5

POWER IN CLOSE RELATIONSHIPS:

A DYADIC POWER THEORY
PERSPECTIVE

Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

Interpersonal power is an integral part of all human relationships. Relational

power dynamics exist irrespective of whether or not they become expressed in
conflict but affect such basic relational elements as conflict topic choice, conflict
strategy choice, and conflict outcomes. Power relations are therefore a crucial
element to understanding the achievement of instrumental, relationship, and
identity goals. Dunbar’s (2004) Dyadic Power Theory was created with precisely
these considerations in mind. Our goal in this chapter is threefold: to examine the
evidence collected thus far about the ability of DPT to explain power in dyads; to
evaluate the methodological challenges facing DPT researchers; and finally to
consider new applications of DPT to dyadic relationships.

Explicating Interpersonal Dyadic Power

In any relationship, partners negotiate outcomes based on the resources at their
disposal, such as the structural authority given to them by society at large (Neal
& Neal, 2011). Several theoretical perspectives have examined the dynamics of
power in dyadic relationships. Among these perspectives are: social exchange
theory (Emerson, 1976; Kelley & Thibaut, 1978; Molm, 1997), inter-
dependence theory (Lennon et al, 2013; Rusbult & Arriaga, 1997), normative
resource theory (Blood & Wolfe, 1960), the chilling effect (Roloff & Cloven,
1990; Solomon & Samp, 1998), relational control approaches (Ellis, 1979;
Rogers & Farace, 1975; Siegel et al, 1992), necessary convergence commu-
nication theory (Miller-Day, 2004; Miller-Day & Jackson, 2012), and sex role
theories which encompass sociological, psychological, and cognitive-develop-
mental approaches to gender and power (Bem, 1983; Connell, 1985; Ferree,
1990; Hare-Mustin, 1988).
76 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

Since the approaches to power have been multidisciplinary and are based on
assumptions that come from very different perspectives, the theories of power can
bear little resemblance to one another (Cromwell & Olson, 1975). A comprehensive
review of these theories is available elsewhere in Berger, 1994. We here present one
theory that developed out of a multidisciplinary view of power and incorporates
concepts from several theories mentioned above. DPT is an attempt to explain the
dynamic interplay of conversational partners as they manage control of an interaction
through dominance behaviors (Dunbar, 2004). It asserts that the relational history and
structural forces that bestow power upon actors affect communication patterns and
behavioral displays. Those communication patterns then predict outcomes such as
satisfaction with the relationship, perceptions of who “wins” an argument, whether
or not goals are achieved, and even the likelihood of conflict itself.
Building upon past work by Rollins and Bahr (1976), Roloff and Cloven
(1990), and others, DPT was first proposed by Dunbar (Dunbar, 2000, 2004) and
its propositions have been tested in several empirical studies by Dunbar and col-
leagues (e.g., Dunbar & Abra, 2010; Dunbar et al, 2008; Dunbar & Burgoon,
2005; Dunbar et al, 2014) and others (Bevan, 2010; Recchia et al, 2010; Smith et
al, 2011). It has been applied to work relationships (Lindsey et al, 2011), families
(Dunbar, 2004; Dunbar & Johnson, 2015), romantic couples (Dunbar et al, 2008;
Dunbar & Burgoon, 2005; Worley & Samp, 2016) and stranger interactions
(Dunbar & Abra, 2010; Dunbar et al, 2014) (see Table 5.1 for a complete list of
known DPT studies).
Typical studies testing DPT have examined verbal and nonverbal power stra-
tegies in the conflict discussions of interaction partners. In some cases, the con-
versants are given a task and then the power strategies that are employed can be
examined as the dyad’s problem-solving unfolds. Dunbar and Burgoon (2005)
invited couples to their lab and gave them a task, to decide how to spend a fic-
tional gift of $1000. Dunbar and Abra (2010) paired strangers in the lab and asked
them to rank order items from the “desert survival task.” Dunbar et al (2012a)
asked participants to role play that they were co-workers selecting a job applicant
from a selection of resumés and to select the best one. In other cases, we asked
couples to identify sources of conflict in their relationship and discuss them as
they normally would at home (Dunbar et al, 2012b; Dunbar et al, 2008). We
have also used survey methods to examine people’s own impressions of their
power and their communication strategies (Dunbar & Johnson, in press; Lindsey
et al, 2011). Regardless of the task or measurement, power strategies are evident
in the participants’ behaviors and responses and can be used to understand the
nature of the relationship between power and communication.

Propositions of the Theory and Empirical Support

DPT was originally proposed by Dunbar (2004) with eight empirically verifiable
propositions. These propositions can be organized into the pre-interactional
 Power in Close Relationships 77

TABLE 5.1 Summary of Known Studies on DPT in Chronological Order

Article or Paper Type of Sample N Methodology Used

1 Dunbar (2004) Theoretical paper n/a
2 Dunbar & Burgoon Married & cohabiting 194 Laboratory observation
(2005) couples
3 Dunbar & Allen Friends & dating 134 Survey
(2005) couples
4 Dunbar (2008) Married & cohabiting 94 Laboratory observation
couples
5 Dunbar & Abra Strangers 150 Laboratory experiment
(2010)
6 Bevan (2010) Family members & 266 Survey
romantic partners
7 Lindsey et al. (2011) Workplace 214 Survey
co-workers
8 Dunbar & Mejia Married & cohabiting 20 Qualitative analysis of
(2012) couples Dunbar & Burgoon
(2005) dataset
9 Dunbar et al. (2012) Strangers 150 Secondary analysis of
Dunbar & Abra (2010)
dataset
10 Smith et al. (2011) Married couples 134 Laboratory observation
(followed by
commentary by
Dunbar & Abra,
2012)
11 Dunbar (2012) Married & cohabiting 148 Laboratory observation
couples
12 Dunbar et al. (2014) Strangers 106 Laboratory experiment
13 Dunbar & Johnson Varied relationships 164 Survey
(2015)
14 Worley & Samp Romantic couples 350 Survey
(2016)

factors that affect the relationship, the interactive process that occurs within the
conversation, and then the post-interactional implications of that process.

