ArticlePDF Available

Abstract

Studies consistently show sex differences in married couples' use of demand and withdraw behavior. The social structure hypothesis proposes that these differences are the result of power differentials between spouses. This study examined the link between 3 aspects of marital power and demanding and withdrawal behavior. Contrary to social structure predictions, results showed that wives did not possess less decision-making ability or access to resources and appeared to exhibit greater situational power (i.e., domineering and dominant behaviors) than did their husbands during problem-solving discussions. Furthermore, the spouse who exhibited the most demands also exhibited the most domineering and dominant behaviors, whereas the spouse who exhibited the most withdrawal exhibited the least domineering and dominant behaviors during problem-solving discussions. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Sex Differences in the Use of Demand and Withdraw Behavior in
Marriage: Examining the Social Structure Hypothesis
David L. Vogel, Megan J. Murphy, Ronald J. Werner-Wilson, Carolyn E. Cutrona, and Joann Seeman
Iowa State University
Studies consistently show sex differences in married couples’ use of demand and withdraw behavior. The
social structure hypothesis proposes that these differences are the result of power differentials between
spouses. This study examined the link between 3 aspects of marital power and demanding and withdrawal
behavior. Contrary to social structure predictions, results showed that wives did not possess less
decision-making ability or access to resources and appeared to exhibit greater situational power (i.e.,
domineering and dominant behaviors) than did their husbands during problem-solving discussions.
Furthermore, the spouse who exhibited the most demands also exhibited the most domineering and
dominant behaviors, whereas the spouse who exhibited the most withdrawal exhibited the least domi-
neering and dominant behaviors during problem-solving discussions.
Keywords: problem solving, demand, withdraw, marital communication, marital power
Currently, the lifetime probability of a marriage ending in di-
vorce is about 50% (Teachman, Tedrow, & Crowder, 2000). This
alarming figure has had negative effects on individuals, families,
and the children of these marriages. For example, research has
shown a link between divorce and the presence of physical and
psychological illness, suicide, violence, depressed immune func-
tion, homicide, and disease mortality rates (Burman & Margolin,
1992). Children of divorced parents tend to experience a range of
negative effects, such as poor academic performance, poor social
skills, and increases in conduct disorders, health problems, depres-
sion, and social isolation (Cowan & Cowan, 1990; Cummings &
Davies, 1994; Forehand, Brody, Long, Slotkin, & Fauber, 1986).
Furthermore, less satisfied marriages have negative effects on
family members, including higher rates of clinical depression
(Beach, Fincham, & Katz, 1998). Distressed marriages are a
source of stress—so much so that a person in a distressed marriage
may have poorer mental health outcomes than a person who is not
married but is socially isolated (Robles & Kiecolt-Glaser, 2003).
Researchers theorize that it is not divorce per se that negatively
affects the individual, but rather the continued marital conflict and
hostility that lead to negative outcomes (Gottman, Katz, &
Hooven, 1997).
Accordingly, the behaviors couples exhibit during conflict sit-
uations have been a central focus of researchers and counselors
developing interventions to alleviate or prevent marital and family
distress and to increase family functioning (Bradbury & Karney,
1993). Within the last 20 years, two behaviors— demand and
withdraw— have become of particular interest (Christensen,
1988). Demanding occurs when one partner pursues changes in the
relationship, whereas withdrawal occurs when one partner at-
tempts to avoid discussing a problematic issue in the relationship.
Demand and withdraw behaviors have been suggested to be com-
mon in intimate relationships (Gottman, 1999), even among rela-
tively satisfied couples (Vogel & Karney, 2002; Vogel, Wester, &
Heesacker, 1999). However, demand and withdraw behaviors have
been of particular concern because of their (a) frequent presence in
distressed couples (Christensen & Shenk, 1991), (b) link with
declines in relationship satisfaction over time (Heavey, Chris-
tensen, & Malamuth, 1995), and (c) link to spousal abuse (Berns,
Jacobson, & Gottman, 1999). In addition, studies consistently
show that women express more demands and men exhibit more
withdrawal during problem-solving discussions (e.g., Christensen
& Heavey, 1990; Gottman & Krokoff, 1989; Heavey, Layne, &
Christensen, 1993; Vogel & Karney, 2002). This sex difference in
problem-solving behaviors spurred interest in why these differ-
ences in behavior exist. If an accurate accounting of why these
differences exist could be found, interventions to assist couples in
changing these behaviors could be developed.
It was initially suggested that sex differences in the use of
demand and withdraw behavior reflect socialized differences in
needs and desires (Christensen, 1988) or biological dispositions
(Gottman & Krokoff, 1989) that women and men bring to a
relationship. Recently, however, researchers have suggested that
demanding and withdrawal behaviors are not due to intrinsic
differences in women and men but to power and resource inequal-
ities within marriages (e.g., Kluwer, Heesink, & Van De Vliert,
2000; Sagrestano, Heavey, & Christensen, 1998). This idea has
been called the social structure hypothesis, which suggests that
demand and withdraw behaviors are most likely to occur in situ-
ations in which there is an imbalance in power such that one
spouse “needs the other’s cooperation for resolution of the conflict
[whereas] the [other] partner can achieve satisfaction without the
David L. Vogel, Carolyn E. Cutrona, and Joann Seeman, Department of
Psychology, Iowa State University; Megan J. Murphy and Ronald J.
Werner-Wilson, Department of Human Development and Family Studies,
Iowa State University.
This research was supported by National Institute of Mental Health
Grant MH068289.
Correspondence concerning this article should be addressed to David L.
Vogel, Department of Psychology, Iowa State University, W112 Lagomar-
cino Hall, Ames, IA 50011-3180. E-mail: dvogel@iastate.edu
Journal of Counseling Psychology Copyright 2007 by the American Psychological Association
2007, Vol. 54, No. 2, 165–177 0022-0167/07/$12.00 DOI: 10.1037/0022-0167.54.2.165
165
other” (Sagrestano et al., 1998, p. 293). In these situations, the
spouse who needs his or her partner’s cooperation is most likely to
use demanding behaviors during the discussion to try to elicit a
change in the other (i.e., equalize the power). The spouse who does
not need his or her partner’s cooperation is most likely to withdraw
from the discussion, to maintain the power and not make any
changes. From this perspective, women are most likely to exhibit
demands and their husbands are most likely to exhibit withdrawal,
because sex-based inequalities in marriage (e.g., men’s control
over household income, women’s responsibility for housework or
child care) lead to an increased likelihood of wives being depen-
dent on their husbands for a successful outcome of a problem-
solving discussion (Christensen & Heavey, 1990; Heavey et al.,
1993; Sagrestano et al., 1998).
Demand–Withdraw Studies and the Social Structure
Hypothesis
Prior research examining sex differences in demand and with-
draw behavior offers some support for the social structure hypoth-
esis. For example, Christensen and colleagues (Christensen &
Heavey, 1990; Heavey et al., 1993; Sagrestano, Heavey, & Chris-
tensen, 1999) asked couples to discuss two problem areas, one
chosen by the husband and one chosen by the wife. They found
that the sex differences in demand and withdraw behavior emerged
during discussions of topics selected by wives. When couples
discussed the wife’s topic, she expressed more demands and he
expressed more withdrawal than when they discussed the hus-
band’s topic. Christensen and colleagues suggested that these
results are consistent with the social structure hypothesis, as the
discussion of an issue in which the wife desires the most change is
a context in which inequality in power is likely to be the greatest
(e.g., a discussion of housework or child care issues), and as such,
the wife is more likely to pressure for change, and the husband is
more likely to withdraw. Also consistent with these findings,
research has generally found that (a) wives desire a greater amount
of change in the relationship than do their husbands and that (b)
relationships in which the wife was dissatisfied with the division of
labor in the relationship report the greatest degree of wife-demand/
husband-withdraw behavior (Gray-Little, Baucom, & Hamby,
1996; Kluwer et al., 2000).
Inconsistencies in the Demand–Withdraw Research
Despite some evidence consistent with the social structure ex-
planation of demand and withdraw behaviors, this literature has
several limitations. First, some of the strongest evidence for the
social structure hypothesis comes from studies of marital conflict
in which the experimenter manipulated the problem-solving issue
discussed by having the couple discuss one topic chosen by the
husband and one topic chosen by the wife. The researchers then
have assumed that each spouse would pick a topic for which the
spouse’s cooperation is needed to resolve the conflict, and thus an
imbalance in power would be present. However, this is not nec-
essarily the case. Just because a spouse chooses a certain topic
does not make it one for which there is an imbalance (Caughlin &
Vangelisti, 1999). A spouse, for example, may choose topics in
which she or he knows her or his partner also desires a change as
a way of preempting or alleviating potential conflict (Vogel &
Karney, 2002). In addition, one spouse may exert more power in
a relationship, in general, or be able to exert greater influence in
certain spheres in which he or she has some expertise (Tichenor,
1999), regardless of who initiated the topic discussion. A woman,
for example, may hold more power in relational domains because
of the perception that women are skilled in handling relationship
issues (McGoldrick, Anderson, & Walsh, 1989). Consistent with
this argument, some studies have reported that men often feel
powerless in certain domains, such as the family (McGoldrick et
al., 1989). Thus, it is only assumed that it is a power imbalance
favoring the husband and not something else that leads to differ-
ences in demanding and withdrawal behavior across wives’ but not
husbands’ topics. To evaluate the precise effect that an imbalance
in the level of power has on the nature of a marital interaction
requires that each partner’s level of power in the relationship be
measured, as well as her or his level of power in relation to that
topic (Sagrestano et al., 1999).
Only a few studies have explicitly examined the relations be-
tween marital power and demand and withdraw behaviors, and
these studies have produced inconsistent findings. Sagrestano et al.
(1999), for example, assessed the role of perceived marital power
(measured by self-reports of the degree to which spouses felt they
could influence their partner) in demand–withdraw behavior. They
found that wives’ and husbands’ self-reported influence over their
partner was unrelated to self-reported demand–withdraw behavior.
Leonard and Senchak (1996) examined the relationship between
power (measured through a questionnaire of marital problem solv-
ing) and the degree of self-reported withdrawal behavior. A pos-
itive correlation (r.26) was found between withdrawal and
perceived power imbalances for wives’ reports but not for hus-
bands’ (r.06; Leonard & Senchak, 1996). Finally, a study by
Babcock, Waltz, Jacobson, and Gottman (1993) found that power
(measured by self-reports of discrepancy in decision making) was
not correlated with self-reported demand–withdraw behavior.
These inconsistencies have led to some concerns about the
accuracy of the social structure hypothesis. Furthermore, certain
studies of marital power in distressed and violent couples have
shown patterns inconsistent with high power leading to withdrawal
behavior and low power leading to demanding behavior. Sprenkle
and Olson (1978), for example, found that wives were more
dominant in clinic populations (which have also been shown to
exhibit greater wife-demand/husband-withdraw behavior), as op-
posed to nonclinic populations (which have been shown to exhibit
less wife-demand/husband-withdraw behavior). In addition, some
types of partner violence (i.e., intimate terrorism) are considered to
be a control tactic used by the dominant partner (see Johnson,
2005, for a discussion). However, violent couples actually have
been found to exhibit more husband-demand/wife-withdraw be-
havior than do their nonviolent counterparts (Babcock et al., 1993;
Holtzworth-Munroe, Smutzler, & Stuart, 1998), who are more
likely to exhibit wife-demand/husband-withdraw behavior.
