The negative secondary transfer effect: Comparing proposed mediation theories

Abstract

The secondary transfer effect proposes that contact with an outgroup impacts attitudes towards another, secondary outgroup. For positive contact, three pathways have been identified for the effect: attitude generalization, multiculturalism, and ingroup reappraisal (deprovincialization hypothesis, operationalized here as national pride). Research on negative secondary transfer effects is still scarce. Using data from a German nationally representative survey, we investigated negative secondary transfer effects from foreigners to refugees. The three pathways were compared while considering positive and refugee contact. Negative and positive secondary transfer effects both occurred (partially) mediated via attitude generalization and multiculturalism but not via national pride. We conclude there might be a risk of generalizing prejudice from unrelated negative experiences via these two mechanisms. Research on forced migration and intergroup contact should further explore them with the ultimate goal of preventing negative secondary transfer effects. Longitudinal or experimental research is needed to address causality, ideally involving various outgroups.

Keywords

attitude generalization deprovincialization intergroup contact negative secondary transfer effect

In recent years, substantial progress has been made regarding the understanding of intergroup contact. Nevertheless, prejudice, stereotypes, and discrimination are still common in everyday intergroup contact, especially in the context of forced migration. Nearly 7 decades have passed since Gordon Allport formulated the contact hypothesis (Allport, 1954) and claimed that positive intergroup contact can reduce prejudice, which has since been widely confirmed (Pettigrew & Tropp, 2006). A possible mechanism that has been discussed is the “secondary transfer effect” (STE; Pettigrew, 2009). This effect has been examined in terms of positive contact, but research on negative STE is scarce (Vezzali et al., 2021). The current study aims to contribute to this knowledge.

Conceptual Description of the STE

Conceptually, the STE describes how positive contact with a first, so-called “primary” outgroup leads to a more positive attitude towards a similar, “secondary” outgroup that was not involved in the contact situation. For example, a German person had a positive interaction with an immigrant from Eastern Europe and thus develops a more favorable view of people with that same migration background, but also of people with other migration backgrounds due to perceiving those outgroups similarly. First attempts to explain this effect included cognitive approaches (e.g., cognitive dissonance, meaning congruence between primary and secondary outgroup attitude due “to the motivational principle of achieving harmony between cognitions”; Ebbeler, 2020, p. 41) and affective approaches (e.g., evaluative conditioning, meaning “affective rather than cognitive transfer” of “lessened anxiety and heightened empathy”; Pettigrew, 2009, pp. 62–63) that were derived from longitudinal German probability samples (Pettigrew, 2009). In his research on STE, Pettigrew assumed that the similarity between groups plays a significant role in the occurrence and strength of the STE.

Proposed Mechanisms Behind the STE

Since then, there have been several theories on the mechanisms of the STE beyond the initial theory. Three main mediational processes have been identified, commonly described as attitude generalization, multiculturalism, and ingroup reappraisal (Ebbeler, 2020). Attitude generalization refers to the initially investigated mechanism that positive contact leads to a more positive attitude towards a primary outgroup that is then generalized to a secondary outgroup. The other two pathways can be summarized under the deprovincialization hypothesis, although there has been some disagreement regarding adequate operationalizations (Lolliot et al., 2013; Vezzali et al., 2021). This hypothesis reflects that intergroup contact challenges an individual’s original views on the in- and outgroup, leading to a broader acceptance of diversity (multiculturalism) and distancing from the ingroup (ingroup reappraisal). In other words, intergroup contact broadens the individual’s beliefs so that differing cultural values can be perceived, understood, and appreciated.

Empirically, different operationalizations of the deprovincialization hypothesis were examined with inconclusive results (Hodson et al., 2018; Lolliot et al., 2013; Sparkman, 2020; Vezzali et al., 2021). Positive results were found for the operationalization via multiculturalism (Verkuyten et al., 2010) or via ingroup reappraisal (Pettigrew, 2009; Tausch et al., 2010, Study 1). Examples of null findings include Studies 2, 3, and 4 of Tausch et al. (2010), who did not find any mediating effects when operationalizing deprovincialization as ingroup attitude. In more recent publications, more empirical evidence has been found for broader operationalizations of the deprovincialization hypothesis, as researchers focused on various parts of Pettigrew’s initial explanation of ingroup reappraisal (Hodson et al., 2018; Lolliot et al., 2013). One of these broader definitions is openness towards new experiences, given by Hodson et al. (2018), who understand Pettigrew’s words as “contact shakes up one’s perspective and encourages novel ways of thinking about how the world works” (Hodson et al., 2018, p. 530).

The Importance of Contact Valence

One of the most crucial factors, which should be taken into account in all kinds of intergroup contact research is contact valence. Yet, for a long time, intergroup contact research was characterized by a positivity bias (Barlow et al., 2012) with few studies investigating negative intergroup contact. The major results from studies on positive intergroup contact suggested contact as an optimistic and feasible way to improve intergroup relations, but were often resting on the crucial key assumption that intergroup contact is at least mainly positive in nature. Only more recently the research paradigm started to shift, following the necessity to research negative intergroup contact (e.g., Barlow et al., 2012; Paolini et al., 2010; Thomsen & Rafiqi, 2017). These studies suggest that negative contact might occur independent from positive contact (Barlow et al., 2012) and can have detrimental consequences for intergroup relations by fostering negative attitudes and increasing prejudice. Thus, a consensus among intergroup contact researchers emerged that one perspective should be accounted for when researching the other (Barlow et al., 2012). This also applies to research on STE (Vezzali et al., 2021).

A Lack of Research on Negative STE

In light of this history, it is unsurprising that comparatively few studies investigated STEs from negative contact (Lissitsa & Kushnirovich, 2020; Vezzali et al., 2021). Although findings vary to a certain extent, these studies provide first empirical evidence for negative STE in direct or parasocial contact (such as exposure to outgroup media portrayals) beyond the scope of positive STEs (Brylka et al., 2016; Jasinskaja-Lahti et al., 2020; Lissitsa & Kushnirovich, 2020; Meleady & Forder, 2018). A large proportion of existing studies took place in the Finnish context with Russian or African immigrant groups, limiting the generalizability of supporting results (Brylka et al., 2016; Jasinskaja-Lahti et al., 2020; Mähönen & Jasinskaja-Lahti, 2016). Additionally, some studies have the limitation that contact with the secondary outgroup was not controlled for (Andrews et al., 2018; Harwood et al., 2011; Joyce & Harwood, 2014; Lissitsa & Kushnirovich, 2018; Mähönen & Jasinskaja-Lahti, 2016). In researching STE, contact with the secondary outgroup should be controlled for, as it might shape primary and/or secondary outgroup attitude. Specifically, it might influence an individual’s attitude towards the secondary group more directly than contact with the primary outgroup (Pettigrew, 2009).

