Personality, Threat and Affective Responses
to Cultural Diversity
KAREN VAN DER ZEE* and INEKE VAN DER GANG
Department of Social and Organizational Psychology, Grote Kruisstraat 2/1, 9712 TS Groningen,
The Netherlands
Abstract
The present study tried to reconcile assumptions from Terror Management Theory that
individual differences in openness to diversity are enhanced by existential threat with own
recent findings suggesting that individual differences are diminished by threat. A model
was supported assuming that it is the nature of the threat that determines which pattern will
hold. We predicted that for stress-related but not for social traits, threat enhances
individual differences in reactions to diversity. Students were confronted with a videotaped
meeting of a homogeneous versus diverse work group. Threat was induced using a
Terror Management Intervention. Indeed, whereas for Emotional Stability individual
differences in responses to diversity were restricted to conditions of threat, for
Social Initiative, individual differences solely occurred under normal circumstances.
Copyright # 2007 John Wiley & Sons, Ltd.
Key words: personality; threat; mortality salience; diversity; teams
PERSONALITY, THREAT AND AFFECTIVE REACTIONS
TO CULTURAL DIVERSITY
The workplace is becoming more culturally diverse, both due to migration and to a growth
in international assignments. Empirical findings suggest that in organizations, differences
among people may result in less attraction and less social integration (O'Reilly, Caldwell,
& Barnett, 1989). Social Identity (Tajfel & Turner, 1986) and Self-Categorization Theory
(Turner, 1982) predict that individuals tend to classify their social environment into
subcategories and that they tend to create a favourable picture of their own category relative
to other categories, in order to retain a positive self-image. Observable differences are
European Journal of Personality
Eur. J. Pers. 21: 453­470 (2007)
Published online 3 January 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/per.619
*Correspondence to: Prof. Dr Karen I. Van Oudenhoven-Van der Zee, Department of Social and Organizational
Psychology, Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands.
E-mail: K.I.van.Oudenhoven-van.der.Zee@rug.nl
Copyright # 2007 John Wiley & Sons, Ltd.
Received 8 August 2005
Revised 29 October 2006
Accepted 5 November 2006
likely to result in subgroup formation, which may have a strong negative impact on group
members' affective responses and ultimately on group functioning (Milliken, Bartel, &
Kurtzberg, 2003; O'Reilly et al., 1989). Not surprisingly, levels of well being in diverse
groups are usually lower compared to homogeneous groups (e.g. Tsui, Egan, & O'Reilly,
1992; Van der Zee, Atsma, & Brodbeck, 2004a; see for a discussion Jackson, Stone, &
Alvarez, 1993). In the present study, we compared affective responses to a diverse versus a
homogeneous work team. We predicted that individuals show more negative and less
positive affective reactions to a diverse as opposed to a homogeneous work team
(Hypothesis 1a).
PERSONALITYAND AFFECTIVE RESPONSES TO CULTURAL DIVERSITY
Diversity may in general inhibit individuals' identification with work groups, but there
appear to be individual differences in the openness to different cultures and hence in the
ability to feel comfortable working in a group composed of members of different cultures
(Van der Zee & Van Oudenhoven, 2000). Recent research has pointed at the relevance of
personality traits to effective functioning in an intercultural work context (e.g. Arthur &
Bennet, 1995; Ones & Viswesvaran, 1997). In the present study, we examined individual
differences in affective responses to a diverse work environment. We focused on two traits
that have been detected as contributing to success in intercultural settings (Mol, Van
Oudenhoven, & Van der Zee, 2001; Van Oudenhoven & Van der Zee, 2002; Van der Zee &
Van Oudenhoven, 2000): Social Initiative and Emotional Stability. Social Initiative can be
defined as a tendency to actively approach social situations and to take initiatives. For
individuals high in Social Initiative intercultural situations may provide an arousing social
experience that they want to explore (e.g. Eysenck & Eysenck, 1985) and we therefore
predicted a positive influence of this trait on reactions to diversity. Hence, a related
construct is Extraversion that seems associated with parts of the brain that are related to
sensitivity for reward or pleasure (Gray, 1991). Second, Emotional Stability refers to a
tendency to remain calm in stressful situations versus a tendency to show strong emotional
reactions under stressful circumstances. It indicates the ability to deal with frustration,
stress, anxiety, pressure to confirm, social alienation and interpersonal conflict that seem
inherent to intercultural situations. A low level of Emotional Stability seems associated
with what in the literature has been referred to as anxiety (e.g. Pickering & Gray, 1999).
Anxiety is linked to high reactivity of the Behavioral Inhibition System (BIS) in the brain
that indicates a high sensitivity to stimuli that signal punishment. Individuals high on this
dimension will be highly responsive to cues that indicate threat and they are therefore
inclined to avoid rather than approach situations of danger and risk. Individuals high in
Emotional Stability seem to be less anxious in intercultural situations because they
appraise less of its threatening aspects (e.g. Van der Zee, Van Oudenhoven, & De Grijs,
2004b). We therefore also expected a positive impact of Emotional Stability on reactions to
diversity. Because both traits are linked to effectiveness in culturally diverse situations, no
individual differences were expected in response to the homogeneous situation. In sum, we
predicted that individuals high in Social Initiative and Emotional Stability show more
positive and less negative affect in response to a diverse team, as compared to low scorers,
whereas they will not differ from low scorers in their responses to the homogeneous team
(Hypothesis 2a).
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
454 K. Van der Zee and I. Van der Gang
THREAT AND AFFECTIVE RESPONSES TO DIVERSITY
Whereas favourable affective reactions to different cultures can be enhanced by
intercultural traits, they may be diminished by threat. The presence of readily detectable
differences within a group may in itself lower the degree of safety that members
experience. Under threat, such differences may be more than groups can cope with. An
important function of culture or a cultural worldview is to create safety. Within a culture,
norms and values emerge which protect those who behave according to these norms and
values. In intercultural situations, both predictability and the sense of sharedness associated
with one's own cultural group membership are threatened. An important aim of this study
was therefore to examine what will happen to individuals' openness to diversity when they
are confronted with threat. This question is relevant, because groups in organizations are
confronted with various kinds of threat on a regular basis, for example due to takeover
attempts, dangerous product failures, employee sabotage or down-sizing (e.g. Turner &
Horvitz, 2001).