Explanations for Inconsistencies in the Previous Research
One reason for the inconsistent results is that power in previous
studies has only been inferred from the topic selected or assessed
through self-reports of decision making or the ability to influence
one’s partner. This limited measurement of power can be prob-
lematic. For example, participants may not recall past attempts to
166 VOGEL, MURPHY, WERNER-WILSON, CUTRONA, AND SEEMAN
exert power accurately, or they may respond in socially desirable
ways. Wife-dominant patterns, for example, tend to counter social
norms, and egalitarian relationships may be seen as more desir-
able; therefore, couples may be less likely to report wife-dominant
behavior and more likely to report egalitarian behavior. Consistent
with this, studies have found that most couples reported them-
selves to be in an egalitarian relationship even when they were not
categorized as such by external observers (Knudson-Martin &
Mahoney, 1996). Couples may also be reluctant to discuss issues
of power in their relationship (Witteman & Fitzpatrick, 1986).
Thus, research on power must include both self-report and obser-
vational measures. Using a multimethod approach would
strengthen the ability of researchers to detect differences and better
represent the complexities of marital interactions (Gottman &
Notarius, 2000).
Furthermore, most of this research has used only a single mea-
surement of marital power (e.g., decision-making ability). How-
ever, marital power has been conceptualized as a multidimensional
concept. Cromwell and Olson (1975), for example, in the intro-
duction of their groundbreaking book Power in Families, sug-
gested that power can be conceptualized as having three domains:
(a) power bases, (b) power processes, and (c) power outcomes.
Power bases consist of “the resources an individual [sic] possesses
which may increase their [sic] ability to exercise control in a given
situation” (p. 6). Power bases traditionally have been measured in
terms of access to resources (e.g., socioeconomic status [SES]
differences between the spouses). The partner with greater SES is
thought to wield more power. Power processes are the interactions
or behaviors used to attempt to gain control over aspects of the
relationship. Power processes have been measured most com-
monly through observing attempts to gain control (named domi-
neeringness) and the success or failure of the specific attempts
(named dominance; Courtright, Millar, & Rogers-Millar, 1979;
Escudero, Rogers, & Gutierrez, 1997). Power outcomes refer to
who ultimately gets his or her way when there is a disagreement.
Power outcomes have been measured largely through self-reports
of who makes decisions. The failure to measure all three of these
domains in the demand–withdraw literature is an important over-
sight, as the role of these power domains in demand–withdraw
behavior is likely complex.
It is interesting that these three domains of power are only
minimally correlated, suggesting that they are tapping unique
aspects of the power in a relationship (Szinovacz, 1987). Few
studies have measured all three domains; however, the notion that
they are measuring unique aspects of power makes sense given
that power bases represent relatively stable characteristics associ-
ated with persons, whereas power processes and power outcomes
represent characteristics of specific interactions that may vary
widely from one situation to another (i.e., how much influence I try
to wield or actually do wield will vary depending on the type of
topic, my interest in the topic, etc.). Therefore, although power
bases and power processes should predict power outcomes, the
strength of the relationships among the three aspects of power will
likely vary from situation to situation. For example, the spouse
who makes the most money in a relationship may have more say
in decisions about money (a high correlation between power bases
and power outcomes), yet because this spouse may work a lot, he
or she may have less say in decisions about child care (a low
correlation between power bases and power outcomes). Thus, each
of these aspects of marital power may need to be measured
separately and their relationship to specific issues explored if
counselors are to better understand the role of power in a relation-
ship.
Purpose
Most researchers have found that demand and withdraw behav-
iors are detrimental to marital satisfaction (Christensen & Heavey,
1990; Jacobson & Margolin, 1979; Noller, Feeney, Bonnell, &
Callan, 1994); however, demand and withdraw behaviors are
found in many marriages, including satisfied ones (Gottman &
Krokoff, 1989; Heavey et al., 1995). Possible explanations for
these results point to the importance of studying marital power:
More egalitarian couples may engage in some demand and with-
draw behaviors but may not rigidly adhere to these behaviors,
whereas less egalitarian couples may engage in more rigid or
higher levels of demand and withdraw behaviors. The goal of this
investigation was to fill in the gaps in the demand–withdraw
literature by directly assessing the relationship among three do-
mains of power and how the domains relate to demand and
withdraw behaviors in married couples. Demand and withdraw
behavior was assessed through both self-report and observational
coding of problem-solving discussions. Power was measured
through differences in occupational status between the spouses
(power bases), observational ratings of the use of domineeringness
and dominance during a problem-solving discussion (power pro-
cesses), and self-report of overall level of power over decisions in
the relationship (power outcomes). Examination of these three
domains of power allows for a thorough analysis of the role of
power in the use of demanding and withdrawal behavior. This is an
important step, as marital therapy programs are often based on the
results of studies examining power processes and power outcomes.
Moreover, demand and withdraw behaviors are frequent in many
couples that counselors see. To our knowledge, this is the first
study to examine marital power and demand–withdraw behavior
with the use of multiple-method assessments. We had three hy-
potheses.
Hypothesis 1: Replication of Prior Research
Married couples typically exhibit sex-differentiated demand–
withdraw patterns during wives’ but not husbands’ topics. There-
fore, we hypothesized that these findings would be replicated in
this sample of married couples across both the self-report and the
observational measures.
Hypothesis 2: Sex Differences in Power Bases, Processes,
and Outcomes
The social structure hypothesis asserts that there are sex differ-
ences in marital power favoring husbands. Therefore, we hypoth-
esized that husbands would possess more power across the three
measured domains of marital power (power bases, power pro-
cesses, and power outcomes).
Hypothesis 3: Marital Power and Demand–Withdraw
Behavior
The main tenet of the social structure hypothesis is that the less
powerful partner is more likely to demand, and the more powerful
167
DEMAND AND WITHDRAW
partner is more likely to withdraw. Therefore, we hypothesized
that the spouse with the least power would exhibit the most
demands and the spouse with the most power would exhibit the
most withdrawal.
Method
Participants
Married, opposite-sex couples (N72) were solicited from an
advertisement placed in a weekly newsletter e-mailed to staff and
faculty and sent through e-mail to university students living in
marriage and family housing at a major university in the Midwest.
Other than asking about marital status, no exclusion criteria were
used. The advertisement offered $60 to couples willing to partic-
ipate in a study of marriage. On average, husbands were 33.5
(SD 8.60) years old; 46% of husbands were employed,
1
51%
were students, and 3% did not report their occupation. Wives
averaged 33.3 (SD 8.84) years old; 47% of wives were em-
ployed (see footnote 1), 40% were students, and 13% were home-
makers. Most participants were European American (66%) fol-
lowed by Asian (22%), Hispanic (5%), African American (4%),
and 3% other nationalities. The average length of the marriage was
84.4 months (SD 85.2).
Measures
Demand and withdraw. To evaluate the extent to which cou-
ples engaged in demand and withdraw behavior during their in-
teractions, we globally coded videotapes of the interactions with
the Interaction Rating System developed by Christensen and
Heavey (1990). The system consists of five dimensions used to
calculate demand and withdraw behaviors (avoidance, discussion,
blame, pressure for change, and withdraws). Each scale asks the
raters to place the spouse’s behavior for that dimension on a
9-point scale. As Heavey et al. (1995) suggested, a final withdraw
score was computed by summing the avoidance and withdrawal
dimensions and then subtracting the discussion score. In turn, a
final demand score was computed by combining the blame and
pressure for change dimensions. Prior research has shown that
these rating dimensions correlate with self-report measures of
demand and withdraw behaviors (Christensen & Heavey, 1990;
Heavey et al., 1993, 1995). Additionally, Holtzworth-Munroe,
Smutzler, Bates, and Vogel (1993) found high correspondence
between this coding system and other coding systems measuring
demand–withdraw behaviors (e.g., Klinetob & Smith, 1996). The
five dimensions that make up the demand–withdraw codes have
been shown to be reliable (average r.85) after independent
raters received at least 3 weeks of training from someone knowl-
edgeable in the use of the rating system (Heavey et al., 1993).
Before ratings of demand and withdraw behaviors were ob-
tained, six independent raters (three male and three female under-
graduate research assistants), who were blind to the research
hypotheses, received approximately 70 hr of training, across 8
weeks, from authors David L. Vogel and Joann Seeman in the use
of the rating system. Training consisted of viewing and rating a
series of videotapes from a separate study of couples interacting.
Raters were instructed to consider frequency, intensity, and dura-
tion of the participants’ verbal and nonverbal behaviors before
making their ratings.
To ensure that interrater reliability remained acceptable
throughout the coding process, each week we assigned two random
subsets of three raters to observe 10 couples (12 couples the final
week), with one set of three raters observing the husband’s dis-
cussion topic and one set of three raters observing the wife’s topic
(i.e., all topics were coded by three independent raters) for a given
couple. The mean of the three raters’ scores was used in the final
analyses. At the end of training, the interrater reliability among the
raters was .83. Each group of three raters continued to achieve
acceptable interrater reliability (i.e., intraclass correlation coeffi-
cients) throughout the study, with an overall reliability of .77
(blame .81, pressure for change .77, withdraw .73, avoid-
ance .78, discussions .79, demand .80, withdraw .75).
Demand and withdraw behaviors were also measured with the
Communication Patterns Questionnaire Short Form (CPQSF;
Christensen, 1988). The CPQSF is designed to assess an individ-
ual’s perception of the way that discussions with his or her partner
are generally conducted. The participant indicates on a 9-point
scale from 1 (very unlikely)to9(very likely) the likelihood that the
couple interacts in a specific manner (e.g., mutual negotiation or
mutual blame) when discussing a specific issue. The current study
asked participants to fill out items on the basis of their identified
topic (i.e., wives filled it out regarding the topic they selected to
discuss, and husbands filled it out regarding the topic they selected
to discuss). All questions on the CPQSF ask about behaviors at the
level of the dyad (e.g., man pressures, nags, or demands while
woman withdraws, becomes silent, or refuses to discuss the matter
further) rather than at the level of the individual (e.g., man pres-
sures, nags, or demands). The measure assesses (a) the likelihood
of the husband demanding while the wife withdraws and (b) the
likelihood of the wife demanding while the husband withdraws, as
well as (c) the total demand–withdraw communication behaviors
of the couple, which is the sum of the first two subscales. Higher
scores on a subscale mean that the person is reporting that the
couple engages in a greater amount of those types of communica-
tion behaviors while discussing a specific issue. The validity of the
scale has been demonstrated through positive correlations with
positive verbal and nonverbal communication behaviors and pos-
itive correlations with negative verbal and nonverbal communica-
tion behaviors (Bodenmann, Kaiser, Hahlweg, & Fehm-Wolfsdorf,
1998). Noller and White (1990) also found that the demand–
withdraw patterns, specifically the wife-demand/husband-
withdraw pattern, discriminated accurately between happy and
unhappy couples. Further validity of the CPQSF has also been
found, as the subscales correlate significantly with observational
measures of demand–withdraw (Bodenmann et al., 1998; Chris-
tensen & Heavey, 1990; Hahlweg, Kaiser, Christensen, Fehm-
Wolfsdorf, & Groth, 2000; Heavey et al., 1993).
