Traditional Masculinities and Men’s Sexism: A Meta-Analysis

Abstract

Traditional masculinities (TM) are a set of meanings related to men. The power attached to these meanings operates to maintain the inequality of gender relations. Previous studies have reported a positive correlation between traditional masculinities and men’s sexism, limiting women’s personal and professional opportunities. To quantify this relationship, a meta-analysis was conducted using 107 independent samples (mainly from the United States) from 21,078 men, reported in 44 articles between 1984 and 2021. Although we observed that TM positively correlated with men’s sexism, there was wide variability in effect sizes. An examination of potential moderators revealed that the relationship was significantly different between different forms of TM. The relationship between sexism and conformity to masculine norms and traditional masculinity ideology was stronger than the relationship with gender role stress. Further, TM were more strongly related to forms of sexism that reflect overt negative attitudes toward women (old-fashioned and hostile sexisms) than to forms that reflect covert sexist attitudes (modern and benevolent sexism). Moreover, the link between TM and sexism was stronger in the general population than in student samples. However, these findings should be interpreted and taken with caution due to the presence of substantial heterogeneity in effect sizes.

Full text

Psychology of Men & Masculinities
Traditional Masculinities and Men’s Sexism: A Meta-Analysis
Vladislav Krivoshchekov, Olga Gulevich, and Ilia Blagov
Online First Publication, July 6, 2023. https://dx.doi.org/10.1037/men0000444
CITATION
Krivoshchekov, V., Gulevich, O., & Blagov, I. (2023, July 6). Traditional Masculinities and Men’s Sexism: A Meta-Analysis.
Psychology of Men & Masculinities. Advance online publication. https://dx.doi.org/10.1037/men0000444

Traditional Masculinities and Men’s Sexism: A Meta-Analysis
Vladislav Krivoshchekov
1
, Olga Gulevich
2
, and Ilia Blagov
2
1
Department of Psychology, University of Bern
2
Department of Psychology, HSE University
Traditional masculinities (TM) are a set of meanings related to men. The power attached to these meanings
operates to maintain the inequality of gender relations. Previous studies have reported a positive correlation
between traditional masculinities and men’s sexism, limiting women’s personal and professional
opportunities. To quantify this relationship, a meta-analysis was conducted using 107 independent samples
(mainly from the United States) from 21,078 men, reported in 44 articles between 1984 and 2021. Although
we observed that TM positively correlated with men’s sexism, there was wide variability in effect sizes. An
examination of potential moderators revealed that the relationship was significantly different between
different forms of TM. The relationship between sexism and conformity to masculine norms and traditional
masculinity ideology was stronger than the relationship with gender role stress. Further, TM were more
strongly related to forms of sexism that reflect overt negative attitudes toward women (old-fashioned and
hostile sexisms) than to forms that reflect covert sexist attitudes (modern and benevolent sexism). Moreover,
the link between TM and sexism was stronger in the general population than in student samples. However,
these findings should be interpreted and taken with caution due to the presence of substantial heterogeneity
in effect sizes.
Public Significance Statement
This study integrates findings from 21,078 respondents across 107 samples of men regarding the
relationship between traditional masculinities and men’s sexism. In general, men’s endorsement of
traditional masculinities was positively associated with sexism against women. However, we found
evidence of considerable variability in the effect sizes. Therefore, the estimates should be taken with
caution and not be interpreted in absolute terms.
Keywords: traditional masculinity ideology, conformity to masculine norms, gender role conflict, sexism
Supplemental materials: https://doi.org/10.1037/men0000444.supp
Psychology of Men & Masculinities defines masculinities
as “the constellation of cultural and individual meanings attached
to men and boys that are attributed to the self as well as to people,
concepts, and objects, embedded in situational cues, performed
as social practices, and distributed through ecological influences”
(Wong & Wang, 2022, p. 2). As such, masculinities do not contain
universal and invariant meanings. Instead, there are multiple,
dynamic meanings associated with men that reflect a society’s
culture and individual constructions of masculinities.
Masculinity scholars identify five areas of analysis of masculi-
nities: (a) meanings that men attribute to themselves, (b) meanings
that people attribute to men in general, (c) the situational
behavior of men, (d) the behavior of men in the presence of other
people, and (e) the meanings associated with men in certain
groups and society as a whole (Wong & Wang, 2022). Among
these areas, a lot of psychological work is dedicated to the first
and second areas—scholars study men’s meanings, they ascribe
to themselves and men in general. Therein, psychologists have
concentrated on studying traditional masculinities (TM), which
reflect the system of individual and cultural meanings that
underlies patriarchy.
A social organization marked by the supremacy of the father in
the clan or family, the legal dependence of wives and children,
and the reckoning of descent and inheritance in the male line; control
by men of a disproportionately large share of power. (Lewis,
2018, p. 382)
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Vladislav Krivoshchekov https://orcid.org/0000-0003-1213-0884
Olga Gulevich https://orcid.org/0000-0003-3806-5064
This article is original. None of its part has been published or presented
before.
Vladislav Krivoshchekov and Olga Gulevich contributed equally and
share the first authorship.
The data are available at https://osf.io/gb8cp/.
Correspondence concerning this article should be addressed to Vladislav
Krivoshchekov, Department of Psychology, University of Bern, Fabrikstrasse
8, 3012 Bern, Switzerland. Email: vladislav.krivoshchekov@unibe.ch
Psychology of Men & Masculinities © 2023 American Psychological Association
ISSN: 1524-9220 https://doi.org/10.1037/men0000444
1

Such a system reproduces itself through, for instance, men’s
endorsement of certain individual and cultural beliefs about
masculinities that are related to their attitudes toward others
(Gerdes et al., 2018;Schwartz et al., 2016).
In particular, the endorsement of traditional masculinities was
found to be associated with sexism. Sexism can be defined as
“individuals’attitudes, beliefs, and behaviors, and organizational,
institutional, and cultural practices that either reflect negative
evaluations of individuals based on their gender or support unequal
status of women and men”(Swim & Hyers, 2009,p.407).A
variety of people can become a target of sexism, but scholars
usually focus on beliefs, attitudes, behaviors, and social practices
that are directed against women. Such sexism is associated with
attitudes toward women–romantic partners (Sibley & Overall,
2011) and work colleagues (Connor et al., 2016), sexual
objectivation of women, and violent behavior against them
(Agadullina et al., 2022). The more people endorse individual-
level sexism, the more their judgments and actions limit the
opportunities of women, and the more physical and psychological
damage they might cause.
Past research has shown that men’s endorsement of traditional
masculinities reinforces sexism (e.g., Gerdes et al., 2018;Schwartz
et al., 2016). The present meta-analysis aims to quantify this
relationship. It was based on the following principles: First, we
looked at the association between traditional masculinities and
sexism in samples of men; second, we included only research that
measured traditional masculinities and sexism with validated
inventories; third, we considered that masculinity is a multifaceted
phenomenon that can take many forms and includes different
domains; and fourth, we included research that measured different
forms of sexism.
Traditional Masculinities
There are multiple masculinities but, in many countries,
traditional masculinities (TM), a system of beliefs about how a
man should behave which dominated Western society prior to the
feminist deconstruction of gender roles and rules, are used by men as
self-guides on how to be a “real”man (Pleck, 1995;E. H. Thompson
et al., 1992). Most studies on TM have been based on research
involving men living in English-speaking countries. Nevertheless,
research within and outside English-speaking countries demon-
strated gender, sexuality, age, ethnic, and national differences
(e.g., Komlenac & Hochleitner, 2022;Komlenac et al., 2023;
Krivoshchekov et al., 2021,2022;R. F. Levant, 2011;R. Levant &
Richmond, 2007).
There are three main theoretical concepts used to describe TM.
The first is traditional masculinity ideology (TMI), which is a system
of beliefs about what men should be in general (R. F. Levant, 2011).
The second is conformity to masculine norms, which is the degree to
which men follow the TMI in their behavior (Mahalik et al., 2003).
Finally, gender role conflict (GRC) refers to the degree to which
conformity to male gender roles restricts, devalues, or violates the
self or others (O’Neil et al., 1986).
TM are theorized to be multifaceted. Psychologists believe that all
forms of TM—TMI, conformity to masculine norms, and GRC—
consist of several interrelated domains (R. F. Levant et al., 2010;
Mahalik et al., 2003;O’Neil, 2015;E. H. Thompson & Bennett,
2015;E. H. Thompson et al., 1992). Despite the interrelated nature
of TM, it is possible to highlight the key domains that are addressed
in different models. In our opinion, these domains reflect two
main ideas.
First, TM entails antifemininity and heterosexuality mandates
(e.g., unwillingness to look and behave like women and gay men,
negative attitudes toward gay men, high activity in the sexual
domain). Second, TM implies that society has a hierarchical
structure, and “real men”should be at the top of the hierarchy
(e.g., control over emotions, independence, risk-taking, striving to
be the best, the primacy of work over personal life, and inclination to
violence). Together, these two ideas function to keep a positive male
identity by helping men maintain their group distinctiveness and
social status.
Researchers theorize that TM has an impact on men’s
beliefs, emotions, and behavior. For example, the authors of
the expectancy–discrepancy–threat model of masculine identity
(Stanaland et al., 2023) consider communities where rigid
masculinity norms are common. They make a distinction between
men who reject these norms and men who see them as self-guides.
Men who accept norms compare themselves with a “real man”and
conclude that they fit or do not fit this image. Men who reject these
norms may find a comparison with a “real man”irrelevant.
Acceptance versus rejection of rigid norms and perceived
conformity versus noncompliance with these norms affects the
reactions of men in the face of threat. Men who either reject TMI or
are confident in their masculinity ignore masculinity threat (i.e., not
being man enough). At the same time, doubt in their masculinity
can result in two outcomes that depend on their motivation: Men
who want to conform to rigid norms feel anxious or guilty and
commit self-destructive acts, whereas men who think they have to
conform to the rigid norms demonstrate sexism and aggression
toward women.
We posit that the elements of this model correspond to three
forms of traditional masculinity. The perception of rigid
masculinity norms widespread in society correspond with TMI.
Aman’sperceptionofhisfit versus noncompliance with
traditionally masculine standards corresponds with high versus
low conformity to masculine norms. A state arising under the
influence of a threat corresponds with GRC. As such, on a
theoretical level, TMI influences conformity to masculine norms,
conformity to masculine norms affects GRC, and GRC causes
internal and external reactions, including sexism.
Sexism Against Women
Sexism (attitudes, beliefs, behaviors, and practices directed
against women) can manifest itself at three levels: individual,
institutional, and cultural. Individual-level sexism refers to
individual reactions or interpersonal interactions designed to
denigrate women, who are viewed as inferior in society.
Institutional-level sexism refers to policies, practices, and norms
that perpetuate inequality by restricting opportunities for women.
Finally, cultural-level sexism refers to symbols and practices that
are used to reinforce the notion that women are inferior to men
(Lewis, 2018).
Different levels of sexism are positively related to each other.
For instance, cross-cultural studies have shown that the more sexist
attitudes are widespread in the country, the lower its level of gender
equality (Bosson et al., 2021;Brandt, 2011;Glick et al., 2000).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2KRIVOSHCHEKOV, GULEVICH, AND BLAGOV

