Journal of Research in Personality 45 (2011) 697–701

Contents lists available at SciVerse ScienceDirect

Journal of Research in Personality

journal homepage: www.elsevier.com/locate/jrp

Brief Report

Gender, jealousy, and attachment: A (more) thorough examination across measures

and samples
Michael J. Tagler ⇑, Rachel H. Gentry
Department of Psychological Science, Ball State University, USA

a r t i c l e i n f o a b s t r a c t

Article history: We examined the role of attachment and gender on responses to hypothetical sexual and emotional infi-
Available online 22 August 2011 delity. Unlike previous studies, both categorical and continuous attachment style and infidelity distress
measures were administered to separate samples of college students and adults. Consistent with previous
Keywords: jealousy research, we found moderate gender differences on forced-choice measures of infidelity distress
Attachment but smaller differences on continuous measures. However, across all analyses, attachment style was not a
Sex differences significant predictor. We discuss this failure to replicate Levy and Kelly (2010) and provide suggestions
Jealousy
for future research.
Romantic relationships
Infidelity distress
Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction (secure, fearful, preoccupied) selected emotional infidelity. These

In a highly influential article, Buss, Larsen, Westen, and
Semmelroth (1992) tested the theory of evolved sex differences
in jealousy by asking college students which scenario they find
more distressing: Imagining their romantic partner forming a deep
emotional attachment to another person, or enjoying passionate
sexual intercourse with that person. The majority of women se-
lected emotional infidelity, whereas men were more likely to select
sexual infidelity. These forced-choice between-gender results have
been replicated many times, primarily with college student sam-
ples. In many studies, however, men exhibit considerable within-
gender response variability. Whereas a clear majority of women
usually select emotional infidelity as most distressing, men are
typically more evenly split in their selection of the two types of
infidelity (Harris (2003)). A rather vigorous debate has ensued
regarding the critical comparisons and pattern of results needed
to test evolutionary predictions of sex differences (e.g., Harris,
2005; Sagarin, 2005). Recent research by Levy and Kelly (2010),
however, suggests that within-gender differences in infidelity dis-
tress can be accounted for by attachment style.
In their study of college students (N = 317 women; 99 men) pub-
lished in a leading psychology journal, Levy and Kelly (2010) re-
ported nearly 100% of men with a dismissing attachment style, and
approximately 55% of dismissing women, selected sexual infidelity
as most distressing. By comparison, the majority of participants
(>60% of men, >70% of women) endorsing all other attachment styles
⇑ Corresponding author. Address: Department of Psychological Science, Ball State
University, Muncie, IN 47306, USA. Fax: +1 765 285 1702.
E-mail address: mjtagler@bsu.edu (M.J. Tagler).
results suggest that gender differences in jealousy are primarily
due to the reaction of dismissing men to the threat of partner sexual
infidelity. Levy and Kelly argue that their dramatic pattern of results
are expected given that men are more likely to have a dismissing
attachment style and that dismissing individuals are more sexually
promiscuous (Bartholomew & Horowitz, 1991; Brennan, Clark, &
Shaver, 1998).
However, administering the same forced-choice infidelity dis-
tress and categorical attachment style measures to a much larger
undergraduate sample (N = 2500 women; 1379 men), Treger and
Sprecher (2010) found a different (and much less pronounced) pat-
tern of results: Preoccupied men were relatively more likely to se-
lect emotional infidelity as most distressing, and avoidant women
were somewhat more likely to select sexual infidelity. But across
all attachment styles, men were more likely to select sexual infi-
delity, and women more likely to select emotional infidelity.
Given these conflicting results, we sought to further examine the
attachment–jealousy relationship. Specifically, we expanded the
methodology of previous studies by including non-forced-choice
measures of infidelity distress, dimensional measures of attach-
ment, and by recruiting a more diverse sample of participants. We
briefly describe the rationale for these changes below.
2. The measurement of jealousy
Levy and Kelly (2010) followed the forced-choice Buss et al.
(1992) procedure of assessing jealousy in which college-aged
participants must choose either sexual or emotional infidelity as
most distressing. As noted, reliable gender differences emerge
when this question format and population are used. However,