The idea that threat diminishes openess to different cultures forms the basic premise of
Terror Management Theory (TMT; Greenberg, Pyszczynski, & Solomon, 1990; Solomon,
Greenberg, & Pyszczynsky, 1991). TMT assumes that humans are born with an instinctive
tendency to survive, but in contrast to animals are conscious of their vulnerability and
unavoidable death. The combination of an instinctive need to survive and the
consciousness of death create anxiety (Greenberg, Solomon, & Pyszczynsky, 1997).
This anxiety enhances the importance of the own cultural values and diminishes openness
to deviant cultural values. Studies within this paradigm have shown that the importance of
one's own cultural values increases, whereas tolerance towards deviant norms or different
cultural groups decreases when an individuaľs basic sense of safety is threatened
(Rosenblatt, Greenberg, Solomon, Pyszczynski, & Lyon, 1989, study 1; Greenberg et al.,
1990). For example, in a series of studies, Greenberg et al. (1990) showed that Christian
subjects rated Christian targets more positively and Jewish targets more negatively, and
that authoritarian subjects more strongly tended to derogate attitudinally dissimilar others,
when mortality was made salient.
The assumption of the present study was that threat has an impact on affective responses
to diversity. Following the TMT-paradigm, a mortality salience intervention was used in
order to introduce threat. Subjects were asked to write a few sentences about what they
think will happen to them when they die, and the emotions that the thought of death arouses
in them (e.g. Greenberg et al., 1993; Lieberman, 1999). We hypothesized that threat
moderates the impact of diversity on affective responses to diversity. More specifically, the
prediction was that individuals will respond with less positive and more negative affect to a
diverse team and with more positive and less negative affect to a homogeneous team in the
high threat than in the low threat condition (Hypothesis 3a).
THREAT, PERSONALITY AND AFFECTIVE RESPONSES TO DIVERSITY
Individuals with high scores on the intercultural traits seem to react more positively to a
diverse work environment. But what happens to individual differences in openness to
diversity when team members are facing additional threat? On the one hand one could
argue that, in response to additional threat, anxiety levels of group members may raise up to
a certain level, which may cause individuals, regardless of their personality to feel
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 455
threatened by diversity. It is reasonable to expect that under those circumstances almost all
humans may experience fear and anxiety, and may strongly need the protection of their
culture in order to retain a positive sense of self and the world around them. Consequently,
individual differences in responses to diversity may diminish (e.g. Van der Zee et al.,
2004b). On the other hand, however, one could argue that when a trait is a reflection of how
well an individual is able to buffer anxiety, individual differences may be enhanced when
additional threat is induced (Strachan, Pyszczysnski, Greenberg, & Solomon, 2001).
Both perspectives have been supported by empirical research. With respect to the first
perspective, in an earlier experimental study, we examined individual differences in
affective responses to descriptions of intercultural situations that varied in potential
stressfulness (Van der Zee et al., 2004b). We found that whereas under normal conditions,
individuals with high scores on intercultural traits showed more positive and less negative
affective reactions to this situation than low scorers, upon a Terror Management
Intervention (TMI) individual differences in emotional reactions to intercultural situations
disappeared. With respect to the second perspective, studies within the TMT paradigm
report higher sensitivity to threat among individuals low in self-esteem (Greenberg et al.,
1993), high in depression (Simon, Harmon-Jones, Greenberg, Solomon, & Pyszczysnski,
1996) and high in authoritarianism (Greenberg et al., 1990, study 2). Greenberg et al.
(1997) argue that particularly the trait of self-esteem provides protection against existential
terror. Individuals high on this trait are assumed to be less prone to respond with increased
worldview defense to reminders of their mortality. In TMT this perspective is referred to as
the `anxiety buffer hypothesis' (Greenberg et al., 1997).
The assumption of the present study was that it is the nature of the trait that determines
which perspective provides the best description of how differential responses to diversity
will be affected by threat. We argued that the first perspective will particularly hold for
social traits, whereas the second perspective provides a more adequate description for traits
related to the regulation of stress. As we argued, individuals high in Social Initiative are not
necessarily equipped with special abilities to deal with threat. They seem to react positively
to diversity, not because they feel capable of dealing with its `dangers', but rather because
they feel attracted to its challenging aspects (see Van der Zee et al., 2004b). To them, threat
to vital aspects of their existence as human beings causes them to experience fear and
anxiety and they do not possess special skills to deal with such feelings. By contrast, for
Emotionally Stable positive responses to diversity seem due to their ability to deal with its
threatening aspects. Additional threat simply means an increase in the amount of anxiety
they have to deal with and they seem well equipped to handle this. Indeed, whereas support
for the first perspective was found for the dimension `Openness' (Van der Zee et al.,
2004b), which clearly is a social dimension, the empirical evidence presented by Terror
Management theorists is particularly based on self-esteem (Greenberg et al., 1993), which
can be regarded as a lower level factor of Emotional Stability (Costa & McCrae, 1992), and
on neuroticism (Goldenberg, Pyszczynski, McCoy, Greenberg, & Solomon, 1999;
Strachan et al., 2001).
In sum, for both traits, we hypothesized a threat by personality by diversity three-way
interaction, but in different directions. For Social Initiative, we predicted that the tendency
of individuals high as compared to low in Social Initiative to show more positive and less
negative affect in response to a diverse team would occur in the low threat but not in the
high threat condition (Hypothesis 4a). Again, for this trait regardless of threat condition, no
individual differences in response to the homogeneous team were predicted. By contrast,
for Emotional Stability, building from the anxiety buffer hypothesis, we predicted that the
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
456 K. Van der Zee and I. Van der Gang
tendency of individuals high on this intercultural trait to show more positive and less
negative affect in response to a diverse team than low scorers would particularly occur
under conditions of high threat and would be less strong in the low threat condition.
Moreover, because individuals high in Emotional Stability seem less in need for the
protection of cultural standards of value in response to threat (Greenberg et al., 1997), we
predicted less positive and more negative affective responses to a homogeneous team
among individuals high as compared to low on this trait in the high threat condition
(Hypothesis 5a).
THREAT, PERSONALITY AND IDENTIFICATION
WITH DIVERSE GROUPS
In the present study, we were not only interested in participants' affective responses to the
experimental conditions, but also in the resulting patterns of identification with the group as
a whole and with their cultural subgroup. Both from Social Identity (Tajfel & Turner, 1986)
and Self-Categorization Theory (Turner, 1982), it can be derived that in culturally diverse
teams individuals may strongly identify themselves with the cultural subgroup, often at the
expense of identification with the team as a whole (see also Van der Zee et al., 2004a).
