Accepted Article

PROF. KIBEOM LEE (Orcid ID : 0000-0003-2775-5596)

PROF. MICHAEL C ASHTON (Orcid ID : 0000-0003-4711-8199)

Article type : Original Manuscript

RUNNING HEAD: Sex Differences in HEXACO Personality

Sex Differences in HEXACO Personality Characteristics Across Countries and Ethnicities

Kibeom Lee
University of Calgary
Michael C. Ashton

Brock University

Author Notes

This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1111/JOPY.12551
This article is protected by copyright. All rights reserved
 Kibeom Lee, Department of Psychology, University of Calgary, Calgary, Alberta, Canada;
Accepted Article
Michael C. Ashton, Department of Psychology, Brock University, St. Catharines, Ontario, Canada.
Correspondence should be addressed to Kibeom Lee, Department of Psychology, University
of Calgary, Calgary, AB T2N 1N4, Canada e-mail: kibeom@[Link], phone: 1-403-210-9469, fax:
1-403-210-8249 or to Michael C. Ashton, Department of Psychology, Brock University, St.
Catharines, ON L2S 3A1 Canada e-mail: mashton@[Link].

This article is protected by copyright. All rights reserved

 Abstract
Accepted Article
Objective: We examined sex differences in the HEXACO Personality Inventory—Revised
(HEXACO-PI-R) factor- and facet-level scales and the associations of national sex differences in
those scales with national characteristics such as wealth and gender equality. Method: HEXACO-
PI-R self-reports were collected online from persons in 48 countries (N=347,192). Results: (1)
Women averaged substantially higher than men in Emotionality and in Honesty-Humility, with
(sample-unweighted) mean differences across countries of d = 0.84 and d = 0.37, respectively; (2) the
HEXACO-PI-R factor scales showed a rather large multivariate sex difference (D > 1 in most
countries), about 16% larger than found in similar samples with the Big Five personality factors, (3)
some facet scales belonging to the same factor showed widely varying sex differences, (4) national-
level sex differences in Emotionality are larger in wealthy and gender-egalitarian countries,
replicating previous counter-intuitive findings, but such a tendency was not clearly observed for
Honesty-Humility, and (5) within several English-speaking countries, sex differences in Emotionality
show little ethnic variation, supporting the suggestion that societal characteristics may influence the
size of sex differences in Emotionality. Conclusion: The HEXACO model of personality structure
provides some new insights in understanding sex differences in personality at the individual and
national levels.

Keywords: HEXACO; Sex Differences in Personality; Honesty-Humility; Emotionality

This article is protected by copyright. All rights reserved

 Sex differences in personality have been examined frequently in the past few decades (e.g.,
Accepted Article
Maccoby & Jacklin, 1974; Zell, Krizan, & Teeter, 2015), and empirical studies have consistently
found that some personality traits show sex differences of moderate to large size, at least within
samples of persons in Western countries. On measures of the Five-Factor Model (FFM) of personality
structure, women have been found to average about half of a standard deviation unit higher than men
do for two dimensions, namely Neuroticism and Agreeableness. On measures of the HEXACO
model of personality structure, women have been found to average about one full standard deviation
unit higher than men do on the Emotionality factor and also nearly one-half standard deviation unit
higher on the Honesty-Humility factor. Although most of these studies were based on self-reports on
personality, very similar findings have also been reported in the studies involving observer ratings
(Lee & Ashton, 2018).
Studies on sex differences in personality traits have been extended to multi-national
investigations since the early 2000s (e.g., Costa, Terracciano, & McCrae, 2001). One counterintuitive
but robust finding from these studies is that such differences are generally larger in more
economically developed and gender-egalitarian countries (Falk & Hermle, 2018; Mac Giolla &
Kajonius, 2019; Schmitt, Realo, Voracek, & Allik, 2008). In the present research, we revisit this
issue using a measure of the HEXACO personality dimensions in a large international online dataset.
Before describing the purposes of the present research, we provide a brief review of studies that have
examined cultural variation in personality sex differences.
Sex Differences in Personality across Cultures
The finding that sex differences in personality tend to be more pronounced in wealthy,
developed, and gender-egalitarian societies appears to contradict predictions from the social role
theory (Eagly & Wood, 1999), which suggests that sex differences in personality are primarily
determined by social norms about traditional sex roles. According to that theory, sex differences in
personality and values are expected to be larger in gender-unequal countries where sex role
stereotypes are generally stronger. Results from previous studies have consistently contradicted this
prediction, and researchers have proposed several different explanations for these counter-intuitive
results.

This article is protected by copyright. All rights reserved

 Falk and Hermle (2018) suggested that people are likely to pursue their preferences without
Accepted Article
restrictions in situations where material and social resources are sufficient. Under these
circumstances, both men and women are more likely to feel comfortable with “being themselves” (i.e.,
with self-expression), which leads to more pronounced sex differences in personality in richer and
more gender-equal countries (see also Schmitt et al. [2008] for an evolutionary-based account for this
counter-intuitive finding). Both of these accounts assume that societal characteristics directly
influence the level of some personality traits for each sex, thereby increasing or reducing personality
sex differences depending on those characteristics.
Costa et al. (2001) provided a somewhat different perspective, suggesting that people’s frame
of reference for personality judgments might differ depending on culture. For example, Costa et al.
suggested that people socialized under traditional gender norms are likely to compare themselves with
other people of the same sex, whereas people socialized under progressive gender norms are likely to
compare themselves with other people of both sexes. Costa et al. also suggested that, in gender-
unequal traditional societies, observed behaviors might be attributed to the gender norms in the
society, rather than to that person’s predisposition per se. Both accounts suggest that societal
characteristics influence people’s judgments about personality, not necessarily affecting true sex
differences in personality per se.
Finally, it may well be that cultural or regional variation in sex differences in personality is
attributable to relatively poor psychometric properties of the personality measures in less
economically developed countries, perhaps reflecting lower education levels in those countries; that is,
if personality scales show poor psychometric properties within a given population, those scales will
tend to underestimate the sizes of sex differences in personality within that population.
Present Research
In this section, we describe some features of the present research that distinguish it from
previous studies of personality sex differences.
Use of the HEXACO Personality Model. Many of the studies reviewed above were based
on measures of the Big Five personality model, but sex differences in personality tend to be somewhat
larger in the HEXACO model of personality structure, which also has an accompanying theoretical
framework that is relevant to sex differences. We therefore conducted our study of personality sex

This article is protected by copyright. All rights reserved

 differences using measures of the HEXACO personality factors. Below, we describe how the
Accepted Article
HEXACO model differs from the Big Five model, and in so doing we also discuss sex differences in
HEXACO personality variables.
As discussed elsewhere (Ashton & Lee, 2001, 2007), three of the six HEXACO personality
factors—Extraversion, Conscientiousness, and Openness to Experience—are very similar to their
counterparts in the Big Five model. The other three factors—Emotionality, Honesty-Humility, and
Agreeableness—reorganize the variance in Neuroticism and Agreeableness in the Big Five, and also
incorporate new variance beyond those Big Five factors. These latter three HEXACO factors have
been theorized to underlie altruistic versus antagonistic tendencies. Below we provide an overview of
this theory, which is of particular relevance to the present research.
The Emotionality factor, which is characterized by such traits as anxiety, physical harm
avoidance, support seeking, and emotional attachment and sensitivity, is thought to represent a trade-
off between maximizing kin survival (as well as the personal survival that helps to promote kin
survival) and maximizing gains that might result from accepting risks to kin and personal survival
(see Ashton & Lee, 2007). Given the greater reproductive investment by women as compared with
men, as well as the more critical role of women in offspring survival and the lower level of parental
uncertainty for women, one would therefore expect that women would show a higher level than men
of the Emotionality factor.
Consistent with this expectation, results of most studies indicate that women average about
one standard deviation higher than men on HEXACO-PI-R Emotionality. This difference is
considerably larger than those typically observed in Big Five Agreeableness or Neuroticism and is
comparable in size to the combined sex differences across all Big Five factors. For example, within
Goldberg’s Oregon dataset, HEXACO-PI-R Emotionality correlated .42 with participant sex, whereas
the five factor-level scales of the NEO-PI-R achieved a multiple correlation of .37 (see, e.g., Ashton,
Lee, & de Vries, 2014).
Honesty-Humility and Agreeableness are interpreted to underlie two different forms of
reciprocal altruistic tendencies. Specifically, Honesty-Humility is a fairness-based reciprocal altruism
(i.e., treating others fairly even when one could successfully exploit them), whereas Agreeableness is
a patience- or forgivingness-based reciprocal altruism (i.e., continuing or resuming cooperation with