Pre-interactional Factors
The first three propositions of DPT, as presented by Dunbar (2004), examine
the pre-interactional factors of authority and resources and predict that
increases in relative authority are related to increases in relative resources,
increases in relative resources produce an increase in relative power, and that
78 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

increases in relative authority produce an increase in relative power. Dunbar

argued that actors’ resource levels relative to their partners and the legitimate
authority to use those resources together create a belief about power relative to
one’s dyadic partner. Relative authority affects relative power in two ways.
First, having the ability to control outcomes confers greater influence. For
example, in marital interactions, the spouse who earns the larger income may
set spending priorities. Secondly, legitimate authority is a cultural sanction of
open power use, and should lead to greater use of power. For example,
employment in the paid labor force garners a cultural expectation of control
over earned income, even though the income is a dyadic resource. Women
who work outside the home may use their extra income to reduce their share
of the housework by utilizing services such as restaurants and domestic workers
(Kamo & Cohen, 1998).
In a test of the first three propositions of DPT, Dunbar (2000) found that
having greater economic resources and having greater authority over household
responsibilities and family spending increased perceptions of power for both
women and men although men and women reported they derived their power
from different resources. Consistent with traditional sex roles, women perceived
they had more control over household decisions but men reported greater eco-
nomic, physical, and psychological resources suggesting that women and men still
have different realms of influence despite gains made by women in the work-
place. Smith et al (2011) argue that, based on DPT, raising a conversational topic
implies legitimate authority over that topic and will then make an actor exhibit
dominance. However, raising a topic as an issue for discussion is probably a signal
of an expected power balance for that topic. The perceived injustice is a result of
an unfulfilled expectation. Wives typically do more of the household labor than
their husbands even when they work the same number of hours outside the
home (Beckwith, 1992). Thus, it is reasonable that presenting the unfair division
of household labor as a topic for discussion comes not out of legitimate authority
(and thus greater perceived power) as Smith et al suggest, but out of an expec-
tation of equality (Dunbar & Abra, 2012). More work is needed to tease out the
first three propositions of the theory to determine if the relationship between
legitimate authority, resources, and perceived power are interconnected in the
way that DPT suggests.

Interactional Factors
The fourth proposition of DPT describes the nonlinear relationship between our
perceived power and our attempt to communicate that power through “control
attempts” within our dyadic interactions. Dunbar (2004) proposed, contrary to
Rollins and Bahr’s (1976) assertion, that the relationship between perceived rela-
tive power and control attempts is curvilinear rather than linear. Partners who
perceive their relative power as extremely high or low will have fewer control
 Power in Close Relationships 79

attempts, while partners who perceive their relative power differences as small or
moderate will have more control attempts. A control attempt is a behavior by
one actor intended to change the behavior of another during a social interaction
(Dunbar, 2004). Although Dunbar equates control attempts with dominance
behaviors, they really can be demonstrated through a wide variety of verbal and
nonverbal messages (Dunbar & Burgoon, 2005; Dunbar & Johnson, in press).
Of all the theory’s propositions, P4 has received the most attention in a variety
of studies, generally focusing on dominance displays within interactions. In tests of
DPT, dominance has been coded in interactions through the use of gestures, facial
expressions, and touch (Dunbar & Burgoon, 2005; Smith et al, 2011), through
ratings made by participants and observers (Dunbar & Abra, 2010; Dunbar et al,
2008), through verbal counts of arguments and linguistic analyses of text (Dunbar
& Burgoon, 2005; Dunbar et al, 2014) and through the use of certain argumenta-
tive strategies like aggressive humor, affection messages, and social support (Dunbar
et al, 2012a; Dunbar et al, 2012b; Dunbar & Johnson, 2015).
The empirical relationship between power and control attempts (specifically
dominance behaviors) has been found to be curvilinear, with an unexpected
asymmetry. In general, individuals in equal power dyads demonstrate the most
dominance followed by those in high power positions and then low power
positions (Dunbar & Abra, 2010; Dunbar et al, 2008; Dunbar & Burgoon, 2005;
Dunbar et al, 2014). In result after result, with a wide variety of dominance
measures, we have found the same asymmetrical “hook” shape emerging in our
studies of the relationship between dominance and power (see Figure 5.1).
Typically, those who either are assigned to the low power position in an
experimental manipulation or report they are in the lowest position of power in
an existing relationship enact the least dominance, by far. Those who are equal in
power demonstrate the most dominance. The high-power individuals appear as
5.15.1
FIGURE
FIGURE

FIGURE 5.1 FIGURE 5.1 FIGURE 5.1

FIGURE 5.1

FIGURE 5.1 The observed relationship between power and dominance.

