84 2 Discussion and conclusion – challenges to descriptive and substantive representation Who favor in-group politicians? In general, I find shared policy is far more important to voters than shared identity (Figure 5) showing that identity compared to policy plays a limited role amongst voters on a whole. Nonetheless, I also reveal differences in voting likelihood based on sharing the same religion and migration backgrounds, but not on gender. Muslim voters in the Netherlands exhibit a preference for Muslim politicians over non-religious and Christian politicians, but this does not hold in France and Germany (Figure 4). The preference of non-religious voters for non-religious over Muslim politicians, however, holds across all three countries, though always with Christian politicians taking an intermediate position (Figure 3). Moreover, individuals with a migration background tend to align themselves more with the high-status outgroup, politicians without a migration background, than with politicians with a migration background (Figure 2) even when that migration background is the same as their own (Figure 1). This research challenges the common assumption that in-group favoritism is specifically a characteristic of minority groups. Contrary to popular beliefs, but in line with the reasoning of Social Identity Theory (Tajfel and Turner, 1979), I reveal that the non-religious majority tends to exhibit favoritism towards their own group and display disfavor towards others slightly more than Muslim minorities favor their in-group, while minorities with a migration background even prefer their out-group. There are several important caveats and implications for further research. Firstly, the eight policy positions randomized in the experiment do not cover the entire policy field, warranting the inclusion of other issues that could better capture Muslim or Islamophobic interests. Secondly, when confronted with a Muslim politician, individuals may rely on heuristics, stereotypes, or cognitive shortcuts to form their opinions about policy positions (Cutler, 2002; Koch, 2000; Lau and Redlawsk, 2001; Portmann, 2022). This suggests that negative perceptions of Muslim politicians, even when they hold the same policy positions as non-religious politicians, may be influenced by additional policy positions and/or stereotypes that come to mind. Thirdly, it is important to understand that there might be a lack of information equivalence between nonreligious and Muslim voters (Dafoe et al., 2018), as voters may not necessarily associate any policy positions with politicians who “do not practice any religion,” potentially leading to heuristics based on projection instead of stereotyping, with more favorable outcomes for politicians who the respondent projected their own policy positions onto. Future research should explore which policy positions come to mind when voters evaluate politicians based on their religious practices or lack thereof. Fourthly, given that I used a novel oversampling method, potential issues could arise that may warrant further exploration using alternative sampling methods to validate results.
RkJQdWJsaXNoZXIy MjY0ODMw