Religious Styles Predict Interreligious Prejudice: A Study of German Adolescents with the Religious Schema Scale

Abstract

Based on a sample of 340 German adolescents age 12 to 25, this article presents an analysis of the effects of religion on two instances of interreligious prejudice: anti-Islamic and anti-Semitic prejudice. Reflecting the emergent interest in implementing a perspective of religious maturity and religious development into research on religion and prejudice, the present study has included the Religious Schema Scale (RSS) which, with its three subscales, Truth of Texts & Teachings (ttt), Fairness, Tolerance & Rational Choice (ftr), and Xenosophia/Interreligious Dialog (xenos), differentiates religious styles. Regression analyses indicate the superior explanatory power of the RSS in comparison to other measures of religiosity. The RSS subscale ttt relates to and predicts anti-Islamic and anti-Semitic prejudice, whereas ftr and xenos relate to and predict disagreement with interreligious prejudice. Results of an analysis of variance using high agreement on ttt, ftr, and xenos for group construction indicate a decrease in interreligious prejudice in relation to religious development.

Notes

¹The model of Leak and Finken has combined the FDS, together with a measure for religious quest (a shortened 13-item version of Altemeyer & Hunsberger's, 1992, Quest scale), into a latent variable called religious openness. Even though the relation of religious development with openness has been evidenced also by other studies (e.g., Streib, Hood, Keller, Csöff, & Silver, 2009), we may raise the question of whether religious development could not offer more than merely religious openness.

²A group of Turkish-Islamic respondents (n = 27) had been excluded from the original database of 367 total cases because we regarded this group to be too small for comparative analyses and because our analyses included a focus specifically on anti-Islamic prejudice in subjects more in the traditional German mainstream.

³Translation of sample items are ours. Unless indicated otherwise, the items and scales are available in German language only and are neither based on an English original nor been translated into English in full length so far.

*Correlation is two-tailed significant on the .05 level.

**Correlation is two-tailed significant on the .01 level.

⁴The nonsignificance of ttt and ftr is not explained by collinearity (VIP < 1.3 for all variables); nevertheless, when ftr is excluded from the regression, xenos has a small but significant regression weight (β = −.14, p = .018), whereas ttt remains insignificant.