127 4 Figure 3. Do voters project their views onto politicians? Figure 3 shows whether projection plays a role in voters’ expectations of politicians. Voters largely expect politicians to share the same viewpoints on same-sex adoption as themselves. 66 percent of voters on the pro-flank expect politicians, irrespective of religion, gender or migration background, to be pro-same-sex adoption as well. Only 13 percent of voters on the anti-flank expect politicians to be in favor of samesex adoption. Moderates maintain intermediate positions: 43 percent of voters with moderate positions on same-sex adoption expect politicians to be in favor of samesex adoption. See the appendix for all exact means and an analysis of respondents who answered “I don’t know” when asked what they expect of politicians. In short, those who have moderate attitudes towards same-sex adoption do not answer “I don’t know” significantly more often than voters on the flanks. In sum, voters indeed project their own views on same-sex adoption onto politicians. Next, I examine whether stereotyping and projection influence each other: do voters only stereotype politicians when their position on same-sex adoption is lukewarm? Figure 4 reveals that 76 percent of voters who are pro-same-sex adoption expect the same of non-religious politicians, whereas 61 percent of them expect this of Muslim politicians. This is indeed a lower percentage, but the confidence intervals overlap, meaning the difference is not statistically significant. Conversely, voters who are antisame-sex adoption tend to expect politicians to think as they do. 21 percent of voters who are anti-same-sex adoption expect non-religious politicians to support samesex adoption, whereas only 8 percent expect Muslim politicians to support it. This is indeed a lower percentage, but again the confidence intervals overlap, meaning the difference is not statistically significant. What is statistically significant is that flankers on both the pro- and anti- sides of the histogram in Figure 1 differ in their expectations of both Muslim and non-religious politicians, leading me to conclude that flankers
RkJQdWJsaXNoZXIy MjY0ODMw