Chapter 22: Cross-cultural issues in personality assessment
Filip De Fruyt & Bart Wille
Ghent University, Belgium
Chapter prepared for: Christiansen & Tett (Eds.) Handbook of Personality at Work
Running head: Cross-cultural issues
Address all correspondence to: Filip De Fruyt, H. Dunantlaan 2, B-9000 Gent, Belgium.
Phone +32 9 264 64 73; Fax +32 9 264 64 99. E-mail: Filip.DeFruyt@ugent.be
IMPORTANCE OF PERSONALITY ASSESSMENT IN A GLOBALIZED ECONOMY
Personality assessment is an established part of many selection procedures in Western
countries (Furnham, 2008; Sackett & Lievens, 2008), despite its questioned predictive validity
throughout the years. Opponents (Morgeson et al., 2007a, 2007b) have mainly inquired the
small magnitude of the predictive correlations and further criticized the fakeability of self-
descriptions in at-stake contexts such as job selection procedures. Proponents (Ones, Dilchert,
Viswesvaran, & Judge, 2007; Tett & Christiansen, 2007) meta-analytically reviewed validity
coefficients and concluded that validities (a) are not trivial, (b) generalize across different
contexts and cultures, with job characteristics acting as a moderator, (c) have demonstrated
utility for selection decisions, and (d) are not necessarily worse than validities obtained with
alternative methods of selection assessment (Rolland & De Fruyt, 2009). Although most
authors agree that many individuals will put their best feet forward when describing their
personality (in a selection context), there are varying opinions on how to handle and consider
impression management. In addition to selection, personality assessment is used more and
more in the context of career development and coaching, so its prominence and impact in the
Industrial and Organizational (IO) field is steadily increasing. Given the range of criteria that
are predicted by traits, it is to be expected that the frequency of personality assessments in IO
professional practice will amplify in a globalized economy, where direct and indirect contacts
with colleagues and customers representing diverse cultural backgrounds will be the norm
rather than the exception. Especially this multi-cultural context generates a series of questions
and challenges beyond the description of personality differences among members of a single
culture. With respect to personality description, questions at stake include: (1) what kind of
trait model (and accompanying operationalization) should one use to describe individual’s
personality within and across cultural contexts: Can one use inventories which are developed
in one culture to assess applicants with a different cultural background? (2) What norms
should one use when comparing individuals from diverse cultural backgrounds applying for
jobs in which they will have to collaborate intensively? (3) Do applicants from diverse
cultural backgrounds perceive assessment contexts differently? In other words, are self-
enhancing strategies in personality descriptions in development or selection contexts
perceived alike across cultural groups? (4) What about the accuracy of personality stereotypes
of cultural groups? Given their potential impact in selection processes, it is important to know
whether such stereotypes reflect a kernel of truth, or do not match observed differences
among cultural groups. With respect to the predictive validity of personality, a key question is
whether culture acts as a moderator of personality-criterion relationships.
The current chapter will first explore the two key constructs, i.e. culture and
personality, examining major models describing basic dimensions of culture and introducing a
model assumed to tap the common core of personality differences observable within and
across cultures respectively. The subsequent section reviews personality findings that showed
to be largely universal across cultures. The next part discusses methodological and
psychometric requirements for comparing personality scores across cultures, followed by an
analysis of the importance of personality dimensions and mean-level personality differences
among cultures. Tett and Burnett’s trait-based interactionist model of job performance (2003)
is subsequently discussed, taking into account the potential impact of culture-level variables.
The implications of these findings for the IO professional practice are discussed in a
practitioner’s window. The chapter closes with a section identifying major knowledge gaps
and perennial issues in the field of cross-cultural personality assessment in IO psychology.
CULTURE AND ITS CORE DIMENSIONS
The definition of culture, how to distinguish among cultural groups, and the kind of
core dimensions that are necessary to describe cultures have been the subject of intensive
debate and research the past decades. Matsumoto (2000) provided an overarching description
integrating different key attributes and defined culture as: “A dynamic system of rules,
explicit and implicit, established by groups in order to ensure their survival, involving
attitudes, values, beliefs, norms, and behaviors, shared by a group but harbored differently by
each specific unit within the group, communicated across generations, relatively stable but
with the potential to change across time” (p. 24). This definition clearly acknowledges that
individuals within a particular culture differ in terms of assimilating and manifesting various
cultural attributes and further underscores that cultures have the potential to change over time.
Both attributes affect how traits will have to be delineated from observable behaviour. There
have been several attempts to investigate core dimensions of cultural differences and cultural
value frameworks in particular [for an excellent review see Nardon and Steers (2009)]. Two
of these models were specifically developed within an IO framework and had considerable
impact on this area, i.e. Geert Hofstede’s four-dimensional model of cultural differences
(Hofstede, 1980, 2001) and the work of the GLOBE group (R. J. House, Hanges, Javidan,
Dorfman, & Gupta, 2004).
In the sixties and seventies Hofstede (Hofstede, 1980, 2001) had access to
international survey-data completed by a large sample of service and marketing personnel
employed in 40 countries of a firm initially referred to as ‘Hermes’ (later on revealed to be
‘IBM’). The survey was intended to assess and compare morale across divisions of IBM
located in multiple countries. Hofstede factor analyzed aggregated scores across employees
within these 40 societies and found that four major dimensions best represented the variance.
The first dimension, ‘Power distance’, reflects how societies find solutions to deal with the
basic problem of human inequality. Cultures characterized by high power distance are
organized very hierarchically often with a set of formal rules on how to navigate within this
hierarchy. In cultures with high power distance, people accept authority and comply with
orders and directions given by those higher in the hierarchy. A second dimension
‘Uncertainty avoidance’ describes how cultures cope with stress in the face of an unknown
future. Societies characterized by high uncertainty avoidance will invest in different
programmes and institutions to deal with harm and disaster; they value stability and do not
tolerate deviant ideas and behaviour. One of the most well-known dimensions of Hofstede’s
model is ‘Individualism-collectivism’, describing how individuals are integrated into primary
groups. In collectivistic cultures, persons’ identity is strongly bound to family relationships
and the ‘in-group’ to which one belongs, whereas in a more individualistic societies a
person‘s identity is more related to individualistic strivings and achievements. In collectivistic
societies, the group will take care of the person, whereas in individualistic societies, people
have to take care of themselves and their direct family. Finally, the fourth factor ‘Masculinity-
femininity’ describes how societies are organised around the division of emotional roles
between men and women. In more masculine societies, assertiveness, making career and
earning money are considered important, whereas more feminine societies value cooperation
and getting along. Hofstede’s research and dimensions were first criticized to have a strong
Western bias, because African and Asian countries were underrepresented among his initial
set of 40 countries. The Chinese Culture Connection (1987) challenged this ethnocentric
viewpoint with a more emic research program, proposing an additional factor: ‘Confucian
work dynamism’. Hofstede (2001) later added this dimension to his model under the label
‘Long-term versus short-term orientation’, reflecting differences among cultures in the choice
of focus for people’s efforts: the future or the present.
Hofstede ranked different cultures in terms of their scores on the cultural values
dimensions (Hofstede, 2001; p. 500; Exhibit A.5.1). For example, the US were ranked as
highly individualistic (rank 1 out of 53), lower on power distance (rank 38), higher on
masculinity (rank 15), and lower on uncertainty avoidance (rank 43) relative to 53 other
countries, whereas Japan was ranked as less individualistic (rank 22-23 out of 53), somewhat
more power distant (rank 33), top in masculinity (rank 1) and higher uncertainty avoidant
(rank 7). These country rankings were used in numerous studies and correlated with other
national-level data such as indicators of economic activity and wealth, health and happiness,
but also aggregate personality and national character ratings. However, as Matsumoto’s
definition of culture underscored, cultures are dynamic entities, and an update of this ranking
of countries as well as a re-examination of the comprehensiveness and the content of
Hofstede’s model may be required after three decades of fast economical, societal and
A second major research effort in the search for the dimensions of cultural values has
been undertaken by a consortium of 160 researchers from many parts of the world under the
direction of Robert J. House. The GLOBE research (Global Leadership and Organizational
Behavior Effectiveness) program (Chokhar, Brodbeck, & House, 2007; R. J. House, et al.,
2004; R. House, Javidan, Hanges, & Dorfman, 2002) was designed to examine implicit
leadership theories and attributes of effective leadership in various cultural contexts. Data of
about 17.000 managers employed in 951 organizations in 62 societies across the world were
examined. In GLOBE, culture is defined as: “the shared motives, values, beliefs, identities,
and interpretations or meanings of significant events that result from common experiences of
members of collectives that are transmitted across generations” (R. J. House & Javidan,
GLOBE defines nine major dimensions of cultural differences, plus an additional six
to describe leadership behavior. These nine dimensions are: institutional collectivism or “the
degree to which organizational and societal institutional practices encourage and reward the
collective distribution of resources and collective action”, in-group collectivism or “the
degree to which individuals express pride, loyalty and cohesiveness in their organizations or
families”, power distance, or “the degree to which members of a society expect and agree that
power should be stratified and concentrated at higher levels of an organization or
government”, performance orientation, or “the degree to which an organization or society
encourages and rewards members for performance improvement and excellence”, gender
egalitarism, or “the degree to which a society minimizes gender role differences while
promoting gender equality”, future orientation, or “the degree to which individuals in
organizations or societies engage in future-oriented behaviors such as planning, investing in
the future, and delaying individual or collective gratification”, humane orientation, or “the
degree to which members of a society encourage and reward individuals for being fair,
altruistic, friendly, generous, caring and kind to others”, assertiveness, or “the degree to which
members of a society are assertive, confrontational or aggressive in social relationships”, and
finally uncertainty avoidance, or “the extent to which members of a society seek certainty in
their environment by relying on established social norms, rituals and bureaucratic practices”
(R. J. House, Quigley, & de Luque, 2010; p. 118; Table 1).
These dimensions were subsequently used to cluster 61 nations participating in
GLOBE according to cultural values and beliefs into 10 a priori proposed clusters: South-
Asia, Anglo, Arab, Germanic Europe, Latin Europe, Eastern Europe, Confucian Asia, Latin
America, Sub-Sahara Africa, and Nordic Europe. This clustering received considerable
empirical support (Gupta, Hanges, & Dorfman, 2002). For example, the Arab cluster includes
Egypt, Morocco, Turkey, Kuwait, and Qatar, and these societies are found to be highly group-
oriented, hierarchical, masculine, and low on future orientation (Kabasakal & Bodur, 2002).