1
Husbands were employed as radiation safety technician, university
staff, manager, system analyst, researcher, electrician, project programmer,
professor, teacher, coach, carpenter, technician, artist, physicist, accoun-
tant, and minister and in advertising, computer support, insurance, and
graphic design. Wives were employed as communications director, uni-
versity staff, manager, researcher, electrician, teacher, accountant, coun-
selor, technical writer, physicist, customer service, secretary, and recep-
tionist.
168 VOGEL, MURPHY, WERNER-WILSON, CUTRONA, AND SEEMAN
Christensen and Heavey (1990) reported reliability of the pat-
terns measured by CPQSF with internal consistencies for wives of
.85 for wife-demand/husband-withdraw and .50 for husband-
demand/wife-withdraw. For the husbands’ data, the alphas were
.71 and .72, respectively. The reliabilities are similar even when
the measure has been used in other countries (Bodenmann et al.,
1998; Hahlweg et al., 2000). In the current sample, the Husband-
Demand/Wife-Withdraw Scale had an alpha of .70 (husbands) and
.60 (wives); the Wife-Demand/Husband-Withdraw Scale had an
alpha of .66 (husbands) and .75 (wives).
Marital power. Following Cromwell and Olson’s (1975) the-
ory, we assessed three domains of marital power (power bases,
processes, and outcomes). Power bases were classified on the basis
of spouses’ current occupation. Two psychology professors were
given the seven rank-ordered occupation categories from Hollings-
head’s Two-Factor Index of Social Position and asked to catego-
rize each couple as (a) being equal in terms of occupation status,
(b) favoring the wife, or (c) favoring the husband on the basis of
their relative ranking of their occupations. For example, if the
husband’s job was in sales and the wife’s job was a manager, the
couple would be categorized as having an SES favoring the wife.
The two professors agreed on all but three cases (two cases were
not classified because 1 partner did not report an occupation), for
96% agreement (␬⫽.93, p.001). In these three cases, a final
code was selected by consensus.
Power processes were assessed with two aspects (domineering-
ness and dominance) of partners’ power during the problem-
solving discussions. Domineeringness, an individual measure of
power, is defined as the number of control attempts exhibited
during the conversations. Dominance, a relational measure of
power, is defined as the number of control attempts by a person
followed by the partner’s submissive and/or accepting moves.
Domineeringness and dominance were assessed with the Family
Relational Communication Control Coding System (FRCCCS;
Heatherington & Friedlander, 1987). To determine whether a
statement was a control attempt, we took into account both the
structure and grammar of the statement and the pragmatic function
of the statement. For example, a statement coded as an assertion
(structure) and an order (pragmatic function) was determined to be
a domineering behavior, according to a system of rules developed
by the coding system’s authors. The FRCCCS has reported inter-
rater reliability (kappa ranging from .75 to .95; Heatherington &
Friedlander, 1987) and is moderately related to other relational
control schemes (Tracey & Ray, 1984). Criterion validity of the
FRCCCS has been established in that family therapists who were
not trained on this coding system were able to match the FRCCCS
codes (ztest of Cohen’s ␬⫽.56, p.001; Gaul, Simon, Fried-
lander, Culter, & Heatherington, 1991). Independent raters (six
female research assistants), who were blind to the research hy-
potheses, received approximately 150 hr of training across 15
weeks from author Megan J. Murphy in the use of the coding
system. Training consisted of coding transcripts of couples’ vid-
eotapes from an unrelated study. Fifty percent of the couples’ tapes
were coded by two raters to ensure that interrater reliability re-
mained acceptable throughout the coding process. Disagreements
were resolved by the training team leader. At the end of training,
the interrater reliability between the raters was .84 (range
.74 –.94). The raters continued to achieve acceptable reliability
throughout the study, with an overall kappa of .80 (range
.68 –.92).
Finally, power outcomes were assessed with the Perceived Mar-
ital Power Scale (Sagrestano et al., 1999). The Perceived Marital
Power Scale was chosen on the basis of its use in other studies
examining marital power and the use of demand and withdraw
behavior (Sagrestano et al., 1999) The Perceived Marital Power
Scale asks participants to respond to four items regarding (a)
potential ability to influence one’s partner, (b) confidence in in-
fluencing one’s partner, (c) likelihood of influencing one’s partner,
and (d) overall rating of whether one can influence the partner. The
items are scored on a 7-point scale, with higher scores representing
more perceived power. Validity of the scale was demonstrated by
its correlation with husbands’ (r.31–.36) and wives’ (r
.31–.32) use of physical violence in the relationship. The internal
reliability of the scale has been reported previously (␣⫽.60 –.67).
For the current study, Item 3, “I am likely to do what my partner
wants,” was dropped because this increased the reliability of the
measure (from .59 to .72 for men and from .61 to .64 for women).
Dropping this item did not change any of the subsequent results.
Marital satisfaction. The Dyadic Adjustment Scale–7
(DAS-7; Hunsley, Best, Lefebvre, & Vito, 2001; Hunsley, Pinsent,
Lefebvre, James-Tanner, & Vito, 1995) was used to measure each
couple’s level of marital adjustment. The DAS-7 is a seven-item
instrument developed from the original DAS (Spanier, 1976) and
was designed to assess the quality of the relationship as perceived
by married or cohabitating couples. The DAS-7 has six items rated
on a 6-point scale and one item rated on a 7-point scale, with
higher scores reflecting greater satisfaction. The DAS-7 has evi-
dence of validity, correlating with the Kansas Marital Satisfaction
Scale (r.67–.73) and the Emotional Self-Disclosure Scale (r
.41–.50; Hunsley et al., 2001). The DAS-7 has shown validity by
discriminating between community and clinic samples (Hunsley et
al., 2001). The DAS-7 also has an internal consistency of .75 to .82
(Hunsley et al., 2001, 1995). For the current study, the internal
consistency was .80 for wives and .68 for husbands.
Procedures
Upon arriving for the initial visit, each spouse was escorted to a
separate room, where informed consent was obtained. While sep-
arated, each participant was asked independently to complete a
survey containing the questionnaires used in the study (e.g., rela-
tionship satisfaction, assessment of overall decision-making ability
in the relationship) and to identify a problem area in the relation-
ship. The problem was identified by asking each spouse to select
an issue in the relationship in which he or she desires the most
change and that cannot be resolved without the spouse’s cooper-
ation. The issues selected by husbands and wives are found in
Table 1. Spouses were then asked to answer some questions about
this topic, including the type of problem-solving behaviors that
generally occur when this topic arises and the importance of the
topic. Subsequently, the spouses were brought together and asked
to discuss each of the problem topics for 10 min. Whose problem
they were asked to discuss first was randomized through a coin
flip. If both spouses chose the same topic, the spouse who lost the
coin flip was asked to choose a second topic. After 10 min, the
experimenter knocked on the door, and the couple was asked to
discuss the second topic (i.e., the other spouse’s topic) for 10 min.
169
DEMAND AND WITHDRAW
Each discussion was videotaped. At the end of the discussions,
couples were separated again. While separated, each spouse was
debriefed, and her or his feelings and reactions to the study were
discussed. Referrals to local mental health and family therapy
clinics were given to all participants.
Results
Descriptive Statistics
Table 2 presents the means and standard deviations for all
variables examined in the analyses. As the means in Table 2
reveal, the married couples in this sample were slightly dissatisfied
(husband: M25.44, SD 3.99; wife: M25.93, SD 5.20
[scale range 7– 43]) and similar to other community samples
(husband: M25.30, SD 4.70; wife: M26.40, SD 4.70;
Hunsley et al., 2001). Couples in the current sample viewed their
selected issues as moderately important (6 on a 9-point scale)
and were generally rated as engaging in low to moderate levels of
demand and withdraw behavior. Their demand and withdraw be-
haviors were very similar in level to other studies that have
measured husbands’ and wives’ demand and withdraw in the same
manner (see mean and standard deviation results from Heavey et
al., 1995, in the parentheses in Table 2). The table also reveals
substantial variability on all of these variables, justifying further
analyses.
Hypothesis 1: Replication of Prior Research
Previous research (e.g., Christensen & Heavey, 1990; Heavey et
al., 1993) has shown that married couples typically exhibit sex-
differentiated demand–withdraw patterns during wives’ but not
husbands’ topics. Therefore, we tested whether this pattern of
results was replicated among this sample of married couples.
These analyses were conducted on both the self-report and the
observational data.
Self-report. Using the CPQSF, participants reported the de-
gree to which wives demand while husbands withdraw and the
degree to which husbands demand while wives withdraw when
discussing a topic of importance to them. To account for the
dependency among the paired husband–wife scores, we conducted
repeated measures analyses of variance (ANOVAs) in which topic
(wives’ report of her topic and husbands’ report of his topic) and
demand–withdraw role (wife-demand/husband-withdraw and
Table 1
Frequency and Percentage of Issues Selected by Husbands and
Wives
Issue
Husbands Wives
Frequency % Frequency %
Friends and family 7 10 14 19
Needs 3400
Sex 3411
Money 13 18 9 13
Intimacy 1111
Time together 9 13 7 10
Feelings and emotions 7 10 9 13
Making decisions 13 18 3 4
Housework 11 15 11 15
Communication 2334
Children 0023
Other relationship changes 3 4 12 17
Table 2
Means and Standard Deviations for All Variables
Variable
Husband Wife
MSDMSD
Self-report
Marital satisfaction (DAS-7) 25.44 (25.30)
a
3.99 (4.70)
a
25.93 (26.40)
a
5.20 (4.70)
a
HDWW (CPQSF) 12.88 5.34 10.24 5.26
WDHW (CPQSF) 12.68 5.35 13.38 6.44
Decision making (PMP) 14.32 3.38 13.68 3.44
Importance of issue 6.33 1.94 6.82 2.04
Observed behavior
Husband’s topic
Demand 3.21 (3.59)
b
1.28 (1.13)
b
3.19 (3.22)
b
1.39 (1.22)
b
Withdraw 0.68 (1.55)
b
3.31 (2.42)
b
0.49 (1.15)
b
3.57 (2.98)
b
Domineeringness 0.35 0.16 0.39 0.18
Dominance 0.13 0.07 0.17 0.08
Wife’s topic
Demand 2.84 (2.68)
b
1.02 (1.03)
b
3.93 (4.29)
b
1.92 (1.23)
b
Withdraw 0.50 (0.03)
b
3.24 (2.42)
b
1.14 (1.98)
b
3.34 (2.13)
b
Domineeringness 0.36 0.15 0.40 0.17
Dominance 0.13 0.07 0.17 0.10
Note. DAS-7 Dyadic Adjustment Scale–7; HDWW Husband demand/Wife withdraw; CPQSF Communication Patterns Questionnaire Short
Form; WDHW Wife demand/Husband withdraw; PMP Perceived Marital Power scale.
a
For comparative purposes, the means and standard deviation results from Hunsley et al. (2001) are presented in the parentheses.
b
For comparative purposes, the means and standard deviation results from Heavey et al. (1995) are presented in the parentheses.