Indeed, it has been shown that, in diverse group contexts, intragroup competition and
negative attitudes towards out-group members may prevent group members from
identifying with the group as a whole (Abrams & Hogg, 1988; Brewer, 1995; see also Van
der Zee et al., 2004a). Therefore, it was predicted that subjects would show lower
identification with the team and higher identification with the own culture in response to a
diverse team than in response to a homogeneous team (Hypothesis 1b).
Second, parallel to the findings for affect, we expected that whereas individuals high on
the intercultural traits show higher identification with the team and lower identification
with the own culture upon confrontation with a diverse team, as compared to low scorers,
less individual differences in identification with the team and with the own culture would
occur in response to the homogeneous situation (Hypothesis 2b). Third, with respect to
threat, advocates of the TMT have argued that upon introduction of threat, identification
with the own cultural norms increases and tolerance towards deviant cultural norms or
different cultural groups decreases, because under those circumstances individuals more
strongly need the protection of the own cultural group (Rosenblatt, Greenberg, Solomon,
Pyszczynski, & Lyon, 1989, study 1; Greenberg et al., 1990; see also Arndt, Greenberg, &
Cook, 2002). Our prediction was therefore that the tendency of individuals to respond with
less team identification and more cultural identification to a diverse team than to a
homogeneous team would be more pronounced in the high threat condition than in the low
threat condition (Hypothesis 3b). Finally, again parallel to the findings for affect, we
predicted a threat by personality by team condition three-way interaction both for Social
Initiative and Emotional Stability but in different directions. For Social Initiative, we
hypothesized that the buffering influence of personality on negative relational responses to
diversity would disappear under the influence of threat. More precisely, we predicted that
the tendency of individuals high on Social Initiative to show higher identification with the
team and lower identification with the own culture upon confrontation with a diverse team,
would disappear under threat (Hypothesis 4b). For Emotional Stability, however, we
predicted that the buffering influence of personality on negative relational responses to
diversity would particularly occur under conditions of high threat, and would be much
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 457
weaker in the low threat condition. More precisely, we predicted that the tendency of
individuals high on Emotional Stability to show higher identification with the team and
lower identification with the own culture upon confrontation with a diverse team would
particularly occur under conditions of high threat (Hypothesis 5b).
METHOD
Participants and procedure
One hundred and sixty-two first-year undergraduate psychology students (80% female,
20% male)1
from the University of Groningen were randomly assigned to a 2 (Threat:
control vs. TMI) Â 2 (Diversity: homogeneous vs. diverse) design. All students received
course credit for participation in the study. In addition, students who were interested
received personal feedback on their personality scores. Students' age varied between 18
and 53 years (M  20.2, SD  3.4).
Figure 1 displays the experimental procedure graphically. First, respondents filled out
indicators of personality and of (pre-manipulation) affect. Second, half of the participants
received a mortality salience intervention. The remaining part of the sample received
a comparable instruction with respect to watching TV. This intervention was followed by a
filler task. Next, a video was presented portraying a meeting of either a homogeneous or a
diverse team. Finally, all participants filled out a questionnaire containing indicators of
identification and affect, and open questions asking them to reflect on the team.
Experimental conditions
TMI
As part of the first experimental manipulation, participants were asked to answer two
questions that were related to their own death. The first question was to write down which
emotions were evoked when they were thinking about their own death. Second, they
received the instruction: `Please give a detailed description of what--in your opinion--
happens to you when you physically die'. In the control condition they were asked to write
Figure 1. Overview experiment.
1
Preliminary analyses revealed no impact of gender on the study results.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
458 K. Van der Zee and I. Van der Gang
down which emotions were evoked when they were watching TV and to give a detailed
description of what happens to them when watching TV.
Diverse versus homogeneous work context
In the second experimental manipulation, participants watched a videotape with a scene
from a fictitious team meeting and were asked to imagine themselves as part of the team. In
order to enhance students' identification with the team, a meeting of a well-known
international fashion company that is regularly visited by students (Hennes & Mauritz) was
simulated. The team consisted of six members, three males and three females. Three native
Dutch members were similar across conditions; the remaining members were either of
native Dutch (homogeneous condition) or of non-native origin (Turkish, Maroccan and
Indonesian, diverse condition). Roles of the team members were played by students who
received a written script in order to prepare for their roles. The script was the same across
both conditions. The group for example discussed how to prepare for a celebrity show for
which Hennes & Mauritz was asked to provide clothes and to take care of dressing the
performers. In order to make both conditions as comparable as possible, scenes that
displayed the three actors that were similar across conditions were included in both
versions of the final tape.
Measures
Prior to the manipulations, subjects filled out indicators of personality and affect. First,
Social Initiative and Emotional Stability were measured using the Multicultural
Personality Questionnaire (Van der Zee & Van Oudenhoven, 2000). Social Initiative
(17 items; a  0.89) was measured by items such as `Is inclined to speak ouť () and `Is
often the driving force behind things' (). Sample items for Emotional Stability (20 items;
a  0.90) are `Can put setbacks in perspective' () and `Keeps calm at ill-luck' ().
Participants could give their answers on a 5-point scale, ranging from not at all applicable
[1] to totally applicable [5].
Pre- and post-manipulation positive and negative affect was measured by taking two
subsamples from an original list of 38 adjectives that described possible feelings (Ybema &
Buunk, 1995). Nine adjectives were used to measure pre-manipulation positive affect:
grateful, energetic, reassured, pleasant, calm, comforted, self-confident, encouraged,
content (a  0.61). The items for negative affect were uneasy, confused, depressed,
anxious, offended, ashamed, tensed, pessimistic and irritated (a  0.68). The adjectives
were in part a translation of the Multi-Affect Adjective Checklist (Zuckerman, 1960;
Zuckerman, Lubin, Vogel, & Valerius, 1964). Participants were asked which of the
described feelings they experienced at the moment. The scores for positive and negative
affect were the number of indicated positive and negative adjectives, respectively. Positive
and negative affect correlated negatively (r  0.51, p < 0.001).
After the first experimental manipulation, participants completed the Positive and
Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988), on which they
reported on 20 items how they felt at the moment. Following previous terror management
studies (see Greenberg et al., 1997; Solomon et al., 1991), the PANAS was included as a
filler task. The PANAS consists of two 10-item subsets (Watson et al., 1988), one
measuring positive affect (e.g. `inspireď) and one measuring negative affect (e.g. `tenseď).
