. What Predicts Divorce? The
Relationship between Marital Processes and Marital Outcomes.
New York: Psychology Press.
Graber, Elana C., Jean-Philippe Laurenceau, Erin Miga, Joanna
Chango, and James Coan. 2011. “Conflict and Love: Predicting
Newlywed Marital Outcomes from Two Interaction Contexts.”
Journal of Family Psychology 25:541–50.
Greenstein, Theodore N. 1996. “Gender Ideology and Perceptions
of Fairness of the Division of Housework: Effects on Marital
Quality.” Social Forces 74(3):1029–42.
Hall, Alastair R., Glenn D. Rudebusch, and David W. Wilcox. 1996.
“Judging Instrument Relevance in Instrumental Variables
Estimation.” International Economic Review 37(2):283–98.
Hawkins, Daniel N., and Alan Booth. 2005. “Unhappily Ever After:
Effects of Long-Term, Low-Quality Marriages on Well-Being.”
Social Forces 84(1):451–71.
Hochschild, Arlie, and Anne Machung. 1989. Working Parents and
the Revolution at Home. New York: Viking.
Homans, George. [1961] 1974. Social Behavior: Its Elementary
Forms. New York: Harcourt Brace.
Johnson, Heather A., Ramon B. Zabriskie, and Brian Hill. 2006.
“The Contribution of Couple Leisure Involvement, Leisure
Time, and Leisure Satisfaction to Marital Satisfaction.”
Marriage & Family Review 40(1):69–91.
Komter, Aafke. 1989. “Hidden Power in Marriage.” Gender &
Society 3(2):187–216.
Kumpfer, Karol L., and Steve Alder. 2003. “Dissemination of
Research-Based Family Interventions for the Prevention of
Substance Abuse.” Pp. 75–119 in Handbook of Drug Abuse
Prevention, edited by Z. Sloboda, and W. J. Bukoski. New York:
Kluwer Academic/Plenum.
Lavee, Yoav, and Ruth Katz. 2002. “Division of Labor, Perceived
Fairness, and Marital Quality: The Effect of Gender Ideology.”
Journal of Marriage and Family 64(1):27–39.
Lavner, Justin A., Benjamin R. Karney, and Thomas N. Bradbury. 2016.
“Does Couples’ Communication Predict Marital Satisfaction, or
Does Marital Satisfaction Predict Communication?” Journal of
Marriage and Family 78(3):680–94.
Ledbetter, Andrew M., Heather M. Stassen-Ferrara, and Megan M.
Dowd. 2013. “Comparing Equity and Self-Expansion Theory
Approaches to Relational Maintenance.” Personal Relation-
ships 20(1):38–51.
Litzinger, Samantha, and Kristina Coop Gordon. 2005. “Exploring
Relationships among Communication, Sexual Satisfaction,
and Marital Satisfaction.” Journal of Sex & Marital Therapy
31(5):409–24.
Lively, Kathryn J., Lala Carr Steelman, and Brian Powell. 2010.
“Equity, Emotion, and Household Division of Labor Response.”
Social Psychology Quarterly 73(4):358–79.
Livingston, Gretchen. 2014. Growing Number of Dads Home with
Kids: Biggest Increase among Those Caring for Family. Pew
Research Center Social and Demographic Trends. https://
www.pewsocialtrends.org/2014/06/05/growing-number-of-
dads-home-with-the-kids/
Miller, Amanda J., and Daniel L. Carlson. 2016. “Great Expec-
tations? Working- and Middle-Class Cohabitors’ Actual and
Expected Divisions of Housework.” Journal of Marriage and
Family 78(2):346–63.
Mirowsky, John. 1985. “Depression and Marital Power: An Equity
Model.” American Journal of Sociology 91(3):557–92.
Navran, Leslie. 1967. “Communication and Adjustment in
Marriage.” Family Process 6(2):173–84.
Nordenmark, Mikael, and Charlott Nyman. 2003. “Fair or Unfair?
Perceived Fairness of Household Division of Labour and
Gender Equality among Women and Men: The Swedish Case.”
European Journal of Women’s Studies 10(2):181–209.
Qian, Yue, and Liana C. Sayer. 2016. “Division of Labor, Gender
Ideology, and Marital Satisfaction in East Asia.” Journal of
Marriage and Family 78(2):383–400.
Risman, Barbara J. 1998. Gender Vertigo: American Families in
Transition. New Haven, CT: Yale University Press.
Ruffieux, Mireille, Fridtjof W. Nussbeck, and Guy Bodenmann.
2014. “Long-Term Prediction of Relationship Satisfaction and
Stability by Stress, Coping, Communication, and Well-Being.”
Journal of Divorce & Remarriage 55(6):485–501.
Ruppanner, Leah, Maria Brandén, and Jani Turunen. 2018. “Does
Unequal Housework Lead to Divorce? Evidence from Sweden.”
Sociology 52(1):75–94.