However, most psychological research has focused on individual-
level sexism, which includes the sexist beliefs, attitudes, and
behaviors that individuals endorse and practice.
Scholars have theorized several forms of sexist attitudes. First,
psychologists distinguish between old-fashioned and modern
sexism (Lewis, 2018;Swim et al., 1995). Old-fashioned sexism is
historically the first form of sexism that reflects an overtly negative
attitude toward women. Old-fashioned sexism refers to the belief
that there are individual differences between women and men
(e.g., in terms of intelligence and leadership skills). Therefore,
women and men should have different roles in society (e.g.,
women should take care of the family and men should pursue a
career) and deserve different treatment. Modern sexism is a form
of sexism that emerged in the second half of the last century, at a
time when regulatory pressure increased in some countries to
suppress old-fashioned sexist attitudes and beliefs about women.
Modern sexism includes the denial of continued discrimination,
antagonism toward women’s demands for additional rights, and
lack of support for policies designed to help women. Thus, this
form of sexism reflects a covertly negative attitude toward women,
caused by the growing similarity in the social status of men
and women.
Second, scholars also distinguish between hostile and benevolent
forms of sexism (Connor et al., 2016;Glick & Fiske, 2001;Glick
et al., 2000). Hostile sexism, overtly misogynistic and competitive
attitudes toward women, is the belief that women are too easily
offended, create problems, and seek to control men. This form of
sexism reflects negative attitudes toward women who pose a threat
to the gender hierarchy (e.g., feminists) but is weakly associated
with attitudes toward women who do not pose a threat. Benevolent
sexism reflects subjectively favorable but patronizing attitudes
toward women. It is the belief that women are more cultured and
elevated than men but need men’s protection and men need women.
This form of sexism reflects a positive attitude toward women who
conform to traditional gender norms and who primarily show purity
and warmth in close relationships (e.g., housewives) but is weakly
associated with attitudes toward women who pose a threat to male
dominance.
Research shows that different forms—old-fashioned versus
modern (Swim et al., 1995) and hostile versus benevolent (Glick
et al., 2000) sexism—are positively related. However, they are
related to other measures (e.g., trait attribution, romantic partner
preference, perceptions of job-related discrimination, and voting
preferences) in different ways. As such, different forms of
individual-level sexism are elements of a single system of gender
representations that justifies and supports gender inequality but
reflects different aspects of these representations.
Relationship Between TM and Sexism
Research over several decades has shown that TM were positively
associated with old-fashioned and modern (Leaper & Van, 2008;
Martínez-Martínez & Paterna-Bleda, 2013;Smiler, 2006), as well as
benevolent and hostile (Amayreh, 2019;Bosson et al., 2021;Covell,
1998) sexisms. Therefore, we hypothesized that endorsement of TM
would be positively associated with men’s sexism (Hypothesis 1).
However, this relationship can be moderated by additional factors.
The first factor is the form of TM. On the one hand, compared to
gender role stress, the content of TMI and conformity to masculine
norms has a greater overlap with sexism (Schwartz et al., 2016);
therefore, one might expect that TMI and conformity to masculine
norms would be more strongly associated with sexism than gender
role stress would be. This is supported by a recent meta-analysis,
which showed that TMI has the strongest and gender role stress
has the weakest association with violence against women
(Krivoshchekov et al., 2023).
On the other hand, the expectancy–discrepancy–threat model of
masculine identity (Stanaland et al., 2023) allows one to consider
different forms of TM as links in the same chain that connects the
presence of rigid masculine norms in society with psychological
reactions to the threat of masculinity. According to this model,
sexism is a direct response to gender role stress; therefore, it can be
assumed that gender role stress would be more strongly associated
with sexism than TMI and conformity to masculine norms would.
As such, Research Question 1 is “How are different forms of TM
related to men’s sexism?”
The second factor is the domain of TM. Some scholars
consider that various elements of TM form a single construct
(i.e., traditional masculinity ideology and GRC; Komlenac &
Hochleitner, 2022;Krivoshchekov et al., 2021;McDermott et al.,
2017;O’Neil, 2015); therefore, it can be assumed that different
domains are similarly associated with sexism. Others believe that
the different elements of TM are relatively independent of each
other and do not form a single construct (conformity to masculine
norms; Komlenac et al., 2023;Krivoshchekov et al., 2022;R. F.
Levant et al., 2020); therefore, one would expect that different
domains are differently associated with sexism. In the second case,
the question arises as to which TM domains are more strongly
associated with sexism.
Traditional masculinities reflect two different characteristics of a
“real man”: The need (a) to be distinct from women and enter
heterosexual relationships and (b) to occupy a high position in the
social hierarchy. Both characteristics imply a certain attitude toward
women: The first case is about romantic relationships and the second
is about dominance over women in the public sphere. However,
the first group of domains seems to have more conceptual overlap
with sexism than the second. Thus, Research Question 2 is “How are
different domains of TM related to men’s sexism?”
The third factor is the form of sexism. On the one hand, sexist
attitudes are elements of one system of gender representations;
therefore, it can be expected that TM is similarly associated with
different forms of sexism. On the other hand, old-fashioned and
hostile sexisms reflect overtly negative attitudes toward women,
while modern and benevolent sexisms are covert attitudes.
Therefore, one might assume that old-fashioned and hostile sexisms
would be more strongly associated with TM than modern and
benevolent sexisms would be. Thus, Research Question 3 is “How is
TM related to different forms of men’s sexism?”
Finally, the relationship between TM and sexism may vary
depending on the characteristics of the sample (e.g., sexuality,
sample size, sample type). Some studies included only young
people, such as pupils from schools and university students,
while the others included more diverse samples. In addition,
some researchers limited themselves to straight respondents,
whereas others included people of different sexual orientations.
Therefore, Research Question 4 is “How is the relationship
between TM and men’s sexism moderated by the characteristics of
the sample?”
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
MASCULINITIES AND MEN’S SEXISM 3