This article is protected by copyright. All rights reserved

 others even when one is perhaps being exploited by them). Interestingly, Honesty-Humility has been
Accepted Article
found to show a moderate level of sex difference (women being higher, meta-analytic uncorrected d =
0.43, Moshagen et al., 2019), but Agreeableness has shown almost none. The sex difference in
Honesty-Humility likely reflects the fact that for men, the potential fitness payoff from exploiting
others is greater than for women, due to the fact that a man can potentially have much higher
reproductive success than can a woman. The lack of sex difference in Agreeableness might reflect the
fact that for men and women alike the potential cost of being exploited is similar, given that the
minimum level of reproductive success (or inclusive fitness more broadly) is zero.
All three of the above HEXACO factors are related to a trait of overall altruism (versus
antagonism), which is also closely related to Big Five Agreeableness (Ashton, Lee, & De Vries, 2014).
This overall altruism construct—like Big Five Agreeableness—shows a moderate sex difference, with
women averaging higher than men, as would be expected given the pattern of sex differences for the
three altruism-related HEXACO factors. The differentiation of altruism-related traits within three
factors of the HEXACO model, along with the theoretical interpretations of those three dimensions
and the sex differences therein, is an important difference between the HEXACO and Big Five
systems.
Factor- and facet-level investigation. As described above, it has been frequently found that
Emotionality and Honesty-Humility show considerable sex differences (women being higher on both),
with their combined sex difference apparently being larger than that shown by the Big Five factors in
combination. (see, e.g., Ashton, Lee, & de Vries, 2014). In contrast, sex differences in the HEXACO
facet scales have not been of primary focus in previous studies. In light of previous results showing
the importance of the specific variance carried by facet-level constructs (e.g., Ashton & Lee, 2019;
Lee, Ashton, Griep, & Edmonds, 2018; Ashton, Paunonen, & Lee, 2014), the pattern of sex
differences in facet-level constructs may not precisely mirror that of the factor domain to which they
belong. As such, in the present research, we report for each culture the sex differences in both factor-
and facet-level scales of the HEXACO-100.
Personality sex differences across ethnicities. In the previous section, we reviewed some
proposed explanations for the observed pattern of cultural variability in sex differences in personality.
All of these explanations assume that such a pattern is caused by certain characteristics of societies,

This article is protected by copyright. All rights reserved

 such as wealth, education, and gender equality. That is, those societal characteristics have been
Accepted Article
thought to produce true sex differences in personality (e.g., Falk & Hermle, 2019; Schmitt et al., 2008)
and/or to allow such sex differences to be expressed in personality self-reports (e.g., Costa et al.,
2001). However, it has not yet been tested whether such causal explanations at the country level are
indeed necessary to explain the cultural variation in sex differences. Below, we describe this problem
further and explain the strategy of the present research.
One may argue that such causal explanations are necessary given the robustly observed
finding that sex differences in personality tend to be larger in countries that are more economically
developed and more gender-equal. However, we should point out that such a finding would also be
obtained if, for whatever reasons, sex differences in personality traits happen to be larger among
persons of European ethnic background than among persons of non-European ethnic background.
This is the case because the countries whose populations are mainly of European background are
relatively wealthy and gender-equal, and also because such countries have typically accounted for
about half of the countries examined in previous studies. That is, in those national-level analyses,
economic development and gender equality have been confounded with European ethnic background,
raising the possibility that the latter variable is instead responsible for the national differences in the
size of personality sex differences. Therefore, it is an empirical question as to whether economic
development and gender equality are even needed as explanations of the pattern of cross-national
variation in the extent to which men and women differ in personality.
In the present research, we first examined the sex differences in the HEXACO personality
variables across 48 countries using self-report data collected online over a period of several years, and
attempted to revisit previous findings as obtained with the Big Five with respect to the HEXACO
variables. We then extended our inquiry to examine variation in sex differences across broad ethnic
groups within a few multi-ethnic countries. Examining ethnic variation in personality sex differences
within the same country (or within some set of similar countries) has some advantages in addressing
the problems described above, because this strategy allows national-level characteristics (such as
economic development, gender equality, and language) to be held constant or nearly constant.
To illustrate the research strategy, we provide Figures 1a and 1b. These graphs indicate sex
differences for imaginary ethnic groups as well as for the European-origin group. White bars show

This article is protected by copyright. All rights reserved

 sex differences in a given trait as observed in groups of people living in the continent of origin for
Accepted Article
their respective broad ethnic groups (e.g., persons of African ethnic origin living in African countries,
or persons of South Asian ethnic origin living in South Asian countries). The sex difference pattern
shown by white bars in these figures is plausible given that larger sex differences have been observed
in wealthier and more gender-equal countries. Gray bars indicate sex differences observed in groups
of people living in several multi-ethnic countries (e.g., US, UK, Australia, etc.) that are highly similar
in terms of societal characteristics (including economic development and gender equality) as well as
their primary language.
Figure 1a shows the case where the observed cultural variation in personality sex differences
is indeed caused by national characteristics, such that within the multi-ethnic countries the
personality sex differences in ethnic groups A, B, and C are equal to that in the European ethnic group,
thus eliminating the variance across ethnic groups in the size of personality sex differences.
Conversely, Figure 1b shows the case where the ethnic groups differ in the size of personality sex
differences even within the multi-ethnic countries; with such a pattern of results, one would not need
to invoke societal characteristics to explain the observed cultural variation in sex differences. Rather,
those results would suggest that the observed cultural variation simply reflects pre-existing variation
across populations in the extent of sex differences in personality (e.g., evolved differences due to local
adaptations or genetic drift, or cultural differences that are highly persistent).
Method
Participants
Data used in the present research were collected through an online survey site
([Link] from October 19, 2014 to October 18, 2018. The survey site was approved by the
University of XXX Conjoint Faculties Research Ethics Board. Participants visited the website
voluntarily and completed the questionnaires primarily for self-exploration purposes. In this study,
we included the participants who (1) completed all the items of the HEXACO-PI-R, (2) passed three
“attention-check” items, and (3) passed two other response-quality checks (i.e., coherence of
responses and use of different response options; for details see Lee & Ashton, 2018). The screening
thresholds in the quality check step were deliberately set very conservatively so that only obviously
non-purposeful responses (e.g., repetitive choice of same response options, quasi-random response,

This article is protected by copyright. All rights reserved

 etc.) were removed (fewer than 1% of respondents were eliminated by this criterion). A total of
Accepted Article
370,857 participants were included in the present research. Of these participants, 57.3% were men,
41.5% were women, and 1.2% did not indicate their sex; their average age was 30.2 years with a
standard deviation of 12.7.
Variables and Measures
Personality. The English-language version of the HEXACO-100 (Lee & Ashton, 2018) was
used to assess the HEXACO personality factors. The order of the items in the online survey is the
same as that of the paper-and-pencil version of the questionnaire. Three attentiveness check items
were embedded evenly throughout the questionnaire (e.g., “This is an attentiveness check; please
indicate ‘neutral’.”). Responses were made on a 5-point scale (1 = strongly disagree, 5 = strongly
agree).
Country. We obtained each respondent’s country, and inspected the ethnic group
composition for each country. We noticed that, for some Asian and Middle Eastern countries, persons
of European ethnic background were numerous, and we removed those respondents on the grounds
that most of them were likely to be expatriates or relatively recent immigrants to those countries. We
also decided a priori to include for our analyses only those countries for which the sample size for
each sex is at least 150. As a result, 347,192 participants from 48 countries were included in the
analyses involving countries. The age and sex distributions of our samples from each of the 48
countries are described in Table 1.1 Internal consistency reliabilities of the HEXACO scales for each
country are shown in Supplementary Table 1.
To evaluate the structure of the inventory within each country, we conducted principal axis
factor analysis with varimax rotation separately for each country. The theoretically expected factor
structure emerged from all of the subsamples, with each factor being defined chiefly by its designated
facet scales (Supplementary Table 2). When we computed congruence coefficients of the six varimax-
rotated factors observed in each country against those found in the US sample (i.e., the largest
sample), the average congruence coefficients across factors and across countries were all above .95,

1 The conditions of our participant consent materials do not allow us to post the dataset to a publicly
available repository. Researchers who would like to examine the data for the variables analyzed in
this article should contact the authors to obtain it.

This article is protected by copyright. All rights reserved

 except for the Conscientiousness factor from the South Korean sample (.89; (see Supplementary
Accepted Article
Table 3 for full matrices of congruence coefficients).
Country-Level Characteristics. We obtained indices of national-level variables of the
Gender Gap Index and Human Development Index (HDI). The Gender Gap Index has been published
by the World Economic Forum since 2006 to gauge a country’s gender equality level based on indices
of health, educational, economic, and political status (see
[Link] Because higher scores indicate higher
levels of gender equality of the country, we refer this variable to Gender Equality Index (GEI) in the
present study. The GEI used in this research is the mean score across three years, from 2015 to 2017.
Within the countries included in the present research, the values ranged from .55 to .84 with a mean
of .72 (SD = 0.06).
The HDI is a composite score based on three key dimensions of human development (a
country’s mean life expectancy at birth, level of education, and per capita income), and is published
by United Nations Development Programme (see [Link]
index-hdi). The HDI used in the present research is also an average across three years, from 2015 to
2017. In the present research, the HDI scores ranged from .56 to .95 with a mean of .84 (SD = 0.10).
Ethnicity. Upon the completion of the HEXACO-100, the participants were provided with a
set of optional questions asking for demographic information. One of these questions assessed
participants’ broad ethnic background:2
Which of the following describes your family background?
(1) East Asian (e.g., Chinese, Japanese, Korean), (2) Southeast Asian (e.g., Vietnamese,
Filipino, Indonesian), (3) South Asian (e.g., Indian, Pakistani, Bangladeshi, Sri Lankan), (4) African,
(5) Arab and/or West Asian (e.g., Turkish, Persian), (6) European, (7) Latin American, (8) Pacific
Islander, (9) Australian Aboriginal, (10) North or South American Aboriginal/Native, (11) Other, (12)
Two or more of the above, and (13) Prefer not to answer.
In our analyses below, we did not include participants indicating themselves as Pacific
Islander, Australian Aboriginal, and North and South American Aboriginal/Native. This was in part

2We use the term “ethnic” to describe broader categories than are often denoted by this term, as each
of our ethnic categories subsumes many narrower ethnic groups and corresponds roughly to what are
often called racial groups.