170 VOGEL, MURPHY, WERNER-WILSON, CUTRONA, AND SEEMAN
husband-demand/wife-withdraw) served as within-subject factors.
The results replicate previous findings, as there was no significant
main effect for topic, F(1, 71) 2.12, p.15, partial
2
.03,
but a significant main effect for demand–withdraw role, F(1,
71) 4.44, p.04, partial
2
.12, qualified by a significant
Topic Role interaction, F(1, 71) 9.68, p.003, partial
2
.06. To determine the nature of the interaction, we performed two
planned comparisons. The first examined the mean differences
between the wife-demand/husband-withdraw and husband-
demand/wife-withdraw patterns during her topic, and the second
examined the same differences during his topic. The mean differ-
ence between the wife-demand/husband-withdraw (M13.38,
SD 6.44) and husband-demand/wife-withdraw (M10.24,
SD 5.26) patterns was significant for wives’ reports about their
topic, t(71) 3.92, p.001, partial
2
.18, but not for the
wife-demand/husband-withdraw (M12.68, SD 5.35) and
husband-demand/wife-withdraw (M12.88, SD 5.34) patterns
reported by husbands about their topic, t(71) ⫽⫺0.20, p.84,
partial
2
.001. Wives reported more wife-demand/husband-
withdraw than husband-demand/wife-withdraw during their topic,
which was consistent with previous research. By contrast, hus-
bands reported no differences in wife-demand/husband-withdraw
or husband-demand/wife-withdraw during their topic.
Observed demand and withdraw behavior. To match the self-
report data, we first examined the observational data by following
Heavey et al.’s (1993) precedent of generating a score indicating
the degree of wife-demand/husband-withdraw behavior by sum-
ming the behavioral ratings of wife’s demanding behavior and
husband’s withdrawal behavior in each problem-solving interac-
tion. Similarly, we generated a score indicating the degree of
husband-demand/wife-withdraw behavior by summing the behav-
ioral ratings of husband’s demanding behavior and wife’s with-
drawal behavior in each interaction. As with the self-report data,
we accounted for the dependency among the paired husband–wife
scores by conducting a repeated measures ANOVA in which topic
(wives’ topic and husbands’ topic) and demand–withdraw role
(wife-demand/husband-withdraw and husband-demand/wife-
withdraw) served as within-subject factors. In addition, to control
for topic order effects, we included whose identified problem was
discussed first (hers first vs. his first) as a between-subjects block-
ing factor. As with the self-report data, there was no significant
main effect for topic, F(1, 70) .02, p.88, partial
2
.00, but
a significant main effect for demand–withdraw role, F(1, 70)
29.0, p.001, partial
2
.29, qualified by a significant Topic
Role interaction, F(1, 70) 5.36, p.02, partial
2
.07. To
determine the nature of the interaction, we performed the same two
planned comparisons, demonstrating significantly more wife-
demand/husband-withdraw (M3.42, SD 3.93) than husband-
demand/wife-withdraw (M1.70, SD 3.48) during discussions
of wives’ topics, t(71) 3.69, p.001, partial
2
.16, but no
significant difference between wife-demand/husband-withdraw
(M2.50, SD 3.41) and husband-demand/wife-withdraw (M
2.72, SD 3.44) during discussions of husbands’ topics, t(71)
0.70, p.49, partial
2
.01. As with the self-report data, more
wife-demand/husband-withdraw than husband-demand/wife-
withdraw behaviors occurred during the wives’ topics, and no
differences in wife-demand/husband-withdraw and husband-
demand/wife-withdraw behaviors occurred during the husbands’
topics.
Next, to clarify whether the above sex differences in the way
husbands and wives engage in the demand–withdraw pattern were
a result of differences in demanding, differences in withdrawal, or
both, we conducted paired sample ttests comparing wives and
husbands in their tendency to engage in demand and withdraw
behavior, separately, during each interaction. These analyses re-
vealed that, during discussions of wives’ topics, wives (M3.93,
SD 1.92) were significantly more demanding than husbands
(M2.84, SD 1.02), t(71) 4.47, p.001, partial
2
.22,
and husbands (M⫽⫺0.50, SD 3.24) were significantly more
withdrawing than wives (M⫽⫺1.14, SD 3.34), t(71) 2.42,
p.02, partial
2
.08. During discussions of husbands’ topics,
wives (M3.19, SD 1.39) were not significantly more de-
manding than husbands (M3.21, SD 1.28), t(71) 0.13, p
.90, partial
2
.00, and husbands (M⫽⫺0.68, SD 3.31) were
not significantly more withdrawing than wives (M⫽⫺0.49, SD
3.57), t(71) ⫽⫺0.93, p.35, partial
2
.01. Thus, the elevated
levels of the wife-demand/husband-withdraw pattern, identified
above, appear to reflect both a high level of wives’ use of demand-
ing behaviors and husbands’ use of withdrawing behavior during
discussions of wives’ issues.
Hypothesis 2: Sex Differences in Power Bases, Processes,
and Outcomes
Thus far, our results replicate prior studies and support the social
structure hypothesis. We next examined whether there would be
the expected sex differences in marital power (i.e., that husbands
would possess more power). To examine this, we first examined
the chi-square difference in wives’ and husbands’ SES (power
bases). We then conducted repeated measures ANOVAs compar-
ing wives’ and husbands’ self-reported ability to make decisions in
the relationship (power outcomes) and observed domineeringness
and dominance behavior during the problem-solving discussions
(power processes). We accounted for the dependency among the
paired husband–wife scores by using spouse sex (husband and
wife) serve as a within-subject factor in the ANOVA analyses. For
the observation data (dominance and domineeringness), we also
controlled for whose identified problem was discussed first (hers
first vs. his first) by having topic order serve as a between-subjects
blocking factor.
The largest group of couples in this sample was classified as
having equal status occupations (43%), followed by couples with
status favoring the husband (32%) and then wives (22%). How-
ever, these differences were not significant,
2
(2, 67) 3.74, p
.15. Furthermore, wives and husbands reported having equal levels
of decision-making ability, F(1, 71) 1.32, p.26, partial
2
.02. Of interest, though, is that when examining the observed
domineering and dominance data, we found that wives, during
discussions of either husbands’ topics, F(1, 70) 7.68, p.007,
partial
2
.10, or wives’ topics, F(1, 70) 6.38, p.014,
partial
2
.08, behaviorally exhibited more domineering at-
tempts than husbands and were more likely to be dominant (i.e.,
have their partner give in) than were husbands during either
husbands’ topics, F(1, 70) 6.68, p.012, partial
2
.09, or
wives’ topics, F(1, 70) 5.80, p.019, partial
2
.08, topics
(see Table 1 for means and standard deviations). These results did
171
DEMAND AND WITHDRAW
not support the social structure hypothesis,
2
as the expected sex
differences in power, favoring men, were not found.
3
Hypothesis 3: Marital Power and Demand and Withdraw
Behavior
Although the expected sex differences in marital power, favor-
ing men, were not found, we still wanted to directly examine the
main tenet of the social structure hypothesis that the less powerful
partner is more likely to use demands and the more powerful
partner is more likely to withdraw. One way this concept has been
examined in previous research (see Vogel & Karney, 2002) has
been to look at the importance of the issue selected. Imbalances in
the relationship could lead to one partner having more need for his
or her topics to be raised. The social structure explanation of
demand and withdraw behavior then suggests spouses ought to
make more demands when discussing topics of greater importance
to them and to withdraw more when discussing topics of less
importance to them. Therefore, we examined the associations
between spouses’ ratings of the importance of their topic and their
spouses’ demand and withdraw behavior during discussions of that
topic. Table 3 presents the zero-order partial correlations between
each spouse’s ratings of his or her issue importance and spouses’
demanding and withdrawal behavior controlling for spouses’ rat-
ing of their issue importance. As the table reveals, and inconsistent
with the social structure predictions, our results showed that during
discussions of husbands’ or wives’ topics, the importance of the
topic to the husband or wife was not related to demand behavior
for either spouse (rs–.08 to .09). Similarly, during discussions
of husbands’ or wives’ topics, ratings of topic importance were
related (rs–.16 to .36) to each spouse withdrawing less, not
more (although only statistically significant in one case for wives).
Furthermore, paired sample ttests revealed that husbands and
wives did not rate their partner’s topic as less important than their
own ( ps.4; see Table 2 for means).
Although topic importance is one way to test the social structure
hypothesis, another way is to examine the relations among the
three domains of marital power (power bases, processes, and
outcomes) and who does the most demanding (difference between
husband and wife demands) and withdrawing (difference between
husband and wife withdrawal) in the relationship. The social
structure hypothesis proposes that the person with least power
should exhibit the most demands and the person with the most
power should exhibit the most withdrawal. To test this hypothesis
in relation to power bases, we conducted separate ANOVAs with
SES classification (1 husband, 0 equal, –1 wife) as the
independent variable and who did the most demanding and who
did the most withdrawal during each problem-solving discussion
as the dependent variables. To test the hypothesis in relation to
power processes, we examined the correlations between who ex-
hibited the most domineeringness during each discussion (differ-
ence between husband and wife domineeringness) and who exhib-
ited the most dominance during each discussion (difference
between husband and wife dominance) and who exhibited the most
demanding and the most withdrawal. Similarly, to examine the
hypothesis in relation to power outcomes, we examined the cor-
relations between who reported the most self-reported decision-
making ability (difference between husband and wife self-reported
decision-making ability) and who exhibited the most demand and
withdraw behavior. The results of all of these analyses were
inconsistent with social structure hypothesis. Differences in SES
between wives and husbands did not differentiate who performed
the most demanding or withdrawal. In addition, as seen in Table 4,
during either the husbands’ or wives’ topic, the spouse who ex-
hibited the most domineering behavior and most dominance ex-
pressed the most demands (rs.32 to .45). Similarly, the spouse
who exhibited the least domineering behavior during the wives’
topic expressed greater withdrawal behavior (r–.29). Further-
more, who withdrew the most was not related to domineering
behavior during the husbands’ topic (r.00) or dominance during
either topic (and the direction of the correlations, r–.17 and
–.21, is consistent with other findings). There was also no rela-
tionship between who performed the most demands or the most
withdrawal and who reported the most self-reported decision-
making ability ( ps.07).
Discussion
The results of this study across both the self-report and obser-
vational data replicate previous research demonstrating a sex dis-
crepancy in the use of demand and withdraw behaviors during
discussions of issues selected by wives but not during discussions
2
One reason these findings regarding power may be opposite of what is
predicted by the social structure hypothesis is that in relatively satisfied
couples, power differences between spouses are less severe than, or even
reversed, in distressed couples. In dissatisfied couples, wives may have less
influence. To address this possibility, we conducted the same repeated
measures ANOVAs but included the couples’ marital satisfaction score as
a factor. If the social structure hypothesis is correct just for distressed
couples, then we should find an interaction between satisfaction and
spouses’ sex (i.e., moderation effect) such that the most distressed couples
would exhibit the most sex discrepancies in marital power in favor of the
husband. However, marital satisfaction did not significantly moderate the
effect of sex on marital power ( ps.13).
