Direct contact and authoritarianism as moderators between extended contact and reduced prejudice Lower threat and greater trust as mediators

TLDR: The authors investigated the moderating influence of direct contact and authoritarianism on the potential of extended contact to reduce prejudice and found that the third-order moderation effect was also significant, revealing that extended contact has the strongest effect among high authoritarians.

Abstract: Using a representative sample of Dutch adults (N = 1238), we investigated the moderating influence of direct contact and authoritarianism on the potential of extended contact to reduce prejudice. As expected, direct contact and authoritarianism moderated the effect of extended contact on prejudice. Moreover, the third-order moderation effect was also significant, revealing that extended contact has the strongest effect among high authoritarians with low levels of direct contact. We identified trust and perceived threat as the mediating processes underlying these moderation effects. The present study thus attests to the theoretical and practical relevance of reducing prejudice via extended contact. The discussion focuses on the role of extended contact in relation to direct contact and authoritarianism as well as on the importance of trust in intergroup contexts.

Figures

Table 2. Means, Standard Deviations, and Correlations between the Predictors, Mediators, and Outcome Variable

Table 1. Coding and proportions for demographic variables in the sample (and in the Netherlands if available) as well as the correlations with RWA, prejudice, trust, and threat

Table 4. Results of the simple slopes analyses testing the effects (β-values) of extended contact on prejudice, trust, and threat at High (+1SD) and Low (-1SD) levels of direct contact and RWA

Table 3. Results of the hierarchical regression analyses: presented values of the main and interaction effects are β-values (the demographic variables are controlled for in step 1).

Full text

Mediated Moderators of Extended Contact 1
Direct contact and authoritarianism as moderators between extended contact and reduced
prejudice: Lower threat and greater trust as mediators
Kristof Dhont & Alain Van Hiel
Ghent University
Department of Developmental, Personality, and Social Psychology
Key words: intergroup contact; authoritarianism; extended contact; prejudice; trust; threat
Running head: Mediated Moderators of Extended Contact
Authors Affiliations: Kristof Dhont and Alain Van Hiel, Department of Developmental,
Personality and Social Psychology, Ghent University, Henri Dunantlaan 2, B-9000, Ghent,
Belgium, e-mail address: Kristof.Dhont@UGent.be
and Alain.VanHiel@UGent.be
Funding Acknowledgements: This research was supported by a PhD research grant from the
Ghent University Research Council (BOF) awarded to Kristof Dhont (#01D23607) under
supervision of Alain Van Hiel

Mediated Moderators of Extended Contact 2
Abstract
Using a representative sample of Dutch adults (N = 1238), we investigated the
moderating influence of direct contact and authoritarianism on the potential of extended
contact to reduce prejudice. As expected, direct contact and authoritarianism moderated the
effect of extended contact on prejudice. Moreover, the third-order moderation effect was also
significant, revealing that extended contact has the strongest effect among high authoritarians
with low levels of direct contact. We identified trust and perceived threat as the mediating
processes underlying these moderation effects. The present study thus attests to the theoretical
and practical relevance of reducing prejudice via extended contact. The discussion focuses on
the role of extended contact in relation to direct contact and authoritarianism as well as on the
importance of trust in intergroup contexts.