Although most studies focused on attitude generalization as a mediating mechanism, two studies have provided evidence for the generalization of outgroup avoidance behavior (Meleady & Forder, 2018) and outgroup threat (Zingora & Graf, 2019). Regardless, research on negative STE is only emerging, calling for future research with robust and extensive study designs (Jasinskaja-Lahti et al., 2020; Vezzali et al., 2021). In particular, further mechanisms of negative STE should be examined and compared. Important mechanisms that have been found in the secondary transfer effect of positive contact, including multiculturalism and ingroup reappraisal, could work in the same or different direction. To gain a deeper understanding of contact valence, it is critical to compare possible mediators and determine their directions and working mechanisms.

Research Context

Studying the effects and mechanisms of negative STE is of particular importance in the context of forced migration. Regular and forced migration have increased due to the ongoing globalization, poverty, and armed conflicts predominantly in Middle Eastern and north African countries such as Syria and the Arab Republic (Cowling et al., 2019). Especially in 2015 and 2016, first asylum applications in Germany increased significantly, causing the public to refer to it as a “refugee crisis” (BAMF, 2021). In this scenario of intergroup contact, many Europeans were presented with the relatively sudden arrival of an outgroup with whom they had not been in contact before. Many promptly held a strong opinion about this new outgroup, often a negative or skeptical one (see e.g., Renner et al., 2017). As Germany admitted approximately 1.5 million forced migrants in the period between 2014 and 2017 alone (Kotzur & Wagner, 2021), refugees make up a substantial minority group in German society. In this context, negative STE from unrelated contact experiences with other migrant outgroups could be especially detrimental for the public perception of forced migrants. To date, little is known about negative STE with regard to refugees and forced migration, except for a single study by Ünver et al. (2021). These researchers found correlational evidence for negative STE (secondary outgroup: Syrian refugees; primary outgroup: other immigrant outgroups) in a sample of Turkish participants as well as a sample of Kurdish participants, a minority group in Turkey (Ünver et al., 2021). More research is needed to see if such results would also emerge in different intergroup contexts. Conducting a similar study in the German context would be especially interesting, because just as in Turkey, refugees are a sizeable minority group in Germany (Kotzur & Wagner, 2021). However, results could also differ in the German context. Compared to the Turkish context, more cultural distance can be expected between host country nationals and the refugee minority group, as evident from the involved countries’ placements on various culture-comparative indices such as the WEIRD framework (Henrich et al., 2010) or the Hofstede dimensions (Hofstede, n.d.).

The Current Study

Therefore, the aim of the current research was to investigate negative STE in the context of refugees and forced migration compared to positive STE, thus adding to the scarce research by looking at data from a representative sample of the German majority group. Since all refugees (in the context of the 2015–2016 European refugee crisis) can be seen as immigrants from a German majority perspective but not all immigrants/people with a foreign background are refugees, an overlap between these two group labels is unavoidable and needs to be approached methodologically. Additionally, the study aimed to compare three mediation mechanisms (attitude generalization, multiculturalism operationalized as acceptance of diversity, and ingroup reappraisal operationalized as national pride) and their role in positive as well as negative STE. Evidence for the latter two mechanisms is so far only based on research on positive STE.

We expect that they are likewise behind negative STE, as extant research suggests that both types of STE rely on the same mechanisms (Vezzali et al., 2021, p. 29). However, we expect them to work in the opposite direction than they work for positive STE: in previous research, negative contact was found to be related to more antiforeigner sentiment (Thomsen & Rafiqi, 2017), increased group salience (Paolini et al., 2010), and increased prejudice (Barlow et al., 2012). Hence, negative contact is theoretically more likely to reduce acceptance of diversity and to emphasize (national) pride by making the ingroup more salient, which in turn would cause more negative attitudes towards a secondary outgroup.

We hypothesize that the aforementioned mechanisms differ in strength, based on previous research that included a combination of the investigated mediators (Lolliot et al., 2013; Vezzali et al., 2021, Tables 1 –3). One study (Pettigrew, 2009) found that deprovincialization (operationalized as ingroup identity) “was a weaker mediator of the secondary transfer effect than was attitude generalization” (Lolliot et al., 2013, p. 89). The same emerged when other operationalizations of deprovincialization were utilized (Tausch et al., 2010, as cited in Lolliot et al., 2013, p. 90). However, all these studies investigated only positive, not negative, STE and, so far, none has investigated all three studied mediation mechanisms jointly. This warrants exploring differences in mediator strength. We acknowledge that more mediators of STE have been identified (Vezzali et al., 2021, Figure 1) but only the selected three mediators were available in the utilized secondary data. Still, we argue that this selection covers all three types of mediators mentioned by Vezzali et al. (2021): one mediator concerning the outgroup (attitude generalization), one mediator concerning the ingroup (national pride), and one mediator involving the self (acceptance of diversity).

Table 1.

Descriptive statistics and correlations for study variables.

	M	SD	1	2	3	4	5	6	7
1	1.97	1.11	1
2	3.59	1.40	.22**	1
3	3.18	0.90	−.08**	.37**	1
4	2.91	0.73	−.08*	.29**	.48**	1
5	1.99	0.79	.00	.06*	.12**	.14**	1
6	2.46	0.72	−.11**	.28**	.55**	.43**	.14**	1
7	0.36	0.48	.19**	.21**	.14**	.12**	.08**	.16**	1

Note. N = 1,553. 1 = negative contact with foreigners; 2 = positive contact with foreigners; 3 = attitude towards foreigners; 4 = acceptance of diversity; 5 = pride in being German; 6 = attitude towards refugees; 7 = contact with refugees. Values given for attitude towards refugees are mean scores of the indicator variables.

p < .05. **p < .01 (two-tailed).

Thus, we proposed the following hypotheses:

Positive contact experiences with the primary outgroup of foreigners are related to more positive attitudes towards the secondary outgroup of refugees (positive STE) when accounting for contact with refugees. Negative contact experiences are related to more negative attitudes towards the secondary outgroup (negative STE) when accounting for contact with refugees.

Both positive and negative STEs are mediated by the attitude towards the primary outgroup as well as by acceptance of diversity and national pride. More specifically, we hypothesize that,

2a. Positive contact is related to more positive attitudes, higher acceptance of diversity, and lower national pride, which are in turn related to more positive attitudes towards the secondary outgroup.

