Research Note Draft

A Research Note: The Differential Impact of Threats on Ethnic Prejudice toward Three Minority Groups in Britain

Key words: Threat, Prejudice, Ethnic Minorities, Intergroup Relations, Survey Experiment, Britain

Abstract Failing to learn from past research, studies that investigate the relationship between threat and prejudice often employ questions that ‘couple’ perceived threats and attitudes to minorities and do not control for attitudes to the native in-group. Using a national survey, in this research note we replicate, update and expand innovative research by Sniderman and colleagues, using the ‘decoupling’ approach to reveal how distinct types of threat motivate prejudice toward three minority groups in Britain. Our results confirm the primacy of cultural over economic threats in motivating hostility toward minority groups, and suggest that safety concerns have gained prominence. By controlling for attitudes toward the native in-group we also reveal the importance of accounting for generalized hostility when studying prejudice.

1

1.

Introduction

A growing number of studies investigate the role of perceived threat in motivating prejudice toward minorities. Commonly a distinction is drawn between two types of threat that are thought to constitute distinct, although not necessarily mutually exclusive, explanations of hostility toward out-groups. The first are economic threats, whether at the individual-level, such as threats to personal finances, or the collective-level, such as threats to the economy (Quillian 1995). The second are cultural threats, such as threats to shared values or ways of life (Sniderman and Hagendoorn 2007). Whereas findings suggest that cultural threats have a consistent and strong effect on hostility to out-groups, the findings for economic threat are mixed (e.g. Gonzalez et al. 2008; Malhotra et al. 2013; McLaren and Johnson 2007). Moreover, fewer studies distinguish a third type of threat: the perception that out-groups are associated with criminality and/or terrorism, and threaten the in-group’s safety (but see, e.g., McLaren and Johnson 2007). While these distinct threats might not be mutually exclusive1, investigating the extent to which they influence attitudes toward minority groups could yield wider insights for the study of prejudice. However, in a pioneering study, Sniderman, Hagendoorn and Prior (2004; henceforth SHP) revealed how much of this research was conflating perceptions of threat with attitudes toward minorities by ‘coupling’ threats and minority groups in the same question (e.g. Gonzalez et al. 2008; McLaren 2003; Schlueter et al. 2008). SHP ‘decoupled’ threat items such as “I am afraid that my economic prospects will get worse because of ethnic minorities” by omitting the reference to ethnic minorities before assessing their impact on prejudice. They thereby ensured that the estimates of the effect of threat on prejudice were no longer a consequence of the measures of threat and prejudice being tautologous.

1

Indeed, recent explanations of prejudice such as the Integrated Threat Theory (ITT) (Stephan and Renfro 2002; Stephan and Stephan 2000) view threats as complementary instead (Riek et al. 2006: 338). 2

In this research note we replicate, update and expand the work of SHP and respond to two problems. First, despite the much-cited findings of SHP2, studies have continued to adopt the ‘coupling’ approach (Dhont and Van Hiel 2011; Pereira et al. 2010; Pettigrew et al. 2010; Wike and Grim 2010). This conflation of threat and prejudice makes it near impossible to distinguish the effects of different threats –economic, cultural or safety– on prejudice.3 Second, studies (including SHP’s) have generally not compared the effect of threat on attitudes toward minorities to attitudes toward the native in-group. This is problematic. To the extent that threat predicts negative attitudes (by members of the in-group) toward the ingroup, threat is not predicting prejudice to minorities but rather a more general feeling of hostility toward others. In short, a baseline effect of threat on general hostility is required to detect the effect of threat on prejudice. Our study examines the role of distinct threats in motivating prejudice toward three minorities in Britain: Black British, Muslims, and EU migrants who settled in Britain in growing numbers following the accession of Central and East European states to the European Union. Using a nationally-representative survey we replicate SHP’s research conducted in the Netherlands in 1997-98 (also: Sniderman and Hagendoorn 2007) in a substantially different economic and political context, namely Britain in 2011, which was grappling with the arrival of a global financial crisis alongside elevated concerns over immigration as a result of rising levels of net migration into the country. We also expand on this research by additionally examining the impact of threat on hostile attitudes toward the 2