This article is protected by copyright. All rights reserved

 because the comparisons across geographical regions (as described below) are not as readily applied
Accepted Article
to these ethnic origins.
As described in the Introduction, we examined ethnic variation in the extent of sex differences
in personality within two groups defined by the countries in which participants are living. First, sex
differences in personality were computed among persons who live in countries within the continental
regions in which their declared ethnic background originated. For example, for persons declaring
ethnic backgrounds as East Asian, Southeast Asian, South Asian, and Arab/West Asian, sex
differences were computed from those who were living in Asian countries.3 For Europeans, sex
differences were computed from those living in continental Europe, excluding the UK and Ireland
(see below ). Hereafter, we will refer this group to the “continent of origin” group. Second, sex
differences in personality were also computed for the subsample of participants living in six multi-
ethnic English-speaking countries (Australia, Canada, New Zealand, Ireland, UK, and US) that have
similar levels of the national-level GEI and HDI variables. Hereafter we refer to this group as the
“English-speaking countries” group.
Results
For each personality scale, we first computed age-corrected means and standard deviations for
each of the sex-by-country cells, and computed the d statistics for each country.4 (See Table 1 for
demographic information, the Gender Equality Index [GEI], and the Human Development Index [HDI]
for 48 countries). Overall results will be discussed with sample-unweighted means of the d statistics
(as opposed to sample-weighted means) to avoid the results being influenced heavily by a few
countries; for example, two countries, US and UK, account for 62.5% of total participants. In
presenting the results, we focus primarily on effect sizes (i.e., d statistics) and do not rely on any
statistical significance tests. Specifically, we mainly discuss the results for which effect sizes

3 We had planned to include Arab/West Asian persons living in northern African countries, but none
of those countries had enough participants for inclusion in our analyses.
4 Scores on each personality variable were regressed on age, and the resulting residualized scores

were used for subsequent analyses. The age-corrected scores for each scale have a mean of zero and a
standard deviation that is very similar to that of raw scores. Means and standard deviations of the age-
corrected scores on the HEXACO-PI-R scales are shown in Supplementary Table 4 for each country.
Cohen’s d statistics were computed as d = (Meanwomen – Meanmen) /[(SDwomen + SDmen )/2] for a given
personality scale.

This article is protected by copyright. All rights reserved

 approach a d of 0.20, which is the threshold of a “small” effect size according to Cohen’s (1988)
Accepted Article
guidelines. Table 2 shows standardized sex differences (i.e., Cohen’s d) in the HEXACO factor and
facet scales for 48 countries; multivariate Mahalanobis D statistics (see Del Giudice, 2013; Del
Giudice, Booth, & Irwing, 2012) are also reported for the set of six factor scales and for the set of 25
facet scales. The bottom row of Table 2 also shows the sample unweighted mean sex differences, and
we first interpret these overall results for personality variables.
For Honesty-Humility, women averaged higher than did men, with a sample-unweighted mean
d of 0.37. At the facet level, similar levels of sex differences were found for Fairness, Modesty, and
Greed Avoidance (sample-unweighted mean ds ranging from 0.26 to 0.36), but a much smaller sex
difference was found for Sincerity (sample-unweighted mean d = 0.09).
For Emotionality, women averaged substantially higher than men did (sample-unweighted
mean d = 0.84). In the current study, the d statistics for Emotionality were noticeably larger in such
countries as the United States (0.98), the United Kingdom (0.97), Canada (1.02), Australia (1.11),
Netherlands (1.14), and Germany (1.01) than in such countries as India (0.67), Philippines (0.60),
South Korea (0.41), Japan, (0.47), and Indonesia (0.48). Unlike the pattern of results observed for the
Honesty-Humility facet scales, the Emotionality facet scales all showed fairly large sex differences,
with sample-unweighted mean ds ranging from Anxiety (0.49) to Sentimentality (0.67).
The remaining HEXACO factors did not show sizeable sex differences, with the largest
sample-unweighted mean d being -0.12, for Agreeableness (i.e., men being slightly higher).
Nevertheless, some facets in these domains showed small to moderate levels of sex differences. First,
men scored almost one third of standard deviation unit higher than women did for Patience in the
Agreeableness domain (sample-unweighted mean d = -0.28). Second, two facets in the
Conscientiousness domain showed sample-unweighted mean ds approaching an absolute value of
0.20, with Organization being higher for women (0.16), and Prudence being higher for men (-0.19).
Third, three facet scales in the Openness domain showed sample-unweighted mean ds exceeding an
absolute value of .20, with two facets being higher for men (Inquisitiveness = -0.35;
Unconventionality = -0.24) and the other higher for women (Aesthetic Appreciation = 0.41). Finally,
the interstitial scale, namely Altruism, showed a moderately strong effect size favoring women
(sample-unweighted mean d = 0.44).

This article is protected by copyright. All rights reserved

 The overall size of sex differences, computed on the basis of the six factor scales and also on
Accepted Article
the basis of the 25 facet scales, is shown in Table 2 by Mahalanobis D statistics. As seen in the table,
the factor-level Ds ranged from 0.47 (South Korea) to 1.33 (Denmark) with an unweighted mean of
0.95, and the facet-level Ds ranged from 0.80 (Japan) to 1.60 (Denmark) with an unweighted mean of
1.24. The HEXACO-PI-R factor- and facet-level Ds for many of these countries can be compared
with those reported by Mac Giolla and Kajonius (2019) for the IPIP-NEO measure of the Big Five
factors in a different online sample. As shown in Supplementary Table 5,5 the HEXACO-based D
values from the present study tended to be considerably larger at the factor level (19% larger) and
slightly larger at the facet level (8% larger) than were the Big Five-based D values in the
corresponding countries as reported by Mac Giolla and Kajonius. If the analysis is confined to those
countries with relatively larger sex differences in the Mac Giolla and Kajonius dataset (those with D >
0.80, all of which are Western countries), the ratios are slightly larger (22% for factor scales and 13%
for facet scales). In those countries, the HEXACO and Big Five factor scale Ds were 1.11 and 0.91,
respectively, and the corresponding facet scale Ds were 1.37 and 1.21, respectively.
Variations in Sex Differences across Countries and Ethnicities
The two national characteristics included in the present research (i.e., GEI and HDI) were
strongly correlated at .52, and to simplify the results, we combined GEI and HDI after standardizing
each index. The GEI was not available for Hong Kong, so we excluded that country from the
national-level analyses.
In the national-level analyses, we focus our discussion primarily on the six factor-level scales,
because we believe that examining the 25 facet-level associations observed from a sample size of 47
would run the risk of over-interpretation. We will mention some of the facet-level findings as they
are related to the factor-level findings.
National sex differences in Emotionality were significantly associated with the GEI/HDI
composite (r = .56, p < .01),6 showing that sex differences in Emotionality tended to be more
pronounced in wealthier and more gender-equal countries (Figure 2). As such, Emotionality showed

5 In this supplementary table, we analyzed raw scores (not age-adjusted scores) and we included only
participants with ages between 19 and 69 years inclusive, to be consistent with the methods used by
Mac Giolla and Kajonius.
6 National sex differences in Emotionality correlated .60 with GEI and .41 with HDI.

This article is protected by copyright. All rights reserved

 a pattern of cultural variation in sex differences similar to that shown for other personality variables in
Accepted Article
previous studies. All four facets in this domain showed similar relations with the GEI/HDI composite
(rs of .60 for Fearfulness, .35 for Anxiety, .56 for Dependence, and .46 for Sentimentality).
In contrast, national sex differences in the remaining factors did not show statistically
significant associations with the GEI/HDI composite (rs of .24 for Honesty-Humility, .08 for
Extraversion, .20 for Agreeableness, .14 for Conscientiousness, and .16 for Openness to Experience).
It is interesting that Honesty-Humility, which showed a moderately large overall sex difference, did
not show a strong association with the GEI/HDI composite. National sex differences for the Honesty-
Humility facet scales showed a rather complex pattern of association with the GEI/HDI composite,
including both positive and negative relationships (rs of -.48 for Sincerity, .01 for Fairness, .45 for
Greed Avoidance, and .38 for Modesty). The results involving the Honesty-Humility factor and its
facets suggest that cultural variation in personality sex differences can show different patterns
depending on the traits being examined. 7
Ethnic variation in sex differences in the Emotionality factor were examined separately for
several broad ethnic groupings, each separated into the two geographically-defined groups of
participants described above: (1) people living in the continent from which their ethnic group
originated (the continent-of-origin group) and (2) people living in one of several multi-ethnic,
primarily English-speaking countries (Australia, Canada, Ireland, New Zealand, UK, and US). Recall
that these English-speaking countries tend to be fairly homogeneous in terms of the focal national
characteristics, namely, GEI and HDI (as seen in Figure 2, red plots). As such, sex differences across
ethnic groups observed in this sample are not likely to be due to differences in these national

7 National sex differences in Altruism also showed a strong association with the GEI/HDI composite
(r = .54), which is consistent with findings from previous reports (e.g., Falk & Hermle, 2019) and also
fits with the tendency for Altruism to load in part on the Emotionality factor. In addition, there are
three other facets for which national-level sex differences were significantly associated with the
GEI/HDI composite (rs of -.34 for Social Self-Esteem, .34 for Sociability, .44 for Flexibility). As
seen in Table 2, the overall sex differences for those scales were very small, and therefore the pattern
of associations suggests “cross-over” interactions (i.e., men being slightly higher than women for
some countries and vice versa for other countries). Considering the lack of plausible theory explaining
such cross-over interactions and the potential for sampling error in national-level analyses based on
47 countries, we decided not to interpret these results.