80 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

though they are lower in dominance than the equal power individuals, but often,
they are not significantly different from that equal-power group. And yet,
because of the reoccurrence of that same pattern across studies and with different
measures of dominance, we do not believe that the difference between equal and
high power individuals is a result of error or chance.
With power dynamics in play, once a control attempt is made the partner may
respond with control attempts of his or her own. The degree to which he or she
allows the other’s control attempts to remain unchallenged will also be related to
his/her own perception of power. Dunbar (2004) proposed (in what she labeled
proposition 6) that as a partner's perception of his (her) own power relative to
that of his (her) partner increases, counter-control attempts will increase as well.
This proposition has rarely has been examined in the current DPT literature but
an analysis by Dunbar and Mejia (2012) speaks to this. They qualitatively ana-
lyzed the interactions of ten couples from the Dunbar and Burgoon (2005) data-
set, five of which were couples with the highest self-reported power imbalance
and five of which had the most similar power. In their detailed analysis of the
couples’ conversations, they found that “challenging” or “collaboration” was a
type of behavior used by equal power types whereas “marginalization” or
“withdrawal” was used more by unequal power types. This suggests that, con-
sistent with proposition 6, those with the greater power equality are also those
that engage in more of the counter-control attempts rather than letting their
partners’ control attempts go unanswered.

Post-interactional Factors
Dunbar’s (2004) final three propositions represent the outcomes of the interac-
tion, including the amount of control one gains and the resulting level of satis-
faction with both the relationship and the interaction. Although Rollins and Bahr
(1976) ended their model with the outcome of a particular interaction, other
communication scholars are interested in the outcome of the conflict for the
relationship more generally. Dunbar proposed that greater control attempts
would lead to more control and counter-control attempts would hinder the
control over the outcomes (called P5 and P7, respectively, in the original state-
ment of the theory). Some DPT studies have examined the outcomes of a parti-
cular argument such as who “wins” or satisfies more of their demands (Dunbar &
Abra, 2010; Dunbar et al, 2008; Recchia et al, 2010), but others have examined
sustained effects such as long-term relational satisfaction (Dunbar & Burgoon,
2005; Dunbar & Johnson, 2015).
It is a relatively uncontroversial finding in the close relationships literature that
power equality tends to be associated with higher satisfaction (Gray-Little &
Burks, 1983; Oyamot et al, 2010). Dunbar (2004) argued (in proposition 8) that
relational satisfaction is a key variable that ought to be examined when investi-
gating the impacts power imbalances have on the long-term effects of a
 Power in Close Relationships 81

relationship. She posited that the relation between perceived relative power and
satisfaction is curvilinear such that partners who perceive their relative power as
extremely high or low will report lower levels of satisfaction compared to part-
ners who perceive the relative power differences as small or moderate.
While P5 and P7 are concerned with immediate outcomes, P8 is concerned
with the long-term effects of power inequality on relational satisfaction. To
investigate the immediate outcomes for interactions, Dunbar et al (2008) investi-
gated whether perceptions of power influenced three different interaction out-
comes (perceptions of conflict escalation, perceptions of progress on the conflict
issue, and “who won” the interaction) and found that perceived power did not
have a direct effect on these perceptions of the conflict’s outcome. Similarly,
Bevan (2010) also found no differences on the types of strategies used in serial
arguments and proposes that goal orientation, rather than power alone, predicted
the outcome of the serial argument interactions. However, in an experimentally
manipulated power relationship, Dunbar and Abra (2010) found that those in the
unequal high-power position changed their previous decision the least after
interacting with their partner while those in unequal low-power changed their
decision the most which suggests power affected control over the decision made
after the interaction was over. Few studies have examined the relationship
between control and counter-control attempts which would require intensive
conversational analysis techniques. Thus, it appears that the “jury is still out” as to
whether power or behavioral dominance will have an effect on the outcome of a
conflict.
In terms of relationship satisfaction, although previous research has suggested
that relational satisfaction is highly correlated with perceptions of power and that
inequality is a major determinant of marital dissatisfaction and dissolution (Cahn,
1992; Clements et al, 1997), our work testing DPT’s proposition that relational
satisfaction is related to power has not found a significant relationship between
power and relational satisfaction in a study of married and cohabiting couples
(Dunbar, 2000). Relationship satisfaction also did not influence how compliance-
gaining messages were reportedly used by the self or the partner in a study of
friends or romantic partner conflicts (Dunbar & Allen, 2005). It may be that the
homogeneity of our samples influenced these results because the participants
overall reported they are highly satisfied, relatively equal in power, and used a
limited range of compliance gaining messages in both studies.
We attempted to remedy this shortcoming in the Dunbar and Abra (2010)
study in which we also explored the link between power and satisfaction in an
experiment between strangers. The results suggest that people in unequal low-
power positions are generally less satisfied than people in equal-power positions,
and people in unequal high-power positions are also less satisfied than people in
equal-power positions (although this difference was not statistically significant).
This result is consistent with proposition 8 which suggests that equal power
people will be the most satisfied. Equity theory suggests that individuals
82 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

experience distress when they are in inequitable relationships (when their inputs
do not correspond to their outputs) and they attempt to restore equity between
the partners. In close relationships, men and women who believe they have been
equitably treated feel more content in their relationships and perceive the rela-
tionship to be more stable than those in inequitable relationships, even when they
benefit from the inequality (Dunbar & Abra, 2010; Sprecher, 2001; Walster &
Walster, 1975). There is discomfort experienced when we are placed in a position
of power because we are sensitive to the apparent discomfort of our lower-power
partners. However, because of the disparate findings reported above, there is
more research needed on what the outcomes are for DPT. We now consider
important challenges to performing that research.