Method
Inclusion Criteria
To be included in the present meta-analysis, each study had
to meet several predefined criteria, namely gender composition
of the sample, inventories for measuring TM, and inventories for
measuring sexism.
Gender Composition of the Sample
We included two types of studies: Studies conducted using men-
only samples and studies conducted using mixed samples that
reported separate data for men and women. In both cases, we used
only the responses of men. We excluded studies with women-only
samples and studies that reported data for men and women together
because the goal of the present meta-analysis is to examine men’s
self-reported sexism.
Inventories to Measure TM
We included studies that measured at least one of the three forms
of TM (i.e., TMI, conformity to masculine norms, and GRC).
Analysis of the papers found during the search identified 10
inventories that were used to study the relationship between TM and
sexism.
To measure TMI, different versions of five inventories were
used: Male Role Norms Inventory (MRNI; R. F. Levant et al.,
2010), Male Role Norms Scale (MRNS; E. Thompson & Pleck,
1986), Brannon Masculinity Scale (BMS; Brannon & Juni, 1984),
Precarious Manhood Beliefs (PMB; Bosson et al., 2021), and
Adolescent Masculinity Ideology in Relationships Scale (AMIRS;
Chu et al., 2005). Three out of the five inventories (MRNI, MRNS,
BMS) are thoroughly described in a review article on TM
measurement (E. H. Thompson & Bennett, 2015), and the
validations of the AMIRS and PMB have been presented in the
original publications.
To measure conformity to masculine norms, various versions of
the Conformity to Masculine Norms Inventory (Mahalik et al.,
2003), Auburn Differential Masculinity Inventory (Burk et al.,
2004), and Hypermasculinity Inventory (Mosher & Sirkin, 1984)
were used, whereas for measuring GRC, Masculine Gender Role
Stress (Eisler & Skidmore, 1987), and Gender-Role Conflict Scale
(GRCS; O’Neil et al., 1986) were used. The Hypermasculinity
Inventory includes statements for measuring both conformity to
masculine norms and TMI, which has raised doubts about which
form of TM it represents. Based on the statements used in the coded
articles, we decided to code this inventory in the present meta-
analysis as conformity to masculine norms.
Inventories to Measure Sexism
During the screening, we identified four groups of inventories that
were used to measure sexism. The first group included ques-
tionnaires to measure old-fashioned sexism: Old-Fashioned Sexism
Scale (OS; Swim et al., 1995), different forms of Attitudes Toward
Women Scale (ATW or AWS; Spence & Helmreich, 1978;Spence
et al., 1973), and Traditional Gender Beliefs Scale (Dasgupta &
Rivera, 2006). The second group included questionnaires for
measuring modern sexism: Modern Sexism (MS; Swim et al., 1995)
scale and Neosexism scale (Tougas et al., 1995).
The third group included one inventory designed to measure
benevolent attitudes toward women: Benevolent Sexism, a subscale
from the Ambivalent Sexism Inventory (BS; Glick & Fiske, 1996).
The fourth group included inventories for measuring hostile
attitudes: Hostile Sexism, a subscale from the Ambivalent Sexism
Inventory (HS; Glick & Fiske, 1996) and Hostility Toward Women
Inventory (Check, 1984).
Literature Search
To identify the eligible studies, we conducted a systematic source
search from April to August 2021. The literature search was
conducted across seven electronic databases: Web of Science,
Scopus, ProQuest, Google Scholar, APA PsycInfo, EBSCO
(Academic Search Ultimate, eBook Collection), and Directory of
Open Access Journals (DOAJ). To ensure that a broad spectrum of
studies was included in our meta-analysis, we placed no restrictions
related to subject area, type of sources, or year of publication. This
search resulted in a list of journal articles, conference abstracts, and
dissertation texts.
To identify relevant studies, we searched using four concept
blocks, three that were designed to identify studies assessing TM
and one designed to identify studies assessing sexism. All terms
within the same concept block were connected with “or.”We ran
three searches in each database using fields of title, abstract, and
keywords, one with each TM concept block paired, using “and,”
with the sexism concept block. Search terms are displayed in
Figure 1.
Our search produced a total of 1,702 search results. We manually
excluded 1,151 duplicate articles. Two authors subsequently
assessed each of the remaining 551 results for relevance (“yes,”
“no,”“maybe”) based on the abstract. Those coded as “maybe”were
discussed by both authors and were considered jointly and
rejected or accepted after discussion. For the resulting 502 records,
we subsequently retrieved the full-text articles for more careful
examination. Following our inclusion criteria, we excluded
additional articles because they did not contain relevant measures
(n=405) or used a nonmale sample (n=41). We further excluded
several articles after careful examination of the method sections
because they did not contain necessary correlations (n=10) or were
papers that had different titles and different statuses (published vs.
unpublished) but belonged to the same author and reported the same
results (n=2). In this case, we coded the published source. Figure 2
contains the PRISMA flow diagram that summarizes the overall
search process.
The final list included 44 sources, including 33 journal articles,
two master’s theses, and nine dissertations. These sources included
107 independent samples. Thirty-nine samples were collected in the
United States, and 68 samples were collected in other countries,
namely the United Kingdom (Archer, 2010;Archer & Rhodes,
1989), Poland (de Zavala & Bierwiaczonek, 2021), South Korea
(Seo et al., 2022), Switzerland (Martínez-Martínez & Paterna-Bleda,
2013), Spain (Larrañaga et al., 2013), and cross-cultural research in
62 countries (Bosson et al., 2021). A list of these studies is reported
in the reference section of this article, as well as online at the Open
Science Framework (Krivoshchekov & Gulevich, 2023).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
4KRIVOSHCHEKOV, GULEVICH, AND BLAGOV

Information Retrieved From the Studies
Each study included in the meta-analysis was coded for a number
of variables. First, we extracted the effect sizes (i.e., correlations)
and associated pvalues for the relationships between TM and
sexism. Most studies did not report exact pvalue, therefore,
we coded them at four levels (i.e., “.001,”“.01,”“.05,”and “ns”
for nonsignificant results).
Second, we coded inventories for measuring TM. If the
researchers only measured the overall score, we coded the effect
size. If the researchers reported correlations both for the overall
score and the scores for separate subscales, we coded all the
reported data.
To analyze the effect of possible moderators, that is, the form and
domains of TM (Research Questions 1 and 2), we coded the overall
scores of inventories for measuring TM. We then investigated the
effect sizes among subscales corresponding to different domains of
TM. A preliminary analysis of the studies indicated that most
researchers provided data on separate subscales. The correspon-
dence among the contents of the subscales is presented in Table 1.
The correspondence between the Conformity to Masculine Norms
Inventory and GRCS was based on R. F. Levant et al. (2016).
Second, we coded the inventories for measuring sexism.
We distinguished between old-fashioned, modern, benevolent,
and hostile sexisms (Research Question 3). Third, we coded the
characteristics of the respondents: the number of respondents, the
average age of respondents, sexual orientation (exclusively
straight sample vs. predominantly straight sample vs. lack of
data on sexual orientation), and the sample type (students vs.
general sample that included men of different ages; Research
Question 4).
Analytical Strategy
The entire analysis was conducted in R (R Core Team, 2022). We
transformed correlation coefficients to Fisher’szscores for the
analysis. To calculate the variances for each effect size, we used the
metafor package (Viechtbauer, 2010). The analysis followed the
guidelines to conduct a high-quality meta-analysis (Pigott &
Polanin, 2020).
Typically, researchers reported the correlations for different
domains of TM and the total scores along with multiple measures
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Figure 1
Search Terms
Figure 2
PRISMA Flowchart
Records identified through
database searching, N= 1702
(Specific results: Web of
Science = 460, Scopus = 256,
ProQuest = 122, Google Scholar
= 784, PsycINFO = 27, EBSCO
= 42, DOAJ = 11)
Records removed before
screening:
Duplicate records removed
(n= 1151)
Abstracts were screened
(n= 551)
Records excluded (n= 49)
Full texts assessed for eligibility
(n= 502).
Full-text articles excluded, with
reasons (n= 446)
No relevant measure = 405
Non-male sample = 41
Method sections were analyzed
(n=56).
Full-text articles excluded,
with reasons (n= 12)
No correlations = 10
Results duplicates = 2
Studies included in review
(n= 44)
Identification
Screening
Included
Note. DOAJ =Directory of Open Access Journals; PRISMA =Preferred
Reporting Items for Systematic Reviews and Meta-Analyses.
MASCULINITIES AND MEN’S SEXISM 5

