Content uploaded by Ruth Gaunt
Author content
All content in this area was uploaded by Ruth Gaunt on Nov 14, 2017
Content may be subject to copyright.
Couple Similarity and Marital Satisfaction: Are
Similar Spouses Happier?
Ruth Gaunt
Bar-Ilan University
ABSTRACT This study examined the role of couple similarity in spous-
es’ marital satisfaction and affect. The associations between spousal sim-
ilarity and relationship measures were examined in a sample of 248
married couples. As hypothesized, greater similarity between partners
was associated with higher levels of marital satisfaction and lower levels
of negative affect. In particular, similarity on the gendered personality
and values domains was more strongly associated with relationship meas-
ures, whereas similarity on the attitudes and religiosity domains showed
weaker and inconsistent patterns of associations. Finally, profile-based
similarity tended to be a stronger and more consistent correlate of rela-
tionship measures than difference score-based similarity. The implica-
tions of these findings for processes underlying intimate relationships are
discussed.
Relationship researchers have long been interested in whether great-
er similarity is associated with better relationship quality (e.g., Hea-
ton, 1984; Meyer & Pepper, 1977; see for review Karney & Bradbury,
1995). Recent advances in measurement and analysis techniques
have prompted new attempts to answer this question using more
elaborated procedures (e.g., Gattis, Berns, Simpson, & Christensen,
2004; Luo & Klohnen, 2005; Watson et al., 2004).
Despite a growing body of literature on the link between couple
similarity and satisfaction, the evidence is equivocal. Whereas some
studies have found that spousal similarity is associated with greater
marital satisfaction (e.g., Blum & Mehrabian, 1999; Caspi & Herbener,
Correspondence concerning this article should be sent to: Ruth Gaunt, Department
of Sociology and Anthropology, Bar-Ilan University, 52900 Ramat-Gan, Israel.
E-mail: gauntr@mail.biu.ac.il
Journal of Personality 74:5, October 2006
r2006, Copyright the Authors
Journal compilation r2006, Blackwell Publishing, Inc.
DOI: 10.1111/j.1467-6494.2006.00414.x
1990; Robins, Caspi, & Moffitt, 2000; Russell & Wells, 1991), other
studies have failed to find such an association (e.g., Gattis et al., 2004;
Glicksohn & Golan, 2001; Watson et al., 2004). These conflicting results
have recently led researchers to conclude that ‘‘the available evidence is
inconsistent and difficult to interpret’’ (Watson et al., 2004, p. 1035) and
that ‘‘the association of similarity and dissimilarity with marital satis-
faction is largely unknown’’ (Gattis et al., 2004, p. 567).
Several flaws have plagued much of the research on the link be-
tween couple similarity and satisfaction. First, most studies on this
issue have focused exclusively on similarity in personality traits (e.g.,
Gattis et al., 2004; Nemechek & Olson, 1999; Robins et al., 2000).
Although personality traits seem particularly important to overall
similarity between partners, other dimensions (e.g. value priorities,
attitudes, religious beliefs) may play an important role as well.
Moreover, it is plausible that some dimensions of similarity contrib-
ute more than others to explaining marital satisfaction. Second,
many studies examining the link between similarity and satisfaction
have been based on relatively small sample sizes (e.g., Glicksohn &
Golan, 2001). This fact adds to the interpretative difficulties of the
findings. Finally, almost all of the studies on this link have imple-
mented the difference scores operationalization of couple similarity.
Although intuitive and easy to grasp, this operationalization has
several major drawbacks, as explained below.
The present study attempts to address these issues and extend
previous research in several important ways. Using a large sample
of married couples, this study examines similarity over a broad range
of domains, including value priorities, gendered personality traits,
family role attitudes, and religiosity. This broad examination makes it
possible to assess the effects of similarity across different domains by
direct comparison. Moreover, this study further explores the effect of
couple similarity on the spouses’ positive and negative affect. Finally,
this study implements a profile-based operationalization of similarity,
which has several advantages over the widely used difference scores
operationalization. The two approaches are then compared to assess
the relative strength of their associations with marital satisfaction.
Dimensions of Similarity Between Partners
As mentioned above, almost all previous studies have focused on the
role of personality traits and have yielded inconsistent results
1402 Gaunt
regarding the link between similarity and satisfaction (e.g., Gattis
et al., 2004; Robins et al., 2000). One exception is the recent study by
Lou and Klohnen (2005), in which similarity measures were
obtained on values, political attitudes, and religiosity, as well as on
personality domains. On the basis of their findings, these researchers
concluded that whereas similarity on personality domains is associ-
ated with marital satisfaction, similarity on values, attitudes, and
religiosity is not.
There are several reasons, however, to reexamine this conclusion
empirically. First, the measure of value priorities that served in Lou
and Klohnen’s study (2005) was a short inventory developed specif-
ically for that study, with no available reliability and validity
indicators. The present study adopts the measure developed by
Schwartz (1992) on the basis of his theory of human values (Schwartz
& Bilsky, 1987). This theory suggests that the primary content aspect
that differentiates values is the type of motivational goal they
express. As a person attributes greater importance to a value, the
attainment of goals to which that value is directed will become
more important. In a series of cross-cultural studies, Schwartz (1992)
showed that people’s basic values are usually organized into 10
distinct motivational types: power, achievement, hedonism, stimu-
lation, self-direction, universalism, benevolence, tradition, conform-
ity, and security. Schwartz (1992) also identified two orthogonal
conflicts between categories of values: values reflecting openness
to change (e.g., stimulation, self-direction) oppose values reflect-
ing conservatism (e.g., tradition, conformity), and values reflecting
self-enhancement (power, achievement) oppose values reflecting self-
transcendence (universalism, benevolence).