Methodological Challenges in Testing DPT

The first difficulty rests in people’s beliefs about power. Many tests of DPT rely
on participants self-reporting power and dominance behaviors. Self-reports of power
provide challenges as participants may not be fully aware of their power. In fact,
participants who report equal power in their romantic relationships have been found
by third-party coders to exhibit characteristics of inequality. Dunbar and Burgoon
(2005) suggest that equally balanced relationships are most comfortable so relational
partners may convince themselves their relationship is equal when it actually may
not be. Or, because U.S. culture emphasizes equality in marriages, couples may be
reluctant to report a marriage that departs from this ideal. Alternately, if partners
recognize their relationship is unequal, they may compensate through their control
attempts to change the power discrepancy (Dunbar & Burgoon, 2005).
Several tests of DPT have found differences in coders’ perceptions of dom-
inance compared to that of the participants themselves (Dunbar et al, 2008;
Dunbar & Burgoon, 2005). Participants’ self-report of power did not influence
their perceptions of dominance which indicates that participants may have a dif-
ferent perception of dominance than objective coders. Self-reports provide a
methodological challenge in examining what is objectively occurring versus what
participants perceive is occurring in an interaction. Participants may not be aware
of their or their partner’s control attempts. DPT researchers defend the use of
self-reports because the mechanism through which power is translated into
behavior requires actors to perceive power differentials. Perceived power simila-
rities and differences can be validly measured with self-reports (Dunbar et al,
2008; Dunbar & Burgoon, 2005). However, participants may be unable to sepa-
rate their perceptions of the current situation from perceptions of their partner
generally (Dunbar & Burgoon, 2005). In this case, third-party observers may
be more useful because they are unaware of the norms of the couple and can be
more objective of the interaction. Research on DPT has shown that participants
may not use individual behaviors to evaluate dominance but actually may
be making more holistic judgments of the interaction (Dunbar & Burgoon, 2005).
 Power in Close Relationships 83

In addition to participants’ inability to recognize power differentials and dom-

inance behaviors, self-reports face a cultural challenge: a stigma may be attached
to reporting low power in a romantic relationship. As previously stated, partici-
pants consistently describe their relationships as equal power, but this could be
due to a reluctance to admit one has low power. Participants may not want to
admit to a researcher (or to themselves) that they fear potential repercussions and
therefore are less likely to report low power in the relationship.
Making this problem even more difficult is the Dunbar and Burgoon (2005)
claim that power-discrepant romantic relationships are probably rare and there-
fore difficult to study in naturally occurring contexts. Relationships that are
extremely power discrepant are unlikely to survive marriage or cohabitation.
Further, members of such relationships may not want to be studied. Dunbar and
colleagues (Dunbar et al, 2012b) attempted to invite couples to the lab who had
clear power discrepancies. Surveys of relationship satisfaction and power were sent
to a large sample of participants and those with the largest self-reported power
discrepancies and a sample of power-equal couples were invited to participate in a
laboratory study where they would discuss an area of conflict on camera for addi-
tional compensation. Compared to the equal-power couples, the power-discrepant
couples were far more likely to refuse the invitation and were more likely to cancel
or no-show the appointment. In the end, the power discrepant sample was only
mildly more power-unequal than the equal-power couples. On the self-report
measure of power, the low power (M = 2.9, SD = .62) and high power (M = 5.1,
SD = .59) individuals self-reported they were only 1.1 point (on a 7-point scale)
away from the equal power group (M = 4.00, SD = .19). Thus, getting unequal
couples to agree to be studied is more difficult than equal power couples.
Indicators of conflict or power can also be suppressed in a laboratory setting.
Caughlin and Scott (2010) claim that a weakness of studies in a laboratory is that
couples might not discuss topics in the laboratory that they actually discuss at
home. Topics couples discuss in private may be more strongly linked to power.
The lack of perceived power differential may be due in part to the inability of
researchers to prompt discussion on strongly differentiating topics.
To address the issues of self-reports of power and dominance, there are prob-
ably at least two options. One is to study naturally occurring power-discrepant
relationships. Some examples are parent–child dyads (Perosa & Perosa, 2001;
Recchia et al, 2010) or workplace dyads (de Reuver, 2006; Lindsey et al, 2011;
Willemyns et al, 2003) in which occupying a low power position carries less
stigma. Researchers can also conduct studies in the laboratory in which power is
experimentally manipulated and use trained confederates (Dunbar & Abra, 2010;
Hall et al, 2006) providing more control over the interaction. When power is
manipulated experimentally, we often use a “layered” approach to ensure that
those who are assigned to a high level of power compared to their partner not
only “feel” powerful but also actually have power, in terms of controlling out-
comes, over their interaction partner (Dunbar & Abra, 2010; Dunbar et al, 2014).
84 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

However, in Dunbar and Abra’s (2010) laboratory study in which participants

and confederates in the high power position could veto their partner’s decision,
they were surprised to discover that only the experimental confederates actually
used this option. Participants in the high power condition rarely used this option
and instead tried to convince their partner to acquiesce. This speaks to the
methodological challenges of studying power and dominance; individuals in
naturally occurring relationships are reluctant to admit low power, but partici-
pants in a laboratory experiment may be reluctant to enact behaviors consistent
with their position. No DPT studies have determined whether or not these
manipulations are isomorphic with the way real couples or dyads determine who
is the more powerful among them.
Another methodological challenge to testing DPT is in the structural expec-
tations for power that cannot be changed. The type of relationship, whether it
is romantic partner, friend, family member, or coworker can influence
dominance behaviors (Dunbar & Johnson, 2015). This could be due to struc-
tural expectations (Neal & Neal, 2011). Bevan (2010) found a difference
between parent–child, sibling, and romantic partnerships and claims this could
be due the inherent characteristics of the relationships such as voluntariness,
interdependence, or simply differing expectations. DPT predicts that dyads with
discrepant power, such as a supervisor/subordinate relationship, are more likely
to use latent power strategies (Dunbar, 2004; Dunbar & Burgoon, 2005).
Powerful individuals do not need to use overt control attempts by nature of
their position (Lindsey et al, 2011). The predetermined structure of the rela-
tionship determines how control attempts are enacted. This provides challenges
in testing DPT.
Dunbar et al (2014) attempted to examine this by having participants deceive
during a role play. Participants took on the roles of manager and employee in
making a hiring decision. Consistent with the findings of a survey of real man-
agers and subordinates (Lindsey et al, 2011), the participants tended to agree that
deception by a superior is more acceptable than deception by a subordinate.
However, powerless individuals may be reluctant to display dominance for fear of
retaliation and so they may resort to deception as a method of minimizing con-
flict and gaining control. To study the use of power and control attempts in
actual workplace situations, especially when involving deception, is a methodo-
logical challenge in testing DPT.
Finally, the study of interactions involving participants who lack relational
history cannot fully recreate sustained power discrepancies. Experimentally
manipulated power differences do not match the timeframe of relational power
dynamics, and it would not be surprising to learn that sustained power equality
and inequality have effects that are stronger than those produced in the lab.
Dunbar and Abra (2010) recognized this, and cautioned that using relationships
with a structural power difference and a relational history may produce different
results.
 Power in Close Relationships 85