Citation count as of April 2016: 223 (Web of Science); 556 (Google Scholar). Measuring threat this way is sometimes justified on the grounds that threat is conceptualized as “intergroup threat”, which “occurs when one group's actions, beliefs, or characteristics challenge the goal attainment or well-being of another group” (Riek et al. 2006: 336). Nonetheless, as Sides and Citrin argue “prejudice does not enter into the perception of threat or the calculation of self-interest”. All that matters is the existence of a threat. If there is a preference of one ethnic group over another, this should reflect the expected impact of these groups on one’s material interests (2007: 479). Thus, we should measure threat without referring to minority groups, even if the threat results from the perceived actions or beliefs of these groups. 3

3

white British in-group. We demonstrate how both feelings of cultural and safety threat trump economic threats in explaining prejudice toward all three minority groups, even amid the Great Recession and inflows of EU migrant workers. This confirms earlier research on the centrality of cultural threats in motivating out-group hostility but suggests that, in a context of global security concerns, threats to safety have gained prominence. Importantly though, the positive impact of collective safety threats on hostility toward white British in-group reveals how it is essential for future academic work to account for a baseline effect of generalized hostility.

2.

SHP’s Decoupling Experiment To gauge the relative importance of economic, cultural and safety threats SHP first

gave one half of a randomly selected sample traditional threat questions that ‘coupled’ objects at risk (i.e. the economic situation, national culture and safety) with minorities. Survey respondents were asked the extent to which they agreed or disagreed with statements that directly associated minorities with threats to economic prospects, neighborhood safety and national identity. Next, for the remaining half of the sample, the objects at risk from minority groups were ‘decoupled’, by modifying the wording of the question, and omitting any reference to minorities. The results demonstrated that in contrast to research that couples objects at risk with minorities, decoupled questions allow for a more robust assessment of the impact of distinct threats on measures of prejudice. The results of the experiment revealed that where threats and minorities were coupled, respondents who felt threatened on one dimension were more likely to feel threatened along all other dimensions. In contrast, when threats were decoupled from minorities, different threats were more distinct and their differential impact on provoking hostility toward each minority group became clear, with cultural threats having ‘by far the largest impact (SHP, p.

4

38). Safety threats had the smallest impact and economic threats occupied a middle ground. These results are often cited as providing compelling evidence that cultural threat is a core predictor of hostility to out-groups and often more so than economic or safety threats (Hjorth 2015; Lucassen and Lubbers 2012; Newman et al. 2012). SHP’s results relate only to the Netherlands at a particular point in time, and so the question of whether cultural threats – when measured independently of the threatening group – are also the strongest predictor of group hostility toward minorities elsewhere and at other times remains unclear. This question is especially intriguing given that many studies rely on data collected before the post-2008 Great Recession, which triggered an increase in the salience of economic concerns (e.g. Valentino et al. 2012; though see Goldstein and Peters 2014; Hainmueller and Hopkins 2015). That the experiments in the Netherlands were undertaken amidst relative economic stability provides one possible explanation as to why economic threats were much weaker predictors of hostility than cultural threats. In contrast our study was conducted amidst the Great Recession, when public concern over the economy dominated the issue agenda, and which may have enhanced the role of economic threats. 4

3.

Data and Measurement

Our data were obtained from a nationally representative online survey of the UK population undertaken between July 11-18 2011 in partnership with YouGov. YouGov relies on a large volunteer opt-in panel of approximately 360,000 British adults, recruited from different sources. From this panel a sub-sample of 1,097 adults was drawn. To further ensure the sample reflects the national adult population, including levels of Internet access, they