One of the major purposes of GLOBE was to examine leadership practices and how good
leadership was perceived within these clusters. For example, in the Arab cluster, mid-level
managers defined outstanding leadership as characterized by team-oriented and charismatic
features, but also that an outstanding leadership style is not reflected by extreme positions on
leadership traits (Kabasakal & Bodur, 2002). This example nicely illustrates how cultures
may differentially value leadership behavior and hence value the personality traits that are
associated with this competency (Bono & Judge, 2004). In addition, evidence is provided for
the distinction between the cultural values and the cultural practices part in the assessment of
the GLOBE dimensions, with only the values, but not the practices, being associated with
features of outstanding leadership behavior. Javidan et al. (2006; p. 903) concluded that: “In
other words, leaders’ reported effectiveness is associated with the society’s cultural values
and aspirations, but the society’s effectiveness is associated with its cultural practices”.
Both Hofstede and the GLOBE consortium have been very influential in alerting IO
psychologists to the notion of cultural differences, providing the field with dimensional
models to denote cultural attributes. Both approaches strongly contrast with the many ‘easy’
operationalizations of culture that simply rely on race or nationality as markers of an
individual’s culture and disregard the cultural heterogeneity beyond directly accessible
markers. Different comparative reviews and mutual criticisms (Hofstede, 2006, 2010;
Javidan, et al., 2006) have further sharpened our thinking about cross-cultural differences, its
applications and its challenges. It is clear now that cultural value dimensions have main
effects on a series of outcome variables, such as emotions, attitudes and perceptions,
behaviors and job performance (Taras, Kirkman, & Steel, 2010), but culture values can also
moderate relationships between other predictors (e.g. personality) and these outcomes.
ONE MODEL FITS ALL? THE UNIVERSAL STRUCTURE OF PERSONALITY
Personality and culture show reciprocal relationships, with the expression of
personality traits affected by culture and individuals’ unique personality affecting and shaping
that culture (Chao & Moon, 2005). The Five-Factor Theory distinguishes between basic
tendencies and characteristic adaptations (McCrae & Costa, 1996). Basic tendencies (i.e. the
traits from the Five-Factor Model) are considered as causal entities that are largely
independent from cultural factors but are assumed to influence various characteristic
adaptations, such as interests, motives, work competencies and values. Individuals’ value
systems are also shaped by cultures’ shared meaning systems. An important assumption of the
Five-Factor Theory is that the structure of personality should be relatively invariant across
different cultures. The question at stake becomes whether this FFM can be used for cross-
cultural personality descriptions?
Cross-cultural replicability of the Big Five
The past decades, personality psychologists reached a relative consensus on the
importance of five major personality dimensions, the so called Big Five, to represent the
major variance among personality descriptions. Lexical studies conducted in different
languages mainly converged on the nature and number of factors suggesting that extraversion,
agreeableness, neuroticism, conscientiousness and intellect were necessary and sufficient to
account for the communalities enclosed in self- and peer descriptions on large sets of
personality descriptive adjectives (Goldberg, 1982). From a different angle, Costa and
McCrae (1992) introduced the Five-Factor Model of personality (FFM), complementing their
initial NEO model, already capturing the domains of neuroticism, extraversion and openness
to experience, with the domains of agreeableness and conscientiousness. Openness to
experience deviates from the lexical Big Five intellect factor because it reflects a broader
content including receptivity to a range of experiences and a fluid and permeable structure of
consciousness that is not well represented in the natural language by trait adjectives (McCrae,
1994). They further demonstrated that this FFM was able to accommodate all main factors
recurrently observable across major personality inventories. Although the terms ‘Big Five’
and ‘FFM’ are often used interchangeably to refer to the consensus on their importance as
major constructs of personality, they have clearly different historical roots.
Cross-cultural psychologists have pointed our attention to the distinction between emic
and etic approaches with respect to the use of psychological constructs in cross-cultural
research (Matsumoto, 2000). Indeed, most FFM research has been etic in origin, examining
the replicability of instruments that were to a large extent originally designed in Western
cultures, in a multitude of countries across the globe. Such investigations have been done
widely with the NEO-PI-R (Costa & McCrae, 1992; Rolland, 2002) and its successor the
NEO-PI-3 (McCrae, Costa, & Martin, 2005). There is massive evidence that the FFM
structure is replicable in self and peer ratings in cultures across all continents, at least when
administered to people with sufficient reading command of the native language. Moreover,
the FFM structure also showed to be valid across different age groups from adolescence
(NEO-PI-3; De Fruyt, De Bolle, McCrae, Terracciano, & Costa, 2009) to adulthood (NEO-PI-
R; McCrae & Terracciano, 2005b), making it the model par excellence to study gender and
developmental trends from a cross-cultural perspective (see further in this chapter). De Fruyt
and colleagues (2006) further illustrated that the factor structure of the NEO-PI-R kept
preserved across the IQ distribution in selection contexts, whereas Marshall and colleagues
(2005) demonstrated that the NEO-structure was replicable across different administration
contexts, including not-at-stake situations, career counseling (mild at-stake) and selection
(high-stakes context). Together, these studies suggest that the FFM is applicable for cross-
cultural assessment of personality in IO applications.
Dimensions beyond the Big Five
There have been also emic, also called indigenous, approaches towards personality
description, where researchers started within a particular culture to comprehensively sample
personality descriptors bottom-up and examine their underlying structure, rather than
importing (top-down) a personality inventory designed in a different culture. For example,
Meiring and colleagues (Meiring, Van de Vijver, Rothmann, & De Bruin, 2008) examined the
structure of personality descriptors in 11 languages spoken in South-Africa, Church and his
team (Katigbak, Church, Guanzon-Lapena, Carlota, & del Pilar, 2002) suggested additional
indigenous dimensions to account for the commonality in Filipino college student personality
ratings, and Benet-Martinez and John (1998) examined the personality structure in the
Spanish language in Hispanic minorities. Overall, these authors demonstrated that there is
evidence for a common cross-cultural personality descriptive vocabulary, next to emic traits
that may have particular relevance and importance for a specific culture. Cheung and Leung
(1998), however, strongly argued in favor of Chinese indigenous personality measures.
Personality psychologists have suggested additional factors beyond the Big Five as
well within Western cultures. For example, Paunonen and Jackson (2000), reconsidering an
initial selection of personality adjectives made by Saucier and Goldberg (1998), suggested 10
possible dimensions that are difficult to position within the Big Five: (1) religious, devout,
reverent, (2) sly, deceptive, manipulative, (3) honest, ethical, moral, (4) sexy, sensual, erotic,
(5) thrifty, frugal, miserly, (6) conservative, traditional, down to earth, (7) masculine-
feminine, (8) egoistical, conceited, snobbish, (9) humorous, witty, amusing, and (10) risk-
taking and thrill-seeking. Likewise, further elaborating within the lexical research paradigm,
Ashton, Lee and Son (2000) suggested ‘honesty-humility’ as a sixth major factor, reflecting
attributes such as fairness and sincerity. Reviewing these supplements, it is unclear whether
some are to be considered as facets or blends of the Big Five or are indeed replicable major
dimensions above and beyond the basic five. For example, a reanalysis of the data initially
used by Ashton and colleagues (2004) as support for the honesty-humility dimension, by the
same group of authors (except, Ashton and Lee), showed that no more than three factors, i.e.
extraversion, agreeableness and conscientiousness, were replicable across 14 datasets from 12
different cultures, with ‘honesty-humility’ turning up as a facet of agreeableness (De Raad et
al., 2010). This re-analysis further demonstrated that also a well-known personality factor like
‘neuroticism/emotional stability’, which is represented in almost every single theory or model
on personality differences, was not replicable. Overall, this work by de Raad and colleagues
(2010) nicely illustrates the limits of the lexical paradigm analyzing the passive personality
descriptive vocabulary to denote the major dimensions of personality.
Up until now, it is unclear whether these additional or emic-derived traits predict
criteria of importance for IO psychology, beyond the dimensions and facets already enclosed
in broad personality taxonomies for which there exists cross-cultural support. Contrary to the
personality field, comprehensiveness is not necessarily the most important requirement for a
personality descriptive taxonomy to be used in IO psychology. For applied purposes, such as
selection assessment, predictive validity is ultimately most important, and rather than being
comprehensive, a personality measure should reflect those traits that are most useful to
understand the criteria of interest such as job performance or leadership emergence for
example. This implies that a personality measure fit for IO applications, should assess several
facets of conscientiousness, such as ‘self-discipline’, ‘achievement’, ‘planning’, but also traits
that form blends between conscientiousness with other broad personality domains such as
‘control’ (forming a blend with neuroticism) and ‘proactivity’ (blending with extraversion)
(Rolland & De Fruyt, 2009), because there is evidence that Conscientiousness and related
traits are predictors of work performance.
General versus contextualized personality inventories in IO psychology
Many personality inventories in the past were developed from a clinical angle, e.g. the
Eysenck Personality Questionnaire (EPQ; Eysenck & Eysenck, 1975) or the Minnesota
Multiphasic Personality Inventory (MMPI; Butcher & Williams, 2000), followed by a
generation of inventories focusing at the description of trait variation observable in the
general population such as the NEO-PI-R (Costa & McCrae, 1992) or the scales from
Goldberg’s International Personality Item Pool (IPIP; Goldberg et al., 2006). The legislation
on job selection assessment in many countries, including the US (Americans with Disabilities
Act) and many European countries (e.g. France; Loi n° 92-1446 du 31 Décembre 1992),
explicitly require that assessments should have demonstrable relevance for the work context.
The implication for personality assessment is that personality inventories administered in the
context of job selection or career coaching should be directly relevant to judge on an
individual’s suitability for a particular job or contribute to an understanding of functioning at
work. General personality inventories, however, often include many items that are not
immediately work-related, making such instruments potentially contestable when used in IO
professional practice. From a different angle, in an attempt to increase validities of personality
assessments for IO applications, Lievens, De Corte and Schollaert (2008) convincingly
demonstrated that the inclusion of a frame-of-reference substantially increases the validity of
personality descriptions to predict performance criteria. They showed that adding a frame-of-
reference to the general instructions for personality description (e.g. “Describe how you
generally behave at work or at school”) or adding word tags to the items (e.g. “I am curious at
work”) leads to higher validities for predicting criteria considered important in the framed
These two evolutions led to an increased use of contextualized personality inventories
specifically designed to assess personality at work either through the administration of work-
related personality descriptive items or through the addition of a ‘work-frame’ to the
instructions or a combination of these. Introducing ‘work context’ in the items, on top of the
personality behavioral descriptive part, makes such inventories inevitably more culture-
bound. For example an item like: “A negative evaluation at work bothers me for days” (as an
indicator of frustration tolerance) (PfPI: Rolland & De Fruyt, 2009) introduces an
organizational and cultural practice into a personality descriptive item. Merging context and
behavioral description introduces extra challenges to demonstrate equivalence of measures
across cultures (see further in this chapter).