This article is protected by copyright. All rights reserved

 characteristics or in the English proficiency levels of the participants. Therefore, if these two
Accepted Article
geographical groupings show a similar pattern of ethnic group differences in the size of personality
sex differences, then it is difficult to attribute the observed cross-national variation in personality sex
differences to societal differences in wealth or gender equality.
Table 3 shows sex differences in Emotionality for each ethnic group separately for the two
geographical groups (i.e., continent-of-origin group and English-speaking countries group); see
Figure 3 for a graphical presentation of the results. For the continent-of-origin participants (i.e., white
bars in Figure 3), the sex difference in Emotionality was noticeably larger for European-origin
participants (d = 0.96) than for those of the other ethnic groups, ranging from d = 0.55 (East Asian-
origin group) to d = 0.84 (Latin American-origin group), an expected finding given the national-level
results described above.
Within the English-speaking countries group, the variation across ethnic groups in the size of
the sex differences was substantially reduced. Specifically, except for the European ethnic group
(which showed nearly identical sex differences across the two geographical groups, 0.96 versus 1.00),
the other ethnic groups living in English-speaking countries showed much larger sex differences
(hence moving toward the mean sex difference for Europeans) than did their counterparts in the
continent-of-origin group. This result is consistent with the claim that the size of sex differences in
large part depends on the characteristics of the country in which those persons live.
Within the English-speaking countries group, African- and Latin American-origin
participants showed sex differences very similar in size (ds = 1.04 and 0.97, respectively) to that of
the European-origin group (d = 1.00). However, some ethnic groups still showed noticeably smaller
sex differences than did European-origin persons. The observed ds were 0.76 for East Asian-origin
persons, 0.82 for Southeast Asian-origin persons, 0.83 for Arab/West Asian-origin persons, and 0.82
South Asian-origin persons.
Post-hoc examinations: Analyses of Means for Women and Men. The analyses presented
above are based on d statistics for sex differences, with no reference to the means for the sex-by-
ethnicity groups. The comparison of the means in self- (and observer) reported personality traits
across different cultural groups must be approached prudently due to the problems of the frame-of-
reference effect on personality assessment (Ashton, 2007; Perugini & Richetin, 2007) and of social

This article is protected by copyright. All rights reserved

 desirability (McCrae, Yik, Trapnell, Bond, & Paulhus, 1998). With this caveat in mind, we examined
Accepted Article
the means on Emotionality for the sex-by-ethnicity cells separately for each of the two geographical
groups (continent-of-origin versus English-speaking countries) to gain some additional insight.
The means and standard deviations of age-corrected Emotionality scores for the sex-by-ethnic
groups are shown in Table 3. In general, for the non-European ethnic groups, women averaged higher
and men averaged lower for Emotionality in English-speaking countries than did their counterparts in
the continents of origin (see Figure 4). The greater sex differences in Emotionality observed within
English speaking countries thus result from a shifting of both sexes’ distributions toward the opposing
ends of the dimension.8
Discussion
Below, we describe and discuss each of the major findings of this study of sex differences in
the HEXACO personality factors.
Individual-Level Findings
The overall multivariate sex differences in the HEXACO-PI-R scales were rather large: the
sample-unweighted mean across the 48 countries was 0.95 at the factor-scale level and 1.24 at the
facet-scale level. These differences are larger than those reported previously in similar online
samples, based on scales of similar length, for the Big Five factors and (to a considerably lesser extent)
their constituent facets. As seen in Supplementary Table 5, when results from the same country are
compared, multivariate sex differences are about 19% larger at the factor level and 8% at the facet
level for the HEXACO than for the Big Five; and when the analysis is confined to Western countries
(which show larger sex differences for both frameworks), the differences are even greater, reaching
22% for factor level and 13% for the facet level. On the one hand, the fact that these differences are
greater at the factor level than at the facet level suggests that they result in part from the different

8 One incidental observation from Figure 4 is a tendency for women to show rather similar average
levels of Emotionality across different ethnic groups, whereas men show noticeably larger variation in
average levels of Emotionality across those ethnic groups. In particular, men in the East Asian-,
Southeast Asian-, South Asian-, and Arab/West Asian-origin groups averaged considerably higher
than did men in the African-, European-, and Latin American-origin groups. It is this variation in
men’s average scores across ethnic groups that is mostly responsible for the observed ethnic group
variability in sex differences in Emotionality.

This article is protected by copyright. All rights reserved

 classification of traits within the two frameworks. On the other hand, the fact that some differences
Accepted Article
remain even at the facet level suggests that they result in part from the different sampling of traits
within the instruments that assess the two frameworks. In any case, the phenomenon of sex
differences in personality appears to be more thoroughly represented within the HEXACO model and
measures than within those of the Big Five.
The overall multivariate sex differences across the HEXACO-PI-R factor scales were in large
part due to two factors, namely Emotionality and Honesty-Humility. The overall sample-unweighted
effect sizes for sex differences were d = 0.37 for Honesty-Humility and d = 0.84 for Emotionality,
which are somewhat smaller differences than those reported in previous studies conducted in Europe
and North America (see Lee & Ashton, 2006, for results from North American samples). This
reflects the fact that the present dataset includes many samples beyond North America and Europe,
and that those samples contributed equally to the overall estimate regardless of their sample size. The
overall sample-weighted ds for Honesty-Humility and Emotionality in the present research were
larger (i.e., 0.40 and 0.95, respectively). These results show that estimates of personality sex
difference could vary widely depending on the regions (or to a lesser extent the ethnic groups) from
which samples were drawn. We discuss this issue in more detail in the next section.
Several narrower personality traits, as assessed by facet-level scales of the HEXACO-PI-R,
also showed appreciable sex differences. First, all of the four facets in Emotionality showed
comparably large sex differences, suggesting that it is probably the common variance in Emotionality
that is implicated in sex differences. Second, while three facet scales in Honesty-Humility showed
similar, moderate-sized sex differences (around three-tenths of a standard deviation unit), Sincerity
showed a small sex difference (about one-tenth of a standard deviation unit). It appears that Sincerity
contains some specific variance for which men average higher than women do, thereby balancing out
women’s higher scores on the common variance in Honesty-Humility. Finally, women averaged
almost half a standard deviation unit higher than men do in the Altruism facet scale, which is not
surprising considering that Altruism is positively associated with Emotionality and Honesty-Humility.
The importance of the facet-level investigation in personality sex differences can be seen even
more prominently in the domains of Conscientiousness and particularly for Openness to Experience.
Within both dimensions, one or more facets showed noticeable sex differences in the opposite

This article is protected by copyright. All rights reserved

 direction of another facet. For example, with respect to Conscientiousness, women showed a
Accepted Article
somewhat higher average for Organization than men did, but the opposite result was found for
Prudence. With respect to Openness to Experience, women averaged higher than did men for
Aesthetic Appreciation, but the opposite result was found for Inquisitiveness and Unconventionality.
Such prominent sex differences at the level of narrower traits would have gone unnoticed had the
present research focused only on the broad factors.
We should also note that personality factor domains that include facets showing sex
differences of at least modest size in opposite directions are very likely to show scalar non-invariance
across two sexes (see Bogaert, Ashton, & Lee, 2018; McCrae, 2015). In the current dataset, when we
conducted an item-level multigroup confirmatory factor analysis for each domain involving the self-
reported men and women groups, the conditions of metric invariance were met for all domains based
on the conventionally-used cutoff criteria (Chen, 2007), but the conditions of scalar invariance were
not met for Emotionality, for Conscientiousness, and particularly for Openness to Experience.
Although there is no consensus among researchers as to how to handle scalar non-invariance cases,
we agree with McCrae (2015) that mean comparisons should not be abandoned simply because the
suggested conditions of scalar invariance are not met, and that sex differences in specific variance
must be regarded as an important substantive result to be interpreted and discussed.
National (and Ethnic)-Level Findings
Sex differences in Emotionality were found to be larger in wealthy and gender-equal countries
as measured by the GEI/HDI composite, replicating the surprising finding observed in studies
adopting the Big Five. However, such a pattern was not clearly observed for the other five
personality dimensions, including Honesty-Humility, which is the only other dimension in the
HEXACO model that shows a sizable sex difference. These results suggest the counter-intuitive
findings about cultural variation in personality sex differences are primarily true for Emotionality, but
not so much for Honesty-Humility.
Why do Honesty-Humility and Emotionality show such different results? We suspect that this
might be due to the fact that traits included in HEXACO Emotionality—anxiety, fearfulness,
sentimentality, dependence—are traditionally considered to be more feminine (versus masculine)
characteristics. Although the various accounts submitted by previous researchers (Costa et al., 2001;