4

At the time of our study 57% of British adults rated ‘the economy’, and 29% unemployment, as one of the most important issues facing the country, making economicrelated issues the two most important for citizens (Ipsos-MORI 2011). 5

were weighted for age, gender, region, social grade, party identification, and newspaper readership. Only white British (88%) respondents were included in the analyses. In formulating our questions we largely relied on SHP’s work, adapted to the British context. Details on the survey design, representativeness, response rates, question wording and descriptive statistics are in the Online Appendix. As our measure of prejudice we use group hostility measures for three prominent minority groups: Black British, Muslims, and East Europeans. Our measure of group hostility is an index, based on a question asking respondents the extent to which they (dis)agree with eight characterizations of a group as: trustworthy, selfish, law-abiding, intrusive, slackers, violent, complainers, and by nature inferior. The 1-4 scores on each item were recoded so that high scores indicate greater hostility, and were averaged for each individual. Hostility toward the white British majority (the in-group) was measured to provide a point of reference for expressed hostility toward the minorities. We would expect individuals with a negative view of their own group to be generally negative in their characterization of and interaction with other groups, and to view the world as a more threatening place. To measure perceived economic, cultural and safety threats, we utilize SHP’s decoupled threat measure and replicate their ‘decoupling’ experiment to validate this measure. A randomly selected 40 per cent of the sample was given the ‘coupled’ question, which asked: ‘To what extent do you agree or disagree with the following statements?’:     

‘I am afraid of increasing violence and vandalism in my neighbourhood [by ethnic minorities]’ ‘I am afraid that my own economic prospects will get worse [because of ethnic minorities]’ ‘I am afraid of increasing violence and vandalism in British society [by ethnic minorities]’ ‘These days, I am afraid that the British culture is threatened [by ethnic minorities]’ ‘I am afraid that the economic prospects of British society will get worse [because of ethnic minorities]’

6

The answer categories were recoded so that high scores indicate higher threat. For the remaining 60 per cent of the sample, the objects at risk and the minority group were ‘decoupled’ by modifying the wording of the statements to omit any reference to minorities. There are a number of socio-demographic and psychological characteristics that are routinely identified as sources of prejudice, and which might also predict threat. Following SHP we include controls for education (terminal age of education), occupational status (social grade), level of employment (work status), authoritarian values and self-esteem. In addition, we control for respondents’ gender, age and whether they were born in the UK.

4.

Results

Following SHP, we first validate our decoupled threat measures. We briefly discuss the results here, but include a full discussion in the Online Appendix (Tables A.4 and A.5). In the coupled condition the different threats are highly correlated with an average correlation of 0.69. In the decoupled condition the different threats are not only substantially less correlated, with a mean correlation of 0.39, but the range of magnitudes of the correlation coefficients is much greater (0.50 vs. 0.18). Next, we show that whereas under the coupled condition the different threats share similar predictors -- in particular strength of British identity, individuals’ evaluation of their personal financial situation, and authoritarianism -- under the decoupled condition the predictors vary depending on the type of threat, with economic threats being more closely connected to judgments about the economy and one's own personal finances, and cultural threat most strongly associated with strength of British identity. Collective safety threat is associated with both identity and economic assessments, while neighborhood safety threat is associated with personal financial assessments. Authoritarianism under the decoupled condition is only a significant predictor of cultural threat and collective safety threat, whereas low self-esteem – not a significant predictor of

7

threat in the coupled condition – significantly predicts neighborhood safety, cultural and economic concerns. These results are consistent with SHP and reinforce the idea that when we omit a reference to ethnic minorities in our formulation of threats, we are able to measure distinct threats. This places us in a good position to evaluate their differential impact on prejudice. We next present results from our analysis of the role of threats in predicting variations in prejudice toward the three minority groups. The coefficient estimates and their 95% confidence intervals derived from Ordinary Least Squares (OLS) regressions of prejudice on each threat -- entered individually -- are presented in Figure 1 for the coupled and decoupled conditions. The regressions also include the socio-demographic and psychological control variables. The results for the coupled condition reveal how each type of threat is estimated to have a highly significant and positive effect on hostility toward all minorities. Moreover, none of the threats are significantly related to hostility toward the white British in-group. Using the decoupled items allows us to identify the differential impact of each type of threat. In contrast to the results found under the coupled condition, in this case only some types of threat emerge as having statistically significant effects on hostility and, furthermore, the effect sizes vary across each of the minorities. Although the effects of threat on hostility are strongest when Muslims are concerned, the results show a similar ordering in the importance of distinct threats for predicting hostility toward Black Britons and East Europeans. Threats to neighborhood and collective safety have significant effects on hostility toward all three minorities; however, cultural threat has the strongest effect. Concerns over individual finances only impact hostility toward Muslim and East European minorities significantly with a p-value ≤ 0.10, whereas concern over the national economy does not significantly predict hostility.