of sexism; therefore, the derived effect sizes are not independent.
To account for the dependency, we applied the robust variance
estimation (RVE; Hedges et al., 2010) available via the robumeta
package (Fisher & Tipton, 2015). This method allows multiple
effect sizes from the same study to be included in a meta-analysis,
even when information on the covariance of these effect sizes is
unavailable.
We used the total scores of the inventories to represent TM in our
analyses. Also, given the multidimensional nature of TM, we
performed the analysis described below separately for total scale
scores and for the separate domains of TM. To estimate the overall
correlation between TM and sexism, we first used an intercept-only
meta-regression model, where the intercept was interpreted as
the precision-weighted average of the observed effect sizes and
corrected for effect-size dependence.
Second, we performed a moderation analysis, where the
moderator variable was included in the meta-regression as a
predictor. For categorical variables with two-factor levels, we used a
ttest for the regression coefficient (i.e., the difference between two
levels) as a test of moderation. For categorical variables with three-
and more-factor levels, we performed cluster wild bootstrapping
(CWB) via the wildmeta package (Joshi & Pustejovsky, 2022;Joshi
et al., 2022). This function allows testing of whether the average
effect size is equal across all levels of the moderator using the F-type
test (note that the function in R only produces a pvalue as an output).
Joshi et al. (2022) recommend using CWB over Approximate
Hotelling test in meta-analyses using RVE. They demonstrated that
compared to extant small-sample correction methods, CWB
maintains adequate Type I error rates and provides more power,
particularly for multiple-contrast hypothesis tests. In contrast, CWB
and Approximate Hotelling test have very similar power in tests of
single-contrast hypotheses (i.e., ttests). As such, we applied small-
sample corrections only to single-contrast tests (Tipton, 2015). To
estimate the weighted mean effect sizes for different levels of
moderators, we used meta-regression models without intercept. To
examine the bivariate relationship, we ran the meta-regression
models for all moderators separately. After that, we used the meta-
regression model that simultaneously included all moderators.
Publication Bias
To investigate the presence of publication bias, we first used
funnel plots and examined the asymmetry via Egger’s regression,
which is a weighted, least squares regression of effect size on
standard errors. The significance of the coefficient associated with
standard error in Egger’s regression can be interpreted as a test of
funnel plot asymmetry (Sterne et al., 2011). Second, we used the
Precision Effect Test–Precision Effect Estimate with Standard
Errors (PET–PEESE; Stanley & Doucouliagos, 2014).
In simulation studies, it has been demonstrated that PET performs
better at identifying true zero effects, while PEESE leads to better
estimates when the true effect size is nonzero. As such, it is
recommended to use both methods. We used the PET method to
test whether there was a significant nonzero effect size. If the PET
analysis showed a significant result, PEESE was then used to
estimate the true effect size. Both Egger’s and PET–PEESE
regression tests were adjusted for correlation between effect
estimates.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 1
The Conceptually Related Content Among the Subscales in the Present Meta-Analysis
HMI ADMI MRNS CMNI GRCS
No directly comparable subscale No directly comparable subscale No directly comparable subscale Disdain for homosexuals Restrictive affectionate behavior between men
No directly comparable subscale No directly comparable subscale Antifemininity No directly comparable subscale No directly comparable subscale
Callous sexual attitudes Sexual identity No directly comparable subscale Playboy No directly comparable subscale
No directly comparable subscale No directly comparable subscale No directly comparable subscale Self-reliance No directly comparable subscale
No directly comparable subscale Antifeminine attitudes Status Power over women
Winning
Pursuit of status
Dominance
Success, power, competition
Need for success and achievement
No directly comparable subscale No directly comparable subscale No directly comparable subscale Primacy of work Conflict between work and family relations
No directly comparable subscale Devaluation of emotion No directly comparable subscale Emotional control Restrictive emotionality
Violence No directly comparable subscale Violence Violence No directly comparable subscale
Danger No directly comparable subscale No directly comparable subscale Risk-taking No directly comparable subscale
Note. We only incorporated the subscales that were available in the data set of the present meta-analysis. For example, HMI has more subscales but only these were available in the studies included in
the present meta-analysis. HMI =Hypermasculinity Inventory; ADMI =Auburn Differential Masculinity Inventory; MRNS =Male Role Norms Scale; CMNI =Conformity to Masculine Norms
Inventory; GRCS =Gender-Role Conflict Scale.
6KRIVOSHCHEKOV, GULEVICH, AND BLAGOV

Our sample of studies contains considerable heterogeneity in
effect size, which we discuss in further detail below. As such, using
the methods above can be problematic, as they have previously
demonstrated poor performance in instances where there is
heterogeneity in effect sizes (Alinaghi & Reed, 2018;Macaskill
et al., 2001;Pustejovsky & Rodgers, 2019).
Results
Characteristics of the Data Set
We identified 44 documents with 107 independent samples and
339 effect sizes, and data from 21,078 respondents. The years of
publication ranged between 1984 and 2021 (the median year was
2013). We found 339 effect sizes (185 were total scores from the
scales and 154—subscales from the scales) for the relationship
between TM and men’s sexism, obtained from 107 independent
samples. The main characteristics of the data set are presented in
Table 2. Studies are described in detail in online Supplemental
Materials. The data and R code are available in the Open Science
Framework at https://osf.io/gb8cp/?view_only=b6069db49e2d43e
f8398656427a92949.
Correlations Between TM and Sexism
The overall correlation between TM and sexism was positive,
Pearson’sr=.307, 95% CI [.278, .334], and significantly different
from zero, t(101) =20.4, p<.001. This result supports Hypothesis 1.
However, we found evidence of considerable heterogeneity in the
effect sizes (prediction interval, PI [.005, .63]; Borenstein et al.,
2017) for the relationship between TM and sexism. This result
indicated that there was greater variability than one would expect
due to sampling error. Such variability might be explained by
differences between studies; therefore, we performed a moderation
analysis for the relationship between self-ascribed TM and sexism.
The results are presented in Table 3.
Correlation with sexism significantly differed between different
forms of TM (Research Question 1), with the strongest between
conformity to masculine norms, Pearson’sr=.388, 95% CI [.342,
.434], followed by TMI, Pearson’sr=.295, 95% CI [.260, .329],
and GRC, Pearson’sr=.28, 95% CI [.218, .339].
Moreover, correlation significantly differed between different
forms of sexism (Research Question 3), old-fashioned: Pearson’s
r=.369, 95% CI [.286, .445]; hostile: Pearson’sr=.308, 95% CI
[.271, .343]; benevolent: Pearson’sr=.247, 95% CI [.219, .273];
modern: Pearson’sr=.449, 95% CI [−.226, .833]; and ambivalent
(general score; Pearson’sr=.484, 95% CI [.027, .773]) forms not
being significantly different from zero.
Finally, we found evidence of significant differences in effect
sizes for the relationship between TM and sexism by characteristics
of the sample (Research Question 4): sample type (students:
Pearson’sr=.281, 95% CI [.253, .309], general population:
Pearson’sr=.41, 95% CI [.335, .479]), sample’s sexual orientation
(straight: Pearson’sr=.368, 95% CI [.295, .436]); mostly straight:
Pearson’sr=.435, 95% CI [.278, .569]; unknown sexual
orientation: Pearson’sr=.283, 95% CI [.254, .310]). The were
no significant differences based on publication status (see Table 3).
We also tested the meta-regression model where multiple
moderators were entered as predictors. As presented in Table 4,
the overall meta-regression model was significant. However, the
only significant predictor of the effect size in this model was the
form of TM, CWB pvalue =.012 (while hostile sexism was
significant on its own, the form of sexism, in general, was not a
significant predictor, CWB pvalue =.21). Thus, only the form of
TM was able to explain a unique proportion of effect-size
heterogeneity in the relationship between TM and men’s sexism
when controlling for other moderators.
Correlations Among the Domains of TM and Sexism
To address the multidimensionality of TM (Research Question
2), we investigated effect sizes for the relationship between
separate domains of TM and sexism. The overall correlation was
positive and significantly different from zero (Pearson’sr=.291,
95% CI [.235, .346]), df =15.8, p<.001) with substantial
heterogeneity (PI [−.057, .657]). As indicated in Table 5, we found
evidence that all but one of the analyzed dimensions of TM (risk-
taking) significantly positively correlated with sexism, and the
formal test indicated that effect sizes were not significantly
different among the dimensions of TM, CWB pvalue =.079. Also,
we found evidence of considerable heterogeneity for all domains
but restrictive emotionality.
Publication Bias
Funnel plot for the effect sizes based on the total scores is
illustrated in Figure 3. Visual inspection of the plots revealed a
certain degree of asymmetry. The Egger’s regression test was
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 2
Description of the Data Set
Characteristic kn
Total 107 339
TM forms
TMI 83 170
CMN 15 125
GRC 12 44
Sexism forms
Benevolence 74 86
Old-fashioned 18 86
Hostility 86 110
Modern 3 53
Ambivalent (total score) 33
Sample type
Students 89 231
General population 19 108
Sample sexual orientation
Exclusively straight 917
Predominantly straight 9 117
Unknown 89 205
Publication status
Published 93 301
Unpublished 14 38
Note. The sum for TM forms and sexism forms does not equal 107
because in some cases authors used the same sample but multiple measures.
The sum for the sample type (students vs. general population) =108 due to
Smiler’s(2006)study using one independent sample but reporting data for
students and general populations separately. k=number of independent
samples; n=number of effect sizes. TM =traditional masculinities; TMI =
traditional masculinity ideology; CMN =conformity to masculine norms;
GRC =gender role conflict.
MASCULINITIES AND MEN’S SEXISM 7