Structure of maladaptive/dark side traits
The past years, IO psychology witnessed a growing attention for the assessment of
aberrant traits and personality dysfunction (De Fruyt & Salgado, 2003; Salgado & De Fruyt,
2005; Wu & Lebreton, 2011). This transfer followed a growing awareness in human resources
to pay more attention to dark side behaviors at work, partly accelerated by the multiple
examples of mismanagement and the economical crisis after the millennium. Before this shift,
human resources as a discipline was heavily under the influence of positive psychology, with
more attention for the bright than the dark side of functioning.
There have been few attempts to assess maladaptive aspects of personality in the work
context, with Robert and Joyce Hogan among the first to call the attention of IO psychologists
to the dark side of personality (R. Hogan, Hogan, & Roberts, 1996). In clinical psychology
and psychiatry, aberrant personality traits are described on Axis II of the Diagnostic and
Statistical Manual of Mental Disorders (DSM-IV-TR; American Psychiatric Association,
2000), which articulates ten specific personality disorders, including the paranoid, schizoid,
schizotypal, antisocial, borderline, histrionic, narcissistic, avoidant, dependent, and obsessive-
compulsive personality disorder. Recent developments in clinical psychology, however,
support the view that personality disorders do not represent qualitatively distinct categories,
but should be conceived as continua of personality tendencies (Van Leeuwen, Mervielde, De
Clercq, & De Fruyt, 2007) that affect broad areas in people’s lives, including behavior at
work (De Fruyt et al., 2009). The validity of general personality descriptive models, such as
the FFM, to understand personality pathology has been extensively investigated (Costa &
Widiger, 2002). This research line has demonstrated that general and maladaptive personality
traits substantially overlap and that personality disorders can be described along the FFM
dimensions, suggesting that differences between normality and abnormality are quantitative
rather than qualitative.
Although well documented in Western countries, this assumption has not been
examined widely outside North-America or Western Europe, except for a study by Rossier,
Rigozzi and the Personality across Culture Research Group (2008) replicating these
associations in 9 French-speaking African countries. The paradigm shift, in which personality
disorders are better understood dimensionally, together with the observation that general
personality traits also capture core features of personality pathology, suggests that these
constructs and assessment methodology might be applied successfully in IO psychology. The
cross-cultural replicability of the FFM, together with the work by Rossier and his group
(2008), are a first step in examining whether the evaluation of personality dysfunction may
extend cross-culturally. Given the results of the GLOBE research group on the perception of
leadership, it is to be expected that narcissistic leadership will be more perceived
dysfunctional in for example the Arab cluster relative to Germanic European countries, where
outstanding leadership is defined by team-oriented and charismatic features, in the absence of
extreme positions on leadership traits (Kabasakal & Bodur, 2002).
REPLICABLE FINDINGS ACROSS CULTURES
Due to the relative consensus on the cross-cultural replicability of more structural
aspects of personality, considerable progress has been made the past decade to examine cross-
cultural patterns of gender and age differences. Data have been accumulated through meta-
analytic summaries of convenience samples, but also via targeted sampling across various
cultures using a single comprehensive personality inventory. The major advantage of this last
approach is that one circumvents the necessity to classify different scales assumed to assess a
similar construct when compiling the meta-analytic database. The Personality Profiles of
Cultures (PPOC; McCrae & Terracciano, 2005a; McCrae & Terracciano, 2005b) and the
Adolescent Personality Profiles of Cultures Project (APPOC; De Fruyt, De Bolle, et al.,
2009), a consortium of international research partners collecting data with the NEO-PI-R
(Costa & McCrae, 1992) or its more reader and adolescent friendly version, the NEO-PI-3
(Costa, McCrae, & Martin, 2008; McCrae, Martin, & Costa, 2005), have considerably
contributed to this field. Given their comprehensive and hierarchical character, and
replicability across a range of cultures, the NEO measures are well suited to examine gender
and age differences across the globe.
Universal gender differences
In a follow-up on previous narrative (Maccoby & Jacklin, 1974) and meta-analytic
reviews (Feingold, 1994) of gender differences on a more narrow set of traits, Costa,
Terracciano and McCrae (2001) investigated gender differences in NEO-PI-R self-ratings
obtained in 24 samples of adults and 14 samples of young adults across the FFM domains and
their 30 facets. They further examined gender differences as a function of social economic
status indicators of cultures, including Hofstede’s (2001) dimensions, in addition to gross
domestic product, female literacy, life expectancy, and fertility rate, indicated by the number
of children. Although convenience samples, largely taken from Western cultures and often
with undergraduates serving as young adult samples, the data lent itself to an examination of
gender differences due to the replicable factor structure of the NEO-PI-R across countries.
Observed gender differences were further compared with gender stereotypes assessed with the
Bem Sex Role Inventory (Bem, 1974) to investigate whether stereotypes have some ‘kernel of
Costa et al.’s (2001) findings can be easily summarized as follows: (1) At the FFM
domain level, females score higher on neuroticism and agreeableness, and the orientation of
these differences also generalizes to their facets. For extraversion and openness, gender
differences seem to cancel out against each other at the domain level, but there are consistent
gender differences at the facet level. Men score higher on E5: excitement-seeking and E3:
assertiveness, whereas women have on average higher scores on E1: warmth, E2:
gregariousness, and E6: positive emotions. Men further obtain higher scores on O5: openness
to ideas, whereas women score higher on O2: aesthetics, O3: feelings and O4: actions.
Negligible gender differences are observed for conscientiousness. Important from the
perspective of the current chapter is that these patterns generalize within (across young and
older adults) and across cultures, suggesting stable cross-cultural patterns. (2) If gender
differences are observed, they are usually limited to half a standard deviation, with most
differences reflecting a quarter standard deviation. (3) There is a strong agreement between
gender stereotypes (Bem, 1974) and observed gender differences, underscoring the ‘kernel of
truth’ hypothesis regarding gender stereotypes. (4) Both nature and size of the differences are
largely consistent with previous literature on a more limited set of traits and meta-analytic
evidence described by Feingold (1994). The findings further suggest that gender differences
also generalize from young to late adulthood. (5) If gender differences show up to some
extent in personality ratings of one trait, the size of these differences generalizes across the
other traits, suggesting that gender differentiation orientation generalizes within a culture.
This finding inspired Costa and colleagues (2001) to rank societies in terms of gender role
differentiation, showing that Zimbabwe had the lowest gender role differentiation, with small
to negligible gender differences among traits, whereas Belgium showed the largest differences
across the FFM. (6) Costa et al. (2001) correlated this ranking of sex role differentiation with
the criteria characterizing cultures and found that if gender differences are observed, they are
more sizeable in countries with a larger gross domestic product, literacy and life expectancy
of women, and lower fertility. These findings are intriguing and surprising, because also the
Scandinavian countries such as Norway, Sweden and Denmark are at the top end of the
observed gender differentiation ranking. These countries were among the first Western
societies in action to reduce gender inequality and glass ceiling effects, and especially in these
countries gender differences are more pronounced. These findings obtained from convenience
samples and self-ratings were largely confirmed in research by the PPOC and APPOC
research teams examining gender differences in a much broader set of cultures (50 different
countries across all continents) in which individuals were requested to describe somebody
they knew well (McCrae & Terracciano, 2005b; Table 4, p.553).
Universal age differences
A parallel route was followed accumulating findings on cross-cultural age trends for
personality ratings, first starting with analyses of mainly convenience samples obtained from
a limited set of societies, followed by a more systematic description of age effects across a
broad range of cultures by the (A)PPOC research teams. McCrae and colleagues (1999)
started to examine whether the age trends observed in the normative NEO-PI-R sample
generalized across five additional cultures (Germany, Italy, Portugal, Croatia, and South-
Korea) further trying to figure out whether these age-trends reflect common maturation
processes (in the case of similar patterns across cultures) or whether age patterns were more
culture-bound (in the case of different age trends). In line with the patterns observed in the
US, neuroticism, extraversion and openness showed average declines with age in adulthood,
whereas agreeableness and conscientiousness showed mean-level increases. The magnitude of
these changes was small to moderate. These age trends were further confirmed at the FFM
domain level for extraversion, openness and conscientiousness in a broad set of 50 cultures by
the PPOC research team, underscoring the notion that these age patterns reflect either
common maturation patterns showing up relatively independent of cultural differences
(McCrae & Costa, 1996) or are bound to common cultural processes that assert relatively
common influences on traits across cultures.
ONE METHOD HITS ALL? MEASUREMENT CHALLENGES WHEN COMPARING
A series of measurement issues and bias have to be taken into account before
constructs and measures can be meaningfully compared across cultures. Cross-cultural
researchers have distinguished among construct, method and item bias (F. Van de Vijver & K.
Leung, 1997) and the absence of bias is referred to as equivalence or invariance. Church
(2010) provides an excellent introduction to the terminology and measurement challenges
within cross-cultural measurement.
Church (2010) describes that construct or conceptual bias occurs: “when the
definitions of the construct only partially overlap across cultures” (p. 154). For example, the
content of a construct like intelligence is in some cultures constrained to cognitive
functioning, whereas it includes more social competences in other cultures. The personality
trait of assertiveness has a more negative connotation in the Netherlands, Belgium and
Germany, though is considered mainly as a desirable and extraverted attribute in the US (De
Fruyt, Mervielde, Hoekstra, & Rolland, 2000). Church (2010) distinguishes among three
forms of method bias: sample, instrument and administration bias. Cross-cultural comparisons
may be distorted through sample differences on possible confounding factors, design
characteristics of the instrument (e.g. the use of Likert scales or the sorting of items across a
Q-sort format may be familiar in one culture, but less frequently adopted in another culture),
and finally the way the data are administered may be experienced differently by cultural
groups and induce response differences (Church, 2010). For example, selection assessments
may be perceived as more threatening in individualistic countries with a high power distance.
A third kind of bias is item bias or differential item functioning (DIF): “DIF occurs when
individuals with the same level or amount of a trait, but from different cultural groups, exhibit
a different probability of answering the item in the keyed direction” (Church, 2010; p. 154).