3
As our sample was more diverse than many studies on marital couples
(i.e., 34% non-Caucasian), we decided to examine the potential effect of
couples’ race/ethnicity on the results. Specifically, we re-examined the
results by including the racial/ethnic groups with sufficient numbers (i.e.,
Caucasian and Asian) in the analyses (all other groups, n3). Therefore,
we conducted the same analyses as before, but this time including race/
ethnicity (Caucasian 1, Asian 2) as a between-subjects factor. How-
ever, race/ethnicity did not have a significant effect on the self-report or
observed degree of power discrepancies between wives and husbands
(ps.12).
Table 3
Correlations Between Topic Importance and Observed Demand–
Withdraw Behaviors During His Topic and Her Topic
Topic importance
Husband
demand
Husband
withdraw
Wife
demand
Wife
withdraw
Husband’s topic
His rating of importance .07 .16 .04 .20
Wife’s topic
Her rating of importance .08 .23 .09 .36
**
**
p.01.
172 VOGEL, MURPHY, WERNER-WILSON, CUTRONA, AND SEEMAN
of issues selected by husbands (Christensen & Heavey, 1990;
Heavey et al., 1993). Although these initial results support the
social structure hypothesis, the other findings of the study do not.
The subsequent analyses revealed that husbands and wives had
equal SES, picked topics of equal importance to them, and self-
reported having equal ability to make decisions in the relationship.
Moreover, wives behaviorally exhibited more domineering at-
tempts and were more dominant (i.e., more likely to have their
partner give in) than husbands during discussions of either
spouse’s topic. Thus, the expected sex differences in marital
power, favoring husbands, were not found. Furthermore, spouses
who expressed the most demands during either husbands’ or
wives’ topics did not appear to exhibit less situational power
during the discussion (power process), as would be hypothesized
by the social structure hypothesis, but in fact exhibited the most
domineering behavior and were most likely to be dominant during
the discussions. Similarly, exhibiting greater withdrawal was not
related to domineering behavior during the husbands’ topic or
dominance during either topic and actually related to less use of
domineering behavior during the wives’ topic. We also found no
statistically significant relationship between differences in SES
status between spouses (power bases) or who self-reported the
most ability to make decisions in the relationship (power out-
comes) and who demanded or withdrew the most. In all, these
results suggest that spouses may exert certain types of marital
power and may exert more or less power across areas of their
relationship. Thus, the relationship between marital power and
demand and withdraw behaviors is more complex than simply who
is able to exert more control in the relationship.
The lack of the clear confirmation of the social structure hy-
pothesis, although somewhat unexpected, is in fact largely consis-
tent with the extant studies. The self-report studies that have
looked specifically at marital power and problem-solving behavior
have shown that husbands’ and wives’ self-reported ability to
make decisions in the relationship was unrelated to self-reported
demand–withdraw behavior (Babcock et al., 1993; Sagrestano et
al., 1999), that wives are more dominant in relational domains
(Sprenkle & Olson, 1978), and that couples exhibiting violent
behaviors are most likely to use male demand–female withdraw
behavior rather than the reverse (Babcock et al., 1993; Holtzworth-
Munroe et al., 1998). The strongest evidence for the social struc-
ture hypothesis came from studies of marital conflict in which the
experimenter manipulated the problem-solving issue discussed. In
this study, when we manipulated the topic discussed in the same
manner, we found the same demand–withdraw difference that
other studies have found. However, when we went further and—
instead of assuming an imbalance in power was present—assessed
marital power from three different domains and used both self-
report and observational data, we found results inconsistent with
the notion that during problem-solving discussions, women have
less marital power and therefore demand more and that men have
greater marital power and thus withdraw more.
One reason for the current findings is that choosing a specific
topic does not necessarily make it one for which there is a true
imbalance in marital power. Many forms of power may extend
themselves across domains. Women may have to address a topic
several times to exert enough influence for their partners to accede
to their wishes, whereas men may only have to address a topic
once to influence wives to change in response to their topic. In
other words, the topic of discussion in itself may not provide
researchers with an “even playing field” to examine power differ-
ences. Some researchers have also noted that many relationship
topics do not neatly fall into “his” and “her” issues and that
couples may often choose to discuss issues for which both desire
an equal amount of change (Caughlin, & Vangelisti, 1999). In fact,
one study by Caughlin and Vangelisti (2000) found no sex differ-
ences in who desires the most change in their partner. Furthermore,
many of the issues discussed by the couples could be considered in
the relational domain, which women may feel more responsible for
managing, possibly because of the perception that women are more
skilled than men in handling relationship issues (McGoldrick et al.,
1989). This assertion is strengthened when one looks to the dif-
ferential findings regarding the observed and self-reported marital
power data. The self-report data asked participants to respond to
questions about their overall ability to make decisions in the
relationship (i.e., was not domain specific), and this may have led
these couples to be more likely to report equal power. However,
power processes (i.e., domineeringness and dominance) were ob-
servationally coded from the problem-solving discussions. These
discussions were focused on “changing or improving the relation-
ship” and thus may have been an area in which women are able to
assert more marital power.
Our inconsistent findings with the social structure hypothesis
lend support to recent alternative hypotheses (e.g., personality
factors; Caughlin & Vangelisti, 2000) as well as the notion that
demand and withdraw behaviors may simply be learned (rein-
forced) patterns over time. However, this does not mean the social
structure hypothesis may not be partially accurate. For example, it
is possible that although the social structure hypothesis does not
apply to nondistressed couples, it does apply to more distressed
couples. In nonclinically distressed couples, power may be equal,
and women may be able to assert their needs (domineeringness)
and have those needs responded to (dominance). Gottman (1999),
for example, noted that marriages in which men accept women’s
influence tend to be more satisfied. In clinically dissatisfied cou-
ples, however, women may have less power and/or may not have
their power attempts responded to. Although our findings did not
show that the couples’ marital satisfaction level was related to
marital power differences between the spouses, previous research
has found that unhappy spouses are less likely to have negotiated
mutually respected roles (Gray-Little et al., 1996). Thus, our
sample, although similar to most community research samples and
Table 4
Correlations Between Who Was Most Demanding, Who Was the
Most Withdrawing, and Who Had the Greater Power During
His Topic and Her Topic
Behavior
Husband topic Wife topic
Most
domineering
Most
dominant
Most
domineering
Most
dominant
Most
demanding .40
***
.33
**
.45
***
.32
**
Most
withdrawing .00 .17 .29
*
.21
Note.
*
p.05.
**
p.01.
***
p.001.
173
DEMAND AND WITHDRAW
slightly distressed, may not have had serious enough problems to
elicit the power imbalances and the demand–withdraw patterns
described by the social structure hypothesis. Therefore, future
researchers may want to replicate the current study with a more
distressed or specifically clinical sample.
Another possible reason that we did not find the expected
support for the social structure hypothesis is how we assessed
marital power. We went further than previous studies in including
an assessment of three domains of marital power; however, marital
power can be measured in a number of ways. In particular, power
bases have been suggested to contain a number of important
elements that could be assessed. Although we measured power
bases in the most traditional way, as personal assets or resources
(i.e., occupational status), power bases could also be assessed
through gendered role expectations and norms, societal sanctions,
personal attributes such as attractiveness, communication skills
and expertise, the desire to be in the relationship, as well as the
perceived number of attractive alternative options (Babcock et al.,
1993). Each of these different aspects may play a role in the overall
amount of power spouses have in their marriage and may need to
be measured separately to understand the relationship of marital
power and the use of demand and withdraw behavior. For exam-
ple, it may be particularly important to assess gender role norms
and expectations. The wife earning more or having a higher status
job does not necessarily translate into a relationship in which the
woman has more marital power, possibly owing to different con-
tributions being weighted differentially, depending on whether the
contribution fits gender role expectations (Tichenor, 1999). Thus,
a woman’s earnings may be weighted to have less impact, and a
man’s earnings weighted more, to fit societal notions and make the
discrepancies acceptable (Komter, 1989). Therefore, true power
discrepancies, from this perspective, may be based less on overt
differences between spouses but more on the expectations (explicit
or unstated) that give one partner greater ability to exert power in
a given situation (Rollins & Bahr, 1976).
This leads us to believe that counselors and researchers may
want to assess for not just a problem area or overt differences in
reported ability to exert power but instead how satisfied the
spouses are with their relative contributions (i.e., income, educa-
tion, commitment, skills, etc.) to the marriage and how the couple
arrives at decisions. It is likely that the demand and withdraw
behaviors arise in couples who are either dissatisfied with their
relative contributions or dissatisfied with how much power each
spouse has in how discrepancies are resolved. Researchers, for
example, have suggested that husbands are satisfied with their
contributions to the marriage because they underestimate their
wives’ contributions to household tasks, responsibilities, and child
care and overestimate their own contributions (Komter, 1989). As
a result, future studies examining spouses’ satisfaction with their
relative inputs to the marriage, and the potential gender role
weightings, may help verify the accuracy of the social structure
hypothesis.
Implications
One of the main implications of this study for counselors is that
marital power may not rest entirely with one spouse or the other.
Spouses may exert more or less marital power in certain areas or
during certain topics, or they may be able to exert different types
of marital power. For counselors and researchers, this means that
it is important to access spouses’ level of power from multiple
domains and across multiple areas of the relationship. When work-
ing with a couple, it is important for counselors to specifically
assess how power bases (SES, education, and perceptions of
gender role), power processes (how the couple discusses the issue
and uses tactics such as domineeringness or dominance), and
outcomes (who ultimately makes the decision) affect the couple’s
decision making and satisfaction across all problematic aspects of
the relationship. Although it seems logical that counselors already
attend to power dynamics in couples relationships, there is some
evidence that counselors are not adept at detecting clear power
abuses such as domestic violence (Avis, 1992), let alone power
imbalances or power dynamics associated with less severe behav-
iors. Our sample of couples from the community, for example,
suggests that domineeringness and dominance behavior are present
in many couples’ relationships. As such, directly attending to these
process issues may be important. Researchers can assist in this
process by focusing on developing tools to better assess these
domains of marital power and start to examine their separate
contributions to marital communication and marital satisfaction.
A second important implication of this study is that it sheds
some new light into how counselors can intervene to reduce the
negative connection between demanding and withdrawal behav-
iors and negative outcomes such as reduced marital satisfaction.
The common notion is that demand and withdraw behaviors are
detrimental to marital satisfaction (Christensen & Heavey, 1990;
Jacobson & Margolin, 1979; Noller et al., 1994); however, in some
cases they are not (Gottman & Krokoff, 1989; Heavey et al.,
1995). One reason for this difference in the relationship between
demand and withdraw behaviors and marital satisfaction may be
the degree to which the spouses share power. Couples who have
nonegalitarian relationships may engage in more rigid demand and
withdraw behaviors and not be able to alter their positions,
whereas egalitarian couples may engage in demand and withdraw
behaviors, but the behaviors may be more flexible or complemen-
tary (e.g., the spouses engage in similar levels of each behavior)
because the couple is more accepting of each others’ position, in
general. Therefore, one of the reasons that couples in this study
may not have exhibited more pronounced demand–withdraw pat-
terns is that wives were able to assert their immediate needs, and
these assertions were responded to. When both spouses are able to
express their needs, problematic communication patterns may be
decreased. Thus, severing the link between demand and withdraw
behavior and negative outcomes may require interventions that not
only address the presence of demand and withdraw behaviors in
couples’ relationships directly but also attend to spouses’ ability to
exert certain types of marital power (i.e., power processes). For
example, one way to address power, in session, is to work with
couples’ moment-to-moment displays or exhibitions of marital
power (power processes) in an effort to make positive changes for
the couple, such as improved marital satisfaction. Trying to negate
demand and withdraw behaviors without also negotiating issues of
power may lead to disappointing results.