Schwartz’s values inventory served to test this theory in more than
200 samples from over 60 countries (Schwartz, 1992; Schwartz &
Sagiv, 1995). This set of 10 value types has also been used to explain
a wide variety of attitudes and behaviors across many countries
(Schwartz & Bardi, 2001). It is therefore possible that couples’ sim-
ilarity in values, as measured by this well-established measure, would
show positive associations with marital satisfaction.
Second, with regard to the attitudes domain, Lou and Klohnen’s
study (2005) focused on political attitudes. Spouses’ political
attitudes are relatively general views that do not relate directly to
their everyday family activities. Similarity on these attitudes may
therefore not be particularly relevant to marital satisfaction. Ajzen
Similarity and Marital Satisfaction 1403
and Fishbein (1977) proposed that whereas general attitudes
can serve as predictors of a set of related behaviors, in order for
an attitude to predict a certain behavior it must be specific to that
behavior. Similarity on attitudes regarding the husband’s and
wife’s family roles may thus be more suitable for testing the effect
of similarity on marital satisfaction. Similarity on these specific at-
titudes seems important for agreement or conflict concerning parent-
ing activities and the allocation of responsibilities. The present study
focuses on attitudes toward family roles in general and toward
parental roles in particular, assuming that spousal similarity on
this dimension may have important implications for relationship
outcomes.
With regard to the personality domain, this study focuses on three
categories of personality traits (Bem, 1974): traits that are perceived
by laypeople as typical of women (feminine traits); traits that are
perceived as typical of men (masculine traits); and neutral person-
ality traits. The use of this classification makes the notion of com-
plementarity (‘‘opposites attracts’’) seem relatively plausible in that
it portrays the feminine wife married to the masculine husband, both
satisfied with their well-defined roles in the relationship. Therefore,
this gendered traits domain seems particularly relevant for testing
the question of whether marital satisfaction is linked to couple sim-
ilarity or complementarity.
Difference-Score-Based Versus Profile-Based Similarity
Most previous research has implemented difference scores opera-
tionalization of couple similarity. Absolute value difference scores
are typically computed by subtracting one spouse’s score on a par-
ticular dimension from the other spouse’s score and then computing
the absolute value of this difference. Difference scores are computed
on the overarching index level rather than on the individual-item
level, therefore discarding a substantial amount of information (Luo
& Klohnen, 2005). In addition, absolute value difference scores can
only range from zero, indicating that spouses have equal levels of an
attribute, to some positive number, indicating that spouses have dif-
ferent levels on that attribute (Luo & Klohnen, 2005).
Profile scores operationalization of couple similarity, on the other
hand, captures each couple’s similarity in terms of organization of
responses. A profile similarity index is computed for every couple by
1404 Gaunt
correlating each husband’s and wife’s responses across all items on a
given domain. Thus, this type of similarity index takes into account
a considerable amount of information, compared to difference score-
based similarity. In addition, profile similarity indices are sensitive to
the varying degrees of agreement and disagreement that may exist
between husbands and wives in terms of any given set of attributes
(Luo & Klohnen, 2005). Similarity scores can range from negative
correlations, suggesting that spouses are opposite in terms of
the profile of their ratings on a given domain, to positive correla-
tions, suggesting that spouses are similar to each other.
To date, only three studies on the link between similarity and
satisfaction have operationalized similarity through profile correla-
tions. Two of these studies focused solely on couple similarity in person-
ality characteristics, with inconsistent results: One of them found a
positive association with marital satisfaction (Caspi & Herbener, 1990),
and the other did not (Glicksohn & Golan, 2001). The third is
Luo and Klohnen’s study (2005) described above, in which simi-
larity in values and attitudes was not related to relationship quality.
However, as the above discussion suggests, extending the measure-
ment of values and shifting the focus of attitudes may broaden
our understanding of the pattern of associations between similarity
and satisfaction.
The Present Study
In light of the above reasoning, the present study examines the role
of couple similarity in spouses’ marital satisfaction and affect. This
study examines similarity in value priorities, gendered personality
traits, role attitudes, and religiosity. In general, it is hypothesized
that the more similar spouses are to each other, the higher their levels
of marital satisfaction and positive affect, and the lower their level of
negative affect. The study further attempts to explore whether
similarity on certain dimensions is more predictive of relationship
outcomes than similarity on other dimensions. Moreover, because
marital satisfaction was found to be related to psychological
well-being (e.g., Gable et al., 2004; Mauno & Kinnunen, 1999;
Suhail & Chaudhry, 2004), this study explores whether couple sim-
ilarity is also linked to spouses’ positive and negative affect. Finally,
this study implements both the widely used difference score-
based similarity as well as the profile-based operationalization of
Similarity and Marital Satisfaction 1405
similarity. The contribution of the two similarity indices is compared
to assess the relative strength of their associations with relationship
satisfaction.
METHOD
Participants and Procedure
The sample for the current study is part of a larger research project on
work and family. Participants were 248 Jewish Israeli heterosexual coup-
les recruited by research assistants through day care centers and commu-
nity child-health facilities.