In a recent study, Church and colleagues (2011) examined DIF in factor loadings and
intercepts from a multigroup confirmatory factor analysis of NEO PI-R data obtained in the
USA, the Philippines, and Mexico, showing that 40 to 50 percent of the items exhibited some
form of DIF. Moreover, DIF at the item level also affected the facet level, suggesting that the
comparison of mean level facet and domain scores across cultural groups should be done with
In addition, Church (2010) defines different forms of equivalence, including
conceptual, linguistic and measurement equivalence. Different degrees of overlap between
how constructs are defined across cultures is indicated by conceptual equivalence, whereas
linguistic equivalence points to the accuracy of translations. For example, the NEO-PI-R item
“I wouldn’t enjoy vacationing in Las Vegas”, as a reverse indicator of E5: Excitement-
seeking, may have to be amended to better fit a local culture, when one would consider to use
the NEO-PI-R in let’s say Iran. Finally, different levels of measurement equivalence or
measurement invariance will have to be demonstrated (Vandenberg & Lance, 2000). In line
with the confirmatory factor analysis (CFA) framework, weak factorial or configural
invariance is demonstrated when the same number of latent constructs and the same pattern of
salient and nonsalient factor loadings is presented across (cultural) groups. Metric invariance
(strong factorial invariance) can be concluded when factor loadings (slopes) can be
constrained to be equal across cultures without significant loss of model fit (Church, 2010).
Finally, scalar invariance can be demonstrated when also the item intercepts are equal across
Steenkamp and Baumgartner (1998) have argued that mean scores of (cultural) groups
are only meaningfully comparable when configural, metric and scalar invariance have been
established, showing that the factorial structure (configural), the scale intervals (metric), and
the zero point of the scale are the same across different groups. Scalar equivalence is about
the meaning of scores for different groups (F. J. R. Van de Vijver & K. Leung, 1997), in other
words: does a particular raw score indicates the same level of a trait in different groups and
has the same interpretation in all cultures? If scalar equivalence is demonstrated, then we can
derive meaningful conclusions from such comparisons. The demonstration of some form of
scalar equivalence is hence a prerequisite for making comparisons among any groups
(McCrae & Terracciano, 2008).
The determination of scalar equivalence is hotly debated among cross-cultural
personality researchers (McCrae & Terracciano, 2005a). A main group of cross-cultural
psychologists uses multigroup CFAs (MCFAs) to examine scalar equivalence, although the
requirements for CFA are very stringent. Alternatively, Item-Response Theory (IRT)-based
methods to examine DIF can be used to establish scalar equivalence (Reise & Henson, 2003),
but large sample sizes are required and analyses become more complex when items with
Likert-scales have to be analyzed. Adopting these methods for comparing sets of personality
descriptive item across cultures learns that many items show DIF, but also that DIF forwards
to the facet level, and does not cancel out across multiple items compiling a facet (Church, et
al., 2011). A second way to demonstrate scalar equivalence is through the use of bilingual
retest studies, in which bilingual respondents administer a personality inventory twice. Under
the condition of equivalence, means for the two language versions of the inventory should be
equal (McCrae & Terracciano, 2005b). The MCFA approach is further criticized because it
generally assumes that the indicators of a trait are interchangeable and this is usually not the
McCrae and colleagues (McCrae & Terracciano, 2005a) and McCrae and Terracciano
(2008) proposed a different route to demonstrate scalar equivalence, and argue that scalar
equivalence is not an absolute property, but a matter of degree for which a pattern of evidence
should be demonstrated, preferably via several of the previously suggested methods, because
all have their specific drawbacks. Instead, they suggest a top-down approach where the group
means are considered as scale scores, and their construct validity is investigated. A potential
difficulty here is that one needs data from a large number of cultures and appropriate criteria
at the culture-level. McCrae and Terracciano (2008) argue that if one is able to pinpoint a
nomological network of convergent and discriminant validity for a culture-level construct, the
mean scores must have some degree of scalar equivalence (see further in this chapter).
Self-reports versus multi-informant ratings
Although observer ratings have been used frequently in personality research (Hofstee,
1994), this source of assessment input has been underresearched and underutilized in IO
psychology (Connelly & Ones, 2010). There are two reasons to assume that reports by
knowledgeable others (peers, supervisors or subordinates) will be used progressively more,
i.e. evidence for increased validity above and beyond self-descriptions and the expanding use
of 180 or 360 degrees feedback in the course of career development and coaching trajectories.
Barrick, Mount and Strauss (1993) were among the first to report in the international
literature that observer ratings of the FFM had validities to predict work performance that
were almost twice the validities of self-ratings in sales people. Similar findings were reported
by Oh and Berry (2009) using 360 degree ratings of managerial performance. Operational
validities for supervisor ratings predicting task and contextual performance were significant
for four of the FFM dimensions, except for agreeableness, and generally increased when
combined with peer and subordinate ratings. When further complemented with self-ratings,
the operational validities ranged from .23 (agreeableness) to .45 (openness to experience) for
task performance and from .37 (openness to experience) to .50 (extraversion) for contextual
performance. The adjusted multiple Rs for the FFM dimensions rated by all raters were .53
and .58 for managerial task and contextual performance respectively. These findings suggest
that the inclusion of observer ratings increases validity coefficients, and that this increase is
also function of the different rater perspectives. In comparison with studies relying on self-
ratings, also the other FFM dimensions show up as significant dimensions explaining facets of
work performance. Oh, Wong and Mount (2010) meta-analytically summarized validity
coefficients available in 16 studies reporting on 20 independent samples, with observer
personality ratings and work performance criteria rated by different sources. For all FFM
dimensions, including openness to experience, validities for observer ratings were higher than
for self-ratings, and increased with the number of raters available. The meta-analytic work by
Connelly and Ones (2010; Table 11) shows convergent results, and also pleas to involve
multiple raters to improve reliability and validity.
Despite evidence that observer ratings have incremental validity beyond self-
descriptions and that validity of the assessments increases with the number of observers, it is
not clear whether subordinate or 180 degree ratings are easy to obtain in all cultures. This is a
highly underresearched area in IO psychology. One can assume, for example, that in cultures
characterized by large power distance, inviting employees to rate the attributes of their
supervisor may be perceived as odd, whereas in more collectivistic cultures, peers may have
difficulty to perceive a target as an independent self, describing the target’s personality more
in terms of fulfilling (work) roles and relationships with significant others in the in-group
(Heine, 2001). Such cultural attributes may have a profound impact on the personality
descriptions and induce different forms of administration, construct and rating bias when
working with observer ratings.
Impression management tendencies and culture
In general, personality psychologists agree that candidates will put their best feet
forward in selection assessments affecting the means of personality scales. To reckon this
phenomenon, De Fruyt et al. (2009) argued to take the assessment context into account, and
compare an individual’s score relative to others’ scores obtained under similar assessment
conditions. For IO applications, the implication is that test developers will have to provide
different norm sets obtained in low, mild or high stakes assessment contexts. Personnel
coaching and development programs are usually considered in Western societies as mild at
stake situations, whereas selection assessment is usually conceived as a high stake situation,
although it remains an empirical question whether these conditions are perceived likewise
across the globe. Anyway, for within culture comparisons, locally built norms are necessary,
and there should be convergence between the context of application and the context in which
the normative data have been collected.
There is further evidence that cultures differ in terms of motivation for self-
enhancement. A recent meta-analysis across 91 cross-cultural comparisons by Heine and
Hamamura (2007) showed an average effect size of .84 in self-enhancement between Western
and East-Asian samples. These differences can be partly explained due to the different cost-
benefit ratio for self-enhancement for North-Americans versus East-Asians. Self-enhancement
contributes to self-esteem and generates positive feelings among North-Americans, but
negatively impacts upon East-Asians threatening their within group integration and
relationships (Heine & Buchtel, 2009). There is further evidence that East-Asians hold more
dialectical views about themselves, including positive and negative views, whereas North-
Americans underscore the positive views. Whether these self-enhancing tendencies also
differentially operate in selection contexts is unclear.
MEAN-LEVEL PERSONALITY DIFFERENCES ACROSS CULTURES
Aggregate personality ratings and geographical patterns
There is a long tradition of speculation about a geographical distribution of personality
traits; in other words: “where one lives reveals what one is like” (Allik & McCrae, 2004; p.
13), although there are hardly empirical studies comparing personality ratings across multiple
cultures. The main reason is that several requirements (see previously in this chapter) must be
fulfilled before mean trait ratings can be meaningfully compared. Although personality traits
have been mainly studied at the level of individuals, the past years witnessed a growing
attention for aggregate ratings of personality, i.e. a mean computed for a trait across a sample
of individuals living in a particular culture that is subsequently used as a variable
characterizing that cultural group. The level of analysis hence shifts from the individual to the
culture-level. If such differences across cultures would be replicable, systematic, and valid,
then such aggregate ratings may be of considerable value for IO psychological applications.
Assume, for example, that there would be systematic differences across cultures in terms of
aggregate ratings of conscientiousness, then one could examine whether such differences are
associated with culture-level variables such as gross domestic product, wealth, productivity,
or absenteeism data. Likewise, the demonstration of meaningful average personality
differences among US states would necessitate the compilation of specific norms per region.
Primary evidence for the existence of regional personality differences in the United
States has been provided already some decades ago by Krug and Kulhavy (1973) using
Cattell’s Sixteen Personality Factor Questionnaire (Cattell, Eber, & Tatsuoka, 1970), and
more recently by Plaut, Markus and Lachman (2002) using a measure of the Big Five.
Corroborating this research line, Rentfrow, Gosling and Potter (2008) examined regional
personality differences in scores on the Big Five Inventory (John & Srivastava, 1999) in an
impressive sample of near to 620.000 internet respondents. A comparative analysis across
these three studies learns that aggregate trait levels are to a considerable extent consistent
across different geographical locations for neuroticism and openness to experience and
somewhat consistent for extraversion and agreeableness, despite differences in sampling,
measures, and a time frame of 30 years (Rentfrow, 2010). No consistent patterns for
conscientiousness were observed. Moreover, regional personality differences were associated
with important culture-level variables, including social connectedness (social capital),
political orientation and health. For example, state-level agreeableness was correlated .35 (p
<.05), conscientiousness -.44 (p <.05), and state-level neuroticism -.52 (p < .05) with social
capital, and people living in left-leaning states were higher in openness and lower in
conscientiousness relative to right-leaning civilians (Rentfrow, 2010; Rentfrow, et al., 2008).