8

Consistent with the work of SHP, and despite emerging amid a markedly different context, our results suggest that hostility to minority groups is motivated chiefly by cultural threats. However, in contrast to SHP our results also point to the importance of concerns over increasing rates of violence and vandalism in national and more proximate contexts. Moreover, undertaken during a period of relative economic stability the SHP study suggests economic concerns occupy a middle ground ‘of more consequence than threats to safety, of less than threats to cultural identity’ (p. 38). Despite a wider context of financial crisis and austerity, when the salience of economic concerns was high, our results suggest neither individual nor collective economic threat are significant predictors of hostility toward minorities -- this also despite media coverage that has framed EU migrants from Central and Eastern Europe as posing a distinct economic threat to native white Britons. Lastly, and in clear contrast to the results reported for the coupled condition, we find that hostility toward the white British in-group is significantly and positively related to threats to collective safety, as well as culture (the latter only with a p-value ≤ 0.10). This might imply that, to a certain extent, concerns about wider society are related to negative views of people more broadly and generalized hostility, not just to views of specific minority groups.

9

Figure 1. Unstandardized OLS Coefficient Estimates and 95% Confidence Intervals of the Impact of Coupled (C) and Decoupled (D) Threats (Entered Individually) on Group Hostility

Note: Models include control variables (see Appendix Table A.6).

10

In Figure 2 we present the OLS coefficient estimates from models that include the significant decoupled threats simultaneously to investigate their independent impact on hostility. The results underscore the dominance of cultural threat in explaining hostility toward all minority groups, confirming the findings of previous research. Irrespective of the dominant role of cultural threat, however, a threat to neighborhood safety is significantly related to hostility toward Black Britons, suggesting that aside from cultural considerations prejudice toward this group may be rooted in concerns over threats to safety. The impact of perceptions of increasing violence and vandalism in society on hostility toward the three groups becomes insignificant when we control for other threats, in particular cultural threat. One possible explanation for the latter finding is that concerns about violence in wider society and feelings of cultural threat might originate from a more general concern about British norms and values. This idea is supported by our previous finding that strength of British identity is positively related to both collective safety and cultural threat. This more general concern about norms and values is not solely restricted to minority groups: the significant and positive effect of collective safety threat on hostility toward the white British in-group (in this instance trumping cultural threat) indicates that this concern relates to a view of any British resident as untrustworthy, selfish and violent -- some of the characterizations used in our group hostility measure.

11

Figure 2. Unstandardized OLS Coefficient Estimates and 95% Confidence Intervals of the Impact of Threats on Group Hostility

Note: Models include control variables (see Appendix Table A.7).

Given the more general concern about British norms and values, does the impact of threat on prejudice remain when we account for hostility toward the white British in-group? In answering this question we extend SHP`s work. The coefficient estimates reported in Figure 3 show that while hostility toward the in-group is positively and significantly related to hostility toward minorities, the impact of cultural threat on hostility toward minorities remains unaffected when attitudes toward the in-group are taken into account. Nor is the impact of neighborhood safety threat on prejudice toward Black Britons affected. So, although we find some evidence for a baseline effect of wider concern about society on hostility toward people generally, this does not negate our finding of the centrality of threats to culture, in particular, in predicting hostility toward three minority groups.

12

Figure 3. Unstandardized OLS Coefficient Estimates and 95% Confidence Intervals of the Impact of Threats on Group Hostility, with White British Hostility

Note: Models include control variables (see Appendix Table A.8).