Participants could respond on a five-point scale ranging from (1) not at all applicable to (5)
totally applicable. Factor analyses indeed provided support for two distinct factors that
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 459
explained 48.4% of variance (eigenvalue > 4.3). One item for positive affect appeared to
load highly on both factors (`exciteď) and was excluded from further analysis. Internal
consistencies of the final scales were 0.87 for positive affect and 0.86 for negative affect,
respectively. Both scales were significantly, but modestly interrelated, r  0.23, p < 0.05.
After the filler task, participants watched the videotape of the team meeting. Preceding
the videotape they received a short instruction in which the context of the video was
explained. Participants were told that they had to watch the video carefully, because later
on they would be asked to answer some questions regarding its content. Following the
videotape, participants were asked questions pertaining to the dependent variables. Affect
was measured with a second subset of the same measure that was described for
pre-manipulation affect. Now the adjectives that were used to measure positive affect were
hopeful, cheerful, relaxed, proud, enthusiastic, relieved, strengthened, optimistic, inspired
and good-humoured (a  0.60). Adjectives for negative affect were angry, sad, insecure,
worried, nervous, frustrated, discouraged and aggressive (a  0.59). Again, the scores for
positive and negative affect were the number of indicated positive and negative adjectives,
respectively. The correlation between positive and negative affect was r  0.46,
p < 0.001. In addition, participants received two open questions regarding the video, asking
them to reflect on the team's functioning with respect to the two issues that were on the
agenda during the team meeting.
Finally, participants filled out questionnaires for identification and out-group attitudes.
First, they received a 7-item questionnaire on identification with the team that was
developed for the purpose of the present study (e.g. `I can easily imagine myself as a
member of the team') A 5-point scale was used, ranging from not at all applicable to (5)
totally applicable. Cronbach's alpha for the scale was high, a  0.83. Next, they received a
5-item questionnaire on identification with the culture that was developed for the purpose
of the present study. A sample item of this scale is: `I am proud of my cultural backgrounď
(a  0.79).
RESULTS
Manipulation check
In order to check whether the participants had indeed been thinking about death and
watching television, two independent raters that were blind to condition were asked to
assign each respondent to condition on the basis of the things they wrote down in response
to both open questions included in the manipulation. This resulted in a `correcť
classification and perfect agreement between raters in all cases but one. One subject in the
control condition associated watching TV with images of war and reflected on feelings of
fear and sadness. This case was excluded from further analyses. On the whole, it seems that
participants had been thinking about death in the Terror Management condition and about
watching TV in the control condition. In addition, the responses to the video clearly
suggested that all subjects had observed the video carefully and had an understanding of the
viewpoints and the final decisions with respect to both issues that were raised and also had
an understanding of the group dynamics.
MANOVA with Threat and Diversity as between-subject factors and pre-manipulation
positive and negative affect as covariates, revealed a significant effect of Threat on the
PANAS-scores, F(2, 151)  4.30, p < 0.05. Univariately, this effect appeared to be
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
460 K. Van der Zee and I. Van der Gang
significant for negative affect (NA), F(1, 152)  5.01, p < 0.05, but not for positive affect
(PA), F(1, 152)  1.90, n.s. Participants in the TMI-condition experienced more negative
affect, M  1.79, than participants in the control condition, M  1.59. In previous terror
management studies (for an overview, see Greenberg et al., 1997) typically no effects of
mortality salience on the PANAS questionnaires were found. In order to make sure that the
results with respect to affect can be attributed to responses to the videotaped information
rather than to differential affective responses to mortality salience, it was decided not only
to control for pre-manipulation affect, but also for the PANAS-scores in subsequent
analyses.
Diversity, threat and affect
All study hypotheses were tested using the regression approach within MANOVA.
Thereby, both manipulations were entered as dichotomous independent variables and the
personality variables were entered as continuous independent variables. It can be argued
that individual differences in affective responses to experimental conditions reflect a priori
differences in affect associated with different personalities. Extraversion and Neuroticism,
traits that are conceptually strongly related to respectively our constructs of Social
Initiative and Emotional Stability, are sometimes even referred to as positive and negative
affectivity in the literature (e.g. Watson & Clark, 1992). The present data revealed that
solely Social Initiative was related to pre-manipulation positive affect (r  0.15); the
PANAS-scores were both related to Social Initiative (PA: r  0.20; NA: r  0.19) and
Emotional Stability (PA: r  0.27; NA: r  0.42). In subsequent analyses, we controlled
for the effects of these three variables.2
First, we examined the impact of Threat and Diversity on affect. Multivariate analyses
revealed no multivariate significant effect of Threat on affect, F < 1, n.s. We did find a
significant effect of Diversity on affect, F(2, 150)  3.13, p < 0.05, h2
 0.04. At the
univariate level, this effect was significant both for positive affect, F(1, 157)  5.07,
p < 0.05, h2
 0.03, and for negative affect, F(1, 157)  3.90, p < 0.05, h2
 0.03,
indicating that, as predicted (Hypothesis 1a) participants perceived more positive affect in
the homogeneous (M  2.40) than in the diverse condition (M  1.80) and consistently less
negative affect in the homogeneous (M  0.99) than in the diverse condition (M  1.45). As
predicted, at the multivariate level, the Threat  Diversity interaction was marginally
significant, F(2, 150)  1.67, p < 0.10. At the univariate level, we did find a
Threat  Diversity interaction effect on negative affect, F(1, 80)  8.24, h2
 0.02. The
pattern of findings worked out slightly different than we predicted, indicating that Diversity
had a clear impact on participants' negative feelings in the control condition, (M  0.54 in
the homogeneous condition, M  1.44 in the diverse condition), whereas it had no
significant impact in the TMI-condition, F < 1 (M  1.35 in the homogeneous condition,
M  1.35 in the diverse condition). It must be noted that the effect was not very strong.
Personality, threat and affective responses to diversity
A main effect of Personality on affect was solely found for Emotional Stability (Table 1).
This trait was significantly positively related to positive affect, b  0.21, p < 0.01, and
significantly negatively to negative affect, b  0.18, p < 0.05. At the multivariate level,
2
Controlling for affect did not appear to have a significant influence on the results.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 461
for neither trait a Personality  Diversity interaction effect (Hypothesis 2a) was found, nor
did we find significant interaction effects between Threat and Personality. However, as
predicted, at the multivariate level, both for Social Initiative and for Emotional Stability we
found a significant Threat  Personality  Diversity three-way interaction effect.