Taking a cross-cultural angle, Allik and McCrae (2004) analyzed NEO-PI-R self-
reports from 27.965 college-age and adult men and women from 36 different cultures. Allik
and McCrae (2004) considered means comparable because scalar equivalence was roughly
demonstrated through a set of bilingual studies showing similar personality profiles across
translations, together with evidence for the construct validity of within-culture aggregate
personality ratings (see further in this chapter). They found that standard deviations for the 30
NEO-PI-R facets were systematically larger among European cultures than among Asian and
Black African cultures. Multidimensional scaling showed personality traits to be
geographically distributed, with neighboring countries exhibiting more similar personality
profiles. A multidimensional scaling plot of 36 cultures, rotated towards a horizontal
dimension positively associated with extraversion and openness and negatively with
agreeableness, and a vertical axis associated with neuroticism and low conscientiousness,
showed a clear separation between European and American cultures on the right from Asian
and African cultures on the left. The US and Canada were located near the bottom on the right
of the plot, together with a Baltic and the Scandinavian countries. Although geographical
proximity grouping was not perfect, it was certainly not random. A similar analysis on data
collected with observer ratings in the course of PPOC (McCrae & Terracciano, 2005a),
followed by a rotation to maximize associations with extraversion (horizontal axis) and
neuroticism (vertical axis) again showed a plot of cultures that were historically and ethnically
related. Summarizing the patterns across these two studies, the first relying on self-reports and
the second on observer ratings, shows that Europeans and Americans are higher in
extraversion and somewhat higher in openness compared to Asians and Africans.
In addition to comparing means across cultures, one can also factor analyze aggregate
personality ratings from multiple cultures, also called ecological factor analysis (EFA).
McCrae and Terracciano (2005a) adopted EFA on aggregated personality ratings of
individuals from 51 cultures and showed that four of the FFM, neuroticism, openness,
agreeableness, and conscientiousness, replicated the individual-level structure, with
extraversion showing close approximation, loaded by five extraversion facets and some other
facets that did no load the individual-level extraversion factor. They concluded that the FFM
is not only applicable at the individual-level, but also that there is a culture-level FFM, with a
specific culture-level Extraversion factor that is somewhat different from the individual-level
Finally, Stankow (2011) used hierarchical linear modeling (HLM) to examine
individual, country and societal cluster differences on big five personality traits, attitudes,
values and social norms in a sample of 2029 students from across the globe. Instead of
computing an average per culture, HLM enables to decompose observed variance across
different nested levels. Individuals were nested under 45 countries (level 2), and these were
nested in 9 societal clusters (level 3) culled from GLOBE. Both personality traits (7.41% of
the variance) and values (7.48%) were only slightly affected by country and societal cluster
differences, and variance was mainly explained at the level of the individual, ranging from
87.23% for Openness to 95.77% for Agreeableness. Social norms were assessed with the 9
GLOBE dimensions. Also their variance was to a large extent explained at the level of the
individual (average of 84.07%), with 5.97% and 9.95% accounted for by the country and
societal cluster level respectively. The results reported by Stankov (2011) suggest that cultural
influences on big five personality trait scores are limited, although results should be
interpreted with caution because sample size is limited, especially at levels two and three of
Do aggregate ratings predict something meaningful?
In the course of the PPOC-project, McCrae et al. (2005a) correlated aggregate
personality observer ratings obtained with the NEO-PI-R (Costa & McCrae, 1992) with
culture-level variables. Aggregate personality ratings turned out to be replicable within
cultures and showed meaningful associations with Hofstede’s dimensions, values (Inglehart &
Norris, 2003; Schwartz, 1994), well-being, gross-domestic product and the human
development index. Several of these associations were replicated in APPOC (McCrae et al.,
2009). Aggregate observer means converged with aggregate self-reports for the domains of
neuroticism, extraversion and openness to experience, but not for agreeableness and
conscientiousness, although significant convergent associations were found for four
agreeableness and four conscientiousness facets.
The validity of aggregate traits and the nature of the previously described significant
associations with culture-level criteria have been subject of intense debate [for a discussion on
culture-level criteria associations with conscientiousness, see: (Heine, Buchtel, &
Norenzayan, 2008)]. As a reply, Mõttus, Allik and Realo (2010) examined associations
between self-reports on conscientiousness facets and a broad range of culture-level criteria
across 42 cultures, including the 36 cultures from McCrae (2002), expanded with 3 African
cultures, Lithuania, Poland, and Finland. Also the associations with the observer-ratings
reported in PPOC (McCrae & Terracciano, 2005a) were examined. They provided clear a
priori hypotheses about the expected relationships, examined associations at the facet instead
of the domain level, and used a range of criteria (17, e.g. atheism, smoking, democracy,
obesity, alcohol consumption, gross domestic product, …) that were really representative of
the culture and its population. Without correcting for gross domestic product, 29% of the
correlations were significant at p <.01. Controlling for national wealth reduced the number of
significant correlations by almost half. Confirmation of hypotheses was different across the
six conscientiousness facets and self- versus observer ratings.
National character ratings
A different type of culture-bound personality ratings are national character ratings, i.e.
descriptions of the personality of a typical individual representing a national or a cultural
group. Such descriptions can be auto or hetero stereotypes with the first reflecting ratings
provided by in-group members, whereas hetero stereotypes are provided by people of a
different culture. These national character ratings are subsequently compared to observed
descriptions of in-group members to examine whether such ratings have validity or are just
stereotypes in the eye of the beholder without a kernel of truth.
Terracciano and colleagues (2005) examined the correspondence between national
character ratings on a measure of the FFM and APPOC observer ratings on the NEO-PI-R
(McCrae & Terracciano, 2005a) across 49 cultures. There was no correspondence between the
two sets of ratings. For example, Indonesia, Nigeria, Turkey, Poland and Japan obtained the
highest national character scores for neuroticism, though the observed means (expressed in T-
scores) on the NEO-PI-R for neuroticism for these countries ranged from 47.8 (Nigeria) to
51.4 (Turkey). The authors concluded that national character ratings appeared to reflect
The inaccuracy of geographical personality stereotypes has been further confirmed in
studies by McCrae, Terracciano, Realo and Allik (2007) with Northern and Southern Italians,
and by Realo and colleagues (2009) comparing Russian self-reported averages with
perceptions by civilians of neighbor countries. Rogers and Wood (2010), however, did find
that Americans’ geographical personality stereotypes for openness to experience and
neuroticism show considerable accuracy when compared with the results reported by
Rentfrow and colleagues (2008), with above chance accuracy for agreeableness and
extraversion. They further show what regional indicators might contribute to accuracy such as
population density and political voting patterns. Rogers and Wood (2010) conclude their work
by saying that geographical personality stereotypes may have some accuracy under certain
DOES ‘BIG’ ALSO REFLECTS UNIVERSAL IMPORTANCE AND VALIDITY?
Although there is considerable support for the factors that are minimally necessary to
structure personality traits and help defining their nomological net, this evidence does not
imply that personality traits as concepts are perceived equally important in all cultures (Heine
& Buchtel, 2009), and all Big Five dimensions are equally important for understanding
personality at work across cultures.
People from different cultures do no equally weigh personality information. There is
evidence that people from more collectivistic societies rely more on situational factors and are
less inclined to use personality information than people from individualistic cultures to
explain differences in behavior (Heine & Buchtel, 2009; Morris & Peng, 1994). Although the
factor structure of traits seems to be roughly replicable across cultures, this does not imply
that all Big Five dimensions are equally important within each single culture. For example,
within Western-industrialized and individualistic societies, getting ahead traits such as
extraversion and conscientiousness may be considered more important, whereas in more
collectivistic cultures more communal and getting along traits like agreeableness may be
valued more. Likewise, it can be hypothesized that more interpersonal traits such as
extraversion and agreeableness will be esteemed differently as a function of the power
distance level of a culture. These examples clearly show that replicability of factor structure
across cultures and importance of factors within specific cultures are two different questions
and actually, there is a dearth of studies examining the importance of the Big Five dimensions
across cultures. Moreover, the significance of Big Five dimensions within a particular culture
may change over time. For example, Western industrialized countries in which traits like
extraversion and conscientiousness were considered important dimensions for adaptation and
functioning, may notice a shift towards increased importance of openness to experience
related traits such as innovation, creativity and self-reflection. Finally, the importance of
personality traits relative to other individual differences such as intelligence, attitudes, skills,
and values may change across time in a rapidly transforming world economy. The current
meta-analyses on predictor-criterion validities summarize validity coefficients reported in
individual studies published across a broad time period, often decades ago. Given the largely
changing economies of the past 20 years, it might be interesting to examine cohort differences
in validity coefficients.
Validity generalization is a crucial issue for IO applications and practices that are
similarly applied in different cultures. The majority of the meta-analyses on the predictive
validity of trait measures relied on individual studies conducted with Westerners (Barrick &
Mount, 1991; Connelly & Ones, 2010; J. Hogan & Holland, 2003; Oh, et al., 2010; Salgado,
1997). As far as we know, there is no meta-analytic evidence that validities of personality
measures generalize to non-Western cultures. Such confirmation is not only absent for the
FFM, but is also deficient for indigenous traits. For example, it would be interesting to
examine whether traits resulting from indigenous personality research in China (Cheung et al.,
1996), such as ‘interpersonal relatedness’, predict aspects of job performance, such as
contextual performance, better than the FFM. At the level of the FFM, it would be interesting
to investigate whether the same traits predict similar criteria across cultures, and whether the
magnitude of these predictive validities is moderated by cultural characteristics. For example,
Heine and Buchtel (2009) recently suggested that personality may be less predictive of
behavior in collectivistic cultures, due to the presumed larger impact of norms, prescribed
roles and pressure from the social network on the person’s behavior.
Tett and Burnett’s (2003; Figure 1, p. 503 ) trait-based interactionist model of job
performance can be used to better understand how culture may impact upon the validity of
traits to predict work behavior and job performance. They distinguish work behavior from job
performance, because the latter involves an evaluation within a specific context, that may be
valued differently across cultures. Work behavior may lead to intrinsic rewards for the
individual, due to the possibility to express his/her personality, whereas job performance leads
to extrinsic rewards such as salary, feedback and recognition from others.
