1. Economic conditions, group relative deprivation and ethnic threat: A cross-national analysis ESS Dissemination Seminar Warsaw Peter Schmidt, University Giessen and Humboldt Research Fellow of PFBR

2. Based on a paper submitted to a journal for a special issue on the immigration module of the ESS by B.Meuleman/K.Abts/P.Schmidt/T.Pettigrew/E.Davidov

3. Table of Content Introduction Theoretical Background Data
Multilevel SEM Model: Specification and Results
Summary and Limitations
Appendix: Mplus Code Multilevel SEM Model
Open Questions

4. 1. Introduction

5. Introduction
Intergroup contact has received considerable attention in explaining ethnic threat
Group relative deprivation (GRD) has been largely ignored in cross- national research as explanation of economic threat
the 7th round of the European Social Survey (ESS) includes questions assessing respondents’ feelings of group deprivation in relation to immigrants
offers for the first time an opportunity to contextualize the threat- inducing effect of GRD across Europe

6. 2. Theoretical Background

7. Theoretical Background
drawing upon group conflict theory, numerous cross- national studies have evidenced that individual indicators of socio-economic positions, such as employment status, income or education (Gorodzeisky 2011; Kunvovich 2004; Raijman, Semyonov, and Schmidt 2003) national-level economic conditions (Quillian 1995; Schneider 2008; Semyonov, Raijman and Gorodzeisky 2006; Meuleman, Davidov and Billet 2009) are predictive of ethnic threat perceptions
most of the empirical studies specify a direct link between individual and contextual economic indicators and exclusionary attitudes

8. Theoretical Background
since Stouffer`s introduction of the concept ‘relative deprivation‘ (Stouffer et al. 1949; Pettigrew 2015) and Blumer`s (1958) seminal paper ‘Race prejudice as a sense of group position`
it is known that the perceived relative positioning of social groups may mediate the effect of absolute positioning on threat perceptions

9. Theoretical Background
research into Group Relative Deprivation (GRD) shows that feelings that one´s group is unfairly deprived of desirable goods have a far more decisive impact on prejudice compared to living conditions per se (Vanneman and Pettigrew 1972; Walker and Pettigrew 1984; Runciman 1966; Smith et al. 2012; Walker and Smith 2002)
the concept of GRD has largely been neglected in cross- national research on ethnic threat

10. Theoretical considerations
We examine the effect of GRD on perceived ethnic threat across various European countries
focus on two complementary theories on the nature of the relationships between ethnic groups, namely Group Conflict Theory (GCT) and Group Relative Deprivation (GRD)
supplement these frameworks by related explanatory models, namely Intergroup Contact Theory (ICT) and basic human values (PVQ)

11. Group Conflict Theory (GCT)
the central proposition of GCT is that real competition over scarce resources between social groups - such as ethnic groups – results in ethnic conflict and negative attitudes towards outgroups
in particular, the Integrated Threat Theory (Stephan and Stephan 1993, 1996) distinguishes realistic (economic) and symbolic (cultural) threat
importantly, distinct out-groups can be perceived either as an economic (realistic) threat, a cultural (symbolic) threat or a combination of both (Hjerm and Nagayoshi 2011; Meuleman et al )

12. Group Relative Deprivation (GRD) theory
Runciman (1966), 33-35) introduced a crucial distinction between individual Relative Deprivation (IRD) and a feeling that one`s in-group is deprived compared to relevant out-groups (GRD)
GRD mediates the effects of social and economic indicators on perceived ethnic threat
We postulate that (1) individual social economic status influence GRD, which in turn (2) affects perceived threat (level 1)
We assume GRD to mediate the impact of economic context on ethnic threat perceptions (level 2)

13. Related theoretical frameworks: Intergroup Contact Theory (ICT) and basic human values (PVQ)
one can expect that GRD and ICT affect threat perceptions
the causal relationship between GRD and ICT is unclear
value priorities are crucial factors shaping individuals` attitudes towards ethnic minorities and immigrants (Beierlein, Kuntz and Davidov 2016; Davidov et al. 2008; Sagiv and Schwartz 1995)
Underlying mechanisms: (gains and losses, RC explanation)

14. 3.Data

15. Data
data from the 7th round of ESS ( , edition 2.1) is a repeat module of the immigration module from the first ESS wave of
the immigration module in the ESS partly replicates the immigration module included in the first wave of ESS in as well as introduces new questions
a review of the module and its theoretical background can be found on the ESS website

16. Indicators
Perceived ethnic threat was measured using five items that inquired about respondents` feeling whether immigrants were a threat:
for the economy, the labor market, the welfare state, the cultural life and the religious customs
confirmatory factor analysis evidenced that it is not possible to distinguish between the two concepts (economic and cultural threat) with the ESS data
Threat and Values had been found to be approximately invariant over the different European countries (Davidov, Ciecuch and Schmidt 2018)

18. Activity Status: Goldthorpe scores

24. 4. Multilevel SEM Model: Specification and Results

25. Figure 1. Country average scores of group relative deprivation (GRD) and perceived ethnic threat

26. s

27. Estimation
Only 20 countries might hamper estimation (Meuleman/Billiet 2009)
Bayes estimator in Mplus was used (Hox et al. 2012, Stegmüller 2013)

29. Results GRD was a significant predictor of threat Across Individuals
Between european countries
Indirect effect of level 2 variable long term employment on
threat is .298
No direct effect

30. 5. Summary and Limitations

31. Summary
a multilevel structural equation model (MLSEM) demonstrates a considerable effect of GRD on ethnic threat both on the individual and country level
the results indicate that GRD is not only an important mediating factor between social structural positions and perceived threat
it fully mediates the impact of contextual economic indicators on ethnic threat

32. Summary
contact also played a role in the explanation of ethnic threat
individuals experiencing positive contact expressed lower ethnic threat
theoretical considerations lead us to expect that GRD and contact influence the level of ethnic threat

33. Limitations
Endogeneity Problem: Threat may influence contacts and group related deprivation: ethnic threat may result in both avoidance of contact with immigrants, perception of such contact as negative and increased feelings of group deprivation that are activated by threat in the first place
it was not possible to examine the cross-country comparability of the single-item measure for GRD
it was not possible to include an indicator of individual rather than group relative deprivation

34. Thank you for your attention!

35. Details of the estimation
We followed the procedures described in van de Schoot et al. (2014).
All prior distributions were specified to be noninformative with the default N(0,∞) for factor loadings and intercepts and IG(-1,0) or IW(0,-3) for (co)variances.
Use of Gelman-Rubin criterion (Gelman et al. 2004) with 0.01 as the cut-off value (Hox, van de Schoot and Matthijsse 2012).
we used a thinning factor of 50, increased the number of effective draws to 10,000.
The Kolmogorov-Smirnov test comparing the posterior distributions for the chains confirmed convergence for all parameters.

36. Summary
the findings demonstrated that GRD has a considerable effect on ethnic threat both on the individual and country levels of analysis
this effect was consistent and significant
economic indicators, such as socio-economic status on the individual level and country long-term unemployment rates, had a consistent effect on GRD
with vulnerable economic conditions associated with higher levels of GRD
the effect of economic conditions on ethnic threat was mediated by feelings of group deprivation both on the individual and country levels of analysis

37. 6. Appendix: Mplus Code Multilevel SEM Model

43. 7. Open Questions?