nonsignificant for the relationship between TM and men’s sexism
(b
1
=.494, 95% CI [−.035, 1.023], p=.067).
The intercept for TM, b
0
=.253, 95% CI [.216, .291], p<.001,
was statistically significant at the conventional level in the PET
regression. Therefore, we used the intercept from the PEESE
regression as the estimate of the true effect sizes. The intercept in the
PEESE regression was significantly different from zero (b
0
=.274,
95% CI [.249, .298], p<.001). Compared to the original estimate
from RVE meta-regression, the estimate was somewhat smaller than
the original effect size (by .04).
Funnel plots for the nine separate domains of TM based on the
scores from subscales are illustrated in Figure 4. Visual inspection
of the plots revealed the noticeable asymmetry for the three
domains: status, violence, and risk-taking. As indicated in Table 6,
Egger’s regression test, however, was nonsignificant for all
domains.
The PET intercepts for antifemininity, self-reliance, playboy, and
risk-taking were nonsignificant (see Table 6). Therefore, we used
them as the estimates of the true effects with the understanding that
they are statistically indistinguishable from zero. Compared to the
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 3
Effect Sizes for the Relationship Between TM and Sexism by Moderators
Moderator ES 95% CI PI df p
Moderation
statistic df p I
2
TM forms CWB .006 81.99
Traditional masculinity
ideology
.304 [.266, .342] [−.007, .61] 76.7 <.001
Conformity to masculine norms .410 [.356, .465] [.01, .72] 13.4 <.001
Gender role conflict .288 [.222, .353] [−.02, .60] 10.4 <.001
Sexism forms CWB <.001 79.57
Benevolence .252 [222, 281] [−.03, .54] 63.14 <.001
Old-fashioned .387 [.294, .479] [.10, .67] 15.71 <.001
Hostility .318 [.278, .358] [.03, .60] 58.39 <.001
Modern .484 [−.230, 1.198] [.20, .77] 1.64 .09
Ambivalent sexism (total scale) .528 [.027, 1.028] [.24, .81] 1.98 .045
Sample type t=3.35 26.2 .002 80.43
Students .289 [.259, .319] [−.002, .58] 82.1 <.001
General population .436 [.349, .522] [.14, .73] 17.5 <.001
Sample’s sexual orientation CWB .002 81.20
Exclusively straight .386 [.304, .467] [.09, .69] 8.91 <.001
Predominantly straight .466 [.285, .646] [.17, .77] 7.90 <.001
Unknown .291 [.260, .321] [−.008, .59] 82.18 <.001
Publication status t=1.68 17.1 .112 81.94
Published .305 [.273, .337] [−.004, .61] 86.8 <.001
Unpublished .391 [.286, .496] [.09, .70] 12.9 <.001
Note.TM=traditional masculinities; ES =effect size (Fisher’sz); PI =prediction interval, a range into which we can expect the effects of future studies
to fall based on present evidence; CI =confidential internal; CWB =cluster wild bootstrapping; df =Satterthwaite degrees of freedom (if the Satterthwaite
degrees of freedom are less than 4, the Type I error rates can be tremendously larger than .05, and, therefore, pvalue should not be trusted); I
2
=ratio of
true heterogeneity to total variance across the observed effect sizes.
Table 4
The Meta-Regression Model Predicting Effect Sizes for the Relationship Between TM and Sexism
Variable b95% CI tdfp
Intercept
Masculinity form (GRC) −.175 [−.292, −.059] −3.13 20.88 .005
Masculinity form (TMI) −.039 [−.155, .077] −.71 17.71 .49
Form of sexism (old-fashioned) .066 [−.087, .219] .89 22.90 .38
Form of sexism (hostile) .041 [.007, .074] 2.44 71.84 .02
Form of sexism (modern) .156 [−.026, .578] 1.42 2.27 .28
Form of sexism (ASI) .201 [−.07, .474] 2.12 3.72 .11
Sample type (general population) .098 [−.009, .205] 1.89 23.24 .07
Sample sexual orientation
(Predominantly straight)
.073 [−.096, .242] 0.93 13.39 .37
Sample sexual orientation (unknown) −.042 [−.184, .098] −.64 15.35 .53
Sample size .00003 [−.0001, .0002] .46 4.28 .67
Publication status (unpublished) .102 [−.008, .212] 1.96 15.88 .07
Model parameters CWB pvalue =.024, I
2
=75.78
Note. df =degrees of freedom (if the Satterthwaite degrees of freedom are less than 4, the Type I error rates can be
tremendously larger than .05, and, therefore, pvalue should not be trusted). TM =traditional masculinities; CI =
confidential internal; GRC =gender role conflict; TMI =traditional masculinity ideology; ASI =ambivalent sexism
inventory total score; CWB =cluster wild bootstrapping.
8KRIVOSHCHEKOV, GULEVICH, AND BLAGOV

original estimates from RVE meta-regressions, all estimates except
for risk-taking were much smaller than the original effect sizes.
The PET intercepts for heterosexual self-presentation, status,
restrictive emotionality, primacy of work, and violence were
significantly different from zero (see Table 6). Therefore, we used an
intercept from the PEESE regression as the estimates of the true
effect sizes. The intercept in the PEESE regression was significantly
different from zero for heterosexual self-presentation, status,
restrictive emotionality, primacy of work, and violence. Compared
to the original estimates from RVE meta-regressions, estimates
for heterosexual self-presentation (larger by .004), status (larger by
.003), and restrictive emotionality (lower by .012) were similar to
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 5
Effect Sizes for the Relationship Between TM and Sexism by Domains
Domain ES 95% CI PI df p n
Heterosexual self-presentation .281 [.168, .394] [.16, .40] 4.21 .002 14
Antifemininity .377 [.188, .565] [.04, .71] 3.99 .005 7
Self-reliance .186 [.186, .186] [.186, .186] 1<.001 6
Status .283 [.192, .375] [.04, .53] 11.6 <.001 48
Restrictive emotionality .208 [.152, .264] [.15, .27] 3.75 <.001 17
Primacy of work .147 [.077, .218] [.08, .22] 3.32 .006 16
Violence .310 [.219, .400] [.09, .53] 6.8 <.001 18
Playboy .329 [.118, .541] [−.08, .74] 4.95 .01 13
Risk-taking .230 [−.100, .561] [.04, .42] 1.9 .09 10
Note. Scales were combined into the domains in the following way based on Table 1: Heterosex ual self-
presentation =“heterosexual self-presentation”and “Restrictive Affectionate Behavior Between Men”;
antifemininity =“antifemininity”and “Avoidance of Femininity”; self-reliance =“self-reliance”and “Facade/
Counterdependence”;status=“status,”“winning,”“pursuit of status,”“success, power, and competition,”and
“Power Over Women;”restrictive emotionality =“restrictive emotionality”and “Emotional Control”;primacyof
work =“primacy of work”and “Conflict Between Work and Family”; violence =“violence”and “Toughness”;
playboy =“playboy”; risk-taking =“risk-taking.”ES =effect size (Fisher’sz); n=number of effect sizes; PI =
prediction interval, a range into which we can expect the effects of future studies to fall based on present
evidence; df =Satterthwaite degrees of freedom (if the Satterthwaite degrees of freedom are less than 4, the Type
I error rates can be tremendously larger than .05, and, therefore, pvalue should not be trusted); I
2
=ratio of true
heterogeneity to total variance across the observed effect sizes. TM =traditional masculinities; CI =confidential
internal.
Figure 3
The Funnel Plot for the Total Score
MASCULINITIES AND MEN’S SEXISM 9

the original effect sizes, in contrast to estimates for work (larger by
.026) and violence (larger by .069).
Discussion
This present research reports findings from the meta-analysis of
the relationship between TM and men’s sexism against women. We
quantified the link and examined whether there were differences
in the correlations due to the form of TM (TMI, conformity to
masculine norms, and gender role stress), the domain of TM,
the form of sexism (old-fashioned vs. modern vs. hostile vs.
benevolent), and sample characteristics.
Relationship Between TM and Sexism
Together these results provide important insights into the
relationship between TM and men’s sexism against women.
Overall, TM was positively related to sexism in male samples
(r=.307). This means that men who endorse traditional
masculinities to a greater extent—that is, that men should be
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Figure 4
Funnel Plots for the Domains of Traditional Masculinities
Table 6
Egger’s Test and PET–PEESE Technique Among TM Domains
Domain
Egger’s test PET PEESE
Estimate [95% CI]
Heterosexual self-presentation .53 [−1.745, 2.815] 259*[.054, .465] .285*** [.168, .403]
Antifemininity 10.57 [−.929, 22.074] −.40 [−.929, 22.074] —
Self-reliance 2.83 [−.642, 6.305] −.125 [−.642, 6.305] —
Status .86 [−.352, 2.074] .228*** [.138, .320] .258*** [.193, .322]
Restrictive emotionality .511 [−.357, 1.380] .186*** [.137, .236] 196*** [.163, .23]
Primacy of work −.15 [−1.081, .773] .179*** [.126, .232] .173*** [.135, .211]
Violence −2.46 [−5.637, .724] .479*** [.219, .739] .379*** [.246, .513]
Playboy 1.11 [−2.671, 4.894] .179 [−2.671, 4.894] —
Risk-taking −.75 [−4.192, 2.681] .226 [−4.192, 2.681] —
Note. PET =Precision Effect Test; PEESE =Precision Effect Estimate with Standard Errors; TM =traditional
masculinities; CI =confidential internal.
*p<.05. *** p<.001.
10 KRIVOSHCHEKOV, GULEVICH, AND BLAGOV