2b. Negative contact is related to less positive attitudes towards the primary outgroup, a lower acceptance of diversity, and more national pride, which are in turn all related to more negative attitudes towards the secondary outgroup.

3. The STE via the mechanism of attitude generalization shows a higher effect size than via acceptance of diversity and national pride for both the positive and negative STEs.

The resulting hypothesized model is depicted in Figure 1.

Figure 1.

The proposed structural equation model.

Methods

Sample and Design

To answer the research questions, data were used from the German general social survey “Allgemeine Bevölkerungs-Umfrage der Sozialwissenschaften,” in short “ALLBUS” (GESIS, 2017, 2018) conducted in 2016, which is representative of the German population. The ALLBUS is conducted every other year, and data sets consist of 3,000 to 3,500 independent respondents randomly drawn from private households. The current study aims to investigate negative STEs from a majority group perspective in a cross-sectional design. Due to the nature of the research questions, steps were taken to filter the target population (German majority sample). Consequently, only respondents without a reported migration background were included in the analyses. This was done by filtering out respondents who reported not to hold a German citizenship since birth or reported that one or both of their parents were not born in Germany. We acknowledge that respondents might unrelatedly differ on their self-perception (and others’ perception of them) as members of the societal majority group, which might not be picked up by this filtering process.

Measures

Measurement of each construct is summarized in the following sections. For additional information on item wording, see Appendix A in the supplemental material.

Positive contact with foreigners

The frequency of positive contact experiences with the primary outgroup of foreigners was measured with a single item: “When you think about all your contacts with foreigners who live in Germany: How often have you had positive experiences?” (Wasmer & Baumann, 2018, p. 23). The original item was rescaled so that answer categories ranged from 1 (never) to 5 (always). The separate answer category “I have never had contact with foreigners” was coded as 0.

Negative contact with foreigners

The frequency of negative contact experiences with the primary outgroup of foreigners was measured with a single item: “And how often have you had negative experiences?” (Wasmer & Baumann, 2018, p. 24). Answer categories were coded in the same fashion as for the measurement of positive contact with foreigners.

Contact with refugees

Whether or not participants had experienced any contact with the secondary outgroup refugees was assessed with a single item: “In recent months, many refugees have come to Germany. Have you ever had direct personal contact with refugees?” (Wasmer & Baumann, 2018, p. 98). Answer categories were recoded (0 = no, 1 = yes).

Attitude towards foreigners

Respondents’ attitude towards the primary outgroup foreigners was assessed via the single item: “Do you think that the presence of foreigners is advantageous or disadvantageous for Germany?” (Wasmer & Baumann, 2018, p. 31). Answer categories ranged from 1 (clearly disadvantageous) to 5 (clearly advantageous).

Attitude towards refugees

Respondents’ attitude towards the secondary outgroup refugees was assessed via four items (α = .82) from which a latent variable was created. These four items concerned evaluations of whether the presence of refugees poses more risks or more opportunities for German society in various domains of life (see Appendix A in the supplemental material). Item answer categories were recoded to range from 1 (considerably more risks) to 5 (considerably more opportunities).

Acceptance of diversity

Acceptance of diversity was measured via two items (r = .44): “It is better for a country if all people belong to a common culture” and “A society with high levels of cultural diversity will be better at tackling new problems” (Wasmer & Baumann, 2018, p. 35). Item answer categories ranged from 1 (completely disagree) to 4 (completely agree). The second, negatively coded item was reverse-scaled and a mean score was created.

National pride

National pride was assessed via a single item: “Would you say you are very proud, fairly proud, not very proud, or not at all proud to be German?” (Wasmer & Baumann, 2018, p. 25). Answer categories were recoded to range from 1 (not at all proud) to 4 (very proud).

Analytic Strategy

Before the ALLBUS data were analyzed, a number of filters were applied to retain the target sample of a German majority group (see Appendix B, Table B1 in the supplemental material). First, missing cases were filtered out in a listwise fashion, which reduced the sample size from originally N = 3,490 to N = 2,961. Second, participants who reported a migration background were filtered out, further reducing the sample size to N = 2,593. Respondents were coded as having a migration background if they did not have the following characteristics: holding a German citizenship since birth (Wasmer & Baumann, 2018, pp. 16–17) and having both parents born in Germany (including those born in former eastern territories; Wasmer & Baumann, 2018, p. 78). The final filtering step aimed to exclude participants that did not conceptually distinguish between the primary outgroup termed “foreigners” and the secondary outgroup termed “refugees.” An open question on the subject was administered prior to the foreigner-related items. The specific wording was, “When you think of the foreigners living in Germany, which groups do you think of?” (Wasmer & Baumann, 2018, p. 22). Respondents were encouraged to name one or more groups that came to mind. In analyzing the open answers to this question, it was possible to identify respondents who thought of refugees in combination with the group label “foreigners” and might have understood both group labels as interchangeable. Using several keywords (e.g., “refugees,” “asylum,” “persecution”), respondents actively naming refugees were excluded. These keywords are reported in Table B1 (supplemental material). This filtering step resulted in N = 1,873 remaining cases. Additionally, those who did not answer the open question and those indicating an all-inclusive understanding of the group label “foreigners” were excluded (e.g., “I think of no particular group since everyone is a foreigner somewhere,” “non-Europeans,” “Those that do not integrate”). Both authors reviewed the selections and differences were solved through discussion. Finally, a sample size of N = 1,553 emerged from the filtering. These respondents still were heterogeneous in their answers to the open question but did not indicate that the group label “foreigners” included refugees from their perspective. Table 1 depicts descriptive statistics and item correlations. We conducted a number of robustness analyses comparing this sample with the rest of the respondents (see Appendix B, supplemental material). Significant (though small) differences on items of interest emerged only for contact with refugees, χ²(df = 1) = 17.93, p < .001, and positive contact with foreigners, t(df = 2959) = 2.19, p = .028.

To answer the research hypotheses, the multiple mediation model depicted in Figure 1 was estimated. Mediation analyses were carried out using Mplus Version 8.6 (Muthén & Muthén, 2017). Assessing model assumptions, we found the linearity assumption reasonably fulfilled (see Figure B1 in Appendix B, supplemental material). To handle a departure from multivariate normality (z = 6.45), Byrne (2016) recommends > 5 as a cut-off value; thus, we employed bias-corrected bootstrapping with 10,000 iterations (Nevitt & Hancock, 2001). All reported coefficients are unstandardized.