This article is protected by copyright. All rights reserved

 Falk & Hermle, 2018; Schmitt et al., 2008) involve different mechanisms, all of those accounts
Accepted Article
implicitly or explicitly focus more on stereotypically “feminine” or “masculine” personality traits.
For example, Schmitt et al primarily discussed sex differences in personality traits arising from sex
differences in parental investment (Trivers, 1972). Other researchers have discussed gender-related
cultural influences (e.g., gender equality) on the expression and development of personality traits
(Falk & Hermle, 2018) and on the judgmental processes of personality assessment (Costa et al., 2001).
One of the salient differences between the HEXACO and Big Five frameworks is that these “feminine
versus masculine” traits—which are interpretable in terms of kin-altruistic and related tendencies (e.g.,
Ashton & Lee, 2007)—are primarily located on the HEXACO Emotionality factor but between the
Big Five Neuroticism and Agreeableness factors, with some of these traits being somewhat poorly
accommodated by the Big Five space. These differences in the location and representation of
traditional feminine-versus-masculine personality traits may have obscured the differential pattern of
cultural variation in sex differentiation in personality traits.
Does the strong national-level association shown here mean that societal characteristics (e.g.,
wealth and gender equality) cause sex differences in Emotionality? To answer this question, we
investigated ethnic variation in Emotionality sex differences in two different geographical groups (i.e.,
people living in the continent of origin for their ethnic group and people living in several multi-ethnic
English-speaking countries).
For persons living in the continent of origin for their ethnic group, we found larger sex
differences Emotionality for European-origin participants than for non-European-origin participants,
which is expected given the national-level results outlined above. However, this contrast between the
European- and non-European-origin groups was much smaller when the comparisons were based on
people living in English-speaking countries only (i.e., the group in which wealth and gender-equality
are largely controlled). These findings provide the first direct evidence supporting the notion that
societal characteristics are indeed responsible for the cultural variation in personality sex differences
reported in the past studies. Interestingly, African- and Latin American-origin persons living in
English speaking countries showed sex differences as large as those for European-origin persons, a
result which suggests that societal characteristics could be almost entirely responsible for the
observed cultural variation in personality sex differences in the previous and present studies.

This article is protected by copyright. All rights reserved

 For the other ethnic groups, sex differences in Emotionality in English-speaking countries
Accepted Article
were still noticeably smaller than for the European-origin group, and it is unclear yet whether this is
because sex differences are inherently smaller for these groups, or because people in these groups are
less acculturated (i.e., still somewhat under the influence of the cultural traditions of their ethnic
groups), or both. It would be useful to examine in future research some people of these ethnic
backgrounds who were adopted by parents with European background. If a similar level of sex
difference is observed from these trans-ethnic adoptive family studies, the smaller sex differences in
Emotionality for East/Southeast Asians are more likely to be biological in origin.
We conducted a post-hoc analysis of means for men and women to better understand the
finding of larger sex differences for the non-European origin groups in English-speaking countries
than for their counterparts in the continents of origin. It would be reasonable to expect that living in a
wealthy and gender-equal society would have a stronger effect on women’s personality than on men’s
personality because such social characteristics may allow women to self-express to a greater extent,
given the historical inequality between women and men, and hence the greater potential for men to
express their preferences even in gender-unequal societies). However, the post-hoc analysis suggested
that this is not the case. The larger sex differences in Emotionality observed in English-speaking
countries reflect not only higher Emotionality for women, but also lower Emotionality for men. As
such, societal characteristics appear to influence the way that Emotionality is developed, expressed,
and judged for both women and men.
Limitations and Future Research
One unanswered question is that although the present findings support the notion that social
characteristics may influence the size of sex differences in Emotionality, we still do not know which
theory, if any, can best explain that influence (e.g., evolutionary, attributional, frames of reference,
etc.). Future research specifically designed to test these theories—and perhaps some alternative
accounts–could eventually clarify this issue.
Second, it is possible that the smaller sex differences in personality observed in less developed,
less gender-equal countries can be attributed to poorer psychometric properties of the personality
measures in those countries. Although we did not find any noticeable psychometric problems with
the data collected in those regions, at least in terms of factor structure and internal consistency

This article is protected by copyright. All rights reserved

 reliability, there are other psychometric aspects that may have been overlooked. Moreover, it may
Accepted Article
well be that each country has a somewhat different level of average English proficiency, which could
have influenced the national-level results of the present study. In future research, the present results
should be compared with those obtained from translated versions of the HEXACO-PI-R.
On a final note, we believe that the overall results of the present research are likely to be
generalizable, given that the total sample size was very large, and that a reliable, valid personality
inventory was used. Nevertheless, we should remind readers that each national sample is not perfectly
representative of the national population, and therefore the estimates of sex differences in any single
country reported in this study must be taken with caution, especially when the sample size for the
country is relatively small.
Declaration of Conflicting of Interests
The authors have received royalties for the commercial use of the HEXACO-PI-R.
Funding
The authors received no financial support for the research, authorship, and/or publication of this
article.

This article is protected by copyright. All rights reserved

 References
Accepted Article
Ashton, M. C. (2007). Self-reports and stereotypes: A comment on McCrae et al. European Journal
of Personality, 21, 983-986.
Ashton, M. C., & Lee, K. (2001). A theoretical basis for the major dimensions of personality.
European Journal of Personality, 15, 327-353.

Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and practical advantages of the HEXACO
model of personality structure. Personality and Social Psychology Review, 11, 150-166.

Ashton, M. C., & Lee, K. (2019). Religiousness and the HEXACO personality factors and facets in a
large online sample. Journal of Personality, 87, 1103-1118.

Ashton, M. C., Lee, K., & De Vries, R. E. (2014). The HEXACO Honesty-Humility, Agreeableness,
and Emotionality factors: A review of research and theory. Personality and Social Psychology
Review, 18, 139-152.

Ashton, M. C., Paunonen, S. V., & Lee, K. (2014). On the validity of narrow and broad personality
traits: A response to Salgado, Moscoso, and Berges (2013). Personality and Individual
Differences, 56, 24-28.

Bogaert, A. F., Ashton, M. C., & Lee, K. (2018). Personality and sexual orientation: extension to
asexuality and the HEXACO model. The Journal of Sex Research, 55, 951-961.

Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance.

Structural Equation Modeling, 14, 464–504.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ:
Erlbaum.

Costa, P. T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality traits across
cultures: Robust and surprising findings. Journal of Personality and Social Psychology, 81,
322–331.
Del Giudice, M. (2013). Multivariate misgivings: Is D a valid measure of group and sex
differences?. Evolutionary Psychology, 11, 1067-1076.

This article is protected by copyright. All rights reserved

 Del Giudice, M., Booth, T., & Irwing, P. (2012). The distance between Mars and Venus: Measuring
Accepted Article global sex differences in personality. PloS one, 7(1): e29265.
Eagly, A. H., & Wood, W. (1999). The origins of sex differences in human behavior: Evolved
dispositions versus social roles. American Psychologist, 54, 408–423.
Falk, A., & Hermle, J. (2018). Relationship of gender differences in preferences to economic
development and gender equality. Science, 362, eaas9899.
Lee, K., & Ashton, M. C. (2006). Further assessment of the HEXACO Personality Inventory: Two
new facet scales and an observer report form. Psychological assessment, 18, 182-191.
Lee, K., & Ashton, M. C. (2018). Psychometric properties of the HEXACO-100. Assessment, 25,
543-558.
Lee, K., Ashton, M. C., Griep, Y., & Edmonds, M. (2018). Personality, religion, and politics: An
investigation in 33 countries. European Journal of Personality, 32, 100-115.

Mac Giolla, E. M., & Kajonius, P. J. (2019). Sex differences in personality are larger in gender equal
countries: Replicating and extending a surprising finding. International Journal of Psychology,
54, 705-711.

Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of sex differences. Stanford, CA: Stanford
University Press.
McCrae, R. R. (2015). A more nuanced view of reliability: Specificity in the trait hierarchy.
Personality and Social Psychology Review, 19, 97–112.
McCrae, R. R., Yik, M. S. M., Trapnell, P. D., Bond, M. H., & Paulhus, D. P. (1998). Interpreting
personality profiles across cultures: Bilingual, acculturation, and peer rating studies of Chinese
undergraduates. Journal of Personality and Social Psychology, 74, 1041–1055.
Moshagen, M., Thielmann, I., Hilbig, B., & Zettler, I. (2019). Meta-analytic investigations of the
HEXACO Personality Inventory(-Revised): Reliability generalization, self-observer agreement,
intercorrelations, and relations to demographic variables. Zeitschrift für Psychologie, 227, 186-
194.
Perugini, M., & Richetin, J. (2007). In the land of the blind, the one-eyed man is king. European
Journal of Personality, 21, 977-981.

This article is protected by copyright. All rights reserved

 Schmitt, D. P., Realo, A., Voracek, M., & Allik, J. (2008). Why can’t a man be more like a woman?
Accepted Article Sex differences in Big Five personality traits across 55 cultures. Journal of Personality and
Social Psychology, 94, 168-182.
Trivers, R. (1972). Parental investment and sexual selection. In R. Campbell (Ed.), Sexual selection
and the descent of man (pp. 136–179). Chicago, IL: Aldine-Atherton.
Zell, E., Krizan, Z., & Teeter, S. R. (2015). Evaluating gender similarities and differences using
metasynthesis. American Psychologist, 70, 10-20.