different from women and occupy a dominant place in the social
hierarchy—are more supportive of beliefs that limit women’s
opportunities in the public sphere than men who endorse traditional
masculinities to a lesser extent.
At the same time, we observed the evidence for a conceptual
difference between TM and sexism. There is a weaker association
between TM and sexism than between TM and endorsement of
violent attitudes toward women (r=.347; Krivoshchekov et al.,
2023) and could be considered a moderate association in the
context of the broader social–psychological literature (Lovakov &
Agadullina, 2021). This result is in line with the proponents of
the conceptual differences between individual forms of TM and
sexism (e.g., Schwartz et al., 2016).
The content of TM targets women to a certain extent and implies
an ambiguous attitude toward them. The desire to be different from
women or to dominate others (including women) is not equivalent
to a negative attitude toward women. At the same time, the desire
to engage in romantic and/or sexual relationships with women
does not equal a positive attitude. This is likely why the link
between traditional masculinity and sexism varies based on
additional conditions.
Sexism was more strongly associated with TMI and conformity to
masculine norms than it was with gender role stress. These findings
are consistent with the results of a previous meta-analysis, according
to which, compared to GRC, TMI, and conformity to masculine
norms were more strongly associated with attitudes toward violence
against women (Krivoshchekov et al., 2023). One explanation
could be that the content of Gender-Role Conflict Scales correspond
less to scales of sexism than scales of TMI and conformity to
masculine norms.
Further, TM were more strongly associated with forms of sexism
that reflect overtly negative attitudes toward women (old-fashioned
and hostile sexisms), than with forms of sexism that reflect more
hidden beliefs (e.g., benevolent sexism). Perhaps this happens
because TM imply that feminine characteristics are less valuable
than masculine ones, and in romantic relationships, a woman is
perceived as a trophy that should be won than a “pure”being who
needs love and care.
The differences were also evident in the sample characteristics.
TM was more strongly associated with sexism in the general
population than in students. As such, one has to be careful when
trying to generalize the patterns found in student samples to the
general population of the country. In addition, TM were more
strongly associated with sexism in predominantly or exclusively
straight samples than in samples with unknown sexual orientation.
However, this finding should be situated in the fact that most of
the research included in the present meta-analysis was done on
men whose sexual orientation was unknown (i.e., not measured
or reported).
Among the domains, we observed that the relationship between
TM and sexism ranged from .147 (primacy of work) to .377
(antifemininity), but there were no significant differences between
them. On the one hand, these results speak in favor of the fact that
different domains of TM could be considered elements of a single
construct. On the other hand, the varying size and statistical
significance of these associations, as well as the relatively small
number of studies, cast doubt on this conclusion.
Finally, the study-to-study variation in true effect sizes was
considerable, as evidenced by the prediction intervals, and we had
only limited success in identifying the possible sources for this
heterogeneity. Only the form of TM was statistically significant.
The metaregression model that aimed to reduce potential difficulties
caused by confounding moderators also was not successful in
the explanation of the variability in the effect sizes. These results
imply the existence of unidentified sources of variation in these
correlations across studies and suggest that researchers should
attempt to identify other contributing variables in the future. Thus,
although one might expect a positive correlation between TM
and sexism against women in samples of men, it is hard to conclude
how strongly these phenomena are related to each other.
Publication Bias
Current methods for detecting publication bias are still in
development. As such, we used three types of tests to detect whether
there was evidence of publication bias present in our research
sample. We used publication status as a moderator, a funnel plot
along with Egger’s regression of funnel plot symmetry, and a
PET–PEESE technique. All three analyses did not indicate
considerable publication bias. We should note, however, that these
methods have previously demonstrated poor performance when
there is heterogeneity in effect sizes (Alinaghi & Reed, 2018;
Macaskill et al., 2001;Pustejovsky & Rodgers, 2019).
When controlling for publication bias, overall effect sizes became
smaller and sometimes did not significantly differ from zero,
especially among different domains of TM. Such results often
suggest that entire studies have gone unpublished or unsupportive
findings have been omitted from published reports. At the same
time, the present set of studies is not characterized by an
overabundance of barely significant results and different methods
yielded conflicting results. This implies that if there is a publication
bias in this meta-analysis, it probably would not substantively alter
our interpretation of the presence and direction of the relationship
between traditional masculinities and men’s sexism against women.
Limitations and Future Directions
As with any meta-analysis, our confidence in the conclusions is
limited by the quality of available data. Therefore, it is crucial to
place the results of the present meta-analysis in context so that
they can be interpreted correctly. Below we highlight the main
limitations of studies included in this meta-analysis addressing
the relationship between TM and sexism and conclude with the
limitations of the present meta-analysis itself.
First, TM and sexism are complex phenomena that manifest
themselves at different levels, from the attitudes and behavior of
individuals to the attitudes and social practices that exist at the level
of society and are embodied in social–educational, economic, and
political institutions (e.g., see Lewis, 2018, for sexism; see Wong &
Wang, 2022, for masculinities). However, most psychological
research looks at the relationship between individual-level
traditional masculinities and sexism. A possible line of future
research is to analyze the relationship between situational
masculinities and sexism (Wong & Wang, 2022). One might
consider how people with different levels of acceptance of TMI
and conformity to masculine norms react to temporary influence
that either suppresses, strengthens, or threatens their masculinity,
including under what conditions they display stronger sexism.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
MASCULINITIES AND MEN’S SEXISM 11

Second, scholars note that traditional masculinities can have
different content. There are multiple, dynamic meanings associated
with men that reflect individual constructions of masculinities
and a society’s culture (Wong & Wang, 2022). This is supported by
studies that demonstrate sexuality, ethnicity, and national differ-
ences in content (E. H. Thompson & Bennett, 2015) and
endorsement (Krivoshchekov et al., 2021;Lease et al., 2013;
R. F. Levant, 2011;R. Levant & Richmond, 2007) of masculinities.
However, most of the studies reviewed in the present meta-analysis
were conducted in the United States and on samples of unknown
sexual orientation. The field of psychology of men and masculinities
would benefit from a closer examination of how endorsement of TM
among gay, bisexual, and transgender men related to sexism against
women. Scholars argue that gay, bisexual, and transgender men may
endorse masculinities to a different extent, their masculinities might
have different content, and motivation to be masculine might differ
(e.g., Sánchez, 2016). Future research with a focus on non-WEIRD
(Western, educated, industrial, rich, and democratic) countries and
gay, bisexual, and transgender men will enrich the understanding of
variability in masculinities and whether the findings are universal.
Third, in the meta-analysis, we considered three forms of TM.
In the course of the analysis, it became evident that researchers
more often measured the endorsement of TMI than conformity to
masculine norms and gender role stress. In addition, the meta-
analysis has shown variability in the associations between different
domains and sexism, although these differences did not reach
statistical significance. In many studies, only the total TM score was
calculated. This reduces the reliability of the conclusions about the
relationship of various forms and domains of TM with sexism. This
meta-analysis showed no significant differences between different
domains of TM and sexism. At the same time, previous meta-
analyses have shown that different domains of TM are differently
associated with health-related outcomes (Wong, Ho, et al., 2017)
and attitudes toward violence against women (Krivoshchekov
et al., 2023). As such, it is warranted to conduct more studies
that will examine separate domains of TM to better understand the
multidimensional nature of TM.
Fourth, according to a broad definition, sexism includes cognitive
and behavioral indicators. We observed that psychologists tend
to concentrate on cognitive indicators. At the same time, there are
practically no studies in the scientific literature that examine
the relationship between the subjective endorsement of TM and
behavioral manifestations of sexism in the economic (e.g., in
organizations) or political (e.g., in elections) domains. Future
research should focus on the analysis of the relationship between
masculinities and behavioral manifestations of sexism in various
areas of public life. For instance, it can be assumed that TM in men is
related to how they interact with men and women who work with
them in the same organization, run for local or federal elections, and
are active within the same Internet community. These studies will
reveal how and under what conditions self-ascribed and other-
ascribed masculinities facilitate or hinter existing social practices.
Fifth, we found a relatively small number of studies looking at the
relationship between TM and sexism. Therefore, we were unable to
analyze the interaction between different forms of TM and sexism.
For example, it can be assumed that domains that reflect the
differences between men and women would be more strongly
associated with explicit negative attitudes toward women, and
domains that reflect the belief that men should occupy a high
position in the social system—with more hidden forms of sexism.
We encourage authors to continue the examination of the
correlations between different forms and domains of TM and
sexism, especially using larger samples to increase the power of the
studies.
Finally, while traditional masculinities have a lot of negative
consequences, scholars call to pay more attention to the positive
aspects of masculinities (Cole et al., 2021). Similar to violence
against women (Pérez-Martínez et al., 2023), future research might
focus on what are the positive aspects of masculinities that could be
used to reduce sexism against women in men.
Practical Implications
On a practical level, our findings demonstrate that in order to
tackle sexism, meanings attached to masculinities should be
addressed to empower men to live lives less constrained by gender
role norms and not only the harmful impacts of sexism on women.
Interventions that combine content on both sexism and traditional
masculinities might yield the most effective results in reducing
sexism in men.
Further, present findings mean that interventions aimed to
reduce sexism against women in men need to tackle different
forms of traditional masculinities, that is, what a “real”man
should be like, the extent to which men conform to masculine
norms, and the extent to which men feel stressed because of
conformity to masculine norms. For instance, educational
interventions to reduce men’s sexism might focus on a gender-
transformative approach to questioning meanings attached to
traditional masculinities and the way men and boys are forced to
conform to them.
While present findings are limited to individual-level masculi-
nities and sexism, they might inform practitioners at the community
level. Our results show that students might not be representative, and
practitioners should be cautious when using the same interventions
with a broader audience, especially with men who are underrepre-
sented in the research (e.g., low-income, gay, bisexual, and
transgender men and men from non-WEIRD countries).
Conclusion
The present meta-analysis aimed to quantify the relationship
between TM and men’s sexism against women. We found evidence
that the observed correlations between TM and sexism were
significant and positive. Nevertheless, we could not conclude the
strength of these relationships due to the substantial heterogeneity of
effect sizes. We strongly recommend researchers use larger samples
in future research to increase the power of their studies and follow
open practices to reduce publication bias.
References
References marked with an asterisk indicate studies included in the
meta-analysis.
Agadullina, E., Lovakov, A., Balezina, M., & Gulevich, O. A. (2022).
Ambivalent sexism and violence toward women: A meta-analysis.
European Journal of Social Psychology,52(5–6), 819–859. https://
doi.org/10.1002/ejsp.2855
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
12 KRIVOSHCHEKOV, GULEVICH, AND BLAGOV