Results

The model shown in Figure 1 was tested to see the extent to which the frequency of respondents’ negative contact experiences with the primary outgroup of foreigners living in Germany had an impact on their attitude towards the secondary outgroup of refugees via (a) a more negative attitude towards the primary outgroup, (b) decreased acceptance of diversity, and (c) increased national pride. Results are depicted in Figure 2. Table 2 summarizes the found indirect effects. Table 3 shows the results of contrast analyses that were conducted based on Hypothesis 3.

Table 2.

Summary of indirect, direct, and total secondary transfer effects.

Hypothesized effects	b	SE	95% CI	p
Negative contact → att. foreigners → att. refugees (M₁₋)	−0.05	0.01	[−0.06, −0.03]	< .001
Positive contact → att. foreigners → att. refugees (M₁+)	0.08	0.01	[0.07, 0.01]	< .001
Negative contact → acc. diversity → att. refugees (M₂₋)	−0.02	0.00	[−0.03, −0.01]	< .001
Positive contact → acc. diversity → att. refugees (M₂+)	0.03	0.01	[0.02, 0.04]	< .001
Negative contact → pride German → att. refugees (M₃₋)	0.00	0.00	[0.00, 0.00]	.418
Positive contact → pride German → att. refugees (M₃+)	0.00	0.00	[0.00, 0.00]	.177
Negative contact → att. refugees (c′2)	−0.06	0.01	[−0.09, −0.04]	< .001
Positive contact → att. refugees (c′1)	0.04	0.01	[0.02, 0.06]	< .001
Total effect of negative contact (negative STE)	–0.13	0.02	[−0.16, −0.01]	< .001
Total effect of positive contact (positive STE)	0.15	0.01	[0.13, 0.18]	< .001

Note. N = 1,553. M₁ = attitude generalization; M₂ = multiculturalism; M₃ = ingroup identification. c′1= direct negative secondary transfer effect (STE); c′2 = direct positive STE. Att. = attitude towards; acc. diversity = acceptance of diversity; pride German = pride in being German.

Positive indirect paths are marked with a plus sign, and negative indirect paths with a minus sign.

Table 3.

Pairwise contrasts of indirect effects.

STE valence	Contrast	Paths	B	95% CI
STE valence	Contrast	Paths	B	LLCI	ULCI
Negative STE	Attitude generalization vs. multiculturalism	a7×b1 vs. a8×b2	−0.03*	−0.04	−0.01
	Attitude generalization vs. ingroup identification	a7×b1 vs. a9×b3	−0.05*	−0.06	−0.03
	Multiculturalism vs. ingroup identification	a8×b2 vs. a9×b3	−0.02*	−0.03	−0.01
Positive STE	Attitude generalization vs. multiculturalism	a1×b1 vs. a2×b2	0.05*	0.03	0.07
	Attitude generalization vs. ingroup identification	a1×b1 vs. a3×b3	0.08*	0.07	0.01
	Multiculturalism vs. ingroup identification	a2×b2 vs. a3×b3	0.03*	0.02	0.04

Note. N = 1,553. STE = secondary transfer effect.

Estimates of significant contrasts (as per bias-corrected 95% confidence interval).

Figure 2.

The empirically tested structural equation model.

Hypothesis 1

Positive contact was associated with a more positive secondary outgroup attitude (b = 0.15, p < .001, 95% CI [0.13, 0.18]), indicating a positive STE. Likewise, negative contact was associated with a more negative secondary outgroup attitude (b = −0.13, p < .001, 95% CI [−0.16, −0.10]), representing negative STE.

Hypothesis 2

Regarding Hypothesis 2a, positive STE occurred via attitude generalization and acceptance of diversity (partial mediation) but not via pride in being German. Positive contact was related to a more positive attitude towards the primary outgroup, which in turn was associated with a more positive secondary outgroup attitude, thus indicating a positive STE via attitude generalization (b = 0.08, p < .001, 95% CI [0.07, 0.10]). Additionally, more positive contact was associated with a higher acceptance of diversity that, in turn, was associated with a more positive attitude towards the secondary outgroup, thus indicating a positive STE via multiculturalism (b = 0.03, p < .001, 95% CI [0.02, 0.04]). No significant indirect effect via pride in being German was found (b = 0.00, p = .188, 95% CI [0.00, 0.00]), since no direct effect from positive contact to pride in being German emerged (b = 0.03, p = .085, 95% CI [0.00, 0.06]). Positive contact also had a significant positive direct effect on secondary outgroup attitude (b = 0.04, p = .002, 95% CI [0.02, 0.06]).

Concerning Hypothesis 2b, negative STE occurred via attitude generalization and acceptance of diversity (partial mediation) but not via national pride. Negative contact was related to a more negative attitude towards the primary outgroup, which, in turn, was associated with a more negative secondary outgroup attitude, thus indicating a negative STE via attitude generalization (b = −0.05, p < .001, 95% CI [−0.06, −0.03]). Negative contact was also related to a lower acceptance of diversity, which in turn was associated with a more negative secondary outgroup attitude, thus indicating a negative STE via multiculturalism (b = −0.02, p < .001, 95% CI [−0.03, −0.01]). No significant indirect effect via national pride was found (b = 0.00, p = .42, 95% CI [0.00, 0.00]), since negative contact was not related to pride (b = −0.02, p = .348, 95% CI [−0.06, 0.02]). Additionally, negative contact was directly negatively related to secondary outgroup attitudes (b = −0.06, p < .001, 95% CI [−0.09, −0.04]).

The general model fit was acceptable. Model fit indices: χ² = 87.30 (df = 20, p < .001), RMSEA = .05, TLI = .96, CFI = 0.98, SRMR = .02. Looking at R² values, the set of equations predicting national pride explained a low share of the variance (R² = .01), as opposed to those predicting attitude towards foreigners (R² = .17), acceptance of diversity (R² = .11), and attitude towards refugees (R² = .43).

Hypothesis 3

Contrast analyses indicated that for negative as well as positive STE, the three indirect paths (attitude generalization, multiculturalism, and pride) differed significantly in strength (see Table 3). Comparisons with the national pride pathways shall not be discussed further since the indirect effects via national pride were found to be nonsignificant in the previous analyses for both negative and positive contact.

The positive attitude generalization pathway was significantly stronger than the positive multiculturalism pathway (b = 0.05, 95% CI [0.03, 0.07]). This indicates that positive STE also works to a significantly larger degree via attitude generalization than via multiculturalism.