This article is protected by copyright. All rights reserved

Accepted Article
Table 1
Participant Age and Sex, Gender Equality Index, and Human Development Index by Country
Ns Age Ns Age
Country Men Women Mean SD GEI HDI GEI/HDI Country Men Women Mean SD GEI HDI GEI/HDI
Norway 1279 515 27.73 10.41 0.84 0.95 1.59 Bulgaria 372 152 26.00 8.65 0.73 0.81 -0.08
Finland 828 446 28.37 9.43 0.84 0.92 1.43 Greece 1724 1908 31.79 10.28 0.69 0.87 -0.11
Sweden 2137 669 27.53 10.25 0.82 0.93 1.31 Philippines 1741 1730 26.61 8.19 0.79 0.70 -0.13
Ireland 1945 1244 30.95 12.02 0.80 0.93 1.14 Croatia 536 296 27.16 8.64 0.71 0.83 -0.15
Switzerland 816 485 30.55 11.25 0.77 0.94 0.94 Japan 241 155 34.42 10.91 0.66 0.91 -0.16
New Zealand 3806 2680 31.20 12.92 0.78 0.92 0.92 Slovakia 384 250 30.04 9.60 0.68 0.85 -0.30
Germany 4711 2399 27.70 9.76 0.77 0.93 0.88 S Korea 379 327 26.51 6.98 0.65 0.90 -0.30
Denmark 1135 656 27.10 10.31 0.77 0.93 0.88 Serbia 533 402 25.77 7.64 0.72 0.78 -0.32
Netherlands 3586 2169 30.58 11.63 0.76 0.93 0.80 Romania 1526 900 25.65 7.25 0.70 0.81 -0.34
UK 30597 23840 31.47 13.69 0.76 0.92 0.75 South Africa 1434 908 31.05 11.64 0.76 0.70 -0.39
Canada 13561 10645 28.91 12.18 0.75 0.92 0.66 Colombia 262 151 26.82 9.23 0.73 0.75 -0.39
France 1742 1096 31.21 13.23 0.76 0.90 0.64 Russia 388 207 26.95 9.63 0.69 0.81 -0.42
Belgium 1108 610 28.85 10.61 0.75 0.91 0.61 Hungary 407 266 28.72 9.76 0.67 0.84 -0.44
Australia 10182 10748 30.08 12.42 0.73 0.94 0.59 Mexico 993 408 27.48 11.19 0.70 0.77 -0.54
US 90800 67510 31.11 13.55 0.73 0.92 0.49 UAE 323 189 26.82 8.78 0.64 0.86 -0.59
Spain 1127 614 31.13 13.44 0.74 0.89 0.42 Brazil 1465 544 25.65 8.72 0.69 0.76 -0.68
Singapore 1740 1416 26.53 8.42 0.71 0.93 0.37 Malaysia 811 614 26.71 8.10 0.66 0.80 -0.73
Austria 710 508 26.36 9.07 0.72 0.91 0.35 Indonesia 501 404 24.48 6.99 0.68 0.69 -1.13
Lithuania 359 167 24.34 6.66 0.74 0.86 0.26 Turkey 295 191 25.80 9.08 0.62 0.79 -1.13
Italy 1791 967 31.24 11.45 0.71 0.88 0.11 Kenya 288 157 26.69 6.57 0.71 0.58 -1.44
Poland 1205 528 26.71 8.08 0.72 0.86 0.09 Lebanon 165 208 23.73 7.56 0.60 0.75 -1.51
Portugal 847 383 27.54 10.01 0.73 0.84 0.08 India 4453 1531 26.69 8.97 0.67 0.63 -1.52
Czech 464 223 26.96 9.31 0.69 0.89 -0.01 Pakistan 734 254 25.41 6.89 0.55 0.56 -2.92

This article is protected by copyright. All rights reserved

Accepted Article
Argentina 370 174 26.72 8.82 0.73 0.82 -0.03 Hong Kong 211 189 26.42 9.65 0.93
Note. GEI = Gender Equality Index; HDI = Human Development Index. The GEI/HDI composite was computed after standardizing each index. Countries are
listed in descending order of GEI/HDI composite scores; no composite was computed for Hong Kong because GEI was not available.

This article is protected by copyright. All rights reserved

Accepted Article
Table 2

Standardized Sex Differences in the HEXACO Personality Scales by Country

Factor-Level Scales Facet-Level Scales

H E X A C O
Alt D6 D25
H E X A C O Si Fa Gr Mo Fe An De Se Ss Sb Sc Li Fo Ge Fl Pa Or Di Pe Pr Ae In Cr Un
Norway 0.45 0.98 0.03 -0.12 0.18 0.10 0.05 0.39 0.46 0.45 0.65 0.61 0.69 0.75 -0.14 0.02 0.16 0.06 0.01 -0.01 -0.03 -0.30 0.26 0.21 0.12 -0.12 0.52 -0.22 0.11 -0.18 0.55 1.12 1.35
Finland 0.37 1.07 0.04 -0.18 0.01 0.07 -0.04 0.34 0.35 0.41 0.76 0.63 0.80 0.78 -0.11 -0.01 0.11 0.10 -0.11 -0.01 0.03 -0.42 0.16 0.07 0.02 -0.26 0.54 -0.37 0.17 -0.23 0.54 1.14 1.41
Sweden 0.51 0.96 0.01 -0.16 0.11 0.13 0.13 0.40 0.45 0.50 0.71 0.59 0.63 0.76 -0.13 0.02 0.04 0.08 -0.03 -0.03 -0.06 -0.32 0.18 0.17 0.08 -0.14 0.40 -0.24 0.24 -0.11 0.60 1.09 1.32
Ireland 0.37 0.92 -0.01 -0.09 0.14 -0.06 0.03 0.38 0.27 0.40 0.72 0.56 0.51 0.75 -0.16 -0.04 0.14 0.01 -0.06 -0.03 -0.01 -0.15 0.17 0.11 0.15 -0.06 0.34 -0.41 0.04 -0.21 0.52 1.00 1.28
Switzerland 0.36 0.95 0.09 -0.11 0.08 -0.18 0.07 0.36 0.23 0.42 0.64 0.57 0.69 0.68 -0.13 -0.07 0.24 0.20 0.01 0.03 0.05 -0.38 0.21 0.16 0.11 -0.32 0.28 -0.45 -0.02 -0.35 0.48 1.04 1.37
New Zealand 0.40 0.94 -0.04 -0.07 0.15 0.01 0.07 0.43 0.27 0.35 0.79 0.55 0.56 0.76 -0.14 -0.10 0.07 0.04 -0.08 0.05 -0.02 -0.15 0.19 0.11 0.15 -0.03 0.37 -0.37 0.09 -0.11 0.62 1.04 1.31
Germany 0.44 1.01 -0.03 -0.17 0.06 -0.01 0.02 0.34 0.40 0.52 0.67 0.60 0.68 0.82 -0.15 -0.08 0.12 -0.01 -0.12 0.03 -0.02 -0.37 0.19 0.12 0.06 -0.24 0.43 -0.35 0.10 -0.28 0.48 1.12 1.41
Denmark 0.58 1.17 0.17 -0.02 0.31 0.04 0.11 0.63 0.38 0.57 0.69 0.60 1.01 0.96 -0.01 -0.02 0.38 0.16 0.03 0.07 0.16 -0.29 0.39 0.31 0.24 -0.08 0.59 -0.39 0.08 -0.26 0.80 1.33 1.60
Netherlands 0.46 1.02 0.00 -0.12 0.18 -0.08 0.06 0.40 0.39 0.46 0.74 0.62 0.65 0.85 -0.15 -0.05 0.14 0.05 -0.08 -0.04 0.08 -0.27 0.23 0.23 0.18 -0.14 0.41 -0.45 0.04 -0.29 0.54 1.14 1.43
UK 0.44 0.97 -0.07 -0.04 0.06 -0.15 0.10 0.38 0.29 0.54 0.77 0.58 0.60 0.80 -0.22 -0.15 0.15 -0.01 -0.04 0.11 0.05 -0.20 0.17 0.03 0.07 -0.12 0.26 -0.49 -0.02 -0.24 0.55 1.05 1.32
Canada 0.43 1.02 -0.04 -0.07 0.12 -0.14 0.02 0.40 0.36 0.43 0.84 0.57 0.65 0.79 -0.04 -0.18 0.04 0.06 -0.06 0.04 0.00 -0.17 0.21 0.05 0.11 -0.02 0.29 -0.39 -0.03 -0.33 0.57 1.12 1.35
France 0.38 0.93 0.02 -0.06 0.04 0.04 0.11 0.36 0.22 0.46 0.59 0.59 0.57 0.80 -0.14 -0.05 0.12 0.12 -0.04 0.08 0.09 -0.26 0.16 0.12 0.08 -0.26 0.43 -0.29 0.12 -0.19 0.52 0.98 1.25
Belgium 0.44 1.06 0.03 -0.16 0.10 -0.14 0.10 0.45 0.32 0.41 0.71 0.62 0.71 0.86 -0.12 -0.06 0.22 0.02 -0.13 0.03 0.08 -0.39 0.19 0.10 0.13 -0.16 0.38 -0.40 -0.01 -0.43 0.51 1.18 1.43
Australia 0.41 1.00 -0.08 -0.13 0.05 -0.26 0.11 0.34 0.26 0.50 0.82 0.57 0.63 0.76 -0.19 -0.19 0.12 0.00 -0.14 0.07 -0.02 -0.27 0.21 -0.04 0.07 -0.12 0.18 -0.49 -0.10 -0.43 0.53 1.11 1.35
US 0.38 0.98 -0.11 -0.13 0.08 -0.13 0.04 0.30 0.34 0.41 0.84 0.58 0.58 0.73 -0.12 -0.18 -0.02 -0.03 -0.12 -0.01 -0.06 -0.19 0.16 0.02 0.11 -0.08 0.26 -0.37 -0.03 -0.29 0.49 1.07 1.29
Spain 0.30 0.85 -0.02 -0.14 0.01 -0.07 0.03 0.23 0.22 0.42 0.61 0.58 0.51 0.67 -0.12 -0.11 0.13 0.02 -0.02 -0.04 -0.02 -0.30 0.13 0.00 0.11 -0.21 0.42 -0.39 0.00 -0.27 0.46 0.91 1.21
Singapore 0.29 0.57 -0.05 -0.19 -0.09 -0.06 0.07 0.30 0.21 0.27 0.52 0.27 0.36 0.44 -0.03 -0.13 -0.05 0.04 -0.07 -0.10 -0.07 -0.31 -0.03 -0.12 0.03 -0.15 0.32 -0.40 0.10 -0.23 0.31 0.68 1.01
Austria 0.46 1.03 0.20 -0.25 0.15 0.03 0.02 0.46 0.45 0.41 0.66 0.55 0.79 0.83 -0.02 0.15 0.31 0.15 -0.10 -0.01 -0.10 -0.50 0.23 0.23 0.15 -0.19 0.43 -0.35 0.15 -0.22 0.51 1.23 1.47
Lithuania 0.21 0.97 0.17 -0.08 0.26 0.14 -0.13 0.29 0.17 0.23 0.57 0.59 0.78 0.73 0.18 0.09 0.13 0.16 -0.07 0.08 0.01 -0.23 0.30 0.30 0.24 -0.14 0.59 -0.24 0.22 -0.27 0.25 1.05 1.39
Italy 0.41 0.79 0.03 -0.18 -0.08 0.07 0.23 0.35 0.22 0.42 0.57 0.47 0.47 0.68 -0.10 -0.02 0.12 0.08 -0.06 0.01 -0.10 -0.36 0.01 0.03 0.08 -0.36 0.48 -0.44 0.16 -0.03 0.47 0.91 1.25
Poland 0.30 0.83 -0.09 -0.20 -0.08 0.12 -0.03 0.29 0.27 0.32 0.59 0.54 0.68 0.55 -0.17 -0.04 0.04 -0.11 -0.11 -0.02 -0.13 -0.31 -0.01 0.05 0.06 -0.34 0.57 -0.35 0.25 -0.16 0.37 0.90 1.21
Portugal 0.31 0.91 -0.05 -0.23 0.04 0.19 0.11 0.32 0.17 0.32 0.62 0.56 0.55 0.76 -0.06 -0.13 0.01 0.00 -0.15 -0.05 0.02 -0.44 0.08 0.15 0.16 -0.28 0.57 -0.12 0.19 -0.21 0.43 0.97 1.25
Czech 0.46 0.89 0.01 -0.07 0.10 0.02 0.03 0.46 0.43 0.44 0.67 0.49 0.56 0.73 -0.01 -0.12 0.04 0.12 0.03 0.08 -0.03 -0.27 0.13 0.20 0.11 -0.14 0.58 -0.45 0.12 -0.28 0.49 1.01 1.33
Argentina 0.46 1.19 -0.11 -0.17 0.06 0.05 0.09 0.38 0.41 0.53 0.69 0.81 0.76 0.83 -0.25 -0.12 0.05 -0.06 -0.05 -0.11 0.01 -0.32 0.08 0.09 0.25 -0.26 0.42 -0.13 0.09 -0.29 0.59 1.25 1.45
Bulgaria 0.20 0.99 0.01 -0.22 -0.10 0.08 -0.13 0.17 0.29 0.21 0.64 0.66 0.77 0.66 -0.01 -0.04 0.15 -0.07 -0.04 -0.13 -0.33 -0.17 -0.10 -0.06 0.05 -0.19 0.47 -0.30 0.17 -0.15 0.39 1.04 1.35