Tett and Burnett’s trait-activation theory (2003) further distinguishes moderators of
the trait-work behavior relationship at the task, social and organizational levels. For example,
orderliness as a trait may be positively related to job performance for accounting tasks (task
level), in a team valuing precision and punctuality (social level), and in a detail and outcome
oriented company (organization level), but fail to predict performance in task, social and
organizational environments with a different focus. Moreover, personality expression may be
further affected by job demands (tasks and duties inherent in the job), distractors (factors
interfering with performance), constraints (factors restricting the manifestation of the trait),
releasers (factors counteracting a constraint), and facilitators (factors making triggers more
salient that are already in the situation). An accounting job includes many tasks demanding
orderliness and methodicalness, as such too much small talk with colleagues during working
hours may distract from the primary tasks; the increased use of information technology may
constrain the impact of personality, whereas an unforeseen bug in a program may counteract
such constraint, making individual differences again more salient; finally, dealing with a file
of a golden customer may act as a facilitator for precision and methodicalness.
Reviewing this model, it is clear that culture may impact upon the task, social, and
organizational moderators affecting the personality trait – work behavior relationship.
Moreover, culture will also affect the evaluation of work behavior and the extrinsic rewards
associated with good work performance. For example, in societies with many strong norms
and low tolerance of deviant behavior, the so-called tight cultures versus loose cultures with
weak social norms and higher tolerance of deviant behavior (Gelfand et al., 2011), one can
expect that bad performance leads to lower extrinsic rewards and negative feedback. This
tendency may be strengthened in individualistic societies, holding persons more accountable
for their individual contribution and strivings. In addition, it may be expected that tight
(Gelfand, et al., 2011) and high uncertainty avoidant and more feminine cultures (Hofstede,
2001) will put more constraints on the expression of individual differences, hence impacting
upon the strength of the trait-work behavior relationship that can be observed.
Finally, validity generalization studies often and correctly pay a lot of attention to the
predictor side of the equation. However, one should be also thoughtful about the nature and
construct validity of the criteria that one wants to predict. Job performance indicators may be
perceived very differently across the globe. For example, ‘waiter service’ in a restaurant is
defined and perceived very differently across cultures due to divergence in the way labor is
organized and multiple cultural expectations. What is considered as good performance in most
restaurants in the US (speed of service, removing plates as soon as one person around the
table has finished her/his meal, asking whether the meal is good and fits your expectations
multiple times in a time frame of 20 minutes), leads to dismissal in Western Europe where
eating is considered as a social event requiring time to enjoy the food and company, and
where you make a reservation for a table for the entire evening. In the US people are lining up
until a table is free and multiple shifts of service have to be completed at a single table on an
evening. This cultural difference is also reflected at the financial level: In many US
restaurants you are financially punished when one has to serve more than 5 people at a table,
whereas you may have a discount in Western-Europe. This example nicely illustrates how
(job) performance criteria may be perceived very differently across cultures.
KNOWLEDGE GAPS AND PERENNIAL ISSUES
The previous review has made clear that considerable progress has been made the past
20 years with respect to cross-cultural personality assessment, though it is also obvious that
these developments emerge at a slow pace and most often follow rather than precede calls and
questions emerging from the applied field of personality assessment. Four major challenges
can be identified requiring immediate attention.
Indigenous versus universal traits
It is clear that a common set of traits, integrated within the FFM, can be used to denote
personality differences across the globe, but it remains to be investigated whether indigenous
traits predict variance in IO criteria above and beyond the more universal traits. Despite the
universality of this trait taxonomy, we know almost nothing about the importance within
particular cultures of the major factors enclosed in the FFM. A similar problem arises with
respect to validity generalization. Most validation studies have been conducted with
Westerners, but validities remain to be demonstrated in for example African and South-
American cultures. For IO applications, comprehensiveness of trait taxonomies will be less
important, though inventories will have to include those traits that are most useful for
predicting IO criteria. Studies examining the moderating role of culture on personality-
criterion relationships could be conducted along Tett and Burnett’s trait-based interactionist
model of job performance (2003) described previously, distinguishing the major variables
affecting this relationship.
General versus work-related personality inventories
Legislation within several countries and recent research recommends the use of work-
related over general personality inventories for IO applications. The addition of a ‘work’
frame-of-reference to instructions and/or items (Lievens, et al., 2008) and complementing
self-descriptions of personality with, preferably multiple, observer ratings (Connelly & Ones,
2010; Oh, et al., 2010) enhances reliability and validity of the assessments in Western
cultures. The adoption of work-related personality inventories, including items referring to
observable work behavior, will facilitate the involvement of multiple raters such as
subordinates, direct colleagues or supervisors. It remains to be examined, however, under
what circumstances such observer ratings add validity, and whether societal and
organizational culture moderate such relationships. In addition, scalar invariance will have to
be demonstrated within cultures, before self- and observer ratings can be meaningfully
compared and integrated.
Heterogeneity within cultures
Although migration and various forms of intercultural transmission have been
universal phenomena throughout history, the way in which cultural differences are perceived
and have to be handled in societies has dramatically changed in the recent past. Whereas
immigrants were previously assumed to adapt and assimilate as quickly and profoundly as
possible to the language, habits and culture of the receiving society, Western societies
nowadays consider diversity and a plethora of cultural backgrounds as a strength that should
be taken into account, respected and sometimes preserved. As a result, societies have
definitely become more heterogeneous in terms of the cultural backgrounds of their members.
In addition, individuals within a particular society may belong to different (cultural) groups at
the same time or cultural boundaries may have become permeable and fuzzy. For example,
children from Moroccan immigrants born in Germany may share characteristics with the host
German culture, but will also resemble features and values from their Moroccan roots.
Moreover, people within a culture, may be member of multiple groups at the same time, i.e.
reflecting a different cultural heritage and background, social-economic status (raised in a low
SES family and via upward mobility moved to a higher class, or the other way around), and
gender. These different group attributes will interact and personality inventory developers and
assessment practitioners will have to face this complex reality. Practitioners and researchers
will have to disentangle, for example, whether poor psychometric problems observed in a
heterogeneous group within a single society are attributable to problems with understanding
particular items (due to insufficient language command of the visiting’s culture native
language) or reflect measurement inequivalence.
Differences between cultures and the feasibility of multi-cultural norms
Several studies were reviewed in this chapter suggesting that personality traits show a
geographical distribution within the US (Rentfrow, 2010) and across the globe (McCrae &
Terracciano, 2005a; Stankov, 2011). There are diverging opinions however, with respect to
the comparability of such means, requiring the demonstration of some form or a degree of
scalar equivalence. The heterogeneity of cultural backgrounds represented within societies
and the fact that individuals often belong to multiple groups (e.g. age, gender, and an ethnic
minority group) at the same time, introduces very complex ‘equivalence’ questions to be dealt
with. Given the increasing cultural diversity of the workforce, the global economy, and
contacts with customers from a broad range of cultures, it is to be expected that the
importance of personality traits will increase and it is not an understatement to conclude that
we are just at the beginning of a flourishing field of research and consulting. The challenge
for academia and research will be to take the lead in this debate and provide the applied field
with recommendations and workable suggestions.
Given the increasing multiculturalism within individual societies and the steadily
growing number of contacts across nations in the global economy, human resources
practitioners will be faced more and more with questions on culture’s consequences for the
description and comparison of personality. The previous overview has made a number of
points clear that may help the practitioner facing these questions.
a) The trait structure represented by the Five-Factor Model is valid to describe general
personality traits across different cultures. More indigenous dimensions may
supplement this description. Major age trends for the FFM traits are largely culturally
universal and gender differences seem to be most pronounced and generalizable in
b) There are a series of FFM or Big Five inventories available in different languages
(academic and commercially), although these are not substitutes for each other and
cannot be used interchangeably. For comparative purposes, practitioners should use
the same Big Five/FFM inventory across cultures, examining whether these
translations/adaptations meet (part of) the requirements for making such comparisons.
c) There are no compelling data on the importance of FFM traits across cultures. For
example, extraversion may be considered more important in individualistic societies,
whereas agreeableness may be valued more in feminine oriented cultures.
d) In addition to culture, one should also take into account the assessment context. There
is massive evidence that the assessment context (low versus medium or high stakes)
impacts upon the personality scores within Western cultures, necessitating specific
norms obtained in similar assessment contexts to make meaningful comparisons.
e) It is inconclusive whether self-enhancement/impression management strategies are
used differently across cultures and across contexts within these cultures.
f) Cultures do not differ dramatically in terms of mean level personality scores.
Differences between varying at stake contexts have probably a larger impact on the
distribution of personality scores than differences between cultures.
g) Individuals’ personality descriptions should be preferable compared against normative
distributions obtained from individuals from the same cultural background
administered the inventory in the same (low, medium or high stake) assessment
context. In the absence of such norms, preference should be given to norms taking into
account the assessment context, given the smaller magnitude of differences between
h) For multicultural selection, such as in the case of ex-patriots, it is also recommended
to compare individuals’ scores to the normative distributions obtained in the host
culture (as a supplement to point g). Likewise, for the selection of applicants from
diverse cultural backgrounds who have to work together, it is recommended to
assemble a cross-cultural normative set, representing the different cultural groups.
i) There are not enough studies in non-Western cultures to conclude that the validity of
personality traits to predict various forms of work behavior and performance is
universal in nature and strength. Tett and Burnett’s (2003) trait-based interactionist
model of job performance provides a valuable framework to understand how culture
may moderate this relationship.
j) In addition to paying attention to the predictor side of the equation, i.e. personality
traits, practitioners should also carefully analyze the nature of the criterion. Like
specified in Tett and Burnett’s model (2003), not all work behavior is valued equally
k) There is an increased use of contextualized and maladaptive personality measures, in
addition to general traits. Also the use of observer ratings in addition to self-ratings is
highly encouraged. Whether these new assessment practices are generalizable across
cultures remains an open question.
l) Finally, aggregate personality ratings make sense and are replicable, although do not
correspond to national character stereotypes. Practitioners should hence be very
cautious relying on stereotypes of cultural groups.
Allik, J., & McCrae, R. R. (2004). Toward a geography of personality traits - Patterns of profiles across
36 cultures. Journal of Cross-Cultural Psychology, 35(1), 13-28.
Ashton, M. C., Lee, K., Perugini, M., Szarota, P., de Vries, R. E., Di Blas, L., . . . De Raad, B. (2004). A six-
factor structure of personality-descriptive adjectives: Solutions from psycholexical studies in
seven languages. Journal of Personality and Social Psychology, 86(2), 356-366. doi:
Ashton, M. C., Lee, K., & Son, C. (2000). Honesty as the sixth factor of personality: Correlations with
Machiavellianism, primary psychopathy, and social adroitness. European Journal of
Personality, 14(4), 359-368. doi: 10.1002/1099-0984(200007/08)14:4<359::aid-
Association, American Psychiatric. (2000). Diagnostic and statistical manual of mental disorders (4th.,
text revision ed.). Washington, DC: Author.