Similarly, the negative attitude generalization pathway was significantly different in strength compared to the negative multiculturalism pathway (b = −0.03, 95% CI [−0.04, −0.01]). This result indicates that negative STE works to a significantly larger degree via attitude generalization than via multiculturalism.

Discussion

The aim of the current research was to investigate negative STE in the context of refugees and forced migration while taking positive STE into account, and thus add to the scarce literature by studying a closely representative sample of the German majority group. Three different mechanisms were examined as mediators of the negative as well as positive STE in this yet unexplored intergroup setting and cultural context: (a) attitude generalization, (b) multiculturalism operationalized as acceptance of diversity, and (c) ingroup reappraisal operationalized as national pride; the latter two stem from the deprovincialization hypothesis. Regarding our first hypothesis, positive contact with foreigners was related to a more positive, and negative contact was related to a more negative, attitude towards foreigners living in Germany (primary outgroup) and also towards refugees (secondary outgroup) beyond actual refugee contact. This provides empirical support for the core component of both positive and negative STEs. For the second hypothesis, results were less clear since we found empirical support for the mediating mechanism of attitude generalization and acceptance of diversity, but not for national pride. This emerged for both positive and negative STE, thereby lending support to the idea that both work via similar mechanisms. To summarize these mechanisms, in line with our hypotheses, positive contact was associated with a more positive primary outgroup attitude and a higher acceptance of diversity, and thus indirectly associated with a more positive secondary outgroup attitude. As hypothesized, negative contact operated via the same pathways, albeit in the opposite direction; it was related to a more negative primary outgroup attitude and a lower acceptance of diversity, and thus indirectly associated with a more negative secondary outgroup attitude. With regard to the third hypothesis, it became clear that both negative contact and positive contact affected attitudes towards refugees to a larger degree via attitude generalization compared to multiculturalism. Effect sizes were small to medium, and thus comparable to those found in meta-analyses (e.g., Pettigrew & Tropp, 2006; Sparkman, 2020).

Our findings indicate that negative STE might occur in an intergroup context of forced migration, whereby members of the majority group generalize from unrelated contact experiences with other migrant outgroups to form an opinion on forced migrants. Especially in cases where actual contact with forced migrants is (yet) absent or unlikely, this could have detrimental effects. It could contribute to a negatively biased baseline stance regarding this newly arriving outgroup before contact might even occur. However, we also found an indication that positive STE might similarly apply. Majority group members might generalize from positive encounters with other migrant groups to form a more positive attitude towards forced migrants. Our results indicate these possibilities but are subjected to limitations from our cross-sectional study design.

Nevertheless, our results regarding negative (and positive) STE via attitude generalization are congruent with those of the (cross-sectional) study by Ünver et al. (2021) that was situated in a different national context (Turkey). Even though the gathered evidence is merely cross-sectional, it emerged in a similar fashion in two intergroup contexts where the host cultures differ between one another and also differ regarding their cultural similarity to the refugee outgroup. Germany is a WEIRD (Western, educated, industrialized, rich, democratic; Henrich et al., 2010) country in Central Europe, with historical ties to Christianity. In comparison, Turkey cannot be counted as a WEIRD country, and has historical ties to Islam, a religion practiced by many Turks and Kurds and also by many Syrian refugees. Cultural differences between the two countries and also in relation to Syria emerge when looking at other comparative indices, like the Hofstede values (Hofstede, n.d.). Based on religious differences as well as differences in appearance alone, refugees likely have a much higher probability of being recognized (and ostracized) as an outgroup in Germany compared to Turkey. On the other hand, Ünver et al. (2021) mention that the potential for being ostracized also exists in the Turkish context, for example, via language barriers.

That negative STE occurs in both contexts speaks to the generalizability of the phenomenon across intergroup settings/national contexts. This does not come as a surprise since the contact hypothesis (which lies at the core of STE) has been replicated across different intergroup settings, including minimal groups. Nevertheless, this is so far speculation and, ultimately, more empirical research on negative STE across intergroup settings and national contexts is needed.

On a more theoretical level, the results suggest that positive and negative STE might operate via the same mechanisms, which include the multiculturalism pathway. As Vezzali et al. (2021) state, past research indicated that both positive and negative STE operate via attitude generalization. Effects via multiculturalism have been shown for positive STE but had not been investigated regarding negative STE. We provide empirical results suggesting that multiculturalism is an additional mechanism by which negative STE might occur. Negative contact experiences with a primary outgroup diminish both positive attitudes towards this outgroup as well as general acceptance of diversity (i.e., multiculturalism). In turn, both lead to a less favorable evaluation of outgroups unrelated to the contact scenario. The contrast analyses conducted in light of Hypothesis 3 indicate that the attitude generalization mechanism exerts a relatively stronger indirect effect than the multiculturalism mechanism. Yet, we have to acknowledge shared method bias as a potential alternative explanation (Lolliot et al., 2013, p. 88). Nevertheless, we identified the multiculturalism pathway as an independent additional route, which should be investigated in future studies on negative STE. It differs from the attitude generalization mechanism in that it can be seen as a mediator involving the self (Vezzali et al., 2021), and a process whereby internal adjustment of the worldview underlies the STE, namely “broaden[ing of] individual horizons . . . [so that] the value of other cultures can be more easily perceived and accepted“ (Ebbeler, 2020, p. 36). In general, more knowledge of alternative working mechanisms beyond attitude generalization is needed, ideally involving various intergroup scenarios or cultural settings. So far it also remains unclear whether our findings would replicate for (negative) STEs from a minority group perspective.