Discussion and Suggestions for the Future of DPT

Taking a cue from the methodological limitations discussed above, we believe
that DPT can be expanded beyond its current scope. We propose three new
directions. The first is the idea that within power-balanced married couples, there
are certain domains of influence in which one spouse has relatively higher power.
The second is the fact that relationships are embedded within larger power
structures that might put limits on the dominance strategies individuals within a
given relationship can use. Finally, we suggest that dyadic power dynamics
operate similarly with both individual people as actors and with larger social
groups (e.g. organizations) as actors.

Expanding DPT to Intra-Dyadic Power Domains

Even within stable and generally power-balanced relationships, there are domain-
bounded asymmetries of power. The theoretic and empirical work on DPT thus
far has been concerned primarily with the relationship between the general
power structure of a relationship and the behavioral expressions of power.
Comparisons across dyads have shown the impact of these relational power
structures on behavior. In general, DPT has been concerned with reducing dyads
to particular “types” (e.g. power balanced vs. unbalanced) and then using that
typology to explain and predict power expressions. We propose here that DPT
can and should be expanded to consider the dynamic nature of dyads.
Researchers over the past few decades uncovered the unequal power dynamics
of traditional marriages, noting that the economic and social resources gained by
men in the paid labor force translated into a power advantage for men within the
relationship. Other researchers suggested that this masked the ability of women
within traditional marriages to create desirable outcomes with respect to particular
issues of importance. The tactic of “micromanipulation” (Lipman-Blumen, 1984)
allows otherwise low-power actors to negotiate favorable conditions. Our goal is
not to validate this tactic, but simply to note its existence and the clever granular
parsing of power those actors can tap when broader power conditions are not
advantageous.
Within a dyad, power relations shift across contexts. The conception of dyadic
power has been that of a monolithic, grand average that creates a particular type
of dyad: an unbalanced dyad (power unequal) or a balanced dyad (power equal).
While useful for comparing across dyads, this template makes it difficult to
explain periodic departures from predicted dyadic behaviors. For example, in a
power-unbalanced dyad, there will still likely be situations in which conflict
occurs (albeit less often than in power-balanced dyads). Why would we expect to
find situations of conflict within power-unbalanced dyads when DPT suggests
that low-power actors will mute their disagreements? Relatedly, why would we
expect to find conflict-free periods within power-balanced dyads when DPT
86 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

suggests that equal-power actors will display control attempts? The answer is that
although a dyad may average to an equal or unequal power structure, the parti-
culars of the dyadic power structure will involve equalities and inequalities across
different aspects of the relationship. We call these aspects “relational domains”. A
relational domain refers to any sphere of activity over which partners may
negotiate. This may be a characteristic over which specialized knowledge or
competence may be claimed, as in the “specific task competence” of Status
Characteristics Theory and Expectation States Theory (c.f. Berger et al, 1972;
Ridgeway, 1987; Ridgeway, 1991). Examples of relational domains in romantic
relationships include cooking, finance, artistry, and child-rearing.
Each domain contributes toward the “average” power dynamic of the rela-
tionship. The degree to which the domain contributes to that power dynamic
depends on the importance of that domain to the relationship. Domains valued
by both partners (e.g., money) will contribute more to the power dynamic than
domains valued by only a single partner (e.g. home decoration). One actor may
have great authority over a particular relational domain, and strong task compe-
tence that the dyadic partner lacks, but if the resource is not of high value to that
partner (e.g., home decoration), then the actor’s relative power within the rela-
tionship is not increased by much. On the other hand, if the domain is valued
highly by both actors (e.g., money), then this increases relative power in the
broader sense.
A power-balanced relation represents a situation in which the domains con-
trolled by one partner are collectively of equal import to the domains controlled
by the other partner. In this respect, the domains themselves can be thought of as
resources in the sense used by Emerson and other power-dependence theorists
(e.g., Cook et al, 1993; Emerson, 1962). When the domains collectively are of
high value, then the power balance can be high-high, representing a high degree
of dependence and strong integration. These dyads are likely to persist. When
the resources are collectively of relatively low value, any power balance would be
low-low, representing a low degree of mutual dependence and low integration.
These dyads are less likely to persist. The number of control attempts is a function
of the length of the relationship, so this consideration of dyadic integration leads
to a new proposition:

P9: Control/counter-control attempts will be more frequent in high-high

dyads than in low-low dyads.