Univariately, the three-way interaction effects were significant for positive affect.
Examining the three-way interaction more closely, for Social Initiative, as predicted
(Hypothesis 4a), a significant Personality by Diversity interaction effect was found in the
control condition, F(1, 80)  11.25, p < 0.001, h2
 0.13 but not in the TMI-condition, F(1,
77)  1.17, n.s., h2
 0.02. Further analysis revealed that in the control condition, Social
Initiative was unrelated to positive affective responses to a homogeneous team, b  0.22,
n.s., whereas it was significantly positively related to positive affect in response to a diverse
team, b  0.47, p < 0.001. In the TMI-condition, Social Initiative appeared to be unrelated
to positive affective responses both to a homogeneous team, b  0.14, n.s., and to a diverse
team, b  0.06, n.s. Figure 2a and 2b show the decomposed simple effects of low (1 SD)
versus high levels (1 SD) of Social Initiative on (standardized) positive affect as a
function of condition.
Also in line with our predictions, for Emotional Stability, a significant Personality by
Diversity interaction effect was found in the TMI-condition, F(1, 80)  7.67, p < 0.01,
h2
 0.10, but not in the control-condition, F < 1, n.s., h2
 0.00. Examining the
beta-weights, we found that in the TMI-condition, Emotional Stability was unrelated to
positive affective responses to a homogeneous team, b  0.15, n.s., but was interestingly
highly significantly related to affective responses to a diverse team, b  0.52, p < 0.01. In
the control condition, Emotional Stability appeared to be unrelated both to positive
affective responses to a homogeneous team, b  0.18, n.s., and to a diverse team,
b  0.02, n.s. Figure 3a and 3b show the decomposed simple effects of low (1 SD)
versus high levels (1 SD) of Emotional Stability on (standardized) positive affect as a
function of condition.
Threat, personality and identification with diverse groups
Next, we were interested in the effects of Personality and our experimental manipulations
on patterns of identification with the team and with one's cultural group. Because
identification with the team and with one's cultural background appeared to be unrelated
(r  0.05, n.s.), it was decided to test our hypotheses univariately. Again, we used the
Table 1. Personality and affect by diversity and mortality salience
Multivariate Positive affect Negative affect
F(2, 142) h2
F(1, 156) h2
F(1, 156) h2
Social Initiative <1 (0.00) <1 (0.01) <1 (0.00)
Emotional Stability 2.52
(0.03) 4.01
(0.03) 3.06
(0.02)
Social Initiative  Diversity 1.35 (0.02) 2.70
(0.02) <1 (0.00)
Emotional Stability  Diversity 1.49 (0.02) 2.64 (0.02) <1 (0.00)
Threat  Social Initiative 1.39 (0.02) <1 (0.01) 2.80
(0.02)
Threat  Emotional Stability 2.16 (0.03) 3.24
(0.02) 2.85
(0.02)
Threat  Social Initiative  Diversity 5.12
(0.07) 9.69
(0.04) <1 (0.00)
Threat  Emotional Stability  Diversity 2.48
(0.03) 4.57
(0.03) <1 (0.00)
Significance levels 
p < 0.05, 
p < 0.01, 
p < 0.001.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
462 K. Van der Zee and I. Van der Gang
regression approach within ANOVA, whereby both manipulations were entered as
dichotomous independent variables and the personality variables as continuous
independent variables.
First, the data revealed a significant main effect of Diversity on identification with the
team, F(1, 157)  2.60, p < 0.05, h2
 0.02, thereby supporting Hypothesis 1b. As
predicted, participants' scores revealed higher identification with the homogeneous team
(M  1.93) than with the diverse team (M  1.73). No effects of Threat, F < 1, n.s., nor a
Threat by Diversity interaction effect, F < 1, n.s., on identification with the team were
found (Hypothesis 3b). With respect to the influence of Personality on identification with
the team, we solely found a main effect of Emotional Stability on identification. The higher
participants scored on Emotional Stability, the more strongly they identified themselves
with the team (r  0.20, p < 0.01). No effect of Social Initiative on team identification was
found, nor did we find interaction effects between Personality and the experimental
manipulations on team identification.
Low threat situation
-0.7
-0.5
-0.3
-0.1
0.1
0.3
0.5
homogeneous
diverse
Low threat
1 SD)low SI (-
Positiveaffect
1 SD)high SI (-
(A)
High threat situation
-0.7
Positiveaffect
-0.5
-0.3
-0.1
0.1
0.3
0.5
(B)
homogeneous
diverse
High threat
high SI (+1 SD)low SI (-1 SD)
Figure 2. Social initiative and positive affect as a function of diversity and threat.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 463
Second, we examined the effects of Threat, Personality and Diversity on identification
with the own culture. Univariate analysis of the effects of both traits and both experimental
manipulations on this variable solely revealed a significant interaction between Social
Initiative and Threat on identification with one's cultural background, F(1, 161)  4.39,
p < 0.05, h2
 0.03. Whereas Social Initiative appeared to be unrelated to identification
with one's cultural background in the control condition (r  0.00, n.s.), this trait was
positively related to identification with one's cultural background in the TMI-condition
(r  0.21, p < 0.05). No effects of both experimental manipulations were found, nor did we
find interaction effects of these manipulations with Emotional Stability on identification
(Hypothesis 4b).
DISCUSSION
Interaction with dissimilar others in culturally diverse teams seems to be difficult and
emotionally less satisfying. The results from the present study were in line with earlier
Low threat situation
-0.7
-0.5
-0.3
-0.1
0.1
0.3
0.5
Positiveaffect (A)
homogeneous
diverse
Low threat
high ES (+1 SD)low ES (-1 SD)
High threat situation(B)
-0.7
-0.5
-0.3
-0.1
0.1
0.3
0.5
Positiveaffect
homogeneous
diverse
High threat
High ES (+ 1 SD)1 SD)Low ES (Figure
3. Emotional stability and positive affect as a function of diversity and threat.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
464 K. Van der Zee and I. Van der Gang
findings that show lower well being associated with cultural diversity (Tsui et al., 1992;
Van der Zee et al., 2004a; see for a discussion Jackson et al., 1993). Our participants clearly
responded with less positive affect to a diverse than to a homogeneous team environment.