A third and related implication is that the current findings
support previous research demonstrating that demand and with-
draw behaviors are present even among relatively satisfied couples
(Noller et al., 1994; Vogel & Karney, 2002; Vogel et al., 1999).
Demand and withdraw behaviors are common to some degree in
174 VOGEL, MURPHY, WERNER-WILSON, CUTRONA, AND SEEMAN
nearly all relationships (Gottman, 1999), and thus, studying non-
clinical samples can inform what behaviors may be problematic
for more distressed couples. In our nonclinical community sample,
demand and withdraw behaviors were common at low to moderate
levels, and in the case of demand during wives’ topics, wife and
husband differences were present with moderate effects. Thus, it
could be that even in relationships that are relatively egalitarian
(i.e., in our study no difference in SES was found), some types of
power differences may exist. For example, women may feel re-
sponsible for managing the relationship and resolving interper-
sonal conflicts and therefore demand more. If counselors want to
intervene to help couples reduce the presence of demand and
withdraw behaviors, they may need to assess the other aspects of
the relationship (e.g., each spouse’s level of responsibility for the
topic) that could play a role in the presence of demand and
withdraw behavior.
Strengths and Limitations
Researchers in the field have called for studies of marital power
that include both self-report and external reports of power (El-
dridge & Christensen, 2002; Gottman & Notarius, 2000). There-
fore, one of the strengths of this study is that it is the first that we
are aware of to include self-report and observed assessments of
marital power and demand and withdraw behaviors in marital
interactions. However, despite this strength, the conclusions that
can be drawn from these results may be limited in several ways.
First, given that this was the first study to examine behavioral links
between demanding and withdrawal behaviors and marital power,
generalizations to other populations should be made with caution
until future research replicates these findings with different mea-
surements of power. Furthermore, although the sample was some-
what distressed, these findings may not apply to all dissatisfied
marriages. There is the possibility that the relationship of various
aspects of marital power and demand and withdraw behaviors may
differ in couples who vary in ethnicity or SES. Furthermore,
although there was some diversity in the sample with regard to age,
relationship length, employment status, and race/ethnicity com-
pared with many studies of married couples, future cross-cultural
research could shed further light on claims about the role of social
structure in marital relationships, especially if sex differences were
found to vary across cultures that differ in levels of gender equal-
ity.
The low reliability of some of the self-report assessments also
limits the understanding of the findings, as low reliability limits the
ability to detect relationships among the variables. When we
started the study, we recognized that some of the self-report
measures had limitations in terms of the reliabilities reported in
previous research. However, we chose to use them to be consistent
with the previous research on demand and withdraw behavior and
because there were few self-report measures of the behaviors of
interest (i.e., demand–withdraw behaviors). We also believed that
having both self-report and observational data would add more to
than hinder the interpretation of the findings. The consistent find-
ings with the use of the self-report and observation data support
this belief, but future studies should attempt to measure power
outcomes with reliable self-report assessments.
It should also be noted that the recorded conversations took
place in a laboratory. Although this is the most common setting for
research examining marital communication, the dynamics that
occur in this type of situation may be different from discussions
that take place at home. Furthermore, given the cross-sectional
nature of these data, any causal statements must be made with
caution. In particular, any strong statements about the way marital
power may influence demand and withdraw behaviors over time
must await longitudinal tests of the patterns that appear to emerge
early in marriage. It has been suggested that the balance of power
in a relationship may change (Blumberg & Coleman, 1989), par-
ticularly during times when the relationship is undergoing change
(i.e., birth, job change; Ferree, 1990; Gerson & Peiss, 1985). An
examination of these processes over time may increase under-
standing of this issue. Thus, there is a need to see how marital
power and demand and withdraw behaviors change over the course
of a relationship. Finally, whereas the size of the current sample
compares favorably to other samples used to examine these issues,
it is possible that a larger sample would have had more statistical
power to detect sex differences in the use of marital power. In all,
these findings highlight the need for more research to further the
field’s understanding of complex power dynamics as they play out
in marital relationships.
References
Avis, J. M. (1992). Where are all the family therapists? Abuse and violence
within families and family therapy’s response. Journal of Marital &
Family Therapy, 18, 225–232.
Babcock, J. C., Waltz, J., Jacobson, N. S., & Gottman, J. M. (1993). Power
and violence: The relation between communication patterns, power
discrepancies, and domestic violence. Journal of Consulting and Clini-
cal Psychology, 61, 40 –50.
Beach, S. R. H., Fincham, F. D., & Katz, J. (1998). Marital therapy in the
treatment of depression: Toward a third generation of therapy and
research. Clinical Psychology Review, 18, 635– 661.
Berns, S. B., Jacobson, N. S., & Gottman, J. M. (1999). Demand–withdraw
interaction in couples with a violent husband. Journal of Consulting and
Clinical Psychology, 67, 666 – 674.
Blumberg, R. L., & Coleman, M. T. (1989). A theoretical look at the
gender balance of power in the American couple. Journal of Family
Issues, 10, 225–250.
Bodenmann, G., Kaiser, A., Hahlweg, K., & Fehm-Wolfsdorf, G. (1998).
Communication patterns during marital conflict: A cross-cultural repre-
sentation. Personal Relationships, 5, 343–356.
Bradbury, T. N., & Karney, B. R. (1993). Longitudinal study of marital
interaction and dysfunction: Review and analysis. Clinical Psychology
Review, 13, 15–27.
Burman, B., & Margolin, G. (1992). Analysis of the association between
marital relationships and health problems: An interactional perspective.
Psychological Bulletin, 112, 39 – 63.
Caughlin, J., & Vangelisti, A. (1999). Desire for change in one’s partner as
a predictor of the demand/withdraw pattern of marital communication.
Communication Monographs, 66, 66 – 89.
Caughlin, J., & Vangelisti, A. (2000). An individual difference explanation
of why married couples engage in the demand/withdraw pattern of
conflict. Journal of Social and Personal Relationships, 17, 523–551.
Christensen, A. (1988). Dysfunctional interaction patterns in couples. In P.
Noller & M. A. Fitzpatrick (Eds.), Perspectives on marital interaction
(pp. 31–52). Clevedon, England: Multilingual Matters.
175
DEMAND AND WITHDRAW
Christensen, A., & Heavey, C. L. (1990). Gender and social structure in the
demand/withdraw pattern of marital interaction. Journal of Personality
and Social Psychology, 59, 73– 81.
Christensen, A., & Shenk, J. L. (1991). Communication, conflict,
and psychological distance in nondistressed, clinic, and divorcing
couples. Journal of Consulting and Clinical Psychology, 59, 458 –
463.
Courtright, J. A., Millar, F. E., & Rogers-Millar, L. E. (1979). Domineer-
ingness and dominance: Replication and expansion. Communication
Monographs, 46, 179 –192.
Cowan, P. A., & Cowan, C. P. (1990). Becoming a family: Research and
intervention. In I. E. Sigel & G. H. Brody (Eds.), Methods of family
research: Biographies of research projects (Vol. 1, pp. 1–51). Hillsdale,
NJ: Erlbaum.
Cromwell, R. E., & Olson, D. H. (1975). Power in families. New York:
Wiley.
Cummings, E. M., & Davies, P. (1994). Children and marital conflict: The
impact of family dispute and resolution. New York: Guilford Press.
Eldridge, K. A., & Christensen, A. (2002). Demand–withdraw communi-
cation during couple conflict: A review and analysis. In P. Noller & J. A.
Feeney (Eds.), Understanding marriage: Developments in the study of
couple interaction. Advances in personal relationships (pp. 289 –322).
New York: Cambridge University Press.
Escudero, V., Rogers, L. E., & Gutierrez, E. (1997). Patterns of relational
control and nonverbal affect in clinic and nonclinic couples. Journal of
Social and Personal Relationships, 14, 5–29.
Ferree, M. M. (1990). Beyond separate spheres: Feminism and family
research. Journal of Marriage and the Family, 52, 866 – 884.
Forehand, R., Brody, G., Long, N., Slotkin, J., & Fauber, R. (1986).
Divorce/divorce potential and interparental conflict: The relationship to
early adolescent social and cognitive functioning. Journal of Adolescent
Research, 1, 389 –397.
Gaul, R., Simon, L., Friedlander, M. L., Culter, C., & Heatherington, L.
(1991). Correspondence of family therapists’ perceptions with FRCCCS
coding rules for triadic interactions. Journal of Marital & Family Ther-
apy, 17, 379 –393.
Gerson, J. M., & Peiss, K. (1985). Boundaries, negotiation, consciousness:
Reconceptualizing gender relations. Social Problems, 32, 317–331.
Gottman, J. M. (1999). The marriage clinic: A scientifically based marital
therapy. New York: Norton.
Gottman, J. M., Katz, L. F., & Hooven, C. (1997). Meta-emotion: How
families communicate emotionally. Mahwah, NJ: Erlbaum.
Gottman, J. M., & Krokoff, L. J. (1989). Marital interaction and satisfac-
tion: A longitudinal view. Journal of Consulting and Clinical Psychol-
ogy, 57, 47–52.
Gottman, J. M., & Notarius, C. I. (2000). Decade review: Observing
marital interaction. Journal of Marriage and the Family, 62, 927–947.
Gray-Little, B., Baucom, D. H., & Hamby, S. H. (1996). Marital power,
marital adjustment, and therapy outcome. Journal of Family Psychology,
10, 292–303.
Hahlweg, K., Kaiser, A., Christensen, A., Fehm-Wolfsdorf, G., & Groth, T.
(2000). Self-report and observational assessment of couples’ conflict:
The concordance between the Communication Patterns Questionnaire
and the KPI observation system. Journal of Marriage and the Family,
62, 61– 67.
Heatherington, L., & Friedlander, M. L. (1987). Family relational com-
munication control coding system. Unpublished manuscript.
Heavey, C. L., Christensen, A., & Malamuth, N. M. (1995). The longitu-
dinal impact of demand and withdrawal during marital conflict. Journal
of Consulting and Clinical Psychology, 63, 797– 801.
Heavey, C. L., Layne, C., & Christensen, A. (1993). Gender and conflict
structure in marital interaction: A replication and extension. Journal of
Consulting and Clinical Psychology, 61, 16 –27.
Holtzworth-Munroe, A., Smutzler, N., Bates, L., & Vogel, D. L. (1993).
Withdraw– demand communication behaviors: Comparing two observa-
tional coding systems. Paper presented at the annual meeting of the
Association for Behavioral and Cognitive Therapies (formerly the As-
sociation for the Advancement of Behavior Therapy), Washington, DC.