For the reasons outlined above, we can also predict the topics of conflict that
will occur within a relationship:

P10: Within generally power-balanced relationships, conflict will occur more

often in domains over which both partners have high power than in domains
in which only one partner has high power.
 Power in Close Relationships 87

Consider the case of an auto mechanic married to a computer network engi-

neer. DPT would not expect conflict to arise about vehicle maintenance, nor
about the deployment of the household Wi-Fi system. The power inequalities
within each of those domains would suppress conflict before it even begins.
Within the same relationship, however, we may expect conflict with respect to
domains over which both partners claim expertise, such as child discipline.
Empirical investigation of these domain-bounded power dynamics will serve to
expand DPT beyond cross-dyad comparisons. On a broader scale, DPT would
benefit from a consideration of the role of cultural prescriptions and prohibitions
on conflict.

Expanding DPT to Cultural and Structural Power

We recognize that the degree of relational conflict is partially contingent on the
cultural acceptability of open conflict and disagreement. In some cultural con-
texts, conflict itself is considered an inappropriate display. DPT can and should be
expanded to integrate our knowledge about cultural differences in conflict and
dominance. We expect that this expansion will demonstrate important cross-cul-
tural components, and important intra-cultural components.
For example, there is a notable difference in the display of conflict when com-
paring Japanese and Italian cultures. It is not true that the power differences exist
only in one cultural context, but rather that overt conflict is expected to be sup-
pressed in Japanese culture more than in Italian culture. When the value of overt
conflict itself varies, then we need to incorporate this new variable into future DPT
analyses. We are suggesting that there may be broad cultural expectations that
prevent conflict even when we may expect disagreement between two actors:

P11: Dyads in cultures that stigmatize open conflict will display less conflict
than dyads in cultures that do not stigmatize open conflict.

A study of marital conflict across four cultures found that the role segregation
and male dominance of the culture affected the decision-making process in cou-
ples (Wagner et al, 1990). For example in Saudi society, women are expected to
defer to men on many important issues. We would expect that the degree of
cross-sex public and private conflict is lower in Saudi society than in Western
societies. It is important to note that we expect DPT to explain this supposed
cultural difference by noting that the cultural suppression of male-female conflict
does not represent a real difference in the way that power and dominance are
related; instead, it represents the cultural imbalance of power between men and
women in Saudi society. In fact, DPT would suggest that this power imbalance is
fully responsible for the suppression of conflict. By comparison, the relatively
high level of male-female conflict in Western societies represents the greater
power balance between men and women in Western societies:
88 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

P12: Heterosexual couples in strongly patriarchal cultures will have less

conflict than couples in less patriarchal cultures.

We expect that these cultural considerations will reveal important cross-cultural

differences in dominance and the expression of power, but we also believe that
within a given cultural context the basic prediction of DPT will hold: power-
equal dyads will display greater conflict than power-unequal dyads.

Expanding DPT to Organizational Actors

Although designed for interpersonal communication, the logic of DPT can be
extended to consider inter-organizational communication. Much of the sociological
and political science work on power recognizes the similarities between individuals
and collectivities as actors. The employer-employee dyad can be thought of as ana-
logous to the colonizer-colony dyad. In both cases, the relatively powerful actor
controls highly valued resources (i.e., money and military) while the relatively pow-
erless actor controls easily substitutable resources (i.e., labor and land). As with human
dyads, power is based on resource control. Some resources will be similar (e.g., status,
money), while other resources may be different, but the power dynamics should play
out similarly. In a power-unequal organizational dyad, one organization controls
more valuable resources than the other, creating the inequality. This dependence
implies an inability of the low-power organization to obtain the resource elsewhere,
and hence the low-power organization will fear loss of the resource. This is a weak
bargaining position. Under these circumstances, the low-power organization will
experience the equivalent of the chilling effect, and will attempt to avoid bargaining,
as it is unlikely to result in a favorable outcome. It will also suppress expressions of
dissatisfaction to its partner, lest the partner end the relationship. Further, it will
proactively ensure the satisfaction of its important partner. Similarly, the relatively
high-power organization will find that its problems are proactively addressed by the
low-power organization’s actions, precluding initiation of conflict.
By contrast, organizations with equal power control equally-valuable resources.
Under these conditions, the organizations are free to express dissatisfaction and to
attempt to bargain for a favorable outcome. We should therefore see the defini-
tive curvilinear relationship between relative power and control attempts:

P13: Equal-power organizational dyads will display more control attempts

than unequal-power organizational dyads.

The parallels between organizational dyads and human dyads can be drawn
further, and may stretch to include the items newly included above for human
couples (e.g., domain-bounded power). We expect nonetheless that the funda-
mental idea of DPT – that power dynamics underlie the behavior of dyadic
partners – will hold true irrespective of the particular characteristics of each actor.
 Power in Close Relationships 89

Conclusion
DPT has received a fair amount of research attention in its short history, drawing
in scholars from both within and outside the discipline of communication. We
believe this is because of DPT’s fundamentally interdisciplinary property: it draws
on the knowledge base of several well-established fields. Our goal here was to
review and broaden the theory and to summarize the empirical evidence. DPT is
gaining momentum but as this review has demonstrated there are important voids
to fill in our understanding of the broader implications. We hope that DPT
remains worthy of investigation as the evidence accumulates.