Consistently, they also identified themselves less strongly with the diverse team
environment. Earlier work has already suggested that negative attitudes towards members
of different cultural groups within the team and intra-group competition between cultural
subgroups may prevent group members from identifying with the group as a whole
(Abrams & Hogg, 1988; Brewer, 1995). In the present study, participants' emotional
reactions as well as their levels of identification with the team were based on imagining
oneself as part of the group. There was no history of negative experiences with the group or
an actual perspective of becoming part of the groups. Moreover, the content of the video
was exactly similar in both conditions, so the amount of negative experiences that could be
inferred from the video was similar for the homogeneous and the diverse team. The
negative reactions have to be explained either from negative attitudes towards diversity or
different cultural groups, or upon expectations of negative experiences associated with
diversity, for example based on their own previous experiences.
As predicted, the impact of diversity on individuaľs negative affective responses was in
turn moderated by threat. It must be noted that the effects went in a slightly different
direction than we predicted. Building upon the assumptions made by TMT, we expected
that the differences in affective responses to the diverse as opposed to the homogeneous
team would be more pronounced in case of threat (Greenberg et al., 1997). The assumption
was that upon introduction of threat, identification with the own cultural norms increases
and tolerance towards deviant norms or different cultural groups decreases, resulting in
more positive reactions to homogeneity and more negative reactions to diversity. The data
revealed that under threat, negative responses occurred regardless of whether individuals
had been confronted with a homogeneous or a diverse situation. Apparently, under
conditions of threat, the introduction of diversity does not make much difference anymore.
This may imply that diversity does not make organizations extra vulnerable to conditions
of threat. We believe that it is too early to draw such a firm conclusion. The present results
were based on subjects observing rather than participating in a diverse team. Actual
interaction with similar as opposed to dissimilar others may provide a much stronger
source of self-esteem and ensurance of safety than observing a homogeneous versus a
diverse group. Future studies need to replicate our findings for real interaction situations.
The central issue of the present study concerned the moderating role of threat affecting
individual differences in responses to diversity. The data supported our explanation for the
contrasting findings in the literature with respect to the interplay between threat and
individual differences in affecting responses to diversity (Greenberg et al., 1997; Van der
Zee et al., 2004b). Individuals differed in their reactions to our experimental manipulations
in the direction that we predicted. Specifically, for Social Initiative, positive affective
responses to a diverse versus a homogeneous team solely occurred under non-threatening
circumstances. Under threat individuals high in Social Initiative responded equally
negative as individuals low in this trait, and they tended to identify themselves strongly
with the own culture. Apparently, among individuals high in Social Initiative, induction of
death-related threat did pose a limit to their openness to diversity. The findings point
against the `anxiety buffer hypothesis' assumption that they would be especially protected
against terror and resulting worldview defense mechanisms, but are in line with our own
earlier study, in which we showed that upon TMI differences between individuals low and
high in Openness in their emotional reactions to intercultural situations disappeared (Van
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 465
der Zee et al., 2004b). Thereby, the results support our assumption that for social traits, the
link to success in a culturally diverse context cannot be explained from special abilities to
deal with threat but has to be understood from their higher attraction to the social
stimulation present in diverse situations and from their higher ability to `read other
cultures'. As expected, for Emotional Stability, the pattern of findings did follow the
predictions from the anxiety buffer hypothesis. Individual differences in responses to
diversity linked to this trait were restricted to conditions of threat. We argued that
Emotionally Stable individuals are well-equipped to handle the increase in threat posed by
mortality salience, even in a context of high diversity.
The present findings for Emotional Stability are consistent with the anxiety buffer
hypothesis as put forward by Terror Management Theorists and with their empirical
findings for neuroticism and self-esteem. However, they are inconsistent with their view
that traits indicative of tolerant attitudes (e.g. a liberal worldview) may also protect
individuals against the impact of mortality salience on attitudes towards dissimilar others
(Greenberg, Simon, Pyszczynski, Solomon, & Chatel, 1992). It must be noted that the
dependent variable in the Greenberg et al. (1992) study concerned attitudes towards
dissimilar others. Individuals who are committed to a worldview that embraces tolerance
and even celebration of differences, are likely to be faced with a conflict when their need
for the protection provided by the worldview is increased. Although for liberals derogating
conservative others may defuse the threat by their different attitudes, doing so runs counter
a central tenet of their own liberal worldview and would thus undermine the protection
from anxiety that the worldview provides. High-tolerant (liberal) respondents may
therefore have expressed more positive attitudes under threat. In our study we focused on
affective responses that are less clearly linked to attitudes. Subjects may be less reluctant to
express negative feelings upon confrontation with diversity than to explicitly express
negative attitudes.
In addition, in the Greenberg et al. (1992) paper, tolerance-effects were much weaker in
an experiment (study 2) were an anti-U.S. foreign student criticized the United States on a
variety of serious grounds than in another experiment (study 1) in which an attitudinally
dissimilar student disagreed on controversial social issues. The former type of dissent may
have been a more explicit and potent threat to subjects' worldviews than the latter. As the
authors argue themselves: `with greater threat to the worldview, individuals may become
relatively more concerned with defense and less concerned with living up to the value of
tolerance' (p. 218). This point of view exactly represents our `threat dominance'
perspective.
Although the research on the role of personality in dealing with diversity in
organizations is still in its infacy (e.g. Van der Zee et al., 2004a), there is an increasing
amount of studies addressing the role of personality traits as a determinant of intercultural
success (see for overviews Deller, 1997; Van der Zee & Van Oudenhoven, 2000). However,
whereas speculations can be found in the literature with respect to the underling processes
of why specific traits are related to success in a diverse context, such speculations are rarely
explicitly tested. The present study shed more light on which traits may particularly protect
against the threatening aspects of diversity as opposed to the social-oriented problems that
diversity may also evoke. Such insights may be helpful in determining which traits are
important in which situations and in which phases of international projects or (diverse)
team formation. For example in an earlier study in which foreign students who attended an
international business school were followed over a 6-month period, we found that, whereas
Emotional Stability was the main predictor in the beginning, social traits such as Cultural
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
466 K. Van der Zee and I. Van der Gang
Empathy gained importance over time (Van Oudenhoven & Van der Zee, 2002). In the first
phase, general stress factors (e.g. having to find a place to live, uncertainty about study
results) seemed to prevail in this situation, whereas over time intercultural issues for
example related to subtleties of communication, increased in importance.