Mediated Moderators of Extended Contact 3
Direct contact and authoritarianism as moderators between extended contact and reduced
prejudice: Lower threat and greater trust as mediators
Originally proposed by Wright, Aron, McLaughlin-Volpe, and Ropp (1997), the
extended contact hypothesis asserts that the mere fact of knowing an ingroup member who
maintains close relations with an outgroup member ameliorates outgroup attitudes. During the
past decade, correlational and (quasi-)experimental support for this hypothesis has increased,
demonstrating that people who witness friendships between in- and outgroup members report
lower levels of outgroup prejudice than those without extended contact experiences (Paolini,
Hewstone, Cairns, & Voci, 2004; Turner, Hewstone, Voci, & Vonofakou, 2008; Wright et al.,
1997; for reviews, see Turner, Hewstone, Voci, Paolini, & Christ, 2007b; Vonofakou et al.,
2008).
However, despite the growing evidence in support of the extended contact hypothesis,
researchers have only recently started to investigate the conditions that may increase or
decrease the effectiveness of extended contact in reducing prejudice or, in other words, the
possible moderators of the extended contact effect (Christ et al., in press; Hodson, Harry, &
Mitchell, 2009). Building on this recent work, the present study investigated both direct
contact (i.e., a contextual variable) and authoritarianism (i.e., an ideological variable) as
moderators of the extended contact effect on prejudice. At the same time, we examined the
mediating role of trust and perceived threat on these moderation effects.
Moderators of Extended Contact Effects
One of the biggest advantages of extended contact over direct contact is that it can
reduce prejudice without being contingent on a person’s opportunities to interact personally
with outgroup members (Christ et al., in press; Turner, Hewstone, & Voci, 2007a; Turner et
al. 2008). Indeed, several circumstances may prevent direct contact, e.g., when people do not

Mediated Moderators of Extended Contact 4
work together, do not attend the same school, or do not live in the same neighborhood. Hence,
especially for those individuals with limited or no opportunities for direct interaction with
outgroup members, extended contact may be a valuable alternative (Turner et al., 2007a;
Vonofakou et al., 2008; Wright et al., 1997). Moreover, Christ et al. (in press) obtained both
cross-sectional and longitudinal evidence supporting the hypothesis that extended contact is
most effective among those people who live in segregated areas having only few or no direct
cross-group friendships. Hence, when people do not benefit from direct contact because of a
segregated context, extended contact seems to have the strongest impact on prejudice.
Whereas Christ et al. (in press) investigated direct contact as a moderator of the
extended contact effect, other researchers recently focused on Right-Wing Authoritarianism
(RWA, Altemeyer, 1981; 1998) as a moderator of both direct and extended contact effects.
RWA is defined as the covariation of conventionalism, authoritarian submission, and
authoritarian aggression and is considered a broad social ideological attitude. Although
authoritarianism is highly predictive of prejudice (Duckitt, 2001; Duckitt & Sibley, 2007; Van
Hiel & Mervielde, 2002; 2005), recent studies have also demonstrated that intergroup contact
works better at reducing prejudice among high rather than low authoritarians (Dhont & Van
Hiel, 2009, for anti-immigrant prejudice; Hodson et al. 2009, for anti-homosexual prejudice;
for a review, see Hodson, in press). Moreover, Hodson et al. (2009) reported that the strongest
beneficial effect of extended contact on anti-homosexual prejudice emerged among high
authoritarians.
The present study combined both moderation perspectives and investigated the three-
way interaction effect between extended contact, direct contact, and RWA. As we argued
above, people who do not personally benefit from positive contact experiences profit the most
from their friends’ or relatives’ contact experiences. However, it could be argued that these
beneficial effects may even be stronger among high authoritarians because this group is likely

Mediated Moderators of Extended Contact 5
to be most influenced by the other members of their group. People learn about other ingroup
members’ attitudes and behavior toward outgroup members by witnessing positive intergroup
interactions. These interactions reflect a group consensus that intergroup contact is positively
valued (Turner et al., 2008; Wright et al., 1997). Driven by their underlying motivation to
conform to others, which satisfies their needs for social order and stability (Duckitt, 2001;
Jugert, Cohrs, & Duckitt, 2009), high authoritarians can be expected to be the least critical of
their ingroup members’ opinions and attitudes. They are therefore more likely to adapt and
adhere to perceived social norms. In sum, extended contact may have the strongest impact on
prejudice among high authoritarians who are isolated from direct positive contact.
The Mediating Role of Threat and Trust
Why should people with little or no direct positive contact benefit more from extended
contact than people who experience positive contact themselves? Extending the study of
Christ et al. (in press), we investigated whether the psychological process behind this
moderator effect resides in the potential of extended contact to reduce perceived outgroup
threat and to build and restore trust in the outgroup. Whereas perceived threat relates to
feelings of fear, anger, insecurity, and uncertainty (Stephan & Renfro, 2002; Riek, Mania, &
Gaertner, 2006), trust is associated with feelings of security and transparency and is based on
confidence in another person’s good intentions or behavior (Lewicky, McAllister, & Bies,
1998; Rousseau, Sitkin, Burt, & Camerer, 1998; Tropp, 2008).
Researchers have considered perceived threat and lack of trust as central determinants
of intergroup conflict and prejudice (Dovidio, Gaertner, Kawakami, & Hodson, 2002; Stephan
& Renfro, 2002; Riek et al., 2006; Tam et al., 2008). However, a growing body of research
has shown that positive contact with outgroup members can reduce threat perceptions
(Pettigrew, Christ, Wagner, & Stellmacher, 2007; Tausch, Tam, Hewstone, Kenworthy, &
Cairns, 2007) and increase outgroup trust (Hewstone, Cairns, Voci, Hamberger, & Niens,