5.

Conclusion

Studies of intergroup relations routinely trace prejudice toward perceived out-groups to feelings of economic, safety or cultural threat. Yet, as past research shows, identifying the relative importance of distinct types of threat in explaining prejudice toward minorities can shed new light on how citizens navigate rising ethnic and cultural diversity. In this research note we have made four contributions to the wider literature. First, by replicating SHP’s ‘decoupling experiment’ in the Netherlands but within the context of Britain, we provide further evidence for how decoupling a reference to minorities from perceived objects at risk provides a far more robust assessment of the impact of different types of threat on prejudice. Though some studies continue to couple minorities with perceived threats, our findings underscore the crucial importance of decoupling minorities from anticipated threats. Second, we have also extended current research by including a measure of hostility toward the native in-group, revealing how the effect of threat on hostility is not always limited to ethnic minority groups, but can be directed to people more generally. 13

This underscores the importance of distinguishing the effect of threat on prejudice from that on general hostility. Third, despite being undertaken amidst a markedly different economic context, our results provide further evidence that cultural threat is the strongest predictor of hostility toward minority groups, and more so than feelings of economic and safety threat. Despite the post-2008 financial crisis, neither threats to individual or collective economic well-being had significant effects on prejudice toward minorities. These findings contrast those of SHP, who found economic concerns occupy a middle ground in predicting prejudice between threats to safety and culture. Yet, similar to their study, our results relate to only one West European state, and therefore the extent to which they apply to other cases is still to be investigated. Fourth, some interesting findings emerge for the different minority groups. The lack of an effect of either individual or collective economic threat on hostility toward Central and East European migrants is interesting given that their presence coincides with an economic downturn. This underscores the conclusion that economic concerns play a highly limited role in explaining negative reactions to minority groups. In addition, although cultural threat is the strongest predictor of hostility toward the three groups – encompassing concerns of increased violence in society – this finding is especially striking with regards to Black Britons. These communities have a long history in Britain, first arriving in the 1950s from Commonwealth states, while recent evidence suggests this group is better integrated and more accepted (Ford 2008). Although this may be the case, hostility toward this group is still chiefly motivated by concerns about threats to British values and ways of life, as is the case with hostility toward Muslims and East Europeans. Concerns about neighborhood safety also have a strong effect, but solely on hostility toward Black British. This may be a result of the over-representation of Black British youths in news stories on criminality and violence; however, it is important to note that without any evidence this remains speculation.

14

We have replicated, updated and extended the work of Sniderman et al. (2004) by examining the relationship between threat and prejudice in a very different economic, cultural and security environment. We have also controlled for the potential confounding of ethnic prejudice with generalized hostility. In doing so we have shown how the environmental context can impact the relative effects of different types of threats and that some individuals do express hostility towards their own in group, and yet overall, our results confirm the primacy of cultural threat as the strongest predictor of hostility toward minority groups.

References Dhont, K. and Van Hiel, A. 2011. ‘Direct Contact and Authoritarianism as Moderators between Extended Contact and Reduced Prejudice: Lower Threat and Greater Trust as Mediators.’ Group Processes & Intergroup Relations, 14(2): 223-237 Ford, R. 2008. ‘Is Racial Prejudice Declining in Britain?’ British Journal of Sociology, 59(4): 609-636 Goldstein, J. L., and M. E. Peters. 2014. ‘Nativism or Economic Threat: Attitudes toward Immigrants during the Great Recession.’ International Interactions 40.3: 376-401 Gonzalez, Velasco K., Verkuyten, M., Weesie, J. and Poppe, E. 2008. ‘Prejudice towards Muslims in the Netherlands: Testing Integrated Threat Theory.’ British Journal of Social Psychology, 47: 667-685. Hainmueller, J. and Hopkins, D.J. 2015. ‘The Hidden American Immigration Consensus: A Conjoint Analysis of Attitudes toward Immigrants.’ American Journal of Political Science, 59(3): 529-548 Hjorth, F. 2015. ‘Who Benefits? Welfare Chauvinism and National Stereotypes.’ European Union Politics, online early view