Alinaghi, N., & Reed, W. R. (2018). Meta-analysis and publication bias:
How well does the FAT-PET-PEESE procedure work? Research Synthesis
Methods,9(2), 285–311. https://doi.org/10.1002/jrsm.1298
*Amayreh, W. M. (2019). The role of social dominance orientation,
acculturation, and gender roles on self-reported sexual aggression in
ethnic minority college student men [Doctoral dissertation, The University
of Texas at Austin]. University of Texas Libraries.
*Archer, J. (2010). Derivation and assessment of a hypermasculine values
questionnaire. British Journal of Social Psychology,49(3), 525–551.
https://doi.org/10.1348/014466609X471525
*Archer, J., & Rhodes, C. (1989). The relationship between gender-related
traits and attitudes. British Journal of Social Psychology,28(2), 149–157.
https://doi.org/10.1111/j.2044-8309.1989.tb00856.x
*Barron, J. M., Struckman-Johnson, C., Quevillon, R., & Banka, S. R.
(2008). Heterosexual men’s attitudes toward gay men: A hierarchical
model including masculinity, openness, and theoretical explanations.
Psychology of Men & Masculinity,9(3), 154–166. https://doi.org/10.1037/
1524-9220.9.3.154
Borenstein, M., Higgins, J. P. T., Hedges, L. V., & Rothstein, H. R. (2017).
Basics of meta-analysis: I
2
is not an absolute measure of heterogeneity.
Research Synthesis Methods,8(1), 5–18. https://doi.org/10.1002/jrsm.
1230
*Bosson, J. K., Jurek, P., Vandello, J. A., Kosakowska-Berezecka, N.,
Olech, M., Besta, T., Bender, M., Hoorens, V., Becker, M., Timur
Sevincer, A., Best, D. L., Safdar, S., Włodarczyk, A., Zawisza, M.,
Żadkowska, M., Abuhamdeh, S., Badu Agyemang, C., Akbaş, G.,
Albayrak-Aydemir, N., …Žukauskienė, R. (2021). Psychometric
properties and correlates of precarious manhood beliefs in 62 nations.
Journal of Cross-Cultural Psychology,52(3), 231–258. https://doi.org/10
.1177/0022022121997997
Brandt, M. J. (2011). Sexism and gender inequality across 57 societies.
Psychological Science,22(11), 1413–1418. https://doi.org/10.1177/0956
797611420445
Brannon, R., & Juni, S. (1984). A scale for measuring attitudes about
masculinity. Psychological Documents,14(1). (Document No. 2612).
*Burk, L. R., Burkhart, B. R., & Sikorski, J. F. (2004). Construction and
preliminary validation of the auburn differential masculinity inventory.
Psychology of Men & Masculinity,5(1), 4–17. https://doi.org/10.1037/
1524-9220.5.1.4
*Caldwell, D. (1992). The impact of a treatment program for male batterers
on their sex role attitudes and abusive behaviors [Doctoral dissertation,
University of Rhode Island], ProQuest Dissertations & Theses Global.
*Cecil, A. G. (1996). Comparing male cadet attitudes for masculinity
ideology within civilian and military college environments [Doctoral
dissertation, Arizona State University], ProQuest Dissertations & Theses
Global.
Check, J. (1984). Hostility toward women scale [Doctoral dissertation,
University of Manitoba]. University of Manitoba Library. http://hdl.ha
ndle.net/1993/23279
Chu, J. Y., Porche, M. V., & Tolman, D. L. (2005). The adolescent
masculinity ideology in relationships scale. Men and Masculinities,8(1),
93–115. https://doi.org/10.1177/1097184X03257453
*Cole, B. P., Brennan, M., Tyler, E., & Willard, R. (2020). Predicting men’s
acceptance of sexual violence myths through conformity to masculine
norms, sexism, and “locker room talk”.Psychology of Men & Masculinity,
21(4), 508–517. https://doi.org/10.1037/men0000248
Cole,B.P.,Moffitt-Carney, K., Patterson, T. P., & Willard, R.
(2021). Psychology of Men and Masculinities’focus on positive
aspects of men’s functioning: A content analysis and call to action.
Psychology of Men & Masculinity,22(1), 39–47. https://doi.org/10
.1037/men0000264
Connor, R. A., Glick, P., & Fiske, S. T. (2016). Ambivalent sexism in
the twenty-first century. In C. G. Sibley & F. K. Barlow (Eds.),
The Cambridge handbook of the psychology of prejudice (1st ed.,
pp. 295–320). Cambridge University Press. https://doi.org/10.1017/
9781316161579.013
*Corprew, C. S., III, & Mitchell, A. D. (2014). Keeping it frat: Exploring the
interaction among fraternity membership, disinhibition, and hypermascu-
linity on sexually aggressive attitudes in college-aged males. Journal of
College Student Development,55(6), 548–562. https://doi.org/10.1353/
csd.2014.0062
*Corprew, C. S., III, Matthews, J. S., & Mitchell, A. D. (2014). Men at the
crossroads: A profile analysis of hypermasculinity in emerging adulthood.
Journal of Men’s Studies,22(2), 105–121. https://doi.org/10.3149/jms
.2202.105
*Covell, A. (1998). Characteristics of college males who are likely
to sexually harass women: A test of a mediated model (Publication
No. 9930486) [Doctoral dissertation, University of Southern California].
ProQuest Dissertations Publishing.
*Danube, C. L., Vescio, T. K., & Davis, K. C. (2014). Male role norm
endorsement and sexism predict heterosexual college men’s attitudes
toward casual sex, intoxicated sexual contact, and casual sex. Sex Roles,
71(5), 219–232. https://doi.org/10.1007/s11199-014-0394-4
Dasgupta, N., & Rivera, L. M. (2006). From automatic antigay prejudice to
behavior: The moderating role of conscious beliefs about gender and
behavioral control. Journal of Personality and Social Psychology,91(2),
268–280. https://doi.org/10.1037/0022-3514.91.2.268
*Davis, T. L. (1997). The effectiveness of a sex role socialization-focused
date rape prevention program in reducing rape-supportive attitudes in
college fraternity men (Publication No. 9731784) [Doctoral Dissertation,
The University of Iowa]. ProQuest Dissertations & Theses Global: The
Humanities and Social Sciences Collection.
*Deady, R. (2017). Funny men: The relationship between humor, sexism,
male gender role, and life satisfaction [Doctoral dissertation, The Wright
Institute]. ProQuest Dissertations & Theses Global.
*de Zavala, A. G., & Bierwiaczonek, K. (2021). Male, national, and religious
collective narcissism predict sexism. Sex Roles,84(11–12), 680–700.
https://doi.org/10.1007/s11199-020-01193-3
*Diener, J. M. (2016). Individual variables related to rape myth acceptance
and bystander intervention: Implications for sexual assault prevention
[Master’s thesis, Western Carolina University], ProQuest Dissertations &
Theses Global.
*Edwards, S. R., Bradshaw, K. A., & Hinsz, V. B. (2014). Denying rape but
endorsing forceful intercourse: Exploring differences among responders.
Violence and Gender,1(4), 188–193. https://doi.org/10.1089/vio.2014.
0022
Eisler, R. M., & Skidmore, J. R. (1987). Masculine gender role stress. Scale
development and component factors in the appraisal of stressful situations.
Behavior Modification,11(2), 123–136. https://doi.org/10.1177/0145445
5870112001
Fisher, Z., & Tipton, E. (2015). robumeta: An R-package for robust variance
estimation in meta-analysis. ArXiv.
*Gage, A. N., & Lease, S. H. (2021). An exploration of the link between
masculinity and endorsement of IPV myths in American men. Journal of
Interpersonal Violence,36(13–14), 6145–6165. https://doi.org/10.1177/
0886260518818430
*Gallagher, K. E., & Parrott, D. J. (2011). What accounts for men’s hostile
attitudes toward women? The influence of hegemonic male role norms and
masculine gender role stress. Violence Against Women,17(5), 568–583.
https://doi.org/10.1177/1077801211407296
Gerdes, Z. T., Alto, K. M., Jadaszewski, S., D’Auria, F., & Levant, R. F.
(2018). A content analysis of research on masculinity ideologies using all
forms of the Male Role Norms Inventory (MRNI). Psychology of Men &
Masculinity,19(4), 584–599. https://doi.org/10.1037/men0000134
Glick, P., & Fiske, S. (1996). The Ambivalent Sexism Inventory:
Differentiating hostile and benevolent sexism. Journal of Personality
and Social Psychology,70(3), 491–512. https://doi.org/10.1037/0022-
3514.70.3.491
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
MASCULINITIES AND MEN’S SEXISM 13