0092-6566/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jrp.2011.08.006
698 M.J. Tagler, R.H. Gentry / Journal of Research in Personality 45 (2011) 697–701
research with other question formats and populations has pro-
duced different results.
In regard to question format, studies employing continuous
scales to separately measure distress to emotional and sexual infi-
delity tend to not find significant gender differences (Harris, 2003).
Researchers favoring social-cognitive over evolutionary explana-
tions for gender differences argue that socialized decision-making
processes differently affect men and women’s forced-choice re-
sponses (DeSteno, Bartlett, Braverman, & Salovey, 2002; DeSteno
& Salovey, 1996; Harris & Christenfeld, 1996). In short, because
men have been socialized to believe that women have sex only
when in love, they may reasonably select sexual infidelity as more
distressing because it implies the co-occurrence of emotional infi-
delity. On the other hand, men do not select emotional infidelity
because it does not necessarily imply the co-occurrence of sexual
infidelity. Women, in contrast, may select emotional infidelity as
more distressing because of gender-role expectations concerning
the sexual promiscuity of men. Because it is expected that men will
readily have sex without love, evidence of sexual infidelity does
not necessarily imply emotional infidelity. But, because it is further
expected that men who have fallen in love are almost certainly
also having sex, women select emotional infidelity as most
distressing. Because this debate concerning the measurement of
jealousy remains unresolved (e.g., see recent evolutionary rebut-
tals by Edlund (2011) and Edlund & Sagarin (2009)), we believe
it is important to measure infidelity distress using both forced-
choice and continuous response formats.
Another unresolved issue in this research area is the overreliance
on college student samples. Although several studies have been
conducted with older samples of adults, the results have been less
consistent (see Tagler, 2010 for a review). Generally, the existing
studies with non-college and older-aged samples find the effect is
less robust (e.g., Green & Sabini, 2006; Harris, 2002; Tagler, 2010).
Thus, we believe it is also important for researchers to go beyond col-
lege student samples when studying gender differences in jealousy.
3. The measurement of attachment
Levy and Kelly (2010) used the Relationship Questionnaire (RQ;
Bartholomew & Horowitz, 1991) to categorize participants into se-
cure, fearful, preoccupied, and dismissing attachment categories.
Although the RQ represented an important advance in attachment
theory (the addition of the dismissive attachment type), contem-
porary researchers have largely moved away from the categorical
approach in favor of a dimensional model with separate, continu-
ous anxiety (fear of rejection) and avoidance (discomfort with oth-
ers) scores. While attachment categorization simplifies data
collection and analysis, it limits the ability to detect important
individual differences. In brief, Brennan et al. (1998) demonstrated
that measuring attachment using multi-item anxiety and avoid-
ance scales is more sensitive to degrees of insecurity and thus ac-
counts for more variance in attachment-related emotions,
thoughts, and behaviors. Moreover, their factor analysis of attach-
ment measures found clear evidence of the two global anxiety and
avoidance dimensions. Likewise, Fraley and Waller’s (1998) taxo-
metric analysis clearly indicated that attachment exists as a latent