This article is protected by copyright. All rights reserved

Accepted Article
Table 2 Continued.

Factor-Level Scales Facet-Level Scales

H E X A C O
Alt D6 D25
H E X A C O Si Fa Gr Mo Fe An De Se Ss Sb Sc Li Fo Ge Fl Pa Or Di Pe Pr Ae In Cr Un
Greece 0.48 0.87 -0.05 0.07 -0.06 0.09 0.24 0.45 0.33 0.42 0.61 0.42 0.57 0.75 -0.10 -0.07 0.06 -0.05 0.23 0.16 0.10 -0.26 0.13 0.00 0.05 -0.35 0.59 -0.46 0.18 -0.11 0.51 0.98 1.34
Philippines 0.27 0.60 -0.14 -0.09 0.01 -0.05 0.24 0.28 0.09 0.18 0.44 0.28 0.33 0.54 -0.08 -0.19 -0.09 -0.09 -0.02 -0.05 -0.04 -0.16 0.10 0.00 0.02 -0.10 0.30 -0.24 -0.01 -0.20 0.29 0.67 0.90
Croatia 0.33 0.77 -0.05 -0.12 0.01 0.06 0.02 0.40 0.22 0.37 0.62 0.48 0.50 0.52 -0.13 -0.10 0.06 0.00 -0.04 -0.01 0.05 -0.29 0.22 0.09 -0.04 -0.29 0.43 -0.37 0.22 -0.18 0.33 0.83 1.20
Japan 0.24 0.47 0.07 0.01 0.06 -0.03 0.04 0.31 0.06 0.27 0.41 0.14 0.32 0.39 0.17 -0.18 0.01 0.21 -0.04 0.07 0.09 -0.06 0.18 0.16 -0.05 -0.15 0.34 -0.23 -0.02 -0.20 0.24 0.53 0.80
Slovakia 0.25 0.91 -0.02 -0.24 -0.11 0.02 -0.03 0.24 0.27 0.22 0.69 0.47 0.70 0.65 -0.16 -0.06 0.16 -0.01 -0.09 -0.03 -0.13 -0.41 0.10 -0.10 0.12 -0.49 0.41 -0.37 0.21 -0.24 0.45 0.95 1.30
South Korea 0.19 0.41 0.00 -0.14 -0.12 0.04 0.07 0.32 0.10 0.02 0.36 0.18 0.34 0.22 0.04 -0.04 0.00 0.00 -0.10 -0.07 0.04 -0.29 -0.11 -0.13 -0.05 -0.07 0.39 -0.40 0.15 -0.03 0.12 0.47 0.95
Serbia 0.32 0.85 0.00 -0.16 0.10 0.04 0.07 0.41 0.23 0.21 0.58 0.44 0.59 0.77 -0.01 0.00 0.00 0.02 0.00 -0.05 -0.13 -0.28 0.17 0.20 0.07 -0.14 0.51 -0.31 0.04 -0.17 0.48 0.91 1.17
Romania 0.35 0.75 0.04 -0.14 0.13 0.11 0.05 0.33 0.34 0.31 0.52 0.46 0.49 0.63 0.07 -0.05 0.03 0.07 -0.01 -0.05 -0.08 -0.27 0.16 0.17 0.18 -0.14 0.52 -0.18 0.15 -0.24 0.34 0.90 1.16
South Africa 0.41 0.69 -0.02 0.00 0.13 -0.11 0.16 0.36 0.30 0.37 0.53 0.36 0.40 0.61 -0.06 -0.04 0.02 0.00 0.09 -0.03 0.10 -0.14 0.22 0.08 0.11 -0.01 0.28 -0.36 -0.01 -0.26 0.40 0.81 1.04
Colombia 0.45 0.86 0.18 -0.09 0.23 -0.22 0.05 0.36 0.44 0.47 0.44 0.59 0.53 0.78 0.14 0.03 0.25 0.13 0.02 -0.04 0.06 -0.29 0.41 0.22 0.30 -0.25 0.18 -0.28 -0.16 -0.36 0.45 1.08 1.23
Russia 0.05 0.68 0.01 -0.19 0.02 0.04 -0.06 0.25 -0.15 0.07 0.61 0.42 0.45 0.41 0.01 -0.13 0.09 0.05 -0.11 -0.04 -0.14 -0.25 0.12 0.03 0.13 -0.21 0.57 -0.21 -0.09 -0.24 0.36 0.68 1.21
Hungary 0.53 0.82 0.05 -0.19 0.02 0.05 0.07 0.48 0.50 0.50 0.53 0.43 0.64 0.72 -0.03 -0.04 0.07 0.15 0.00 -0.05 -0.10 -0.36 0.19 0.11 0.07 -0.35 0.61 -0.30 0.07 -0.33 0.47 0.96 1.41
Mexico 0.38 0.73 -0.07 -0.09 -0.06 -0.30 0.02 0.23 0.45 0.39 0.55 0.44 0.38 0.63 -0.13 -0.10 0.03 -0.03 -0.08 0.12 0.01 -0.28 0.02 -0.02 0.13 -0.30 0.16 -0.43 -0.15 -0.45 0.42 0.87 1.11
UAE 0.41 0.69 0.00 -0.07 0.15 0.00 0.28 0.44 0.10 0.40 0.50 0.43 0.37 0.55 0.02 -0.09 -0.02 0.09 -0.05 0.06 -0.05 -0.14 0.26 0.12 0.17 -0.13 0.47 -0.34 0.12 -0.26 0.44 0.84 1.14
Brazil 0.33 0.87 0.04 -0.13 -0.04 -0.08 0.12 0.20 0.20 0.47 0.57 0.52 0.50 0.81 -0.03 -0.01 0.12 0.05 0.01 0.00 -0.02 -0.34 0.02 -0.03 0.08 -0.22 0.32 -0.29 0.02 -0.33 0.54 0.92 1.15
Malaysia 0.41 0.70 -0.06 -0.11 0.13 -0.17 0.20 0.49 0.21 0.28 0.52 0.40 0.44 0.51 -0.04 -0.17 0.03 -0.01 -0.01 -0.05 -0.01 -0.24 0.19 0.07 0.16 -0.05 0.30 -0.36 -0.03 -0.44 0.27 0.85 1.09
Indonesia 0.22 0.48 0.04 -0.14 0.04 -0.07 0.17 0.28 0.02 0.16 0.52 0.23 0.30 0.30 0.04 -0.06 0.00 0.13 -0.09 -0.11 0.02 -0.21 0.09 0.05 0.07 -0.10 0.44 -0.33 0.02 -0.32 0.26 0.55 0.96
Turkey 0.33 0.73 0.23 -0.08 0.17 0.12 0.14 0.39 0.13 0.26 0.52 0.47 0.42 0.66 0.14 0.19 0.25 0.13 -0.03 0.03 0.01 -0.22 0.29 0.15 0.11 -0.11 0.62 -0.28 0.09 -0.14 0.52 0.82 1.21
Kenya 0.21 0.94 -0.14 -0.30 0.10 -0.15 0.20 0.36 -0.07 0.11 0.53 0.66 0.57 0.78 -0.07 -0.23 -0.04 -0.07 -0.11 -0.20 -0.20 -0.37 0.31 -0.01 0.15 -0.18 0.35 -0.38 -0.11 -0.30 0.24 1.02 1.25
Lebanon 0.62 0.70 0.12 0.15 -0.03 -0.30 0.34 0.72 0.29 0.47 0.37 0.46 0.44 0.71 0.02 -0.02 0.18 0.16 0.21 0.38 0.06 -0.16 0.13 -0.01 -0.05 -0.20 0.31 -0.70 -0.07 -0.44 0.53 0.93 1.35
India 0.42 0.67 -0.06 -0.24 0.03 -0.02 0.25 0.40 0.26 0.31 0.44 0.40 0.45 0.51 -0.06 -0.02 0.00 -0.10 -0.02 -0.12 -0.19 -0.38 0.21 0.01 0.07 -0.20 0.32 -0.34 0.10 -0.12 0.30 0.87 1.06
Pakistan 0.27 0.61 -0.07 -0.31 0.07 -0.04 0.23 0.15 0.07 0.37 0.50 0.36 0.36 0.50 -0.12 0.03 -0.02 -0.11 -0.12 -0.07 -0.33 -0.39 0.18 0.08 0.16 -0.20 0.23 -0.36 0.03 0.03 0.24 0.79 1.10
Hong Kong 0.28 0.65 0.08 0.09 -0.17 -0.05 0.04 0.37 0.18 0.27 0.39 0.11 0.60 0.66 -0.01 -0.03 0.11 0.15 0.18 0.15 0.13 -0.14 -0.03 -0.03 -0.08 -0.36 0.37 -0.43 0.18 -0.33 0.39 0.68 1.02
Sample
Unweighted 0.37 0.84 0.01 -0.12 0.06 -0.03 0.09 0.36 0.26 0.36 0.60 0.49 0.56 0.67 -0.06 -0.06 0.09 0.04 -0.04 0.00 -0.02 -0.28 0.16 0.08 0.10 -0.19 0.41 -0.35 0.07 -0.24 0.44 0.95 1.24