References
Beckwith, J. B. (1992). Stereotypes and reality in the division of household labor. Social
Behavior and Personality, 20(4), 283–288. doi: 10.2224/sbp.1992.20.4.283
Bem, S. L. (1983). Gender schema theory and its implications for child development:
Raising gender-aschematic children in a gender-schematic society. Signs, 8(4), 598–616.
doi: 10.1086/493998
Berger, J., Cohen, B. P., & Zelditch, M. (1972). Status characteristics and social interac-
tion. American Sociological Review, 37(3), 241–255. doi: 10.2307/2093465
Bevan, J. L. (2010). Serial Argument Goals and Conflict Strategies: A Comparison between
Romantic Partners and Family Members. Communication Reports, 23, 52–64. doi:
10.1080/08934211003598734
Blood, R. O., & Wolfe, D. M. (1960). Husbands and wives. Glencoe, IL: Free Press.
Cahn, D. D. (1992). Conflict in intimate relationships. New York: Guilford.
Clements, M. L., Cordova, A. D., Markman, H. J., & Laurenceau, J. P. (1997). The
erosion of marital satisfaction over time and how to prevent it. In R. J. Sternberg
& M. Hojjat (Eds.), Satisfaction in close relationships (pp335–355). New York:
Guilford.
Connell, R. W. (1985). Theorising gender. Sociology, 19(2), 260–272. doi: 10.1177/
0038038585019002008
Cook, K. S., Molm, L. D., & Yamagishi, T. (1993). Exchange relations and exchange
networks: Recent developments in social exchange theory. In J. Berger & M. Jr. Zel-
ditch (Eds.), Theoretical Research Programs: Studies in the growth of theory (pp296–322).
Stanford, CA: Stanford University Press.
Cromwell, R. E., & Olson, D. H. (1975). Multidisciplinary perspectives of power. In R. E.
Cromwell & D. H. Olson (Eds.), Power in families (pp15–37). Beverly Hills, CA: Sage.
de Reuver, R. (2006). The influence of organizational power on conflict dynamics. Per-
sonnel Review, 35(5), 589–603. doi: 10.1108/00483480610682307
Dunbar, N. E. (2000). Explication and initial test of dyadic power theory. PhD thesis, University
of Arizona, Tucson, AZ.
Dunbar, N. E. (2004). Dyadic Power Theory: Constructing a communication–based
theory of relational power. Journal of Family Communication, 4(3 & 4), 235–248. doi:
10.1207/s15327698jfc0403&4_8
Dunbar, N. E., & Abra, G. (2010). Observations of dyadic power in interpersonal
interaction. Communication Monographs, 77(4), 657–684. doi: 10.1080/03637751.2010.
520018
90 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

Dunbar, N. E., & Abra, G. (2012). Dyadic power theory, touch, and counseling
psychology: A response to Smith, Vogel, Madon, and Edwards (2011). The Counseling
Psychologist, 40(7), 1085–1093. doi: 10.1177/0011000012456883
Dunbar, N. E., & Allen, T. H. (2005). Power, conflict, and strategic communication: A
dyadic view of compliance–gaining during conflict. Paper presented at the National
Communication Association Annual Convention, Boston, MA.
Dunbar, N. E., Banas, J. A., Rodriguez, D., Liu, S.-J., & Abra, G. (2012a). Humor use in
power–differentiated interactions. Humor: International Journal of Humor Research, 24(4),
469–489. doi: 10.1515/humor-2012-0025
Dunbar, N. E., Bippus, A. M., Allums, A., & King, S. (2012b). The Dark Side of Humor:
The Use of Aggressive Humor in Conflicts in Close Relationships. Paper presented at
the International Communication Association Annual Meeting, Phoenix, AZ.
Dunbar, N. E., Bippus, A. M., & Young, S. L. (2008). Interpersonal dominance in rela-
tional conflict: A view from dyadic power theory. Interpersona, 2(1), 1–33. doi:
10.5964/ijpr.v2i1.16
Dunbar, N. E., & Burgoon, J. K. (2005). Perceptions of power and interactional dominance
in interpersonal relationships. Journal of Social and Personal Relationships, 22, 207–233.
Dunbar, N. E., Jensen, M. L., Bessarabova, E., Burgoon, J. K., Bernard, D. R., Harrison,
K. J., & Eckstein, J. M. (2014). Empowered by persuasive deception: The effects of
power and deception on interactional dominance, credibility, and decision-making.
Communication Research, 41(6), 869–893. doi: 10.1177/0093650212447099
Dunbar, N. E. & Johnson, A. J. (2015). A Test of Dyadic Power Theory: Control
Attempts Recalled From Interpersonal Interactions with Romantic Partners, Family
Members, and Friends. Journal of Argumentation in Context, 4(1), 42-62. DOI: 10.1075/
jaic.4.1.03dun
Dunbar, N. E., & Mejia, R. (2012). A qualitative analysis of power–based entrainment and
interactional synchrony in couples. Personal Relationships. doi: DOI: 10.1111/j.1475-
6811.2012.01414.x
Ellis, D. G. (1979). Relational control in two group systems. Communication Monographs,
46(3), 153–167. doi: 10.1080/03637757909376003
Emerson, R. M. (1962). Power-dependence relations. American Sociological Review, 27, 31–41.
doi: 10.2307/2089716
Emerson, R. M. (1976). Social exchange theory. Annual review of sociology, 2, 335–362. doi:
10.1146/annurev.so.02.080176.002003
Ferree, M. M. (1990). Beyond separate spheres: Feminism and family research. Journal of
Marriage and the Family, 52, 866–884. doi: 10.2307/353307
Gray-Little, B., & Burks, N. (1983). Power and satisfaction in marriage: A review and
critique. Psychological Bulletin, 93(3), 513–538. doi: 10.1037/0033–2909.93.3.513
Hall, J. A., Rosip, J. C., Smith LeBeau, L., Horgan, T. G., & Carter, J. D. (2006). Attri-
buting the sources of accuracy in unequal–power dyadic communication: Who is better
and why? Journal of Experimental Social Psychology, 42, 18–27. doi: 10.1016/j.
jesp.2005.01.005
Hare-Mustin, R. T. (1988). Family change and gender differences: Implications for theory
and practice. Family Relations, 37, 36–41. doi: 10.2307/584427
Kamo, Y., & Cohen, E. L. (1998). Division of household work between partners: A
comparison of Black and White couples. Journal of Comparative Family Studies, 29(1),
131–145.
 Power in Close Relationships 91

Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory of interdependence.

New York: Wiley.
Lennon, C. A., Stewart, A. L., & Ledermann, T. (2013). The role of power in intimate
relationships. Journal of Social and Personal Relationships, 30(1), 95–114. doi: 10.1177/
0265407512452990
Lindsey, L. L. M., Dunbar, N. E., & Russell, J. (2011). Risky business or managed event?
Perceptions of power and deception in the workplace. Journal of Organizational Culture,
Communications and Conflict, 15(1), 55–79.
Lipman-Blumen, J. (1984). Gender roles and power. Englewood Cliffs, NJ: Prentice-Hall.
Miller-Day, M. (2004). Communication among grandmothers, mothers, and adult daughters: A
qualitative study of maternal relationships. Mahwah, NJ: Lawrence Erlbaum.
Miller-Day, M., & Jackson, A. W. (2012). The convergence communication scale:
Development and evaluation of an assessment of interpersonal submission. Journal of
Social and Personal Relationships, 29(8), 1036–1057. doi: 10.1177/0265407512443617
Molm, L. D. (1997). Coercive Power in Social Exchange. Cambridge: Cambridge University
Press.
Neal, J. W., & Neal, Z. P. (2011). Power as a structural phenomenon. American Journal of
Community Psychology, 48(3–4), 157–167. doi: 10.1007/s10464-010-9356-3
Oyamot, C. M., Fuglestad, P. T., & Snyder, M. (2010). Balance of power and influence in
relationships: The role of self–monitoring. Journal of Social and Personal Relationships,
27(1), 23–46. doi: 10.1177/0265407509347302
Perosa, L. M., & Perosa, S. L. (2001). Adolescent perceptions of cohesion, adaptability, and
communication: Revisiting the circumplex model. Family Journal: Counseling and Ther-
apy for Couples and Families, 9, 407–419. doi: 10.1177/1066480701094008
Recchia, H. E., Ross, H. S., & Vickar, M. (2010). Power and conflict resolution in sibling,
parent–child, and spousal negotiations. Journal of Family Psychology, 24(5), 605–615. doi:
10.1037/a0020871
Ridgeway, C. L. (1987). Nonverbal behavior, dominance, and the basis of status in task
groups. American Sociological Review, 52, 683–694. doi: 10.2307/2095603
Ridgeway, C. L. (1991). The social construction of status value: Gender and other
nominal characteristics. Social Forces, 70(2), 367–386. doi: 10.2307/2580244
Rogers, L. E., & Farace, R. V. (1975). Analysis of relational communication in dyads:
New measurement procedures. Human Communication Research, 1, 222–239. doi:
10.1111/j.1468–2958.1975.tb00270.x
Rollins, B. C., & Bahr, S. J. (1976). A theory of power relationships in marriage. Journal of
Marriage and the Family, 38(4), 619–627. doi: 10.2307/350682
Roloff, M. E., & Cloven, D. H. (1990). The chilling effect in interpersonal relationships:
The reluctance to speak one's mind. In D. D. Cahn (Ed.), Intimates in conflict: A com-
munication perspective (pp49–76). Hillsdale, NJ: Erlbaum.
Rusbult, C. E., & Arriaga, X. B. (1997). Interdependence theory. In S. Duck (Ed.),
Handbook of Personal Relationships (2nd ed., pp221–250). Chichester: John Wiley & Sons.
Siegel, S. M., Friedlander, M. L., & Heatherington, L. (1992). Nonverbal relational control
in family communication. Journal of Nonverbal Behavior, 16(2), 117–139. doi: 10.1007/
BF00990326
Smith, J. C. S., Vogel, D. L., Madon, S., & Edwards, S. R. (2011). The power of touch:
Nonverbal communication within married dyads. The Counseling Psychologist, 39(5),
764–787. doi: 10.1177/0011000010385849
92 Norah E. Dunbar, Brianna L. Lane, and Gordon Abra

Solomon, D. H., & Samp, J. A. (1998). Power and problem appraisal: Perceptual founda-
tions of the chilling effect in dating relationships. Journal of Social and Personal Relationships,
15(2), 191–209. doi: 10.1177/0265407598152004
Sprecher, S. (2001). Equity and social exchange in dating couples: Associations with satis-
faction, commitment, and stability. Journal of Marriage and the Family, 63, 599–613. doi:
10.1111/j.1741–3737.2001.00599.x
Wagner, W., Kirchler, E., Clack, F., Tekarslan, E., & Verma, J. (1990). Male dominance,
role segregation, and spouses’ interdependence in conflict: A cross-cultural study. Journal
of Cross–Cultural Psychology, 21(1), 48–70. doi: 10.1177/0022022190211002
Walster, E., & Walster, G. W. (1975). Equity and social justice. Journal of Social Issues, 31(3),
21–43.
Willemyns, M., Gallois, C., & Callan, V. J. (2003). Trust me, I'm your boss: Trust and
power in supervisor–supervisee communication. The International Journal of Human
Resource Management, 14(1), 117–127. doi: 10.1080/09585190210158547
Worley, T. R., & Samp, J. A. (2016). Complaint avoidance and complaint-related
appraisals in close relationships: A Dyadic Power Theory perspective. Communication
Research, 43, 391–413. doi:10.1177/0093650214538447