Although we did not find support for deprovincialization via national pride, there are many different operationalizations of deprovincialization, and thus further mechanisms that need to be explored. Since the original work by Pettigrew (2009), researchers have criticized the operationalization of ingroup reappraisal via pride (Ebbeler, 2020). The idea behind the proposed working mechanism is that positive contact experiences with outgroup members should broaden the perspective by making ingroup membership less important. A less important ingroup membership should reduce outgroup devaluation stemming from the fact that they are not part of the ingroup (Lolliot et al., 2013; Zick et al., 2011). We argued that negative contact could operate this mechanism in the opposite direction. However, national pride might not be a good indicator for ingroup reappraisal from a majority perspective. Prior criticism of Pettigrew’s (2009) deprovincialization hypothesis includes that the theoretical description of said process remained rather vague (Vezzali et al., 2021). In the current study, the measure of national pride was a single-item measure, while other facets of ingroup identification were not covered. Especially in Germany, national pride can be interpreted in different ways (see e.g., Wagner et al., 2012). Wagner et al. (2012) “assume that being proud to be a German is an indicator of generalized positive assessment of the nation” (p. 322), as opposed to “a positive evaluation of German history [which] may reflect idealization of the Holocaust” (p. 322). We can try to empirically approach this question, regressing political ideology (left to right) and shame for the Holocaust (Wasmer & Baumann, 2018, p. 40) on our measure of national pride. For the final sample (N = 1,553) as well as the sample of listwise valid cases (N = 2,961), a more left-wing political orientation is associated with higher national pride, whereas shame for the Holocaust does not exert a significant effect (see Table B6, supplemental material). This implies that our measure of national pride, at the very least, does not seem to reflect the second component mentioned by Wagner et al. (2012). Our lack of empirical support for deprovincialization through national pride could yet be rooted in the methodology of the measurement of national pride rather than in its content. First, ingroup reappraisal describes a longitudinal process, which is obviously not something that can be depicted with the current cross-sectional research design. Second, our operationalization via national pride covers only one aspect of the theorized phenomenon but other aspects also have to be taken into account, for example, it would also be conceivable to operationalize ingroup reappraisal as (change in) ingroup attitude. Future research should focus on all these aspects, preferably with a longitudinal design, and could thus very well uncover empirical support for Pettigrew’s (2009) deprovincialization hypothesis.

Our findings have several practical implications. First, they suggest that interventions could reduce the consequences of negative STEs in different ways. They could target specific outgroup attitudes but they could also target diversity beliefs directly. Targeting diversity beliefs in general might be a more effective intervention strategy for avoiding spillover of prejudice via negative STE if attitudes towards a primary outgroup are deeply ingrained and thus potentially harder to change. In the end, a rigorously designed intervention study across various intergroup contexts would be needed to generate empirical support for these suggestions.

Additional practical implications can be drawn. If both positive and negative STEs work through similar mechanisms, it might be possible to counteract the consequences of negative STE by fostering positive intergroup contact. There is an ongoing scientific debate about whether positive and negative contact differ in terms of effect strength. From visual inspection alone, it becomes clear that our effects from negative STE paths seem smaller (in absolute terms) than those of positive STE paths. These findings are in line with results from Lissitsa and Kushnirovich (2018), who found stronger positive, rather than negative, STE via attitude generalization. However, they contradict findings from Graf et al. (2014) and Árnadóttir et al. (2018) that suggest stronger influence from negative contact as opposed to positive contact (valence asymmetry). A formal statistical test of this idea was beyond the scope of the current article but could be attempted in future research. It would be equally important to explore the potential for interaction between positive and negative contact (see e.g., Árnadóttir et al., 2018) in light of its consequences for secondary transfer effects.

Finally, it should be noted that naming the newly arrived outgroup “refugees” within the groups of people associated with the label “foreigners” could in itself be seen as evidence for STE. In categorizing these outgroups together, attitude change towards the novel outgroup via evaluative conditioning should become more likely (Ebbeler, 2020, p. 38).

The current findings are subject to several limitations. Possibly the most important is that all results stem from cross-sectional analyses. While STE is understood and needs to be treated in terms of processes, cross-sectional data can only give hints regarding interrelations, controlling for other factors. The issue of disentangling the direction of such interrelations has been termed secondary contact problem (Tausch et al., 2010). In our model, the pathways were assumed in a theory-based fashion, but empirical causalities might differ from these assumptions. This is a problem that is common to intergroup contact research (Kotzur & Wagner, 2021), presumably, at least in part, due to practical reasons. Previous research on negative STEs has mostly been cross-sectional (Lissitsa & Kushnirovich, 2020; Vezzali et al., 2021). Therefore, more research is needed with longitudinal and experimental designs with an effective manipulation control. As experimental manipulations of negative contact carry with them a certain ethical issue, observational longitudinal data (in terms of naturally occurring quasi-experiments) could be a valuable alternative source.

Another limitation of the current study concerns the measurement of constructs, many of which were single items. Single-item measures need to be treated cautiously because they might have a lower reliability and might not capture all aspects of complex multifaceted constructs. Using secondary survey data in this study, we are subjected to this limitation, which is common to large-scale survey research. Especially our operationalization of ingroup reappraisal as national pride and the cross-sectional measurement might be criticized. The negative connotation of national pride (in the German context) could be an explanation for why ingroup reappraisal did not mediate the effect of contact with foreigners on attitude towards refugees. It is possible that our finding (that both negative and positive STE work to a significantly larger degree via attitude generalization than via multiculturalism) is due to our operationalization. Measuring attitude generalization might simply be easier and also has a shared method bias with the final outcome (both measuring attitudes towards outgroups). The actual process may be just as strong for multiculturalism or deprovincialization, but the operationalization of acceptance of diversity and national pride, and our single-item measurement of both might not have been suitable.¹

While our choice to analyze secondary survey data generated the mentioned limitations, it also gave us the possibility to conduct sufficiently powered analyses. From Vezzali et al.’s (2021) overview of previous studies on negative STE, it becomes apparent that sufficiently powered studies are needed in this emerging research field.

Similar to bias stemming from the measurement of our constructs, one could argue that omitted variable bias should be considered. As a result, we conducted additional robustness analyses where we estimated our model while controlling for other constructs that might relate to outgroup attitudes (see Appendix B, Table B7, supplemental material). While controlling for political ideology (Thomsen & Rafiqi, 2019), social trust (van der Linden et al., 2017), and anomie (Hövermann et al., 2015), the results still substantially offered the same implications (Appendix B, Table B7, supplemental material).

Another limitation concerns our measurement of secondary outgroup contact. Previous research has called for contact with secondary outgroups to be included as a control variable (e.g., Lolliot et al., 2013; Pettigrew, 2009; Tausch et al., 2010). In the current study, we could only measure the existence of prior contact with refugees but not its frequency or valence.

On a related note, we acknowledge that the primary outgroup (foreigners living in Germany) is an arbitrary group that can imply different meanings for different people. We tried to take this into account by filtering based on the self-reported understanding of the group label “foreigners.” However, this involved the interpretation of qualitative data, which is always subject to personal judgement. Repeating our analyses in an intergroup framework with an easier group distinction would be desirable.