Some authors have stressed the fact that general approaches to personality are adequate
in explaining adjustment to other cultures. In a number of studies we have shown evidence
for the predictive value of specific intercultural traits above the Big Five (e.g. Van der Zee
& Van Oudenhoven, 2000; Van der Zee, Zaal, & Piekstra, 2003). However, the present
support for the anxiety buffer hypothesis for Emotional Stability suggests that this trait may
be less exclusively linked to intercultural situations than a trait such as Social Initiative.
This argument is further supported by the finding that Emotional Stability appeared
predictive of negative affect and identification with the team, regardless of diversity (see
also Van Oudenhoven & Van der Zee, 2002). Although threats due to unpredictability and
unfamiliarity certainly capture important aspect of culturally diverse situations, the ability
to buffer threat-related anxiety may be equally helpful in response to examination stress or
a first date. The homogeneous situation was not stressful in the sense of being culturally
diverse, but there were some underlying tensions in the group session displayed on the
video that may have caused frustration in the observers, regardless of team composition.
Interpersonal skills in understanding the thoughts and feelings of individuals of other
cultures and in interacting with them seem to be more specifically linked to intercultural
situations.
The study has a number of limitations. First, although we relied on well-validated
indicators of affect, the reliability of the Multi-Affect Adjective Checklist (Zuckerman,
1960; Zuckerman et al., 1964) appeared to be low. Future studies may rely on
intensity-based dependent measures of affect, such as the PANAS, rather than on indicators
of the number of different affects. Second, the present study was performed among a
sample of psychology students. Psychology students are by no means representative of the
general population. In general they are higher on traits that indicate tolerance and empathy,
and their attitudes towards diversity are more positive than in the general population. The
present findings need to be replicated among real groups of employees who are actually
faced with an intercultural context at work. Next, the present study relied on a scenario
method to measure cognitions and affective reactions to intercultural situations and a
mortality salience intervention to induce threat. Scenario's have been successfully applied
in earlier studies that examined attitudes towards acculturation strategies in adjustment to a
different culture (Luijters, Van der Zee, & Otten, 2006; Van Oudenhoven, Prins, & Buunk,
1998). In comparison to direct self-descriptions, this methodology is less susceptible to
social desirability influences. However, as we stated earlier the scenario method relies on
subjects observing rather than participating in a diverse team, which limits the ecological
validity of our study. Concerning threat, future studies may focus on the effect of realistic
sources of threat in organizations, such as mergers, downsizing, failure, serious accidents
or competition. Thereby, it seems important to distinguish between different sources of
threat. In the present study, we focused on existential vital threat to an individual team
member, which seems to lower open-ness towards group members with a different cultural
background. However, cohesion among members of culturally diverse teams may be
enhanced in case of threat to the identity of the team as a whole (collective threat), for
example due to inter-group competition (Turner & Pratkanis, 1998). It seems important to
examine the influence of threats at different levels (individual, team, organizational) on
reactions to diversity. There is a clear lack of scientific knowledge on how diverse groups
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 467
perform under threat. Future research may further examine the underlying processes that
link properties of individuals and properties of threat with performance of diverse groups
(see also Turner & Horvitz, 2001). In this way, we can gain insight into how diverse groups
function under threat and how to prevent its detrimental consequences. The present study
was a first step in that direction.
ACKNOWLEDGEMENTS
The authors thank Jan Pieter van Oudenhoven, Kyra Luijers and Menno Vos for their
valuable comments on an earlier version of this paper.
REFERENCES
Abrams, D., & Hogg, M. A. (1988). Social psychology of intergroup relations and group processes.
Florence, KY, US: Taylor & Frances/Routledge.
Arndt, J., Greenberg, J., & Cook, A. (2002). Mortality salience and the spreading activation of
worldview-relevant constructs exploring the cognitive architecture of terror management. Journal
of Experimental Psychology, 131, 307­324.
Arthur, W., & Bennett, W. (1995). The international assignee: The relative importance of factors
perceived to contribute to success. Personnel Psychology, 48, 99­114.
Brewer, M. (1995). Managing diversity: The role of social identities. In S. E. Jackson, & M. N.
Ruderman (Eds.), Diversity in work teams: Research paradigms for a changing workplace (pp.
47­68). Washington, DC, US: American Psychological Association.
Costa, P. T. Jr, & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI- R) and NEO
Five Factor Inventory (NEO-FFI). Professional manual. Odessa, Fl: Psychological Assessment
Recources.
Deller, J. (1997). Expatriate selection: Possibilities and limitations of using personality scales.
In D. M. Saunders, & Z. Aycan (Eds.), New approaches to employee management (Vol. 4,
pp. 93­116). Greenwich, CT: JAI Press.
Eysenck, H. J., & Eysenck, M. (1985). Personality and individual differences: A natural science
approach. New York: Plenum Press.
Goldenberg, J. L., Pyszczynski, T., McCoy, S. K., Greenberg, J., & Solomon, S. (1999). Death, sex,
love, and neuroticism: Why is sex such a problem? Journal of Personality and Social, Psychology,
6, 1173­1187.
Gray, J. A. (1991). Neural systems, emotion, and personality. In J. Madden IV (Ed.), Neurobiology of
learning, emotion, and affect (pp. 273­306). New York: Raven Press.
Greenberg, J., Pyszczynski, T., & Solomon, S. (1990). Anxiety concerning social exclusion: Innate
response or one consequence of the need for terror management? Journal of Social and Clinical
Psychology, 9, 202­213.
Greenberg, J., Pyszczynski, T., Solomon, S., Pinel, E., Simon, L., & Jordan, K. (1993). Effects of
self-esteem on vulnerability-denying defensive distortions: Further evidence of an anxietybuffering
function of self-esteem. Journal of Experimental Social Psychology, 29, 229­251.
Greenberg, J., Solomon, S., & Pyszczynski, T. (1997). Terror management theory of self-esteem and
cultural worldviews: Empirical assessments and conceptual refinements. Advances in Experimental
Social, Psychology, 29, 61­139.
Greenberg, J., Simon, L., Pyszczynski, T., Solomon, S., & Chatel, D. (1992). Terror management and
tolerance: Does mortality salience always intensify negative reactions to others who threaten one's
worldvies? Journal of Personality and Social, Psychology, 63, 212­220.