dimensional rather than categorical construct. Thus, because of
concerns regarding the reliability and validity of measuring attach-
ment categorically, it is important to examine the attachment–infi-
delity distress relationship using dimensional measures.
4. The present study
The present study was designed as a more thorough examina-
tion of the attachment–infidelity distress relationship. To address
the methodological limitations of previous studies, we conducted
a study using both forced-choice and continuous measures of infi-
delity distress, both categorical and dimensional measures of
attachment, and recruited both college students and older adults.
Consistent with previous studies of college students, we expected
significant gender differences on the forced-choice infidelity dis-
tress measure, but small to no gender differences on continuous
scales. Moreover, we expected adults to show less pronounced
gender differences on the distress measures. Given the very limited
and conflicting previous attachment results, we made no hypothe-
ses regarding the attachment–infidelity distress relationship.
5. Method
5.1. Participants
5.1.1. College students
Undergraduates (357 women, 132 men), ranging from 18 to
23 years old (M = 18.91, SD = 0.99) were recruited from a psychol-
ogy participant pool website at a Midwestern (USA.) university.
The majority was White (90.8%), heterosexual (96.1%), and approx-
imately half (53.6%) were currently in a committed romantic rela-
tionship with a mean length of 1.49 years (SD = 1.21).
5.1.2. Adults
We simultaneously recruited a sample of middle-aged
(M = 43.52 years, SD = 12.06) employees (124 women, 64 men) of
the same university that the college study sample was collected.
These participants were recruited via an e-mail sent to all employ-
ees (including administration, faculty, and staff). Similar to the stu-
dent sample, they were predominately White (95.2%) and
heterosexual (88.8%). Most were currently in a committed roman-
tic relationship (78.2%) with a mean length of 15.01 years
(SD = 12.02).1
5.2. Measures
Participants followed a hyperlink to complete the following
infidelity distress and attachment measures, presented in counter-
balanced order.
5.2.1. Infidelity distress
Participants were asked to think of a serious committed rela-
tionship they had in the past, are currently having, or would like
to have, and then to imagine they discover their partner becomes
interested in someone else. Following the Buss et al. (1992)
forced-choice format, they selected which scenario would be most
distressing: imagining their partner (a) has formed a deep emo-
tional attachment with someone else, or (b) engaging in sexual
intercourse with someone else. Separately, participants also rated
their distress for each scenario on 5-point scales (1 = not distressing,
5 = very distressing).
5.2.2. Attachment
We used the Relationship Questionnaire (Bartholomew &
Horowitz, 1991) to categorize participants into secure, fearful, pre-
occupied, and dismissing attachment styles and the Experiences in
Close Relationships-Revised Questionnaire (ECR; Fraley, Waller, &
Brennan, 2000) to measure the attachment anxiety and avoidance
dimensions. In the present study, the ECR demonstrated excellent
reliability (a = .93 for both dimensions).
1
When included as covariates in the regressions, neither age, ethnicity, sexual
orientation or relationship status/length were significant predictors.
 M.J. Tagler, R.H. Gentry / Journal of Research in Personality 45 (2011) 697–701 699