Sargan, John D. 1958. “The Estimation of Economic Relationships
Using Instrumental Variables.” Econometrica 26(3):393–415.
Sassler, Sharon, and Amanda J. Miller. 2017. Cohabitation Nation:
Gender, Class, and the Remaking of Relationships. Berkeley:
University of California Press.
Schieman, Scott, Leah Ruppanner, and Melissa A. Milkie. 2018.
“Who Helps with Homework? Parenting Inequality and
Relationship Quality among Employed Mothers and Fathers.”
Journal of Family and Economic Issues 39(1):49–65.
Carlson et al. 17
Schober, Pia S. 2012. “Paternal Child Care and Relationship
Quality: A Longitudinal Analysis of Reciprocal Associations.”
Journal of Marriage and Family 74(2):281–96.
Sprecher, Susan. 1986. “The Relation between Inequity and
Emotions in Close Relationships.” Social Psychology Quarterly
49:309–21. doi:10.2307/2786770
Stafford, Laura, and Daniel J. Canary. 2006. “Equity and Interdepen-
dence as Predictors of Relational Maintenance Strategies.”
Journal of Family Communication 6:227–54. doi:10.1207/
s15327698jfc0604_1
Staiger, Douglas, and James H. Stock. 1997. “Instrumental Variables
Regression with Weak Instruments.” Econometrica 65:557–86.
Sullivan, Kieran T., Lauri A. Pasch, Matthew D. Johnson, and
Thomas N. Bradbury. 2010. “Social Support, Problem Solving,
and the Longitudinal Course of Newlywed Marriage.” Journal
of Personality and Social Psychology 98:631–44.
Thrane, Christer. 2000. “Men, Women, and Leisure Time: Scandina-
vian Evidence of Gender Inequality.” Leisure Sciences 22(2):
109–22.
Tichenor, Veronica Jaris. 1999. “Status and Income as Gendered
Resources: The Case of Marital Power.” Journal of Marriage
and Family 61(3):638–50.
Voorpostel, Marieke, Tanja Van Der Lippe, and Jonathan Gershuny.
2010. “Spending Time Together—Changes over Four Decades
in Leisure Time Spent with a Spouse.” Journal of Leisure
Research 42(2):243–65.
Walster, Elaine, G. William Walster, and Ellen Berscheid. 1978.
Equity: Theory and Research. Boston: Allyn and Bacon.
Wilkie, Jane Riblett, Myra Marx Ferree, and Kathryn Strother
Ratcliff. 1998. “Gender and Fairness: Marital Satisfaction
in Two-Earner Couples.” Journal of Marriage and Family
60(3):577–94.
Williams, Kristi. 2003. “Has the Future of Marriage Arrived? A
Contemporary Examination of Gender, Marriage, and Psycho-
logical Well-Being.” Journal of Health and Social Behavior
44(4):470–87.
Woodin, Erica M. 2011. “A Two-Dimensional Approach to
Relationship Conflict: Meta-Analytic Findings.” Journal of
Family Psychology 25:325–35.
Zamir, Osnat, Abigail H. Gewirtz, Madelyn Labella, David S.
DeGarmo, and James Snyder. 2018. “Experiential Avoidance,
Dyadic Interaction and Relationship Quality in the Lives
of Veterans and Their Partners.” Journal of Family Issues
39(5):1191–1212.
Author Biographies
Daniel L. Carlson is an associate professor of family, health, and
policy in the Department of Family and Consumer Studies at the
University of Utah. He is a family demographer with expertise in
gender, life course, and health whose research focuses on the con-
sequences of the Second Demographic Transition (SDT) for health
disparities and social inequalities. To date, his general interest in the
consequences of changing family patterns for health disparities and
social inequalities has resulted in two integrated streams of research.
The first stream of research involves understanding how families
have changed during the SDT in the United States and what conse-
quences this has had for children’s and couples’ well-being. The
second research stream explores how the SDT has exacerbated
racial/ethnic and socioeconomic differences in family formation
and how this is related to health disparities and inequalities across
groups.
Amanda J. Miller is an associate professor of sociology and
director of faculty development at the University of Indianapolis.
Her work focuses on the intersections of family, gender, and social
class through examining couples’ divisions of household labor,
relationship progression, and fertility decisions. Furthermore, she
examines best practices for supporting underrepresented students
inside and outside of the college classroom. Most recently, she and
Sharon Sassler published the William J. Goode Award–winning
book Cohabitation Nation: Gender, Class, and the Remaking of
Relationships.
Stephanie Rudd is an assistant professor of social work at the
University of Indianapolis. Her work focuses on equity, oppression,
incivility, and dismantling racism in social work education. Prior to
her role as assistant professor, she worked as a social worker in dif-
ferent capacities for nearly 20 years with clients with serious mental
illnesses. Her teaching focuses on mental health and addictions, and
therefore she regularly presents on motivational interviewing and
mindfulness. She also remains in the clinical field as a clinical con-
sultant and licensure supervisor.