The age of women in this sample ranged from 20 to 45 years (M530,
SD 54.5). The age of men ranged from 22 to 59 years (M532, SD 55.7).
The number of children per couple ranged from 1 to 5 (M51.78,
SD 5.96). Approximately 71% of the husbands and 78% of the wives
had some college-level education, and 5% of the participants had not
finished high school.
Trained interviewers visited the couples in their homes. During
the visit, participants completed comprehensive self-report question-
naires, which took approximately 1 hour to complete. In two families,
only one spouse filled out the questionnaire, and that family’s data
were eliminated from further use. Other missing data were subjected to
listwise deletion.
Measures
Values. The importance that the participants attribute to each of 44
single values as guiding principles in their life was measured with the
Schwartz (1992) value inventory. Each value is accompanied
by a short descriptive phrase, and the participants used a 9-point rating
scale ranging from 1 to 7 to rate the importance of each value as a
guiding principle in their life. The rating scale was labeled as follows: 1
(opposed to my values), 0 (not important), 3 (important), 6 (very important),
and 7 (of supreme importance). The average score for the items in the
standard indexes was computed to measure the priority given to each of
the 10 value types (Schwartz, 1992). Cronbach’s alphas for the value in-
dexes were as follows: power .68, achievement .76, hedonism .65, stim-
ulation .70, self-direction .65, universalism .74, benevolence .66, tradition
.61, conformity .66, and security .60. These reliabilities were within the
range of variation commonly observed for these value types (e.g., Roccas,
Sagiv, Schwartz, & Knafo, 2002).
1406 Gaunt
Gendered personality traits. Participants’ gendered personality was
measured via the Bem Sex-Role Inventory (BSRI; Bem, 1974). This
instrument consists of 60 personality items that were selected to
represent 20 masculine characteristics (e.g., self-reliant, analytical),
20 feminine characteristics (e.g., compassionate, tender), and 20 neutral
characteristics (e.g., friendly, sincere). The respondent is asked to indicate
on a 7-point scale, ranging from 1 5never true to 7 5always true, the
extent to which the characteristic is self-descriptive. The average score for
the items in each index was computed, and Cronbach’s alphas for
the indexes were .87 for masculine traits, .77 for feminine traits, and .64
for neutral traits.
Family role attitudes. A 10-item scale was designed to measure
participants’ attitudes toward the roles of men and women in the
family. Sample items include ‘‘Fathers should be as intensively involved
in childcare as mothers’’ and ‘‘Mothers are inherently better caregivers
than fathers.’’ The five response categories extended on a continuum
from 1 (strongly disagree)to5(strongly agree) and were recoded so that a
high score reflected more egalitarian attitudes toward family roles. The
average score for the 10 items was computed in order to obtain the re-
spondent’s score on family role attitudes. Cronbach’s alpha for this
measure was .83.
Sociodemographic characteristics. Participants completed extensive de-
mographic questionnaires. Demographic variables used in the current
analyses include age, education level, number of children in the family,
and level of religiosity. Religiosity was indicated on a 4-point scale, labe-
led as follows: 1 (secular), 2 (traditional), 3 (orthodox), 4 (ultra-orthodox).
Marital satisfaction. Participants’ marital satisfaction was measured via
the short version of Enriching Relationship Issues, Communication, and
Happiness (ENRICH; Fowers & Olson, 1993). This is a 10-item Likert-
type scale that assesses the respondent’s perceived quality of marriage
across 10 dimensions of the relationship (spouse’s personal traits, com-
munication, conflict resolution, financial management, leisure activities,
sexuality, child rearing, relationship with the extended family, division of
labor, and trust). Responses are indicated on a 7-point scale, ranging
from 1 5fully disagree to 7 5fully agree. An additional item asked par-
ticipants to indicate their overall satisfaction with their marital relation-
ship, on a 7-point scale that range from 1 5dissatisfied to 7 5extremely
satisfied. An average of the 11 items was calculated to create a
measure of overall marital satisfaction. Cronbach’s alpha for this meas-
ure was .78.
Similarity and Marital Satisfaction 1407
Positive and negative affect. Participants’ affect was measured using the
Bradburn Affect Balance Scale (Bradburn, 1969). Participants indicated
whether they had experienced any of five positive feelings (e.g., excited,
proud) and five negative feelings (e.g., depressed, bored) during the past 2
weeks. Responses were indicated on a 4-point scale labeled as follows:
15never, 25rarely,35sometimes, and 4 5frequently. The average
scores for the 5 positive items and the 5 negative items were computed
in order to obtain the respondent’s positive and negative affect scores.
Cronbach’s alphas for these measures were .66 and .60 respectively.
Computing Couple Similarity
Spouses’ similarity was operationalized both in terms of the profile of the
partners’ ratings and of the difference between their scores. To obtain
profile-based similarity, I computed the correlations between each
husband’s and wife’s self-ratings on all items for each of the three
domains (values, traits, and attitudes). Specifically, each spouse was
treated as if he or she were a separate variable, and each item
was treated as an individual who provided answers on both variables.
A correlation was then calculated between those two spouses (variables),
and the number of items comprising that correlation represented the
number of observations or the sample size. The correlations are descrip-
tive indices of similarity between two persons, with possible values
ranging from 1.00 (complete opposites) to 1.00 (complete similarity).