Frequently asked questions

Q1. What are the contributions in this paper?

Using a representative sample of Dutch adults ( N = 1238 ), the authors investigated the moderating influence of direct contact and authoritarianism on the potential of extended contact to reduce prejudice. Moreover, the third-order moderation effect was also significant, revealing that extended contact has the strongest effect among high authoritarians with low levels of direct contact. The present study thus attests to the theoretical and practical relevance of reducing prejudice via extended contact. The discussion focuses on the role of extended contact in relation to direct contact and authoritarianism as well as on the importance of trust in intergroup contexts. During the past decade, correlational and ( quasi- ) experimental support for this hypothesis has increased, demonstrating that people who witness friendships between inand outgroup members report lower levels of outgroup prejudice than those without extended contact experiences ( Paolini, Hewstone, Cairns, & Voci, 2004 ; Turner, Hewstone, Voci, & Vonofakou, 2008 ; Wright et al., 1997 ; for reviews, see Turner, Hewstone, Voci, Paolini, & Christ, 2007b ; Vonofakou et al., 2008 ). Building on this recent work, the present study investigated both direct contact ( i. e., a contextual variable ) and authoritarianism ( i. e., an ideological variable ) as moderators of the extended contact effect on prejudice. At the same time, the authors examined the mediating role of trust and perceived threat on these moderation effects. Whereas Christ et al. ( in press ) investigated direct contact as a moderator of the extended contact effect, other researchers recently focused on Right-Wing Authoritarianism ( RWA, Altemeyer, 1981 ; 1998 ) as a moderator of both direct and extended contact effects. RWA is defined as the covariation of conventionalism, authoritarian submission, and authoritarian aggression and is considered a broad social ideological attitude. Moreover, Hodson et al. ( 2009 ) reported that the strongest beneficial effect of extended contact on anti-homosexual prejudice emerged among high authoritarians. The present study combined both moderation perspectives and investigated the threeway interaction effect between extended contact, direct contact, and RWA. In sum, extended contact may have the strongest impact on prejudice among high authoritarians who are isolated from direct positive contact. Extending the study of Christ et al. ( in press ), the authors investigated whether the psychological process behind this moderator effect resides in the potential of extended contact to reduce perceived outgroup threat and to build and restore trust in the outgroup. The authors also argue that the processes of reducing threat perceptions and establishing trust are also likely to explain why high scorers on RWA are most sensitive to the influence of extended contact. The Present Study In a large, representative sample of Dutch adults, the present study investigated contact with and prejudice toward immigrants from the Turkish and Moroccan populations, which constitute the two largest Muslim communities in the Netherlands. Moving beyond previous studies demonstrating that the impact of extended contact is most pronounced among people with low levels of direct contact ( Christ et al., in press ) as well as among high scorers on RWA ( Hodson et al., 2009 ), the authors expected a three-way interaction effect between extended contact, direct contact, and RWA on prejudice, where high authoritarians with low levels of direct positive contact would benefit most from extended contact. Moreover, the authors conducted mediated moderation analyses ( Muller, Judd, & Yzerbyt, 2005 ) to test whether trust and perceived threat mediate the moderator effects of Mediated Moderators of Extended Contact 8 direct contact and authoritarianism on the relationship between extended contact and reduced prejudice. The authors collected the data for this study in a nationally representative sample of Dutch adults ( non-Muslim and non-Jewish ) without a migration background as part of a larger research project on intergroup contact and attitudes. A total of 1850 people were invited by a survey company to participate in the study and to complete a questionnaire that was administered online in October 2009. Of this sample, 1440 respondents initially agreed to participate, but 202 persons did not complete the full survey. To assess the quality of intergroup contact, participants answered the stem question, ‘ How often do the following characteristics typify your contact with immigrants ? ’ which was followed by eight adjectives: pleasant, annoying ( reverse coded ( R ) ), on an equal footing, nice, distant ( R ), forced ( R ), friendly, and hostile ( R ). The authors administered the 12-item RWA3D scale ( Funke, 2005 ; see Van Hiel, Cornelis, Roets, & De Clercq, 2007 ) to assess the participants ’ levels of Mediated Moderators of Extended Contact 10 authoritarianism ( Cronbach ’ s α =. 