Holtzworth-Munroe, A., Smutzler, N., & Stuart, G. L. (1998). Demand and
withdraw communication among couples experiencing husband vio-
lence. Journal of Consulting and Clinical Psychology, 66, 731–743.
Hunsley, J., Best, M., Lefebvre, M., & Vito, D. (2001). The seven-item
short form of the Dyadic Adjustment Scale: Further evidence for con-
struct validity. American Journal of Family Therapy, 29, 325–335.
Hunsley, J., Pinsent, C., Lefebvre, M., James-Tanner, S., & Vito, D.
(1995). Construct validity of the short forms of the Dyadic Adjustment
Scale. Family Relations, 44, 231–237.
Jacobson, N. S., & Margolin, G. (1979). Marital therapy: Strategies based
on social learning and behavior exchange principles. New York: Brun-
ner/Mazel.
Johnson, M. P. (2005). Domestic violence: It’s not about gender—Or is it?
Journal of Marriage and Family, 67, 1126 –1130.
Klinetob, N. A., & Smith, D. A. (1996). Demand–withdraw communica-
tion in marital interaction: Tests of interspousal contingency and gender
role hypothesis. Journal of Marriage and the Family, 58, 945–957.
Kluwer, E. S., Heesink, J. A. M., & Van De Vliert, E. (2000). The division
of labor in close relationships: An asymmetrical conflict issue. Personal
Relationships, 7, 263–282.
Knudson-Martin, C., & Mahoney, A. R. (1996). Gender dilemmas and the
myth in the construction of marital bargains: Issues for marital therapy.
Family Process, 35, 137–153.
Komter, A. (1989). Hidden power in marriage. Gender and Society, 3,
187–216.
Leonard, K. E., & Senchak, M. (1996). Prospective prediction of husband
marital aggression within newlywed couples. Journal of Abnormal Psy-
chology, 105, 369 –380.
McGoldrick, M., Anderson, C. M., & Walsh, F. (1989). Women in families:
A framework for family therapy. New York: Norton.
Noller, P., Feeney, J. A., Bonnell, D., & Callan, V. (1994). A longitudinal
study of conflict in early marriage. Journal of Social and Personal
Relationships, 11, 233–252.
Noller, P., & White, A. (1990). The validity of the communication pattern
questionnaire. Psychological Assessment, 2, 478 – 482.
Robles, T. F., & Kiecolt-Glaser, J. K. (2003). The physiology of marriage:
Pathways to health. Physiology & Behavior, 79, 409 – 416.
Rollins, B. C., & Bahr, S. J. (1976). A theory of power relationships in
marriage. Journal of Marriage and the Family, 38, 619 – 627.
Sagrestano, L. M., Heavey, C. L., & Christensen, A. (1998). Theoretical
approaches to understanding sex differences and similarities in conflict
behavior. In D. J. Canary & K. Dindia (Eds.), Sex differences and
similarities in communication: Critical essays and empirical investiga-
tions of sex and gender in interaction (pp. 287–302). Mahwah, NJ:
Erlbaum.
Sagrestano, L. M., Heavey, C. L., & Christensen, A. (1999). Perceived
power and physical violence in marital conflict. Journal of Social Issues,
55, 65–79.
Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for
assessing the quality of marriage and similar dyads. Journal of Marriage
and the Family, 38, 15–28.
Sprenkle, D. H., & Olson, D. H. (1978). Circumplex model of marital
systems: An empirical study of clinic and non-clinic couples. Journal of
Marriage and Family Counseling, 4, 59 –74.
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 Press.
176 VOGEL, MURPHY, WERNER-WILSON, CUTRONA, AND SEEMAN
Teachman, J. D., Tedrow, L. M., & Crowder, K. D. (2000). The changing
demography of America’s families. Journal of Marriage and the Family,
62, 1234 –1246.
Tichenor, V. J. (1999). Status and income as gendered resources: The case
of marital power. Journal of Marriage and the Family, 61, 638 – 650.
Tracey, T. J., & Ray, P. B. (1984). Stages of successful time-limited
counseling: An interactional examination. Journal of Counseling Psy-
chology, 31, 13–27.
Vogel, D. L., & Karney, B. (2002). Demands and withdrawal in newly-
weds: Elaborating on the social-structure hypothesis. Journal of Social
and Personal Relationships, 19, 685–701.
Vogel, D. L., Wester, S. R., & Heesacker, M. (1999). Dating relationships
and the demand/withdraw pattern of communication. Sex Roles, 41,
297–306.
Witteman, H., & Fitzpatrick, M. A. (1986). Compliance-gaining in marital
interaction: Power bases, processes, and outcomes. Communication
Monographs, 53, 130 –143.
Received May 16, 2006
Revision received December 18, 2006
Accepted December 19, 2006
Members of Underrepresented Groups:
Reviewers for Journal Manuscripts Wanted
If you are interested in reviewing manuscripts for APA journals, the APA Publications and
Communications Board would like to invite your participation. Manuscript reviewers are vital to the
publications process. As a reviewer, you would gain valuable experience in publishing. The P&C
Board is particularly interested in encouraging members of underrepresented groups to participate
more in this process.
If you are interested in reviewing manuscripts, please write to the address below. Please note the
following important points:
To be selected as a reviewer, you must have published articles in peer-reviewed journals. The
experience of publishing provides a reviewer with the basis for preparing a thorough, objective
review.
To be selected, it is critical to be a regular reader of the five to six empirical journals that are most
central to the area or journal for which you would like to review. Current knowledge of recently
published research provides a reviewer with the knowledge base to evaluate a new submission
within the context of existing research.
To select the appropriate reviewers for each manuscript, the editor needs detailed information.
Please include with your letter your vita. In the letter, please identify which APA journal(s) you
are interested in, and describe your area of expertise. Be as specific as possible. For example,
“social psychology” is not sufficient—you would need to specify “social cognition” or “attitude
change” as well.
Reviewing a manuscript takes time (1–4 hours per manuscript reviewed). If you are selected to
review a manuscript, be prepared to invest the necessary time to evaluate the manuscript
thoroughly.
Write to Journals Office, American Psychological Association, 750 First Street, NE, Washington,
DC 20002-4242.
177
DEMAND AND WITHDRAW
... Resources have been most often defined as discrepancies between partners in income, education, and socioeconomic status (Babcock, et al., 1993;Brezsnyak & Whisman, 2004;Crosbie-Burnett & Giles-Sims, 1991;Vogel et al., 2007;Vogler et al., 2008), but they have also included other factors, such as age difference (Crosbie-Burnett, & Giles-Sims, 1991) and communication skills (Babcock et al., 1993). However, research has failed to consistently support the theoretical hypothesis that valued resources by a partner are predictive of greater power outcomes. ...
... However, research has failed to consistently support the theoretical hypothesis that valued resources by a partner are predictive of greater power outcomes. For example, discrepancies in income, education, and socioeconomic status generally have not been found to be associated with power outcome (Babcock et al., 1993;Brezsnyak & Whisman, 2004;Crosbie-Burnett & Giles-Sims, 1991;Tichenor, 1999Tichenor, , 2005Vogel et al., 2007;Vogler et al., 2008). ...
... The only self-report measure that has been consistently used to assess power processes is the Communication Patterns Questionnaire (Christensen & Sullaway, 1984), which measures demand/withdraw patterns of interaction. These researchers conceptualize the demand/withdraw pattern as reflecting an imbalance of power process (Babcock et al., 1993;Vogel et al., 2007). ...
Article
Full-text available
Relationship power, which refers to the ability to influence one’s romantic partner, is an important organizing principle in the assessment and treatment of couples. Power imbalance is predictive of various dimensions of marital quality, which explains why it is often a central focus of couple therapy. Despite the importance of relationship power in couple therapy, assessing power in clinical settings has been hindered by the lack of a validated measure of power that has high clinical utility. Data from 640 married couples associated with the Flourishing Families Project were used to develop the Perceived Power Imbalance Scale by conducting exploratory factor analysis, confirmatory factor analysis, dyadic predictive validity analysis, and measurement equivalence analysis. Although power has been conceptualized as consisting of both outcome power and process power, results indicated that the final scale consisted of four items that only tapped aspects of process power. The scale demonstrated good reliability and was a significant predictor of marital quality, marital instability, and depression.
... Although men hold greater societal-level or structural power based on traditional gender roles, women's caregiving roles do afford women dyadic power within romantic relationships (Glick & Fiske, 1999;Hammond & Overall, 2017). For example, women are more likely to take an active, guiding role in discussing areas of relationship conflict (Christensen & Heavey, 1990;Kelley et al., 1978;Vogel, Murphy, Werner-Wilson, Cutrona, & Seeman, 2007), and in making decisions related to the home and children (Pew Research Center, 2008;Williams & Chen, 2014). However, women's dyadic power is precarious because their relationship power can be usurped by male partners exerting the broader authority assigned by their dominant social role, which provides a basis to resist and reduce women's power even in relationship domains (Glick & Fiske, 1999;Hammond & Overall, 2017). ...
Article
Lower power during marital interactions predicts greater aggression by men, but no research has identified women's response to lower power. We tested whether women who experienced lower situational power during conflict exhibited greater submission, especially if they held traditional gender role beliefs and thus accepted structural gender differences in power. Newlywed couples (Time 1, N = 204 couples) completed questionnaires and discussed an area of conflict 3 times over 3 years. Individuals who perceived lower power during couples' discussions evidenced greater submission, but this effect was more pronounced for wives, especially wives who held traditional gender role beliefs. Among those with traditional gender role beliefs, greater submission together with lower power predicted lower marital adjustment over time. These results highlight that the low power-submission link, and associated implications for marital adjustment, need to be evaluated in the context of power-relevant situations, gender, and broader power-related beliefs about gender roles.
... Research suggests women are more likely to have 'relational power', which Fox and Blanton (1994) define as the influence one has over emotional resources such as support, relational information, trust, attention, and love. Women often possess these resources because they are stereotyped as being higher in warmth and affection than men and because relationships are traditionally considered to be the domain of women (Knudson-Martin, 2013;Vogel, Murphy, Werner-Wilson, Cutrona, & Seeman, 2007). For example, despite holding more positional power, research shows that men can often feel powerless, especially in the domain of the family (Blanton & Vandergriff-Avery, 2001;Walsh, 1989) and husbands feel less effective than their wives in maintaining family relationships and taking care of children (McGoldrick, 1991). ...
Chapter
Full-text available
Power in Close Relationships - edited by Christopher R. Agnew February 2019
... It is not men but women who typically try to prevent their partner from straying. Vogel et al (2007) find that the woman partner has complete charge of the relationship, and that "wives behaviorally exhibited more domineering attempts and were more dominant (ie, more likely to have their partner give in) than husbands during discussions of either spouse's topic" (p173). In line with this, Coleman & Straus (1986) long ago found that the woman is the controlling partner in 90% of couples. ...
