The relation between religiosity and Muslims’ social integration: a two-wave study of recent immigrants in three European countries

ABSTRACT

Does their degree of religiosity affect how successfully recent Muslim migrants integrate socially into the host society in terms of their social contacts with the majority population and their ethno-religious group? And/or do these co-ethnic and interethnic social contacts affect the religiosity of Muslim migrants over time? On the basis of a two-wave study among recent migrants in Germany, the Netherlands and the U.K., the relationship between religious practices and social contact is studied. Findings indicate that Muslims who were more religious before migration develop more co-ethnic contacts shortly after moving to Europe, which in turn strengthens their religious practice over time. In contrast to popular discourse, more religious Muslims do not develop fewer social contacts with the majority over time. Thus, social integration into the host society is largely decoupled from religiosity, highlighting the importance of differentiating between co-ethnic and majority contacts when studying religiosity and social integration.

KEYWORDS:

• Religion
• social integration
• contact
• Islam
• attendance
• recent migrant

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. In Turkey, an estimated 98 per cent of the population is Muslim; 97 per cent believe in Allah being the One God and Mohammed the Prophet; 84 per cent participate in Ramadan, 44 per cent attends a mosque weekly and 42 per cent pray daily. For Pakistan, these figures are 96 per cent (Muslim), 98 per cent (Allah the One God), and 59 per cent (weekly attendance) (PEW 2012). Although not all countries have reliable figures on religious membership, practices and attitudes among minority populations, available data indicate similarly high rates of membership among Turkish and Pakistani migrants in Germany, the Netherlands and the U.K. (Haug, Müssig, and Stichs 2009; Maliepaard and Gijsberts 2012; ONS 2013), where the number of “native” Muslims is negligible (estimated <1%), large parts of the host populations are irreligious (NL: 42%; GER: 25%; U.K.: 28%; PEW 2010).

2. In multi-ethnic societies, migrants may also develop ties to members of other ethnic minority groups. However, these are not considered in the current paper.

3. We compared our results with two other models commonly used with two-wave panels, simple CS and LDV models. The CS model treats the change in Y (Y2–Y1) as the dependent variable; in LDV models the dependent variable is Y2, which is regressed on Y1 and other predictors. Since LDV models are sensitive to measurement errors and to cofounding effects of third variables, CS models are usually preferable in two-wave studies (Allison 1990; Johnson 2005). However, the emphasis in two-wave CS models is typically on a simultaneous change in dependent and independent variables (Johnson 2005), whereas the aim of the current study is to estimate the effects of the independent variables at t1 on the changes in the dependent variables between t1 and t2. Therefore, we opted for the method described above. Nevertheless, we also estimated CS and LDV models. Since the findings across models were consistent, we argue that it is not likely that our results are biased because of measurement errors or omitted variables (Johnson 2005, 1074).

4. There are no collinearity issues that might explain the contradictory finding of pre-migration and post-migration prayer (VIF < 5).

Additional information

Funding

Part of this research was funded by the Netherlands Organization for Scientific Research (NWO) [grant number 327-25-005] and within the funding program IPID4all by the University of Bamberg promoted by the German Academic Exchange Service (DAAD) and funded by the Federal Ministry of Education and Research (BMBF).