Conclusion

This study is, to our knowledge, one of very few studies on negative STE, and the first to compare three mechanisms in a large German majority sample and in the context of forced migration. We show two parallel pathways by which negative STE might work: via attitude generalization and also via multiculturalism. We further show that attitude generalization seems to be the stronger mechanism, a finding that applies to both negative and positive STE. Generally, our findings imply that negative STE works in a very similar manner as its positive counterpart. The present research opens several avenues for future research. First, longitudinal or experimental studies should be conducted to investigate whether the hypothesized processes can indeed be observed longitudinally and/or from experimental manipulation. Second, knowledge on facilitating conditions as well as on contextual and individual differences regarding negative STEs is yet scarce (Vezzali et al., 2021). Future research should focus on these research gaps and investigate negative STEs for different types of intergroup scenarios and different forms of intergroup contact.

Supplemental Material

sj-docx-1-gpi-10.1177_13684302221105822 – Supplemental material for The negative secondary transfer effect: Comparing proposed mediation theories

Supplemental material, sj-docx-1-gpi-10.1177_13684302221105822 for The negative secondary transfer effect: Comparing proposed mediation theories by Nils T. Henschel and Christina Derksen in Group Processes & Intergroup Relations

Supplemental Material

sj-docx-2-gpi-10.1177_13684302221105822 – Supplemental material for The negative secondary transfer effect: Comparing proposed mediation theories

Supplemental material, sj-docx-2-gpi-10.1177_13684302221105822 for The negative secondary transfer effect: Comparing proposed mediation theories by Nils T. Henschel and Christina Derksen in Group Processes & Intergroup Relations

Footnotes

Acknowledgements

We thank Prof. Dr. Klaus Boehnke, Prof. Dr. Ulrich Kuehnen, and Prof. Dr. Ulrich Wagner for their feedback and general support.

Data accessibility statement

The syntax of the analyses that support the findings of this study is available from the corresponding author, Nils T. Henschel, upon reasonable request. The general data set (without any filtering of open answers) can be accessed at the GESIS website ().

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the German Research Foundation (DFG) as well as matching funds by Jacobs University Bremen as part of the research project “Towards Realistic Computational Models of Social Influence Dynamics” (ToRealSim).

ORCID iD

Nils T. Henschel

Supplemental material

Supplemental material for this article is available online.

Notes

References

Allport

G. W.

(1954). The nature of prejudice. Addison-Wesley.

Andrews

N. P.

Yogeeswaran

Walker

M. J.

Hewstone

(2018). Effect of valenced vicarious online contact on out-group prejudice and perceived out-group variability: A study of online poker. Journal of Applied Social Psychology, 48, 571–581. https://doi.org/10.1111/jasp.12548

Árnadóttir

Lolliot

Brown

Hewstone

(2018). Positive and negative intergroup contact: Interaction not asymmetry. European Journal of Social Psychology, 48, 784–800. https://doi.org/10.1002/ejsp.2365

Barlow

F. K.

Paolini

Pedersen

Hornsey

M. J.

Radke

H. R. M.

Harwood

Sibley

C. G.

(2012). The contact caveat: Negative contact predicts increased prejudice more than positive contact predicts reduced prejudice. Personality and Social Psychology Bulletin, 38, 1629–1643. https://doi.org/10.1177/0146167212457953

Brylka

Jasinskaja-Lahti

Mähönen

T. A.

(2016). The majority influence on interminority attitudes: The secondary transfer effect on interminority attitudes. International Journal of Intercultural Relations, 50, 76–88. https://doi.org/10.1016/j.ijintrel.2015.12.007

Bundesamt für Migration und Flüchtlinge (BAMF). (2021). Schlüsselzahlen Asyl 2021. [Key figures Asylum 2021]. https://www.bamf.de/SharedDocs/Anlagen/DE/Statistik/SchluesselzahlenAsyl/flyer-schluesselzahlen-asyl-2021.pdf?__blob=publicationFile&v=3

Byrne

B. M.

(2016). Structural equation modeling with AMOS: Basic concepts, applications, and programming (3rd ed.). Routledge. https://doi.org/10.4324/9781315757421

Cowling

M. M.

Anderson

J. R.

Ferguson

(2019). Prejudice-relevant correlates of attitudes towards refugees: A meta-analysis. Journal of Refugee Studies, 32, 502–524. https://doi.org/10.1093/jrs/fey062

Ebbeler

(2020). Lateral attitude change in social groups. The role of group similarity and individual differences [Doctoral dissertation]. Rhenish Friedrich Wilhelm University of Bonn. https://nbn-resolving.org/urn:nbn:de:hbz:5-63557

10.

GESIS – Leibniz-Institut für Sozialwissenschaften. (2017). Allgemeine Bevölkerungsumfrage der Sozialwissenschaften ALLBUS 2016. [German General Social Survey 2016] (GESIS Datenarchiv, Köln. ZA5250 Datenfile Version 2.1.0. https://doi.org/10.4232/1.12796

11.

GESIS – Leibniz-Institut für Sozialwissenschaften. (2018). Allgemeine Bevölkerungsumfrage der Sozialwissenschaften ALLBUS 2016 – Zusatzdatensatz offene Frage “Ausländer.” [German General Social Survey 2016 Supplementary Data Set Open Question ‘Foreigners’] (GESIS Datenarchiv, Köln. ZA5254 Datenfile Version 1.0.0. https://doi.org/10.4232/1.13035)

12.

Graf

Paolini

Rubin

(2014). Negative intergroup contact is more influential, but positive intergroup contact is more common: Assessing contact prominence and contact prevalence in five Central European countries. European Journal of Social Psychology, 44, 536–547. https://doi.org/10.1002/ejsp.2052

13.

Harwood

Paolini

Joyce

Rubin

Arroyo

(2011). Secondary transfer effects from imagined contact: Group similarity affects the generalization gradient. British Journal of Social Psychology, 50, 180–189. https://doi.org/10.1348/014466610X524263

14.

Henrich

Heine

S. J.

Norenzayan

(2010). The weirdest people in the world? Behavioral & Brain Sciences, 33, 61–83. https://doi.org/10.1017/S0140525X0999152X

15.

Hodson

Crisp

R. J.

Meleady

Earle

(2018). Intergroup contact as an agent of cognitive liberalization. Perspectives on Psychological Science, 13, 523–548. https://doi.org/10.1177/1745691617752324

16.

Hofstede

(n.d.). Country comparison graphs. GeertHofstede.com. https://geerthofstede.com/country-comparison-graphs/

17.

Hövermann

Messner

S. F.

Zick

(2015). Anomie, marketization, and prejudice toward purportedly unprofitable groups: Elaborating a theoretical approach on anomie-driven prejudices. Acta Sociologica, 58, 215–231. https://doi.org/10.1177/0001699315587988

18.