Article
Full-text available
There is no published science paper demonstrating the existence of ‘misogyny’ defined as the general hostile or negative attitude of men towards women. On the contrary, the notion is not just unsupported but contradicted by all pertinent findings. Data on either implicit or explicit ‘gender attitudes’ show consideration towards females by males either substantially greater than they afford other males—philogyny—or, at worst, neutrality; this even in special conditions in which it might be most expected. The failure to find evidence of ‘misogyny’ has been hidden by eliding the notion with the wider one of ‘sexism’, and in the absence of support for ‘hostile sexism’, ‘benevolent sexism’ was hypothesised; but this also has no basis, given the deeply flawed operational definitions of ‘sexism’ employed in studies. Taken together as ‘ambivalent sexism’, the mode in which this supposedly causes harm—stereotyping according to sex (dubbed ‘stereotype threat’)—has been comprehensively debunked. ‘Misogyny’ and ‘sexism’ have become defined circularly and are, therefore, entirely non-scientific notions in being unfalsifiable. However weakly defined, ‘misogyny’ evidently is nothing but ideological invention: a counter-factual, thereby itself an actually (anti-male) ‘hostile sexist’ notion and term that should never be employed in a scientific discourse other than in its investigation as at best a questionable, highly pejorative construct. Indeed, the use of the label should be investigated as an expression of misandry.
... Males are not 'in charge' even within their own pair bonds (Vogel, 2007;Coleman & Straus, 1986;Bates, Graham-Kevan & Archer, 2014). ...
Article
Full-text available
Genital modification functions proximally in both sexes to denude sexual sensitivity, reducing propensity to engage in sex; impacting specifically extra-pair sex. Here distal function diverges: male GMo is controlling (lowering young males' competitiveness with high-status males for young females), whereas female GMo is 'honest signaling' of future fidelity (in contest for high-genetic-quality pair-bond partners). Only FGMo originated as a benefit for 'cut' individuals. FGMo is both performed and advocated overwhelmingly by females, and does not serve alpha males (in that they can have few concerns about partner fidelity), actually dis-benefiting them (because of impaired sexual receptivity of current and potential pair-bond and extra-pair sex partners). With no basis for male imposition ('male control') to explain FGMo, it can only be intra-sexual.
... So it is that women themselves report perpetrating significantly more 'controlling' behaviour than do men [Bates, Graham-Kevan & Archer 2014], echoing findings that the woman typically is the 'controlling' partner [Vogel et al 2007], and that this is the case in 90% of couples [Coleman & Strauss 1986]. The Bates, Graham-Kevan & Archer study reveals a telling contradiction between selfreported perpetration and victimisation: whereas there was a considerable sex differential in respect of self-reported perpetration, self-reported victimisation was sex-symmetric. ...
Article
Full-text available
Male-specific self-inhibition of violence towards women, corresponding to an evolutionarily highly-conserved male-specific dedicated neural pathway, and a female-specific actual preference in a couple context for physically violent expression of aggression, prompted by oxytocin (the very hormone underpinning pair-bonding): these findings together indicate that a new theory of partner violence [PV] is required, with a female-specific aetiology. This anyway is apparent from the great disparity between the actual and predicted sex-differential in PV injury rates; the only plausible inference from which is overwhelming female compared to male perpetration. It is conceivable that what male-perpetrated PV there may be is by aberrant (psycho-pathological) individuals, with the remainder of male-on-female violence rather than being PV per se-directed as such, with intent to cause harm-is better understood as by displacement from male intra-sexual aggression. The basis of a female-specific PV aetiology is that pair-bonding is now known to have evolved in the female interest to maximise female fertility, and therefore at root women have a stronger interest in preventing partner defection-man
... That the human female rather than the male is the 'controlling' partner is the case in 90% of couples [Coleman & Straus 1986], and even in terms of what are perceived to be male sex-typical forms of 'control' , women at least match men [Graham-Kevan & Archer 2009] -so if there were also factored in any distinctively female forms of 'control' , then presumably women would be revealed as the more 'controlling' partners. Other research reveals that men are obliged simply to give in and agree in the face of women taking overall charge of the relationship in a 'domineering' manner [Vogel & Murphy 2007]. ...
Article
Full-text available
Though human pair-bonding generally is considered to be male proprietorial control and provisioning, the evidence does not support either of these assumptions. Not only is provisioning relatively lacking and of little impact, but in any case is antecedent to the evolution of human pair-bonding; and the male (in pair-bonding species generically) does not prevent the female partner from engaging in chosen extra-pair sex. In thus contradicting the standard biological model of male mate-guarding as preventing partner defection, its function instead appears to be to displace social/sexual access to the female by lower (but not by higher) mate-value males; thereby indirectly facilitating the female part-ner's extra-pair sex with males of her choice or acceptance (males of higher mate-value than the pair-bond partner). Furthermore, by producing successive offspring with the same male, the pair-bond in effect allows the female to project forwards in time her early peak in fertility (her own mate-value). [Reproducing instead through promiscuous sex would entail progressively lower fitness of offspring
Article
The aim of this study was to investigate the mediating role of sexual function in the relationship between sexual schemas and psychological well-being in married women. The descriptive research method was a constraint type which was performed using structural equation modelling method. The statistical population included married women from Shiraz. Using the available sampling method, 150 people were selected. Assessment tools included women's sexual performance index questionnaires (Rosen et al., 2000), sexual schema (Anderson and Sirvanowski, 1994), and psychological well-being (Reef, 1995). The data collected were analyzed in two descriptive levels (using mean, standard deviation) and inferential (using correlation coefficient and structural equation modelling with amos-21 software). The results showed that passionate-romantic and explicit-comfortable schemas had a positive and significant relationship and shy-cautious schemas had a negative and significant relationship with psychological well-being in married women. Sexual function was also a positive and significant predictor of psychological well-being. In addition, the results indicated that sexual function was able to mediate between passionate and romantic schemas and psychological well-being. Based on the research findings, it can be concluded that sexual schemas both directly and through sexual function can affect the psychological well-being of married women.Keywords:sexual schemas, psychological well-being, sexual function, married women
Article
Full-text available
Abstract New Male Studies: An International Journal (NMS) inaugurates its monograph series with the essay In SEX DIFFERENCE EXPLAINED From DNA to Society: Purging Gene Copy Errors, by independent British researcher Steve Moxon. In this monograph Steve argues that all major aspects of male-female human sociality necessarily stem from biological principles; which all arise in solving the core problem faced by all life-forms: the relentless build-up of mistakes in the repeated copying of genes. The 'genetic filtering' to deal with this is the function of the male: The female contribution is carefully to choose only the most dominant / prestigious males, cross-checking that indeed they do possess the best gene sets. This ensures genetic mutations and other errors that would seriously compromise reproduction are purged from the local gene pool. The detailed treatise is solidly grounded in the biological and other sciences and is offered in the interests of stimulating lively, informed discussion. We hope you enjoy the first in the NMS series of Monographs. Miles Groth, Editor NMS
Article
Full-text available
ABSTRACT The sexes properly understood from biological principle interact not in terms of dominance / submission but male deference to females, in recognition of the female as the ‘limiting factor’ in reproduction through being the principal direct investor in offspring. The corresponding function of the male is to act as the ‘genetic filter’ to purge accumulated deleterious genetic material. This entails males necessarily contesting intra-sexually for dominance rank as a measure of ‘good genes’ (the criterion of male attractiveness); the relative lack of which leaves most males subject to ‘policing’ by both sexes in order to restrict their sexual access. Added to the costs incurred in fierce competition, this amounts to serious disadvantage for most males. The polarisation in specialism stems originally from a small size difference in gametes that polarised through this differential being mutually fitness enhancing. This reveals the root of the distinction between the sexes not to be ‘sexual conflict’, which in this and all its applications is a misconception. This underpinning framework of major sex-dichotomy explains why competitiveness is entirely intra-, not inter-sexual, and much the stronger in males; and why the sexes have starkly contrasting though fully complementary sociality. Culture encapsulates this biology in being in itself part of it, thereby obscuring our perception. The implications of this analysis for the social sciences and politics cannot be overstated.
Article
Even though the demand/withdraw pattern of marital communication has been implicated as a contributor to marital dissatisfaction and divorce, relatively little research has examined why couples engage in demand/withdraw. Why would individuals engage in this seemingly negative communication pattern ? To investigate this issue, a study of married couples was conducted in which the association between spouses' desire for change in their partner and couples' demand/withdraw communication was examined The results indicated that individuals' desire for change in their partner was positively related to both husband demand/wife withdraw communication and wife demand/husband withdraw communication. This suggested that spouses' desire for change was associated with them enacting both demanding and withdrawing communication. The link between desire for change and demand/withdraw seemed particularly strong for topics that were salient to the participants, The results imply that no current model of demand/withdraw sufficiently explains why this communication pattern occurs. Suggestions for a more complete explanation of the demand/withdraw pattern are discussed.
Conference Paper
This study examined the relationship between demand-withdraw interaction and battering in couples with a violent husband. The authors compared the interaction patterns of 47 couples with a violent husband with the interaction patterns of 28 distressed but nonviolent couples and 16 happily married nonviolent couples. All couples engaged in videotaped discussions of problem areas in their marriage. Both batterers and battered women showed less positive communication and more negative communication than did their nonviolent counterparts. Additionally, batterers showed significantly higher levels of both demanding and withdrawing than did other men. Battered women demanded more change than did women in nonviolent marriages but were significantly less inclined to withdraw than were their husbands. The discussion of these findings focuses on the interactional dynamics between batterers and battered women and how these interactions might be understood.
Article
Interaction patterns are repetitious communication sequences between members of a couple. When these patterns are dysfunctional, they are often the focus of clinical intervention. Yet pattern descriptions, while occurring frequently in the clinical literature, tend to be anecdotal and unsystematically studied. A questionnaire containing descriptions of 12 common interaction patterns was independently administered to both members of 55 undergraduate couples. Agreement on pattern occurrence between couples was low but significantly greater than chance. Three of the interaction patterns were significantly related to standard measures of relationship satisfaction. Furthermore, evidence indicated the superiority of the pattern questionnaire over an alternate measurement instrument, an adjective checklist. The potential usefulness of this questionnaire in clinical assessment, intervention, and research is discussed.
Article
Marital Conflict and Child Development. Conflict in the Marital Dyad. Children's Reactions to Marital Conflict. Effects of Specific Aspects of Marital Conflict on Children. Interparental Conflict and the Family. Methodology and Message. Conclusions, Implications, and Guidelines.
Article
We investigated which of 4 proposed short form substitutes conserves the construct validity of the full Dyadic Adjustment Scale (DAS). Using data from 196 married or cohabiting Canadians, we examined internal consistency values, corrected item-total correlations, and the pattern of relations of the full- and short-form versions with a range of relationship measures. Both the Satisfaction subscale and a 7-item version appear to have promise as adequate substitutes for the full DAS.
Article
This article examines the demand-withdraw communication dynamic during which one spouse requests change and the other disengages from the topic. We evaluated two problem-solving discussions by each of 50 married couples using both self-report questionnaires and microanalytic observational coding. Working under the hypothesis that demand-withdraw results from a disparity in the motivation to change, we asked each couple to discuss an issue about which the husband wanted the wife to change and an issue about which the wife wanted the husband to change. Data showed that wives demanded and husbands withdrew during discussions of her issue, whereas husbands demanded and wives withdrew during discussions of his issue. Time-series analyses of observational data confirmed that demand and withdraw behaviors are temporally associated during the course of discussion. We classified couples as bidirectional, wife-dominant, husband-dominant, and nondependent, based on the pattern of interdependency they exhibited.