67 ). The following four items measured respondents ’ levels of outgroup trust ( Cronbach ’ s α =. Mediated Moderation Analyses Series of hierarchical regression analyses ( Aiken & West, 1991 ) were conducted to test the hypothesized moderation effects, followed by additional regression analyses to test for mediated moderation effects ( Muller et al., 2005 ). Additionally, the authors were able to show an analogous pattern of results for trust and threat. In a second series of hierarchical regression analyses, the authors tested whether the two-way interaction effects reported above were further qualified by a three-way interaction effect between extended contact, direct contact, and RWA. In the following analyses the authors tested whether trust and threat are mediating variables that account for the twoand three-way moderation effects on prejudice. The present results demonstrated that the effects of extended contact on prejudice were stronger when people reported low, rather than high levels of direct contact ( Christ et al., in press ) as well as among high scorers rather than low scorers on RWA ( Hodson et al., 2009 ). Their results also complement Hodson et al. ( 2009 ), who demonstrated that perceived threat mediated the relationship between extended contact and prejudice among high authoritarians, by showing mediation effects for both threat and trust. Moving beyond previous work, the authors were the first to demonstrate that a three-way interaction effect further qualified the moderation effects of direct contact and authoritarianism on prejudice. In particular, the authors found that extended contact has the greatest effects on prejudice among high authoritarians who do not benefit from direct positive contact via the process of generating trust and reducing threat. The authors obtained support for their hypotheses with a large representative sample that reflects the adult population of the Netherlands. The high external validity is thus an important strength of the present study and contributes to the development of theoretical principles and practical implications that can be applied to a broad population. In the following sections, the authors first discuss the role of trust and perceived threat and reflect on their finding that extended contact is an effective means of combating prejudice among authoritarians. Before concluding, the authors highlight some limitations of the present study. The present study extends these studies and shows that when people are indirectly connected with outgroup members though trusted ingroup members, this Mediated Moderators of Extended Contact 17 connection increases outgroup trust, and especially among those who do not benefit from direct positive contact. By increasing trust and decreasing threat, extended contact also affects the basic motivations that underlie authoritarianism ( Duckitt, 2001 ; Duckitt & Sibley, 2007 ) which explains why the positive effect of extended contact among people who are cut off from positive influences of direct contact is even stronger among high authoritarians. It is important to note that, although a vast amount of research is available on the relationship between authoritarianism and intergroup threat ( e. g., Cohrs & Asbrock, 2009 ; Duckitt, 2006 ; Hodson, Hogg, & MacInnis, 2009 ), studies on the relationship between Mediated Moderators of Extended Contact 18 authoritarianism and trust are scarce. Future research is thus required to further investigate the processes that may be involved in establishing trust and reducing prejudice among high authoritarians. Furthermore, because the authors only used self-report scales, common method variance may have influenced the strength of the relationships between the studied variables. The finding that extended contact has the strongest effect among people who do not experience high quality direct contact with outgroup members and/or high authoritarians attests to the practical relevance of applying strategies based on extended contact. 3. the authors also conducted separate series of follow-up regression analyses with the direct contact quality and direct contact quantity measure ( replacing the combined measure ). These analyses mainly followed the reported findings. Mediated Moderators of Extended Contact 21 4. The present results also clarified that the extended contact x RWA moderation effect is not merely an artifact of floor effects on prejudice and threat or a ceiling effect on trust among low authoritarians. The other “ authoritarian personality ”. The authors must not be enemies: Interracial contact and the reduction of prejudice among authoritarians. The dimensionality of right-wing authoritarianism: Lessons from the dilemma between theory and measurement. Furthermore, similar interaction patterns were present for trust and threat. As reported in Table 3, the authors found significant three-way interaction effects on prejudice and threat, and a marginally significant three-way interaction effect on trust. Whereas Hodson et al. ( 2009 ) found that that the strong effect of extended contact among high scorers on RWA was mediated through perceived threat, the authors can conclude that both trust and threat mediate the moderation effect between extended contact and RWA on prejudice. Furthermore, a regression analysis to test the mediating role of threat separately revealed that the inclusion of threat, β =. 49, p <. 001, also reduced the three-way interaction effect, β =. In sum, the potential of extended contact to increase trust and to decrease threat is the underlying mechanism that explains why extended contact most strongly reduces prejudice among high scorers on RWA with low levels of direct contact. Their results not only show that extended contact has the potential to substantially decrease threat perceptions, but that this is particularly so in settings where the influence of direct positive contact is absent. The potential for extended contact to reduce threat perceptions and establish trust increases its applications beyond prejudice reduction. Furthermore, explanations for this moderation effect of authoritarianism in terms of floor or ceiling effects also do not seem to hold in previous studies. 