means

This article is protected by copyright. All rights reserved

Accepted Article
Note. Si=Sincerity, Fa=Fairness, Gr=Greed Avoidance, Mo=Modesty, Fe=Fearfulness, An = Anxiety, De =Dependence, Se=Sentimental, Ss=Social Self-Esteem, Sb=Social
Boldness, Sc=Society, Li=Liveliness, Fo=Forgivingness, Ge=Gentleness, Fl=Flexibility, Pa=Patience, Or=Organization, Di=Diligence, Pe=Perfectionism; Pr=Prudence,
Ae=Aesthetic Appreciation, In=Inquisitiveness, Cr=Creativity, Un=Unconventionality, Alt=Altruism, D6 = Ds computed on 6 factor scales, D25 = Ds computed on 25 facet scales.

Table 3

Age-Corrected Emotionality Scores and Standardized Differences in Emotionality by Ethnic Group

Age-corrected scores
Ethnic groups Ns Men’s age Women’s age Cohen’s 95% conf.
Men Women
d (W-M) Interval
Continent of origin Men Women M SD M SD M SD M SD
East Asian 2301 1923 27.5 9.1 26.6 8.7 -0.03 0.57 0.27 0.53 0.55 0.520--0.586
Southeast Asian 3362 3189 26.2 7.7 25.8 7.6 -0.02 0.54 0.28 0.54 0.56 0.537--0.589
South Asian 5868 2015 26.5 8.6 26.4 8.4 -0.10 0.56 0.28 0.56 0.67 0.637--0.694
African 1438 690 26.7 7.2 27.9 7.7 -0.17 0.54 0.23 0.56 0.74 0.689--0.789
Arab/West Asian 948 634 26.0 7.5 25.0 7.8 -0.09 0.56 0.27 0.56 0.64 0.587--0.700
European 27094 15281 28.5 10.5 29.9 10.9 -0.26 0.58 0.29 0.56 0.96 0.950--0.973
Latin American 1723 763 26.1 8.3 26.9 9.4 -0.20 0.58 0.29 0.58 0.84 0.788--0.887

English speaking countries

East Asian 5621 5460 24.7 8.4 24.7 8.6 -0.09 0.57 0.34 0.57 0.76 0.734--0.777
Southeast Asian 3070 2482 24.4 7.8 24.9 8.6 -0.08 0.58 0.39 0.57 0.82 0.794--0.855
South Asian 6340 3462 25.4 9.2 24.8 8.8 -0.17 0.59 0.34 0.58 0.87 0.848--0.897
African 5523 2980 25.8 8.5 26.5 9.8 -0.33 0.57 0.27 0.59 1.04 1.015--1.066
Arab/West Asian 2410 1209 25.6 8.9 26.0 9.5 -0.18 0.59 0.31 0.60 0.83 0.788--0.870
European 84772 63321 32.2 13.9 34.1 14.5 -0.25 0.60 0.34 0.59 1.00 0.991--1.003

This article is protected by copyright. All rights reserved

Accepted Article
Latin American 9415 5340 25.4 8.7 26.9 10.0 -0.23 0.56 0.33 0.58 0.97 0.950--0.989

This article is protected by copyright. All rights reserved

 jopy_12551_f1-[Link]

English speaking countries

Accepted Article 1.20 Continents of origin

1.00

0.80
Sex Difference

0.60

0.40

0.20

0.00
A B C Europeans
-0.20
Ethnic groups

-0.40 (A)

English speaking countries

1.20 Continents of origin

1.00

0.80
Sex Difference

0.60

0.40

0.20

0.00
A B C Europeans
-0.20 Ethnic groups

-0.40 (B)

Figure 1. Illustrative examples: Personality sex differences for ethnic groups in two
geographic regions.

This article is protected by copyright. All rights reserved

Accepted Article

Figure 2. Relation between the GEI/HDI composite and national sex differences in Emotionality.
Red plots indicate English-Speaking Countries that were combined in the analyses involving ethnic
group differences (see the Method section)

This article is protected by copyright. All rights reserved

Accepted Article 1.20
English speaking countries
Continents of origin

1.00

0.80
Sex Difference (d)

0.60

0.40

0.20

0.00
East Southest South African Arabs European Latin
Asian Asian Asian West Asian American

-0.20

Figure 3. Sex differences in Emotionality for seven ethnic groups in two geographic regions.
Errors bars indicate 95% confidence intervals.

This article is protected by copyright. All rights reserved

Accepted Article 0.5
Men, Continent of origin
Men, English speaking countries
Women, Continent of origin
0.4 Women, English speaking countries

0.3

0.2
Mean on Emotionality

0.1

0
1 2 3 4 5 6 7

-0.1

-0.2

-0.3

-0.4

Figure 4. Mean age-corrected Emotionality scores (sex by ethnicity by geographic

region)

This article is protected by copyright. All rights reserved