Barrick, M. R., & Mount, M. K. (1991). The Big 5 Personality Dimensions and Job-Performance - a
Metaanalysis. Personnel Psychology, 44(1), 1-26.
Barrick, M. R., Mount, M. K., & Strauss, J. P. (1993). Conscientiousness and Performance of Sales
Representatives - Test of the Mediating Effects of Goal-Setting. Journal of Applied
Psychology, 78(5), 715-722.
Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical
Benet-Martinez, V., & John, O. P. (1998). Los Cinco Grandes across cultures and ethnic groups:
Multitrait multimethod analyses of the Big Five in Spanish and English. Journal of Personality
and Social Psychology, 75(3), 729-750. doi: 10.1037/0022-35188.8.131.529
Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional leadership: A
meta-analysis. Journal of Applied Psychology, 89(5), 901-910. doi: 10.1037/0021-
Butcher, J. N., & Williams, C. L. (2000). Essentials of MMPI-2 and MMPI-A interpretation (2nd edition
ed.). Minneapolis: University of Minnesota Press.
Cattell, R. B, Eber, H. W., & Tatsuoka, M. M. (1970). Handbook for the Sixteen Personality Factor
Questionnaire (16PF). Champaign, IL: Institute for Personality and Ability.
Chao, G. T., & Moon, H. (2005). The cultural mosaic: A metatheory for understanding the complexity
of culture. Journal of Applied Psychology(90), 1128-1140.
Cheung, F. M., & Leung, K. (1998). Indigenous personality measures: Chinese examples. Journal of
Cross-Cultural Psychology(29), 233-248.
Cheung, F. M., Leung, K., Fan, R. M., Song, W., Zhang, J-X, & Zhang, J-P. (1996). Development of the
Chinese Personality Assessment Inventory. Journal of Cross-Cultural Psychology(27), 181-199.
Chokhar, J. S., Brodbeck, F. C., & House, R. J. (2007). Culture and Leadership across the World: The
GLOBE Book of In-depth Studies of 25 Societies. Mahwah, NJ: Lawrence Erlbaum.
Church, A. T. (2010). Measurement issues in cross-cultural research. In G. Walford, E. Tucker & M.
Viswanathan (Eds.), The SAGE handbook of measurement (pp. 151-177). Los Angeles: Sage.
Church, A. T., Alvarez, J. M., Mai, N. T. Q., French, B. F., Katigbak, M. S., & Ortiz, F. A. (2011). Are
Cross-Cultural Comparisons of Personality Profiles Meaningful? Differential Item and Facet
Functioning in the Revised NEO Personality Inventory. Journal of Personality and Social
Psychology, 101(5), 1068-1089. doi: 10.1037/a0025290
Connection, Chinese Culture. (1987). Chinese values and the search for culture-free dimensions of
culture. Journal of Cross-Cultural Psychology(18), 143-174.
Connelly, B. S., & Ones, D. S. (2010). An Other Perspective on Personality: Meta-Analytic Integration
of Observers' Accuracy and Predictive Validity. Psychological Bulletin, 136(6), 1092-1122. doi:
Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory and Five-Factor Inventory
Professional Manual. Odessa, FL: Psychological Assessment Resources.
Costa, P. T., McCrae, R. R., & Martin, T. A. (2008). Incipient adult personality: The NEO-PI-3 in middle-
school-aged children. British Journal of Developmental Psychology, 26, 71-89. doi:
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(2),
Costa, P. T., & Widiger, T. A. (2002). Personality disorders and the Five-Factor Model of personality
(2nd ed.). Washington D. C.: American Psychological Association.
De Fruyt, F., Aluja, A., Garcia, L. F., Rolland, J. P., & Jung, S. C. (2006). Positive presentation
management and intelligence and the personality differentiation by intelligence hypothesis
in job applicants. International Journal of Selection and Assessment, 14(2), 101-112. doi:
De Fruyt, F., De Bolle, M., McCrae, R. R., Terracciano, A., & Costa, P. T. (2009). Assessing the Universal
Structure of Personality in Early Adolescence The NEO-PI-R and NEO-PI-3 in 24 Cultures.
Assessment, 16(3), 301-311. doi: 10.1177/1073191109333760
De Fruyt, F., De Clercq, B. J., Miller, J., Rolland, J. P., Jung, S. C., Taris, R., . . . Van Hiel, A. (2009).
Assessing Personality at Risk in Personnel Selection and Development. European Journal of
Personality, 23(1), 51-69. doi: 10.1002/per.703
De Fruyt, F., Mervielde, I., Hoekstra, H. A., & Rolland, J. P. (2000). Assessing adolescents' personality
with the NEO PI-R. Assessment, 7(4), 329-345.
De Fruyt, F., & Salgado, J. F. (2003). Editorial: Personality and IWO applications: Introducing
personality at work. European Journal of Personality, 17, S1-S3.
De Raad, B., Barelds, D. P. H., Levert, E., Ostendorf, F., Mlacic, B., Di Blas, L., . . . Katigbak, M. S.
(2010). Only Three Factors of Personality Description Are Fully Replicable Across Languages:
A Comparison of 14 Trait Taxonomies. Journal of Personality and Social Psychology, 98(1),
160-173. doi: 10.1037/a0017184
Eysenck, H. J., & Eysenck, S. B. G. (1975). The Eysenck Personality Questionnaire. Sevenoaks: Hodder
Feingold, A. (1994). Gender differences in personality - a metaanalysis. Psychological Bulletin, 116(3),
Furnham, A. (2008). HR professionals' beliefs about, and knowledge of, assessment techniques and
psychometric tests. International Journal of Selection and Assessment, 16(3), 300-305. doi:
Gelfand, M. J., Raver, J. L., Nishii, L., Leslie, L. M., Lun, J., Lim, B. C., . . . Yamaguchi, S. (2011).
Differences Between Tight and Loose Cultures: A 33-Nation Study. Science, 332(6033), 1100-
1104. doi: 10.1126/science.1197754
Goldberg, L. R. (1982). From Ace to Zombie: Some explorations in the language of personality. In C. D.
Spielberger & J. D. Butcher (Eds.), Advances in personality assessment (Vol. 1, pp. 203-234).
Hillsdale NJ: Erlbaum.
Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton, M. C., Cloninger, C. R., & Gough, H. G.
(2006). The international personality item pool and the future of public-domain personality
measures. Journal of Research in Personality, 40(1), 84-96.
Gupta, V., Hanges, P. J., & Dorfman, P. (2002). Cultural clusters: methodology and findings. Journal of
World Business, 37(1), 11-15. doi: 10.1016/s1090-9516(01)00070-0
Heine, S. J. (2001). Self as a cultural product: an examination of East Asian and North American
selves. Journal of Personality(69), 881-906.
Heine, S. J., & Buchtel, E. E. (2009). Personality: The Universal and the Culturally Specific. Annual
Review of Psychology, 60, 369-394. doi: 10.1146/annurev.psych.60.110707.163655
Heine, S. J., Buchtel, E. E., & Norenzayan, A. (2008). What do cross-national comparisons of
personality traits tell us? The case of conscientiousness. Psychological Science, 19(4), 309-
313. doi: 10.1111/j.1467-9280.2008.02085.x
Heine, S. J., & Hamamura, T. (2007). In searh of East Asian self-enhancement. Personality and Social
Psychology Review(11), 4-27.
Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Beverly
Hills, CA: Sage.
Hofstede, G. (2001). Culture's consequences: Comparing values, behaviors, institutions, and
organizations across nations (2nd ed.). Thousands Oaks, CA: Sage.
Hofstede, G. (2006). What did GLOBE really measure? Researchers' minds versus respondents' minds.
[Article]. Journal of International Business Studies, 37(6), 882-896. doi:
Hofstede, G. (2010). The GLOBE debate: Back to relevance. [Article]. Journal of International Business
Studies, 41(8), 1339-1346. doi: 10.1057/jibs.2010.31
Hofstee, W. K. B. (1994). Who should own the definition of personality? European Journal of
Personality, 8(3), 149-162.
Hogan, J., & Holland, B. (2003). Using theory to evaluate personality and job-performance relations:
A socioanalytic perspective. Journal of Applied Psychology, 88(1), 100-112.
Hogan, R., Hogan, J., & Roberts, B. W. (1996). Personality measurement and employment decisions -
Questions and answers. American Psychologist, 51(5), 469-477.
House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V. (2004). Culture, Leadership, and
Organization: The GLOBE Study of 62 Societies. Thousands Oaks, CA: Sage.
House, R. J., & Javidan, M. (2004). Overview of GLOBE. In R. J. House, P. J. Hanges, M. Javidan, P. W.
Dorfman & V. Gupta (Eds.), Culture, Leadership, and Organizations: The GLOBE Study of 62
Societies (pp. 9-26). Thousand Oaks, CA: Sage.
House, R. J., Quigley, N. R., & de Luque, M. S. (2010). Insights from Project GLOBE: Extending global
advertising research through a contemporary framewok. International Journal of Advertising,
House, R., Javidan, M., Hanges, P., & Dorfman, P. (2002). Understanding cultures and implicit
leadership theories across the globe: an introduction to project GLOBE. Journal of World
Business, 37(1), 3-10. doi: 10.1016/s1090-9516(01)00069-4
Inglehart, R., & Norris, P. (2003). Rising tide: Gender equality and cultural change around the world.
New York: Cambridge University Press.
Javidan, M., House, R. J., Dorfman, P. W., Hanges, P. J., & de Luque, M. S. (2006). Conceptualizing and
measuring cultures and their consequences: a comparative review of GLOBE's and Hofstede's
approaches. Journal of International Business Studies, 37(6), 897-914. doi:
John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement, and
theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory
and research (2nd ed., pp. 102-139). New York: NY: Guilford Press.
Kabasakal, H., & Bodur, M. (2002). Arabic cluster: a bridge between East and West. Journal of World
Business, 37(1), 40-54. doi: 10.1016/s1090-9516(01)00073-6
Katigbak, M. S., Church, A. T., Guanzon-Lapena, M. A., Carlota, A. J., & del Pilar, G. H. (2002). Are
indigenous personality dimensions culture specific? Philippine inventories and the five-factor
model. Journal of Personality and Social Psychology, 82(1), 89-101. doi: 10.1037//0022-
Krug, S. E., & Kulhavy, R. W. (1973). Personality differences across regions of the United States.
Journal of Social Psychology(91), 73-79.