To obtain difference score-based similarity, I computed the absolute
value of the difference between each husband’s and wive’s ratings on a
certain index (e.g., each of the 10 value type indexes). I then averaged the
difference scores on the indexes included in each of the three domains.
RESULTS
The hypotheses were evaluated in three steps. First, I examined the
correlations between wives’ and husbands’ original measures of
value priorities, gendered traits, family role attitudes, marital satis-
faction, and affect. Second, I examined the correlations between
profile-based and difference score–based similarity and satisfaction
measures separately for husbands and wives. Third, to determine
which domains of spouse similarity contribute more to explaining
each of the satisfaction and affect measures, a series of multiple re-
gression analyses was conducted for husbands and wives separately.
In each of these analyses, a variable pertaining to one dependent
measure was regressed on the set of similarity domains.
1408 Gaunt
Preliminary Analyses
Intercorrelations between individual measures. Table 1 presents
Pearson correlations between wives’ and husbands’ original meas-
ures of value priorities, gendered traits, family role attitudes, marital
satisfaction, and affect. The intercorrelations among individual
measures followed the previously documented patterns (e.g., Gaunt,
2006; Lindeman & Verkasalo, 2005): More religious individuals tend
to hold more traditional family role attitudes and attribute
greater importance to conservation values. The correlation between
religiosity and the importance attributed to tradition values was
particularly high among both husbands and wives. Similarly, attrib-
uting importance to self-enhancement values correlated with
personality traits that are perceived as masculine, whereas
attributing importance to self-transcendent values correlated with
personality traits that are perceived as feminine.
Correlations between spouses’ measures. The correlations between
wives’ measures and those of their husbands are presented in Table 1.
The correlations between husbands’ and wives’ values were generally
low, except for the relatively high correlation between the importance
attributed to tradition values by the spouses. There were weak-
to-moderate correlations between husbands’ and wives’ traits and
between their attitudes. Exceptionally strong was the association
between the levels of husbands’ and wives’ religiosity.
Correlations with marital satisfaction and affect. Finally, Table 1
also shows the correlations between the original measures and hus-
bands’ and wives’ marital satisfaction and affect. In general, marital
satisfaction and negative affect showed weak or no correlations with
the individual measures. Positive affect was moderately linked to
some values and traits. Finally, husbands’ and wives’ marital satis-
faction strongly correlated, whereas their positive and negative affect
measures showed weak correlations with each other.
Intercorrelations among similarity measures. The intercorrelations
among the measures of spousal similarity on the four domains were
generally low, suggesting that they reflect relatively independent
aspects of similarity (Table 2). Similarity on the traits and the
religiosity domains was related to similarity on the values domain.
Similarity and Marital Satisfaction 1409
Table 1
Pearson Correlations Between Value Priorities, Gendered Traits, Family Role Attitudes, Religiosity, Marital
Satisfaction and Affect
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
1. Hedonism values .35
nnn
.48
nnn
.35
nnn
.27
nnn
.11 .11 .12
n
.32
nnn
.32
nnn
.33
nnn
.20
nn
.17
nn
.21
nnn
.11 .36
nnn
.01 .25
nnn
.06
2. Stimulation values .52
nnn
.11 .43
nnn
.39
nnn
.08 .09 .05 .18
nn
.24
nnn
.33
nnn
.33
nnn
.06 .10 .16
n
.33
nnn
.02 .18
nn
.14
n
3. Self-direction values .40
nnn
.55
nnn
.17
nn
.46
nnn
.28
nnn
.03 .15
n
.32
nnn
.32
nnn
.51
nnn
.34
nnn
.12 .14
n
.20
nn
.20
nn
.14
n
.27
nnn
.03
4. Universalism values .41
nnn
.51
nnn
.58
nnn
.20
nn
.39
nnn
.20
nn
.42
nnn
.42
nnn
.15
n
.24
nnn
.11 .32
nnn
.20
nn
.23
nnn
.15
n
.17
nn
.16
nn
.07
5. Benevolence values .20
nnn
.17
nn
.26
nnn
.37
nnn
.10 .39
nnn
.51
nnn
.46
nnn
.20
nn
.36
nnn
.23
nnn
.41
nnn
.33
nnn
.01 .17
nn
.20
nn
.22
nnn
.03
6. Tradition values .01 .10 .05 .21
nnn
.50
nnn
.47
nnn
.53
nnn
.36
nnn
.12
n
.11 .06 .24
nnn
.22
nnn
.25
nnn
.62
nnn
.12 .12
n
.04
7. Conformity values .21
nnn
.23
nnn
.18
nn
.33
nnn
.60
nnn
.62
nnn
.32
nnn
.59
nnn
.30
nnn
.33
nnn
.21
nn
.28
nnn
.39
nnn
.11 .16
nn
.12
n
.22
nnn
.12
8. Security values .26
nnn
.24
nnn
.22
nnn
.39
nnn
.50
nnn
.55
nnn
.69
nnn
.23
nnn
.39
nnn
.45
nnn
.22
nnn
.25
nnn
.33
nnn
.03 .11 .18
nn
.25
nnn
.01
9. Power values .30
nnn
.29
nnn
.29
nnn
.28
nnn
.09 .17
nn
.32
nnn
.39
nnn
.17
nn
.65
nnn
.42
nnn
.03 .25
nnn
.03 .07 .11 .20
nn
.22
nnn
10. Achievement values .33
nnn
.46
nnn
.59
nnn
.38
nnn
.30
nnn
.17
nn
.32
nnn
.27
nnn
.40
nnn
.13
n
.53
nnn
.09 .27
nnn
.07 .07 .06 .28
nnn
.07
11. Masculine traits .33
nnn
.42
nnn
.50
nnn
.20
nnn
.12 .02 .12 .05 .34
nnn
.59
nnn
.12 .11 .39
nnn
.02 .07 .04 .36
nnn
.02
12. Feminine traits .24
nnn
.13
n
.12 .24
nnn
.46
nnn
.40
nnn
.49
nnn
.42
nnn
.10 .22
nnn
.02 .03 .60
nnn
.13
n
.04 .16
n
.29
nnn
.01
13. Neutral traits .26
nnn
.22
nnn
.22
nnn
.20
nnn
.34
nnn
.32
nnn
.46
nnn
.38
nnn
.28
nnn
.36
nnn
.37
nnn
.59
nnn
.27
nnn
.01 .02 .03 .21
nn
.18
nn
14. Family role attitudes .18
nn
.21
nnn
.19
nn
.18
nn
.06 .22
nnn
.19
nn
.12 .01 .18
nn
.14
n
.06 .01 .36
nnn
.27
nnn
.01 .06 .03
15. Religiosity .36
nnn
.20
nnn
.26
nnn
.19
nn
.18
nn
.56
nnn
.15
n
.12 .12 .11 .18
nn
.08 .02 .21
nnn
.90
nnn
.16
n
.01 .03
16. Marital satisfaction .11 .06 .06 .05 .18
nn
.01 .01 .01 .13
n
.09 .06 .09 .01 .12
n
.01 .49
nnn
.27
nnn
.24
nnn
17. Positive affect .25
nnn
.19
nn
.30
nnn
.18
nn
.19
nn
.04 .11 .08 .01 .30
nnn
.31
nnn
.25
nnn
.20
nnn
.07 .09 .33
nnn
.20
nn
.03
18. Negative affect .10 .01 .03 .01 .07 .12 .06 .01 .12
n
.01 .01 .10 .23
nnn
.01 .08 .26
nnn
.13
n
.13
n
Note: Tests of significance were two-tailed. Correlations above the diagonal are for wives and those below the diagonal are for husbands.
The values on the diagonal are correlations between spouses’ scores.
n
po.05.
nn
po.01.
nnn
po.001.
The two indices of values similarity were moderately correlated, as
well as the two indices of gendered traits similarity. However, the
two indices of couple similarity on the attitudes domain were
unrelated to each other or to the other similarity measures.
Correlation Analyses
Table 3 presents Pearson correlations between profile-based and
difference score–based similarity and husbands’ and wives’ satisfac-
tion and affect. Overall and consistent with the hypotheses, greater
similarity between partners was associated with higher levels of mar-
ital satisfaction and with lower levels of negative affect. The evidence
regarding levels of positive affect is weaker.
For both husbands and wives, similarity on the gendered person-
ality traits and values domains was more strongly and more consist-
ently associated with satisfaction and affect. Similarity on the attitudes
and religiosity domains showed weaker and inconsistent patterns
of associations. Among the subindexes of the difference score–based
similarity in values, similarity on self-direction, conformity, and
achievement values was the most important for wives’ satisfaction
Table 2
Intercorrelations Among Similarity Measures
1234567
1. Values similarity—diff –
2. Values similarity—prof .40
nnn
–
3. Gendered traits
similarity—diff
.08 .18
nn
–
4. Gendered traits
similarity—prof
.20
nn
.28
nnn
.43
nnn
–
5. Family role attitudes
similarity—diff
.04 .07 .01 .01 –
6. Family role attitudes
similarity—prof
.02 .08 .01 .10 .04 –
7. Religiosity—diff .16
n
.21
nn
.05 .09 .01 .08 –
M1.03 .43 .63 .35 .57 .51 .14
SD .35 .15 .33 .18 .48 .30 .37
Note: Tests of significance were two-tailed.
n
po.05.
nn
po.01.
nnn
po.001.
Similarity and Marital Satisfaction 1411
and affect, whereas similarity on benevolence values was particularly
important for husbands’ satisfaction and negative affect.
In general, marital satisfaction showed the strongest and most
consistent associations with couple similarity on almost all domains.
Negative affect showed somewhat weaker associations with similar-
ity in values and traits (although there were no significant differences
between the correlations) and no associations with similarity in at-
titudes and religiosity. Positive affect was not related to almost any
of the couple similarity domains.
Regression Analyses
To test more specifically which similarity domains were most im-
portant in explaining relationship measures, I ran a set of multiple
regression analyses for husbands and wives separately. In each anal-
ysis, a variable pertaining to one dependent measure was regressed
on the set of similarity domains.
Husbands’ measures. Table 4 indicates that the regression equation
of marital satisfaction on the set of spousal similarity domains
was significant overall and accounted for 18% of the variance
in the husband’s satisfaction with marriage. Three of the similarity
domains were significant predictors in the analysis: partners’
profile-based similarity on the gendered personality domain,
their profiled-based similarity on the values domain, and their dif-
ference score–based similarity on the attitudes domain. The greater
the spousal similarity on traits, values, and attitudes, the more sat-
isfied the husband was with his marital relationship.