Q2. What is the effect of extended contact on the motivational processes underlying RWA-based prejudice?

by reducing threat perceptions and establishing trust, extended contact can alter the motivational processes underlying RWA-based prejudice. 

Q3. What are the main areas of interest for kristof van hiel?

Other areas of interest are personality, procedural fairness, information dissemination in groups, need for closure, social cognition, and intergroup contact. 

Q4. What is the effect of extended contact on the motivational processes underlying authoritarianism?

laboratory experiments (Wright et al., 1997) and experimental field studies (Cameron & Rutland, 2006; Liebkind & McAlister, 1999) have demonstrated that extended contact promotes more positive outgroup attitudes. 

Q5. What is the effect of extended contact on the motivational processes that underlie authoritarianism?

Prejudice Reduction among AuthoritariansBy increasing trust and decreasing threat, extended contact also affects the basicmotivations that underlie authoritarianism (Duckitt, 2001; Duckitt & Sibley, 2007) which explains why the positive effect of extended contact among people who are cut off from positive influences of direct contact is even stronger among high authoritarians. 

Q6. What was the recoded score of quantity of intergroup contact?

Prior to multiplication, the scores of quantity of intergroup contact were recoded so that 0 corresponded to no contact and 6 to very frequent contact, and the quality scores were recoded so that the scores ranged from -3 to +3. 

Q7. What is the effect of extended contact on prejudice?

In particular, the authors found that extended contact has the greatest effects on prejudice among high authoritarians who do not benefit from direct positive contact via the process of generating trust and reducing threat. 

Q8. What is the relationship between perceived threat and intergroup conflict?

The reduction of perceived threat has been related to a decrease in hostile or aggressive actions against outgroup members and may thus help to resolve forms of intergroup conflict (Stephan & Renfro, 2002). 

Q9. What is the significance of trust building?

Trust building may be even more far-reaching because trust promotes cooperation across many forms of social interaction (Dawes, 1980; De Cremer & Tyler, 2005) and thus paves the way for a positive intergroup climate characterized by mutual cooperation. 

Q10. What were the items that were likely to be negative for immigrants?

The items (Cronbach’s α = .77) were: ‘Marrying an immigrant is like asking for trouble’; ‘Generally speaking, immigrants are not as smart as Dutch people’; and ‘the Dutch should never have allowed immigrants into their country’. 

Q11. How many people in the study have a migration background?

The authors collected the data for this study in a nationally representative sample of Dutchadults (non-Muslim and non-Jewish) without a migration background as part of a larger research project on intergroup contact and attitudes.