15

Ipsos-MORI. 2011. Issues Index. Available online: http://www.ipsos-mori.com (accessed November 24 2011) Lucassen, G. and Lubbers, M. 2012. ‘Who Fears What? Explaining Far Right-Wing Preference in Europe by Distinguishing Perceived Cultural and Economic Ethnic Threats.’ Comparative Political Studies 45(5): 547-574 Malhotra, N., Margalit, Y. and Hyunjung Mo, C. 2013. ‘Economic Explanations for Opposition to Immigration: Distinguishing between Prevalence and Conditional Impact.’ American Journal of Political Science, 57(2): 391-410 McLaren, L.M. 2003. ‘Anti-Immigration Prejudice in Europe: Contact, Threat Perception, and Preferences for the Exclusion of Migrants.’ Social Forces 81 (3): 909-36 McLaren, L.M. and Johnson, M. 2007. ‘Resources, Group Conflict and Symbols: Explaining Anti-Immigrant Hostility in Britain.’ Political Studies 55(4): 709-732 Newman, B.J., Hartman, T.K. and Taber, C.S. 2012. ‘Foreign Language Exposure, Cultural Threat, and Opposition to Immigration.’ Political Psychology, 33(5): 635-657 Pereira, C., Vala, J. and Costa-Lopes, R. 2010. ‘From Prejudice to Discrimination: The Legitimizing Role of Perceived Threat in Discrimination against Immigrants.’ European Journal of Social Psychology, 40: 1231-1250 Pettigrew, T.F., Wagner, U. and Christ, O. 2010. ‘Population Ratios and Prejudice: Modelling Both Contact and Threat Effects.’ Journal and Ethnic and Migration Studies, 36(4): 635-650 Quillian, L. 1995. ‘Prejudice as a Response to Perceived Group Threat: Population Composition and Anti-Immigrant and Racial Prejudice in Europe.’ American Sociological Review, 60: 586-611. Riek, B.M., Mania, E.W. and Gaertner, S.L. 2006. ‘Intergroup Threat and Outgroup Attitudes: A Meta-Analytic Review’, Personality and Social Psychology Review, 10:

16

336-353 Schlueter, E., Schmidt, P. and Wagner, U. 2008. ‘Disentangling the Causal Relations of Perceived group Threat and Outgroup Derogation: Cross-National Evidence from German and Russian Panel Surveys.’ European Sociological Review, 24: 567-581. Sides, J. and Citrin, J. 2007. ‘European Opinion about Immigration: The Role of Identities, Interests and Information.’ British Journal of Political Science, 37: 477-504. Sniderman, P.M., Hagendoorn, L. and Prior, M. 2004. ‘Predisposing Factors and Situational Triggers: Exclusionary Reactions to Immigrant Minorities.’ American Political Science Review, 98(1): 35-49. Sniderman, P.M. and Hagendoorn, L. 2007. When Ways of Life Collide: Multiculturalism and Its Discontents in the Netherlands. Princeton, NJ: Princeton University Press Stephan, W.G. and Renfro, C.L. 2002. ‘The Role of Threat in Intergroup Relations’, in D.M. Mackie and E.R. Smith (eds.), From Prejudice to Intergroup Emotions: Differentiated Reactions to Social Groups, New York: Psychology Press, pp. 191-207 Stephan, W.G. and Stephan, C.W. 2000. ‘An Integrated Threat Theory of Prejudice’, in S. Oskamp (ed.), Claremont Symposium on Applied Social Psychology, Hillsdale, NJ: Erlbaum, pp. 23-45 Valentino, N.A., Brader, T. and Jardina, A.E. 2012. ‘Immigration Opposition among U.S. Whites: General Ethnocentrism or Media Priming of Attitudes about Latinos?’ Political Psychology, 34(2): 149-166 Wike, R. and Grim, B.J. 2010. ‘Western Views towards Muslims: Evidence from a 2006 Cross-National Survey.’ International Journal of Public Opinion Research, 22(1): 425

17