Lievens, F., De Corte, W., & Schollaert, E. (2008). A closer look at the frame-of-reference effect in
personality scale scores and validity. Journal of Applied Psychology, 93(2), 268-279. doi:
Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of sex differences. Stanford, CA: Stanford
Marshall, M. B., De Fruyt, F., Rolland, J. P., & Bagby, R. M. (2005). Socially desirable responding and
the factorial stability of the NEO PI-R. Psychological Assessment, 17(3), 379-384.
Matsumoto, D. (2000). Culture and psychology: People around the world. Belmont, CA:
McCrae, R. R. (1994). Openness to experience - expanding the boundaries of factor V. European
Journal of Personality, 8(4), 251-272. doi: 10.1002/per.2410080404
McCrae, R. R. (2002). NEO-PI-R: data from 36 cultures: Further intercultural comparisons In R. R.
McCrae & J. Allik (Ed.), The five-factor model of personality across cultures (pp. 105-125).
New York: Kluwer Academic/Plenum Publishers.
McCrae, R. R., & Costa, P. T. (1996). Toward a new generation of personality inventories: Theoretical
contexts for the five-factor model. In J. S. Wiggins (Ed.), The five-factor model of personality:
Theoretical perspectives (pp. 51-87). New York/London: Guilford Press.
McCrae, R. R., Costa, P. T., de Lima, M. P., Simoes, A., Ostendorf, F., Angleitner, A., . . . Piedmont, R. L.
(1999). Age differences in personality across the adult life span: Parallels in five cultures.
Developmental Psychology, 35(2), 466-477.
McCrae, R. R., Costa, P. T., & Martin, T. A. (2005). The NEO-PI-3: A more readable revised NEO
Personality Inventory. Journal of Personality Assessment, 84(3), 261-270.
McCrae, R. R., Martin, T. A., & Costa, P. T. (2005). Age trends and age norms for the NEO Personality
Inventory-3 in adolescents and adults. Assessment, 12(4), 363-373.
McCrae, R. R., & Terracciano, A. (2005a). Personality profiles of cultures: Aggregate personality traits.
Journal of Personality and Social Psychology, 89(3), 407-425.
McCrae, R. R., & Terracciano, A. (2005b). Universal features of personality traits from the observer's
perspective: Data from 50 cultures. Journal of Personality and Social Psychology, 88(3), 547-
McCrae, R. R., & Terracciano, A. (2008). The Five-Factor Model and its correlates in individuals and
cultures. In F.J.R. Van de Vijver, D. A. Van Hemert & Y. Poortinga (Eds.), Multilevel analysis of
individuals and cultures (pp. 249-283). Mahwah, NJ: Erlbaum.
McCrae, R. R., Terracciano, A., De Fruyt, F., De Bolle, M., Gelfand, M. J., & Costa, P. T. (2009). The
Validity and Structure of Culture-Level Personality Scores: Data From Ratings of Young
Adolescents. Journal of Personality, 78(3), 815-838. doi: 10.1111/j.1467-6494.2010.00634.x
McCrae, R. R., Terracciano, A., Realo, A., & Allik, H. (2007). Climatic warmth and national wealth:
Some culture-level determinants of national character stereotypes European Journal of
Meiring, D., Van de Vijver, F., Rothmann, I., & De Bruin, D. (2008). Uncovering the personality
structure of the 11 language groups in South Africa: SAPI project. International Journal of
Psychology, 43(3-4), 364-364.
Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K., & Schmitt, N. (2007a).
Are we getting fooled again? Coming to terms with limitations in the use of personality tests
for personnel selection. Personnel Psychology, 60(4), 1029-1049. doi: 10.1111/j.1744-
Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K., & Schmitt, N. (2007b).
Reconsidering the use of personality tests in personnel selection contexts. Personnel
Psychology, 60(3), 683-729. doi: 10.1111/j.1744-6570.2007.00089.x
Morris, M. w., & Peng, K. (1994). Culture and cause: American and Chinese attributions for social and
physical events. Journal of Personality and Social Psychology(67), 949-971.
Mottus, R., Allik, J., & Realo, A. (2010). An attempt to validate national mean scores of
Conscientiousness: No necessarily paradoxical findings. Journal of Research in Personality,
44(5), 630-640. doi: 10.1016/j.jrp.2010.08.005
Nardon, L., & Steers, R. M. (2009). The culture theory jungle: divergence and convergence in models
of national culture. In R. S. Bhagat & R. M. Steers (Eds.), Cambridge handbook of culture,
organizations, and work (pp. 3-22). Cambridge, England: University Press.
Oh, I. S., & Berry, C. M. (2009). The Five-Factor Model of Personality and Managerial Performance:
Validity Gains Through the Use of 360 Degree Performance Ratings. Journal of Applied
Psychology, 94(6), 1498-1513. doi: 10.1037/a0017221
Oh, I. S., Wang, G., & Mount, M. K. (2010). Validity of Observer Ratings of the Five-Factor Model of
Personality Traits: A Meta-Analysis. Journal of Applied Psychology, 96(4), 762-773. doi:
Ones, D. S., Dilchert, S., Viswesvaran, C., & Judge, T. A. (2007). In support of personality assessment
in organizational settings. Personnel Psychology, 60(4), 995-1027.
Paunonen, S. V., & Jackson, D. N. (2000). What is beyond the Big Five? Plenty! Journal of Personality,
68(5), 821-835. doi: 10.1111/1467-6494.00117
Plaut, V. C., Markus, H. R., & Lachman, M. E. (2002). Place matters: Consensual features and regional
variation in American well-being and self. Journal of Personality and Social Psychology(83),
Realo, A., Allik, J., Lonnqvist, J. E., Verkasalo, M., Kwiatkowska, A., Koots, L., . . . Renge, V. (2009).
Mechanisms of the National Character Stereotype: How People in Six Neighbouring Countries
of Russia Describe Themselves and the Typical Russian. European Journal of Personality,
23(3), 229-249. doi: 10.1002/per.719
Reise, S. P., & Henson, J. M. (2003). A discussion of modern versus traditional psychometrics as
applied to personality assessment scales. Journal of Personality Assessment(81), 93-103.
Rentfrow, P. J. (2010). Statewide Differences in Personality Toward a Psychological Geography of the
United States. American Psychologist, 65(6), 548-558. doi: 10.1037/a0018194
Rentfrow, P. J., Gosling, S. D., & Potter, J. (2008). A theory of the emergence, persistence, and
expression of geographic variation in psychological characteristics. Perspectives on
Psychological Science(3), 339-369.
Rogers, K. H., & Wood, D. (2010). Accuracy of United States regional personality stereotypes. Journal
of Research in Personality, 44(6), 704-713. doi: 10.1016/j.jrp.2010.09.006
Rolland, J. P. (2002). The cross-cultural generizability of the Five-Factor Model. In R. R. McCrae & J.
Allik (Eds.), The Five Factor Model of Personality Across Cultures (pp. 7-28). New York: Kluwer
Rolland, J. P., & De Fruyt, F. (2009). PfPI: Inventaire de Personalité au Travail. Paris: ECPA.
Rossier, J., Rigozzi, C., & Group, Personality across Culture Research. (2008). Personality disorders
and the Five-Factor Model among French speakers in Africa and Europe. La Revue
Canadienne de psychiatrie 53(8), 534-544.
Sackett, P. R., & Lievens, F. (2008). Personnel selection. Annual Review of Psychology, 59, 419-450.
Salgado, J. F. (1997). The five factor model of personality and job performance in the European
Community. Journal of Applied Psychology, 82(1), 30-43.
Salgado, J. F., & De Fruyt, F. (2005). Personality in personnel selection. In A. Evers, N. Anderson & O.
Voskuijl (Eds.), The Blackwell handbook of personnel selection (pp. 174-198). Oxford, UK:
Blackwell Publishing Ltd.
Saucier, G., & Goldberg, L. R. (1998). What is beyond the big five? Journal of Personality, 66(4), 495-
Schwartz, S. H. (1994). Beyond individualism/collectivism: New cultural dimensions of values. In U.
Kim, H. C. Triandis, C. Kagitcibasi, S.-C. Choi & G. Yoon (Eds.), Individualism and collectivism:
Theory, method, and applications. Thousand Oaks, CA: Sage.
Stankov, L. (2011). Individual, country and societal cluster differences on measures of personality,
attitudes, values, and social norms. Learning and Individual Differences, 21(1), 55-66. doi:
Steenkamp, J.-B. E. M., & Baumgartner, H. (1998). Assessing measurement invariance in cross-
national consumer research. Journal of Consumer Research(25), 78-90.
Taras, V., Kirkman, B. L., & Steel, P. (2010). Examining the Impact of Culture's Consequences: A Three-
Decade, Multilevel, Meta-Analytic Review of Hofstede's Cultural Value Dimensions (vol 95, pg
405, 2010). Journal of Applied Psychology, 95(5), 888-888. doi: 10.1037/a0020939
Terracciano, A., Abdel-Khalek, A. M., Adam, N., Adamovova, L., Ahn, C., Ahn, H. N., . . . McCrae, R. R.
(2005). National character does not reflect mean personality trait levels in 49 cultures.
Science, 310(5745), 96-100.
Tett, R. P., & Burnett, D. D. (2003). A personality trait-based interactionist model of job performance.
Journal of Applied Psychology, 88(3), 500-517.
Tett, R. P., & Christiansen, N. D. (2007). Personality tests at the crossroads: A response to Morgeson,
Campion, Dipboye, Hollenbeck, Murphy, and Schmitt (2007). Personnel Psychology, 60(4),
Van de Vijver, F. J. R., & Leung, K. (1997). Methods and data analysis of comparative research. In J. W.
Berry, Y. H. Poortinga & J. Pandey (Eds.), Handbook of cross-cultural psychology: Vol 1:
Theory and method (Vol. 1, pp. 257-300). Boston: Allyn and Bacon.
Van de Vijver, F., & Leung, K. (1997). Methods and data analysis for cross-cultural research.
Thousands Oaks, CA: Sage.
Van Leeuwen, K. G., Mervielde, I., De Clercq, B. J., & De Fruyt, F. (2007). Extending the spectrum idea:
Child personality, parenting and psychopathology. [Article]. European Journal of Personality,
Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance
literature: Suggestions, practices, and recommendations for roganizational research
Organizational Research Methods(3), 4-70.
Wu, J., & Lebreton, J. M. (2011). Reconsidering the dispositional basis of counterproductive work
behavior: the role of aberrant personality. Personnel Psychology, 64(3), 593-626. doi: