Keywords: Methodological challenges and improvements, including in the areas of sampling, measurement, survey design and survey response or non-response

Populist radical-right parties have been extremely successful in regional, national and supranational elections during the last years, also gaining a lot of scholarly attention lately. However, most attention so far has been given to citizens’ party preferences and voting behavior, while research on the development and structure of PRR attitudes still is in its infancy. However, first promising studies deal with the structure of populist attitudes in the explicit realm (e.g. Akkerman et al. 2014; Schulz et al., 2017), and some suggest that PRR attitudes are a latent construct with the three dimensions nationalism, authoritarianism and populism (Rooduijn, 2014). However, building on so-called two-process-models which propose that attitudes are represented, on the one hand, as propositional evaluations of objects that are deliberately accessible through self-reflection (explicit attitudes), and, on the other hand, as automatic associations of objects and valences that are not intentionally formed and often unconscious (implicit attitudes), we suggest to also conceptualize PRR attitudes in the implicit realm and, as a first step, to develop and test an implicit measure for populist attitudes. We believe that such an approach is especially promising for two reasons: First, research has shown the great relevance of implicit conceptualizations of attitudes towards foreigners and minorities (e.g. Arendt & Northup, 2015) which should be closely related to populist attitudes. And second, research has shown that implicit attitudes have great predictive power in contexts in which voicing respective attitudes explicitly is considered as socially undesirable (e.g., Maier et al., 2014), which – according to recent polls – for populist attitudes is still the case in Germany.
In this presentation, we therefore report about the development of an implicit measure for populist attitudes using so-called Implicit Association Tests (IAT, Greenwald et al., 1998). This IAT was presented in an online survey with N = 1,340 participants representative for German online users together with two explicit measures of populist attitudes based on current literature (Rooduijn, 2014; Schulz et al., 2017) as well as an explicit and an implicit measure for anti-immigrant attitudes.
First findings show an only weak correlation between the implicit and the explicit measure, suggesting that the two measurements indeed tackle different cognitive processes. Implicit populist attitudes show a significant impact on the probability to vote for a populist radical right party in Germany (AfD), however effects of explicit populist attitudes as well as anti-immigrant attitudes turn out as even stronger predictors in the overall sample. In our presentation, we will show additional analyses regarding the predictive power of the implicit measure in parts of the sample which show ambivalent attitudes towards the AfD or even strictly exclude to vote for this party as we expect strongest effects of social desirability in these groups.