Jasinskaja-Lahti

Vezzali

Ranta

Pacilli

M. G.

Giacomantonio

Pagliaro

(2020). Conditional secondary transfer effect: The moderating role of moral credentials and prejudice. Group Processes & Intergroup Relations. Advance online publication. https://doi.org/10.1177/1368430220940401

19.

Joyce

Harwood

(2014). Improving intergroup attitudes through televised vicarious intergroup contact: Social cognitive processing of ingroup and outgroup information. Communication Research, 41, 627–643. https://doi.org/10.1177/0093650212447944

20.

Kotzur

P. F.

Wagner

(2021). The dynamic relationship between contact opportunities, positive and negative intergroup contact, and prejudice: A longitudinal investigation. Journal of Personality and Social Psychology, 120, 418–442. https://doi.org/10.1037/pspi0000258

21.

Lissitsa

Kushnirovich

(2018). Secondary transfer effect of positive and negative online contact between groups involved in high-intensity conflict. International Journal of Intercultural Relations, 67, 71–80. https://doi.org/10.1016/j.ijintrel.2018.10.001

22.

Lissitsa

Kushnirovich

(2020). Is negative the new positive? Secondary transfer effect of exposure to LGBT portrayals in TV entertainment programs. Journal of Applied Social Psychology, 50, 115–130. https://doi.org/10.1111/jasp.12644

23.

Lolliot

Schmid

Hewstone

Al Ramiah

Tausch

Swart

(2013). Generalized effects of intergroup contact: The secondary transfer effect. In Hodson

Hewstone

(Eds.), Advances in intergroup contact (pp. 81–112). Psychology Press.

24.

Mähönen

T. A.

Jasinskaja-Lahti

(2016). Ramifications of positive and negative contact experiences among remigrants from Russia to Finland. Cultural Diversity and Ethnic Minority Psychology, 22, 247–255. https://doi.org/10.1037/cdp0000059

25.

Meleady

Forder

(2018). When contact goes wrong: Negative intergroup contact promotes generalized outgroup avoidance. Group Processes & Intergroup Relations, 22, 688–707. https://doi.org/10.1177/1368430218761568

26.

Muthén

L. K.

Muthén

B. O.

(2017). Mplus: Statistical analysis with latent variables: User’s guide (Version 8). Muthén & Muthén.

27.

Nevitt

Hancock

G. R.

(2001). Performance of bootstrapping approaches to model test statistics and parameter standard error estimation in structural equation modeling. Structural equation modeling, 8(3), 353–377. https://doi.org/10.1207/S15328007SEM0803_2

28.

Paolini

Harwood

Rubin

(2010). Negative intergroup contact makes group memberships salient: Explaining why intergroup conflict endures. Personality and Social Psychology Bulletin, 36, 1723–1738. https://doi.org/10.1177/0146167210388667

29.

Pettigrew

T. F.

(2009). Secondary transfer effect of contact: Do intergroup contact effects spread to noncontacted outgroups? Social Psychology, 40, 55–65. https://doi.org/10.1027/1864-9335.40.2.55

30.

Pettigrew

T. F.

Tropp

L. R.

(2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90, 751–783. https://doi.org/10.1037/0022-3514.90.5.751

31.

Renner

Thomas

Mikulajova

Newman

(2017). Threat perception and modern racism as possible predictors of attitudes towards asylum seekers: Comparative findings from Austria, Germany, and Slovakia. International Journal of Business and Social Research, 7. https://doi.org/10.18533/ijbsr.v7i12.1081

32.

Sparkman

(2020). Multicultural experiences and the secondary transfer effect of intercultural attitudes. Social Psychology, 51, 267–283. https://doi.org/10.1027/1864-9335/a000414

33.

Tausch

Hewstone

Kenworthy

Psaltis

Schmid

Popan

Cairns

Hughes

(2010). Secondary transfer effects of intergroup contact: Alternative accounts and underlying processes. Journal of Personality and Social Psychology, 99, 282–302. https://doi.org/10.1037/a0018553

34.

Thomsen

J. P. F.

Rafiqi

(2017). When does superficial inter-group contact reduce antiforeigner sentiment? Negative contact as an essential condition. International Journal of Comparative Sociology, 59, 25–43. https://doi.org/10.1177/0020715217744598

35.

Thomsen

J. P. F.

Rafiqi

(2019). Intergroup contact and its right-wing ideological constraint. Journal of Ethnic and Migration Studies, 45, 2739–2757. https://doi.org/10.1080/1369183X.2018.1493915

36.

Ünver

Cakal

Guler

Tropp

(2021). Support for rights of Syrian refugees in Turkey: The role of secondary transfer effects in intergroup contact. Journal of Community & Applied Social Psychology. Advance online publication. https://doi.org/10.1002/casp.2562

37.

Van der Linden

Hooghe

de Vroome

van Laar

(2017). Extending trust to immigrants: Generalized trust, cross-group friendship and anti-immigrant sentiments in 21 European societies. PLoS One, 12, Article e0177369. https://doi.org/10.1371/journal.pone.0177369

38.

Verkuyten

Thijs

Bekhuis

(2010). Intergroup contact and ingroup reappraisal: Examining the deprovincialization thesis. Social Psychology Quarterly, 73, 398–416. https://doi.org/10.1177/0190272510389015

39.

Vezzali

Di Bernardo

G. A.

Cocco

V. M.

Stathi

Capozza

(2021). Reducing prejudice in the society at large: A review of the secondary transfer effect and directions for future research. Social and Personality Psychology Compass, 15, Article e12583. https://doi.org/10.1111/spc3.12583

40.

Wagner

Becker

J. C.

Christ

Pettigrew

T. F.

Schmidt

(2012). A longitudinal test of the relation between German nationalism, patriotism, and outgroup derogation. European Sociological Review, 28, 319–332. https://doi.org/10.1093/esr/jcq066

41.

Wasmer

Baumann

(2018). German general social survey 2016 (English translation of the German “ALLBUS” Questionnaire). GESIS – Leibniz-Institut für Sozialwissenschaften. https://www.gesis.org/allbus/inhalte-suche/frageboegen

42.

Zick

Küpper

Hövermann

(2011). Intolerance, prejudice and discrimination. A European report. Friedrich-Ebert-Foundation.

43.

Zingora

Graf

(2019). Marry who you love: Intergroup contact with gay people and another stigmatized minority is related to voting on the restriction of gay rights through threat. Journal of Applied Social Psychology, 49, 684–703. https://doi.org/10.1111/jasp.12627

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.27 MB