Glick, P., & Fiske, S. T. (2001). An ambivalent alliance. Hostile and
benevolent sexism as complementary justifications for gender inequality.
American Psychologist,56(2), 109–118. https://doi.org/10.1037/0003-
066X.56.2.109
Glick, P., Fiske, S. T., Mladinic, A., Saiz, J. L., Abrams, D., Masser, B.,
Adetoun, B., Osagie, J. E., Akande, A., Alao, A., Brunner, A., Willemsen,
T. M., Chipeta, K., Dardenne, B., Dijksterhuis, A., Wigboldus, D., Eckes,
T., Six-Materna, I., Expo´sito, F., …Lo´pez, W. (2000). Beyond prejudice
as simple antipathy: Hostile and benevolent sexism across cultures.
Journal of Personality and Social Psychology,79(5), 763–775. https://
doi.org/10.1037/0022-3514.79.5.763
*Grose, R. G., Grabe, S., & Kohfeldt, D. (2014). Sexual education, gender
ideology, and youth sexual empowerment. Journal of Sex Research,51(7),
742–753. https://doi.org/10.1080/00224499.2013.809511
*Hamburger, M. E., Hogben, M., McGowan, S., & Dawson, L. J. (1996).
Assessing hypergender ideologies: Development and initialvalidation of a
gender-neutral measure of adherence to extreme gender-role beliefs.
Journal of Research in Personality,30(2), 157–178. https://doi.org/10
.1006/jrpe.1996.0011
*Harrington, A. G., Overall, N. C., & Cross, E. J. (2021). Masculine gender
role stress, low relationship power, and aggression toward intimate
partners. Psychology of Men & Masculinity,22(1), 48–62. https://doi.org/
10.1037/men0000262
Hedges, L. V., Tipton, E., & Johnson, M. C. (2010). Robust variance
estimation in meta-regression with dependent effect size estimates.
Research Synthesis Methods,1(1), 39–65. https://doi.org/10.1002/jrsm.5
*Hyatt, C. S., Berke, D. S., Miller, J. D., & Zeichner, A. (2017).
Do beliefs about gender roles moderate the relationship between
exposure to misogynistic song lyrics and men’s female-directed
aggression? Aggressive Behavior,43(2), 123–132. https://doi.org/10
.1002/ab.21668
*Jacobs, J. R. (1996). Psychological and demographic correlates of men’s
perceptions of and attitudes toward sexual harassment [Doctoral
dissertation, University of Southern California]. ProQuest Dissertations
& Theses Global.
Joshi, M., & Pustejovsky, J. (2022). Wildmeta: Cluster wild bootstrapping
for meta-analysis (R package Version 0.1.0). https://cran.r-project.org/we
b/packages/wildmeta/
Joshi, M., Pustejovsky, J. E., & Beretvas, S. N. (2022). Cluster wild
bootstrapping to handle dependent effect sizes in meta-analysis with a
small number of studies. Research Synthesis Methods,13(4), 457–477.
https://doi.org/10.1002/jrsm.1554
*Kilianski, S. E. (2003). Explaining heterosexual men’s attitudes toward
women and gay men: The theory of exclusively masculine identity.
Psychology of Men & Masculinity,4(1), 37–56. https://doi.org/10.1037/
1524-9220.4.1.37
*Klann, E. M., Wong, Y. J., & Rydell, R. J. (2018). Firm father figures:
A moderated mediation model of perceived authoritarianism and the
intergenerational transmission of gender messages from fathers to sons.
Journal of Counseling Psychology,65(4), 500–511. https://doi.org/10
.1037/cou0000296
Komlenac, N., Eggenberger, L., Walther, A., Maresch, F., Lamp, E., &
Hochleitner, M. (2023). Measurement invariance and psychometric
properties of a German-language Conformity to Masculine Norms
Inventory among cisgender sexual minority and heterosexually identified
women and men. Psychology of Men & Masculinity. Advance online
publication. https://doi.org/10.1037/men0000429
Komlenac, N., & Hochleitner, M. (2022). Internal structure and invariance
analyses across gender of the German-language version of the Male Role
Norm Inventory-Short Form. Psychology of Men & Masculinity,23(1),
86–98. https://doi.org/10.1037/men0000356
*Koralewski, M. A., & Conger, J. C. (1992). The assessment of social skills
among sexually coercive college males. Journal of Sex Research,29(2),
169–188. https://doi.org/10.1080/00224499209551641
Krivoshchekov, V., & Gulevich, O. (2023). Meta-analysis of the relationship
between traditional masculinity and sexism against women.https://osf
.io/gb8cp/
Krivoshchekov, V., Gulevich, O., & Blagov, I.(2023). Traditional masculinity
and male violence against women: A meta-analytic examination.
Psychology of Men & Masculinity. Advance online publication. https://
doi.org/10.1037/men0000426
Krivoshchekov, V., Gulevich, O., & Ostroverkhova, M. (2022). The
Conformity to Masculine Norms Inventory-30: Validity and measurement
invariance of a Russian-language version. Psychology of Men &
Masculinity,23(1), 59–73. https://doi.org/10.1037/men0000379
Krivoshchekov, V., Gulevich, O., & Sorokina, A. (2021). Russian version of
the Male Role Norms Inventory-Short Form: Structure, validity, and
measurement invariance. Psychology of Men & Masculinity,22(4),
821–837. https://doi.org/10.1037/men0000346
*Larrañaga, E., Yubero, S., Ovejero, A., & Navarro, R. (2013). Validacio´n
de la versio´n española de la Gender Role Conflict Scale for Adolescents
(GRCS-A). Anales de Psicología,29(2), 549–557. https://doi.org/10
.6018/analesps.29.2.143341
*Leaper, C., & Van, S. R. (2008). Masculinity ideology, covert sexism,
and perceived gender typicality in relation to young men’s academic
motivation and choices in college. Psychology of Men & Masculinity,
9(3), 139–153. https://doi.org/10.1037/1524-9220.9.3.139
Lease, S. H., Montes, S. H., Baggett, L. R., Sawyer, R. J., II, Fleming-
Norwood, K. M., Hampton, A. B., Ovrebo, E., Çiftçi, A., & Boyraz, G.
(2013). A cross-cultural exploration of masculinity and relationships in
men from Turkey, Norway, and the United States. Journal of Cross-
Cultural Psychology,44(1), 84–105. https://doi.org/10.1177/002202211
1432293
*Lease, S. H., Shuman, W. A., & Gage, A. N. (2020). Female and male
coworkers: Masculinity, sexism, and interpersonal competence at work.
Psychology of Men & Masculinity,21(1), 139–147. https://doi.org/10
.1037/men0000218
*Leone, R. M., Schipani-McLaughlin, A. M., Haikalis, M., & Parrott, D. J.
(2020). The “White Knight”effect: Benevolent sexism accounts for
bystander intervention in party situations among high status men.
Psychology of Men & Masculinity,21(4), 704–709. https://doi.org/10
.1037/men0000314
Levant, R., & Richmond, K. (2007). A review of research on masculinity
ideologies using the Male Role Norms Inventory. Journal of Men’s
Studies,15(2), 130–146. https://doi.org/10.3149/jms.1502.130
Levant, R. F. (2011). Research in the psychology of men and masculinity
using the gender role strain paradigm as a framework. American
Psychologist,66(8), 765–776. https://doi.org/10.1037/a0025034
Levant, R. F., Hall, R. J., Weigold, I. K., & McCurdy, E. R. (2016). Construct
validity evidence for the Male Role Norms Inventory-Short Form: A
structural equation modeling approach using the bifactor model. Journal
of Counseling Psychology,63(5), 534–542. https://doi.org/10.1037/cou
0000171
Levant, R. F., McDermott, R., Parent, M. C., Alshabani, N., Mahalik, J. R., &
Hammer, J. H. (2020). Development and evaluation of a new short form of
the Conformity to Masculine Norms Inventory (CMNI-30). Journal
of Counseling Psychology,67(5), 622–636. https://doi.org/10.1037/cou
0000414
Levant, R. F., Rankin, T. J., Williams, C. M., Hasan, N. T., & Smalley, K. B.
(2010). Evaluation of the Factor Structure and Construct Validity of
Scores on the Male Role Norms Inventory-Revised (MRNI-R).
Psychology of Men & Masculinity,11(1), 25–37. https://doi.org/10
.1037/a0017637
Lewis, J. A. (2018). From modern sexism to gender microaggressions:
Understanding contemporary forms of sexism and their influence on
diverse women. In C. B. Travis, J. W. White, A. Rutherford, W. S.
Williams, S. L. Cook, & K. F. Wyche (Eds.), APA handbook of the
psychology of women: History, theory, and battlegrounds (pp. 381–397).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
14 KRIVOSHCHEKOV, GULEVICH, AND BLAGOV