The regression of a husband’s positive affect on the set of spousal
similarity domains was not significant. The only significant predictor
of the husband’s positive affect was couple similarity in religiosity:
The greater the spousal similarity on religious beliefs, the more the
husband experienced positive emotions.
The regression of the husband’s negative affect on the set of spou-
sal similarity domains was significant overall and accounted for 7%
of the variance in negative affect. Two of the similarity domains were
significant predictors in the analysis: partners’ profile-based similarity
on the values domain, and their profile-based similarity on the gende-
red personality domain. The greater the spousal similarity on values
and traits, the less the husband experienced negative emotions.
1412 Gaunt
Table 3
Correlations Between Profile-Based (Prof) and Difference Score–Based
(Diff) Similarity, Marital Satisfaction, and Affect
Wives’ measures Husbands’ measures
Marital
satisfaction
Positive
affect
Negative
affect
Marital
satisfaction
Positive
affect
Negative
affect
Values similarity
Hedonism
values—diff
.03 .14
n
.12 .01 .06 .08
Stimulation
values—diff
.04 .11 .03 .06 .01 .01
Self-direction
values—diff
.19
nn
.15
n
.19
nn
.12 .04 .01
Universalism
values—diff
.01 .01 .06 .02 .04 .06
Benevolence
values—diff
.15
n
.04 .04 .16
n
.06 .15
n
Tradition
values—diff
.09 .04 .03 .06 .01 .12
Conformity
values—diff
.13
n
.09 .15
n
.08 .01 .11
Security
values—diff
.10 .01 .10 .09 .04 .06
Power
values—diff
.01 .05 .07 .01 .02 .11
Achievement
values—diff
.12
n
.10 .21
nn
.02 .05 .01
Overall values
similarity—diff
.19
nn
.12
n
.19
nn
.13
n
.04 .08
Values
similarity—prof
.19
nn
.01 .11 .27
nnn
.01 .21
nnn
Gendered traits similarity
Masculine
traits—diff
.13
n
.12
n
.05 .03 .16
nn
.07
Feminine
traits—diff
.12
n
.07 .13
n
.11 .13
n
.01
Neutral
traits—diff
.10 .02 .09 .19
nn
.13
n
.06
(Continued)
Similarity and Marital Satisfaction 1413
Wives’ measures. The regression equation of a wife’s marital
satisfaction on the set of spousal similarity domains was significant
overall and accounted for 13% of the variance in the wife’s
satisfaction with marriage (Table 4). Two of the similarity domains
were significant predictors in the analysis: partners’ profile-based
similarity on the gendered personality domain, and their similarity
on the religiosity domain. The greater the spousal similarity in
traits and religious beliefs, the more satisfied the wife was with her
marriage.
The regression of a wife’s positive affect on the set of spousal
similarity domains was not significant. The only significant predictor
of the wife’s positive affect was partners’ difference-based similarity
on the values domain: the greater the spousal similarity in values, the
more the wife experienced positive emotions.
The regression of a wife’s negative affect on the set of spousal
similarity domains was significant overall and accounted for 11% of
Table 3 (Contd.)
Wives’ measures Husbands’ measures
Marital
satisfaction
Positive
affect
Negative
affect
Marital
satisfaction
Positive
affect
Negative
affect
Overall
gendered traits
similarity—diff
.18
nn
.12
n
.13
n
.15
n
.01 .07
Gendered traits
similarity—prof
.30
nnn
.10 .25
nnn
.33
nnn
.09 .17
nn
Family role
attitudes
similarity – diff
.05 .09 .05 .16
n
.09 .09
Family role
attitudes
similarity—prof
.01 .02 .08 .09 .04 .01
Religiosity
a
—
diff
.16
n
.06 .10 .14
n
.13
n
.03
Note: Tests of significance were two-tailed.
a
Religiosity was measured by only one question, and therefore a profile-based sim-
ilarity score could not be computed.
n
po.05.
nn
po.01.
nnn
po.001.
1414 Gaunt
Table 4
Regression Analyses Predicting Marital Satisfaction and Affect From Profile-Based (Prof) and Difference
Score–Based (Diff) Similarity
Values similarity
Gendered traits
Similarity
Family role
attitudes
similarity
Religiosity
similarity
a
Diff Prof Diff Prof Diff Prof Diff Prof R
2
F(7, 216)
Husbands
Marital satisfaction .02 .17
n
.01 .26
nnn
.17
n
.04 .10 – .18 6.96
nnn
Positive affect .05 .01 .02 .10 .05 .02 .14
n
– .03 1.11
Negative affect .01 .19
nn
.03 .13
n
.11 .03 .04 – .07 2.66
nn
Wives
Marital satisfaction .10 .02 .05 .24
nnn
.05 .02 .13
n
– .13 4.76
nnn
Positive affect .14
n
.07 .11 .03 .07 .03 .03 – .04 1.52
Negative affect .17
n
.01 .02 .21
nn
.05 .10 .03 – .11 3.94
nnn
Note: Standardized beta coefficients are reported.
a
Religiosity was measured by only one question, and therefore a profile-based similarity score could not be computed.
n
po.05.
nn
po.01.
nnn
po.001.
the variance in negative affect. Two of the similarity domains
were significant predictors in the analysis: partners’ difference-based
similarity on the values domain and their profiled-based similarity
on the gendered personality domain. The greater the spousal simi-
larity in values and traits, the less the wife experienced negative
emotions.
Tests of significant differences in coefficients between difference
score–based and profile-based similarity indices were conducted
applying the Fisher ztransformation (Cohen & Cohen, 1984).
Results indicated that the effects of profile-based similarity in
values on the husband’s marital satisfaction and negative affect
were significantly larger than those of the difference score–based
similarity in values. In addition, the effects of profile-based similarity
in gendered traits on the husband’s and wife’s marital satisfaction
and on the wife’s negative affect were significantly larger than
those of the difference score–based similarity in gendered traits.
However, the effect of difference score–based similarity in role
attitudes on the husband’s marital satisfaction was significantly
larger than that of the profile-based similarity in attitudes, and the
effect of difference score–based similarity in values on the wife’s
negative affect was significantly larger than that of the profile-based
similarity in values.
All in all, profile-based similarity seemed a somewhat stronger
and more consistent correlate of relationship measures than differ-
ence score–based similarity, although the pattern of differences be-
tween coefficients was not overwhelming.
DISCUSSION
Results from the current study indicate important relationships be-
tween spousal similarity and marital satisfaction. Consistent with the
hypotheses, greater similarity between partners was associated with
higher levels of marital satisfaction and lower levels of negative af-
fect. In particular, similarity on the gendered personality and values
domains was strongly associated with relationship measures, where-
as similarity on the role attitudes and religiosity domains showed
weaker and inconsistent patterns of associations. Finally, the results
indicated that profile-based similarity tends to be a stronger and
more consistent correlate of relationship measures than difference
score-based similarity.
1416 Gaunt
The findings regarding the strong associations between similarity
in gendered personality traits and marital measures are consistent
with previous studies (e.g., Gattis et al., 2004; Robins et al., 2000;
Watson et al., 2004). However, the findings relating similarity on the
values domain to marital measures are at odds with previous findings
(Lou & Klohnen, 2005). This inconsistency may stem from the use of
different measurement methods. The current study adopted a well-
validated measure of value priorities, which was developed on the
basis of the theory of human values (Schwartz, 1992; Schwartz &
Bilsky, 1987). It is plausible that this highly valid and sensitive
measure revealed a pattern of associations that previously was hid-
den due to insufficient measurement.
The focus on attitudes toward family roles was less fruitful. In
comparison to political views, attitudes toward parenting and family
roles seem more relevant to everyday functioning as a family. In-
deed, for the husbands in this study, similarity on the attitudes do-
main was related to marital satisfaction but not to positive and
negative affect. For the wives in the study, similarity on the attitudes
domain was unrelated to marital satisfaction and affect. Therefore,
the hypothesis regarding the role of specific attitudes has gained little
support. Future research could benefit from considering similarity
on a wider range of specific attitudes, including attitudes regarding
family roles, childrearing, financial management, etc.
This study also highlights the implications of couple similarity for
partners’ negative affect. The association between marital satisfac-
tion and spouses’ affect is not new to researchers (e.g., Mauno &
Kinnunen, 1999; Suhail & Chaudhry, 2004). The findings from the
current study indicate that couple similarity is linked to negative af-
fect but not so much to positive affect. Presumably, discrepancies in
values and traits between spouses may stimulate marital conflicts
and negative emotions of anger and resentment. Future studies
should further explore the contribution of couple’s similarity to
spouses’ affect and well-being, examining the possible mediating role
of marital conflict.
Finally, this study compared two operationalizations of couple
similarity: the widely used difference-scores operationalization and
the rarely used profile-based operationalization. Consistent with the
rationale regarding the greater amount of information taken into
account in the latter (cf. Luo & Klohnen, 2005), the profile-based
similarity appears to be a stronger and more consistent correlate of
Similarity and Marital Satisfaction 1417
relationship measures. Despite its obvious advantages, only three
studies to date have implemented profile-based operationalization to
study the link between similarity and marital satisfaction (Caspi &
Herbener, 1990; Glicksohn & Golan, 2001; Luo & Klohnen, 2005).
The comparison between the results of these two similarity indices,
therefore, suggests that the associations of similarity with relation-
ship measures may be stronger than implied by the findings from
previous studies.
The limitations of this study should be noted. The sample was
restricted to married couples with young children. This restriction
may limit the implications that can be derived from the findings.
Couple similarity may have different implications for relationships
in other stages of life, as it may for couples without children or for
parents of older children. In addition, the majority of the partici-
pants were fairly well-educated, a fact that further restricts the gen-
eralizability of the findings to other populations.
Another limitation was the cross-sectional design of this study,
which implied the simultaneous measurement of couple similarity
and relationship measures. Like previous research in this area (e.g.
Glicksohn & Golan, 2001; Robins et al., 2000; Russell & Wells,
1991), this design cannot be used to address issues regarding the di-
rection of causal relations. It is possible that satisfied couples become
increasingly similar with time, so that couple similarity is the result,
and not the cause, of marital satisfaction. Nonetheless, support for
the current rationale regarding the causal direction comes from
studies that examined whether spouses already were similar at the
time of their marriage (initial assortment), or became more similar
over time (convergence). These studies found that initial assortment
is primarily responsible for couple similarity (e.g., Glickson &
Golan, 2001; Watson et al., 2004). Still, future research is needed
to replicate the present findings using a longitudinal design. A design
that includes a measurement of similarity in the beginning of the
relationship and a measurement of marital outcomes and affect
several years later would successfully address the issue of causal
relations.
Finally, further broadening the range of domains on which sim-
ilarity is examined may advance our understanding of the role of
couple similarity in a relationship. Similarity in socioeconomic back-
ground, family history, cognitive skills, and a wide variety of atti-
tudes and traits may all contribute to explaining relationship quality
1418 Gaunt
and satisfaction. The findings of this study suggest that spousal sim-
ilarity has important implications for marital satisfaction and part-
ners’ affect. Further research exploring how similarity on a wide
range of domains affects various dimensions of relationships and
psychological functioning will deepen our understanding of the
processes underlying intimate relationships and their consequences
for individuals’ happiness.
REFERENCES
Ajzen, I., & Fishbein, M. (1977). Attitude behavior relations: A theoretical anal-
ysis and review of empirical research. Psychological Bulletin,84, 888–918.
Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Con-
sulting and Clinical Psychology,42, 155–162.
Blum, J. S., & Mehrabian, A. (1999). Personality and temperament correlates of
marital satisfaction. Journal of Personality,67, 93–125.
Bradburn, N. M. (1969). The structure of psychological well-being. Chicago:
Aldine.
Caspi, A., & Herbener, E. S. (1990). Continuity and change: Assortative marriage
and the consistency of personality in adulthood. Journal of Personality and
Social Psychology,58, 250–258.
Cohen, J., & Cohen, P. (1984). Applied multiple regression/correlation analysis for
the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Fowers, B. J., & Olson, D. H. (1993). ENRICH Marital Satisfaction Scale: A brief
research clinical tool. Journal of Family Psychology,7, 176–185.
Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R. (2004). What do
you do when things go right? The intrapersonal and interpersonal benefits of
sharing positive events. Journal of Personality and Social Psychology,87,
228–245.
Gattis, K. S., Berns, S., Simpson, L. E., & Christensen, A. (2004). Birds of a
feather or strange birds? Ties among personality dimensions, similarity, and
marital quality. Journal of Family Psychology,18, 564–574.
Gaunt, R. (2006). Maternal gatekeeping: Antecedents and consequences. Unpub-
lished manuscript.
Glicksohn, J., & Golan, H. (2001). Personality, cognitive style, and assortative
mating. Personality and Individual Differences,30, 1199–1209.
Heaton, T. (1984). Religious homogamy and marital satisfaction reconsidered.
Journal of Marriage and the Family,46, 729–733.
Karney, B. R., & Bradbury, T. N. (1995). The longitudinal course of marital
quality and stability: A review of theory, method, and research. Psychological
Bulletin,118, 3–34.
Lindeman, M., & Verkasalo, M. (2005). Measuring values with the short
Schwartz’s Value Survey. Journal of Personality Assessment,85, 170–178.
Similarity and Marital Satisfaction 1419
Luo, S., & Klohnen, E. C. (2005). Assortative mating and marital quality in new-
lywed: A couple-centered approach. Journal of Personality and Social Psychol-
ogy,88, 304–326.
Mauno, S., & Kinnunen, U. (1999). The effects of job stressors on marital sat-
isfaction in Finnish dual-earner couples. Journal of Organizational Behavior,
20, 879–889.
Meyer, J. P., & Pepper, S. (1977). Need compatibility and marital adjustment in
young married couples. Journal of Personality and Social Psychology,35, 331–
342.
Nemechek, S., & Olson, K. R. (1999). Five-factor personality similarity and mar-
ital adjustment. Social Behavior and Personality,27, 309–318.
Roccas, S., Sagiv, L., Schwartz, S. H., & Knafo, A. (2002). The big five person-
ality factors and personal values. Personality and Social Psychology Bulletin,
28, 789–801.
Robins, R. W., Caspi, A., & Moffitt, T. E. (2000). Two personalities, one rela-
tionship: Both partners’ personality traits shape the quality of their relation-
ship. Journal of Personality and Social Psychology,79, 251–259.
Russell, R. J., & Wells, P. A. (1991). Personality similarity and quality of mar-
riage. Personality and Individual Differences,12, 407–412.
Schwartz, S. H. (1992). Universals in the content and structure of values: The-
oretical advances and empirical tests in 20 countries. In M. Zanna (Ed.), Ad-
vances in Experimental Social Psychology (Vol. 25, pp. 1–65). New York:
Academic Press.
Schwartz, S. H., & Bardi, A. (2001). Value hierarchies across cultures: Taking a
similarities perspective. Journal of Cross-Cultural Psychology,32, 268–290.
Schwartz, S. H., & Bilsky, W. (1987). Toward a psychological structure of human
values. Journal of Personality and Social Psychology,53, 550–562.
Schwartz, S. H., & Sagiv, L. (1995). Identifying culture specifics in the content and
structure of values. Journal of Cross-Cultural Psychology,26, 92–116.
Suhail, K., & Chaudhry, H. R. (2004). Predictors of subjective well-being in an
eastern Muslim culture. Journal of Social and Clinical Psychology,23, 359–365.
Watson, D., Klohnen, E. C., Casillas, A., Nus Simms, E., Haig, J., & Berry, D. S.
(2004). Match makers and deal breakers: Analyses of assortative mating in
newlywed couples. Journal of Personality,72, 1029–1068.
1420 Gaunt