Table 1 Table 2
Categorical attachment style frequencies by gender and sample. Logistic regression predicting forced-choice infidelity distress responses.

Secure Fearful Preoccupied Dismissing Variable b SE Wald OR

**
Students Gender (1 = Men) 1.13 0.19 37.06 3.09
Women 138 147 38 34 Sample (1 = Students) 0.51 0.21 6.16* 1.67
Men 61 27 27 17 Fearful 0.07 0.21 0.10 1.07
Preoccupied 0.21 0.28 0.59 1.24
Adults
Dismissive 0.45 0.27 2.74*** 1.57
Women 58 34 13 19
Constant 1.75 0.23 59.15** 0.17
Men 24 18 6 16
Total Note: The odds ratio (OR) indicates the likelihood of selecting sexual infidelity as
Women 196 181 51 53 more distressing than emotional infidelity.
*
Men 85 45 33 33 p < .05.
**
p < .001.
***
p < .10.

6. Results2
6.1. Association between gender and categorical attachment style
We first explored the association between gender and categor-
ical attachment style (see Table 1 for frequencies). Across the sam-
ples, there was a significant gender–attachment association,
v2(3, N = 677) = 17.28, p < .001, / = .16. Consistent with previous
studies using the RQ (Schmitt et al., 2003), men (16.8%) were more
likely than women (11.0%) to endorse the dismissing style. We also
found women (37.6%) more likely than men (23.0%) to endorse
fearful attachment. Examining these results separately by sample
revealed the same pattern of results among college students,
v2(3, N = 489) = 21.00, p < .001, / = .21. Although the gender–
attachment association did not reach significance in the adult sam-
ple, v2(3, N = 188) = 3.02, p = .39, / = .13, consistent with previous
research and our college student sample, more adult men (25.0%)
than women (15.3%) endorsed the dismissive style.
6.2. Forced-choice infidelity distress and categorical attachment
A hierarchical logistic regression was conducted to predict
forced-choice infidelity distress responses with sample (college
student or adult), gender, and attachment style (dummy coded
with secure as the reference category) entered in the first step,
all 2-way interactions entered in the second step, and the 3-way
interactions in the final step. No significant effects emerged from
the models with interactions. The step 1 model was significant,
v2(5, N = 677) = 46.53, p < .001, Nagelkerke R2 = .10, and correctly
classified 71.3% of the cases. As shown in Table 2, both gender
and sample were significant predictors: Men (46.4%) were much
more likely than women (22.2%) to select sexual infidelity as more
distressing, v2(1, N = 677) = 39.36, p < .001, / = .24, and college
students (31.3%) were somewhat more likely than adults (23.9%)
to select sexual infidelity, v2(1, N = 677) = 3.55, p = .06, / = .07. Cat-
egorical attachment was a weak predictor, with dismissive partic-
ipants (37.2%) somewhat more likely to select sexual infidelity
than secure participants (27.4%), v2(1, N = 367) = 3.03, p = .10, /
= .09. Fig. 1 displays the attachment style results separately for
men and women to emphasize the key findings: Gender was a sig-
nificant predictor, but attachment style did not predict nor moder-
ate the gender difference in infidelity distress.
6.3. Continuous infidelity distress and attachment dimensions
We conducted a series of hierarchical multiple regressions to
investigate if gender, sample, attachment dimensions (mean cen-
tered), and their interactions predict continuous infidelity distress
responses. We first separately analyzed distress to sexual infidelity
and distress to emotional infidelity. Because theorists have argued
that the proper test of evolutionary theory is a gender comparison
of the distress difference to sexual versus emotional infidelity
(Sagarin, 2005), we also created a relative distress index by sub-
tracting each participant’s emotional distress rating from their sex-
ual distress rating. Higher scores on this index indicate greater
relative distress to sexual infidelity.
No significant predictors emerged from the models with 3-way
or 4-way interactions.3 The only significant predictor of distress to
sexual infidelity was sample, b = 0.25, SE = 0.09, p < .01. Specifically,
college students reported more distress to sexual infidelity
(M = 4.39, SD = 0.96) than adults (M = 4.15, SD = 1.13), d = .23. The
only significant predictor of distress to emotional infidelity was gen-
der, b = 0.21, SE = 0.06, p < .001. Specifically, women (M = 4.70,
SD = 0.60) reported greater distress to emotional infidelity than
men (M = 4.49, SD = 0.90), d = .27. Analysis of the sexual-emotional
relative distress index revealed a significant constant (intercept),
b = 0.46, SE = 0.15, p < .01, indicating that participants reported
more distress to emotional than sexual infidelity (see Judd, Kenny,
& McClelland, 2001). This tendency, however, was qualified by a sig-
nificant gender  sample interaction, b = 0.39, SE = 0.17, p < .05.
Among college students, men (M = 0.03, SD = 0.76) were more dis-
tressed than women (M = 0.37, SD = 0.84) by sexual infidelity (rel-
ative to emotional infidelity), d = .50. However, adult men
(M = 0.47, SD = 1.00) and women’s (M = 0.47, SD = 1.05) relative
distress scores did not differ (d = 0.00). In sum, these results are con-
sistent with previous studies investigating gender differences in infi-
delity distress. No significant effects of attachment were found.
6.4. Forced-choice infidelity distress and attachment dimensions
Because there remains debate about whether forced-choice or
continuous measures of jealousy are more appropriate (Edlund,
2011), in the spirit of performing a thorough examination we con-
ducted a final analysis using the attachment dimensions, gender,
sample, and their interactions to predict forced-choice infidelity
distress. This hierarchical logistic regression, however, only pro-
duced the same gender (b = 1.15, p < .001, OR = 3.17) and sample
(b = 0.45, p = .04, OR = 1.56) effects previously reported in the logis-
tic regression with categorical attachment. Neither attachment-re-
lated anxiety or avoidance dimensions, nor any interactions, were
significant.
7. Discussion
Levy and Kelly (2010) reported that an overwhelming majority
of dismissing men and a slight majority of dismissing women
select sexual infidelity as most distressing, but the majority of