Jackson, S. E., Stone, V., & Alvarez, E. B. (1993). Socialization amidst diversity: The impact of
demographics on work team oldtimers and newcomers. In B. M. Staw (Ed.), Research in
organizational behavior: An annual series of analytical essays and critical reviews (Vol. 15,
pp. 45­109). Greenwhich, CN: CAI press.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
468 K. Van der Zee and I. Van der Gang
Lieberman, J. D. (1999). Terror Management, illusory correlation, and perceptions of minority
groups. Basic and Applied Social Psychology, 21, 13­23.
Luijters, K., Van der Zee, K. I., & Otten, S. (2006). Acculturation strategies among ethnic minority
workers and the role of intercultural personality traits. Group Processes and Intergroup Relations,
9(4), 561­575.
Milliken, F. J., Bartel, C. A., & Kurzberg, T. R. (2003). Diversity and creativity in work groups: A
dynamic perspective on the affective and cognitive processes that link diversity and performance.
In P. Paulus, & B. A. Nijstad (Eds.), Group creativity: Innovation through collaboration (pp.
75­106). New York: Oxford University Press.
Mol, S., Van Oudenhoven, J. P., & Van der Zee, K. I. (2001). Validation of the M.P.Q. amongst an
Internationally Oriented Student Population in Taiwan. In F. Salili, & R. Hoosain (Eds.), Research
in multicultural education and international perspectives (pp. 167­186). Greenwich, CT: Information
Age Publishing Inc.
Ones, D. S., & Viswesvaran, C. (1997). Personality determinants in the prediction of aspects of
expatriate job success. In D. M. Saunders, & Z. Aycan (Eds.), New approaches to employee
management (Vol. 4, pp. 63­92). Greenwich, CT: JAI Press.
O'Reilly, C. A., Caldwell, D. F., & Barnett, W. P. (1989). Work group demography, social integration,
and turnover. Administrative Science Quarterly, 34, 21­37.
Pickering, A. D., & Gray, J. A. (1999). The neuroscience of personality. In L. A. Pervin, & O. R. John
(Eds.), Handbook of personality: Theory and research. New York: The Guilford Press.
Rosenblatt, A., Greenberg, J., Solomon, S., Pyszczynski, T., & Lyon, D. (1989). Evidence for terror
management theory I: The effects of mortality salience on reactions to those who violate or uphold
cultural values. Journal of Personality and Social, Psychology, 57, 681­690.
Simon, L., Harmon-Jones, E., Greenberg, J., Solomon, S., & Pyszczynski, T. (1996). Mild depression,
mortality salience, and defense of the worldview: Evidence of intensified terror management in the
mildly depressed. Personality and Social Psychology Bulletin, 22, 81­90.
Solomon, S., Greenberg, J., & Pyszczynski, T. (1991). Terror management theory of self-esteem. In
C. R. Snyder, & D. Forsyth (Eds.), Handbook of social and clinical psychology: The health
perspective (pp. 21­40). New York: Pergamon.
Strachan, E., Pyszczynski, T., Greenberg, J., & Solomon, S. (2001). Coping with the inevitability of
death: Terror management and mismanagement. In C. R. Snyder (Ed.), Coping with stress:
Effective people and processes (pp. 114­136). London: Oxford University Press.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel, &
W. G. Austin (Eds.), The psychology of intergroup relations (pp. 7­24). Chicago: Nelson-Hall.
Tsui, A. S., Egan, T. D., & O'Reilly, C. A. (1992). Being different: Relational demography and
organizational attachment. Administrative Science Quarterly, 37, 549­579.
Turner, J. C. (1982). Towards a cognitive redefinition of the social group. In H. Tajfel (Ed.), Social
identity and intergroup relations (pp. 15­40). Cambridge: Cambridge University Press.
Turner, M. E., & Horvitz, T. (2001). The dilemma of threat: Group effectiveness and ineffectiveness
under adversity. In M. E. Turner (Ed.), Groups at work: Theory and research (pp. 445­470).
Mahwah, NJ, US: Lawrence Erlbaum Associates.
Turner, J. C., & Pratkanis, A. R. (1998). A social identity maintenance model of group think.
Organizational Behavior and Human, Decision Processes, 73, 210­235.
Van der Zee, K. I., Atsma, N., & Brodbeck, F. C. (2004a). The influence of social identity and
personality on outcomes of cultural diversity in teams. Journal of Cross-Cultural Psychology, 35,
283­303.
Van der Zee, K. I., & Van Oudenhoven, J. P. (2000). The Multicultural Personality Questionnaire: A
multidimensional instrument of multicultural effectiveness. European Journal of Personality, 14,
291­309.
Van der Zee, K. I., Van Oudenhoven, J. P., & De Grijs, E. (2004b). Personality, threat and cognitive
and emotional reactions to intercultural situations. Journal of Personality, 72, 1069­1096.
Van der Zee, K. I., Zaal, J. N., & Piekstra, J. (2003). Validation of the Multicultural Personality
Questionnaire in the context of personnel selection. European Journal of Personality, 17, 77­
100.
Van Oudenhoven, J. P., Prins, K., & Buunk, A. P. (1998). Attitudes of minority and majority members
towards adaptation of immigrants. European Journal of Social Psychology, 28, 995­1013.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
Personality, threat and responses to diversity 469
Van Oudenhoven, J. P., & Van der Zee, K. I. (2002). Predicting multicultural effectiveness of
international students: The Multicultural Personality Questionnaire. International Journal of
Intercultural, Relations, 26, 679­694.
Watson, D., & Clark, L. A. (1992). On traits and temperament: General and specific factors of
emotional experience and their relation to the five factor model. Journal of Personality, 60,
441­476.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of
positive and negative affect: The PANAS scales. Journal of Personality and Social, Psychology, 54,
1063­1070.
Ybema, J. F., & Buunk, B. P. (1995). Affective responses to social comparison: A study among
disabled individuals. British Journal of Social Psychology, 34, 279­292.
Zuckerman, M. (1960). The development of an affect adjective checklist for the measurement of
anxiety. Journal of Consulting Psychology, 24, 457­462.
Zuckerman, M., Lubin, B., Vogel, L., & Valerius, E. (1964). Measurement of experimentally induced
affects. Journal of Consulting Psychology, 28, 418­425.
Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 453­470 (2007)
DOI: 10.1002/per
470 K. Van der Zee and I. Van der Gang