2 3
Bivariate correlations among all variables are presented in Supplementary Table 1. See Supplementary Table 2 for full hierarchical multiple regression results.
700 M.J. Tagler, R.H. Gentry / Journal of Research in Personality 45 (2011) 697–701
Fig. 1. Percentage of men and women who were more distressed by sexual than emotional infidelity, as a function of attachment style.
secure, fearful, and preoccupied individuals select emotional infi-
delity. However, both our results and those reported by Treger
and Sprecher (2010) indicate a pattern of gender differences con-
sistent with previous research: The clear majority of women select
emotional infidelity, while men are significantly more likely to se-
lect sexual infidelity (across all attachment styles). It is important
to note that Treger and Sprecher found a handful of statistically
significant differences among attachment styles, but the effect
sizes for these were relatively small and their statistical power
was very high. The only marginally significant effect of attachment
we found indicated an increased tendency for dismissive partici-
pants, relative to secure participants, to select sexual infidelity as
more distressing. As such, the collective evidence for a prominent
role of attachment in forced-choice distress responses is
unconvincing.
Moreover, we failed to find a significant role of attachment
using the dimensional approach, continuous measures of infidelity
distress, and across both college student and middle-aged adults.
But, consistent with previous research on infidelity distress
(DeSteno et al., 2002; Edlund & Sagarin, 2009; Harris, 2003), we
found small gender differences on continuous infidelity distress
measures. The fact that our infidelity distress findings (both
forced-choice and continuous) successfully replicated previous
studies increases confidence in the quality of our investigation.
What might explain why Levy and Kelly (2010) found such a
striking effect of attachment? Their small sample size, particularly
for men, may have played a role. Unfortunately a detailed gender
by attachment style sample size breakdown is not reported, but
a careful reading of their results indicates N = 68 men endorsed
either the secure or dismissing attachment style, and only N = 31
men endorsed either the fearful or preoccupied style. By compari-
son, our study included twice as many men. Moreover, Treger
and Sprecher’s (2010) examined over 1300 men, with results sim-
ilar to ours.
Although a key feature and strength of our study was the inclu-
sion of both forced-choice and continuous measures of infidelity
distress, both approaches concern purely hypothetical infidelity
scenarios. Because people are poor at predicting their emotional
reactions (affective forecasting, e.g., Hoerger, Quirk, Lucas, & Carr,
2009), responses to hypothetical infidelity may not accurately rep-
resent how they would respond to actual infidelity (DeSteno, 2010;
Harris, 2002). Given this threat to validity, improved methods to
the study of jealousy are needed. Laboratory manipulations of
jealousy have been developed (e.g., Desteno, Valdesolo, & Bartlett,
2006; Harmon-Jones, Peterson, & Harris, 2009), but in the interest
of experimental control these have only examined novel, scripted
relationships (with confederates). To examine preexisting relation-
ships, we suggest the design of naturalistic studies to capture jeal-
ousy and attachment behaviors in the real world (e.g., Fraley &
Shaver, 1998). Because both attachment and romantic jealousy
are fundamentally relational in nature, future research should also
examine the attachment dyad (i.e., both partners) and the role of
specific relationship experiences (e.g., actual infidelity). Although
self-report measures still have the potential to play an important
role in these efforts, exclusive reliance on forced-choice and cate-
gorical versions is likely to slow progress.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jrp.2011.08.006.
References
Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults:
A test of a four-category model. Journal of Personality and Social Psychology, 61,
226–244.
Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult
attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.),
Attachment theory and close relationships (pp. 46–76). New York: Guilford Press.
Buss, D. M., Larsen, R. J., Westen, D., & Semmelroth, J. (1992). Sex differences in
jealousy: Evolution, physiology, and psychology. Psychological Science, 3,
251–255.
DeSteno, D. (2010). Mismeasuring jealousy: A cautionary comment on Levy and
Kelly (2010). Psychological Science, 2, 1355–1356.
DeSteno, D., Bartlett, M. Y., Braverman, J., & Salovey, P. (2002). Sex differences in
jealousy: Evolutionary mechanism or artifact of measurement? Journal of
Personality and Social Psychology, 83, 1103–1116.
DeSteno, D. A., & Salovey, P. (1996). Evolutionary origins of sex differences in
jealousy? Questioning the ‘‘fitness’’ of the model. Psychological Science, 7,
367–372.
Desteno, D., Valdesolo, P., & Bartlett, M. Y. (2006). Jealousy and the threatened self:
Getting to the heart of the green-eyed monster. Journal of Personality and Social
Psychology, 91, 626–641.
Edlund, J. E. (2011). Jealousy reconsidered: A reply to DeSteno (2010). Evolutionary
Psychology, 9, 116–117.
Edlund, J. E., & Sagarin, B. J. (2009). Sex differences in jealousy: Misinterpretation of
nonsignificant results as refuting the theory. Personal Relationships, 16, 67–78.
Fraley, R. C., & Shaver, P. R. (1998). Airport separations: A naturalistic study of adult
attachment dynamics in separating couples. Journal of Personality and Social
Psychology, 75, 1198–1212.
 M.J. Tagler, R.H. Gentry / Journal of Research in Personality 45 (2011) 697–701
Fraley, R. C., & Waller, N. G. (1998). Adult attachment patterns: A test of the
typological model. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and
close relationships (pp. 77–114). New York: Guilford Press.
Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item-response theory analysis
of self-report measures of adult attachment. Journal of Personality and Social
Psychology, 78, 350–365.
Green, M. C., & Sabini, J. (2006). Gender, socioeconomic status, age, and jealousy:
Emotional responses to infidelity in a national sample. Emotion, 6, 330–334.
Harmon-Jones, E., Peterson, C. K., & Harris, C. R. (2009). Jealousy: Novel methods
and neural correlates. Emotion, 9, 113–117.
Harris, C. R. (2002). Sexual and romantic jealousy in heterosexual and homosexual
adults. Psychological Science, 13, 7–12.
Harris, C. R. (2003). A review of sex differences in sexual jealousy, including self-
report data, psychophysiological responses, interpersonal violence, and morbid
jealousy. Personality and Social Psychology Review, 7, 102–128.
Harris, C. R. (2005). Male and female jealousy, still more similar than different:
Reply to Sagarin (2005). Personality and Social Psychology Review, 9, 76–86.
Harris, C. R., & Christenfeld, N. (1996). Gender, jealousy, and reason. Psychological
Science, 7, 364–366.
701
Hoerger, M., Quirk, S. W., Lucas, R. E., & Carr, T. H. (2009). Immune neglect in
affective forecasting. Journal of Research in Personality, 43, 91–94.
Judd, C. M., Kenny, D. A., & McClelland, G. H. (2001). Estimating and testing
mediation and moderation in within-subject designs. Psychological Methods, 6,
115–134.
Levy, K. N., & Kelly, K. M. (2010). Sex differences in jealousy: A contribution from
attachment theory. Psychological Science, 21, 168–173.
Sagarin, B. J. (2005). Reconsidering evolved sex differences in jealousy: Comment on
Harris (2003). Personality and Social Psychology Review, 9, 62–75.
Schmitt, D. P., Alcalay, L., Allensworth, M., Allik, J., Ault, L., Austers, I., et al. (2003).
Are men universally more dismissing than women? Gender differences in
romantic attachment across 62 cultural regions. Personal Relationships, 10,
307–331.
Tagler, M. J. (2010). Sex differences in jealousy: Comparing the influence of previous
partner infidelity among college students and adults. Social Psychological and
Personality Science, 1, 353–360.
Treger, S., & Sprecher, S. (2010). The influences of sociosexuality and attachment
style on reactions to emotional versus sexual infidelity. Journal of Sex Research,
48, 1–10.