Intra-EU Labour Migration and Attitudes Towards the Welfare State 105 I use several survey questions with categorical responses as our dependent variables, which comes with a number of caveats. It is important to point out that the distance between the categories, for example the gap between disagree and strongly disagree and the gap between agree and strongly agree, may not be equal. Additionally, I attach values to these categories in order to use them within the regression, but the values themselves do not mean anything, only the order in which they are placed matters (Cappelen & Peters, 2018). So, when the value of 5 is assigned to strongly agree and the value of 1 to strongly disagree, a positive coefficient will mean a greater probability of agreeing with the question in the dependent variable. 5.4 Data 5.4.1 Dependent Variables The dependent variables are all drawn from survey questions in the European Social Survey (ESS), round 8 (2016). The ESS is a high-quality, harmonised cross-national individual level survey that provides us with a number of useful questions for measuring attitudes towards redistribution. I follow precedent and primarily use the survey question “Please say to what extent you agree or disagree with each of the following statement: The government should take measures to reduce differences in income levels” (ESS, 2008a: A3, Q.B30: 26) to denote an individual’s attitude towards redistribution. The respondents can answer on a scale from 1 to 5, which I recoded so that 1 = strongly disagree; 2 = somewhat disagree; 3 = neither agree nor disagree; 4 = somewhat agree; and 5 = strongly agree. The proxy is not perfect, the question is broad, unspecific and does not indicate how redistribution should be achieved, for example through higher taxation, direct transfers, or more services (Dallinger, 2022, Margalit & Raviv, 2024). Consequently, our interpretation follows that of the researchers before us who argue that it is a good question to assess attitudes towards a general principle and ideal of fairness and equality rather than a focus on specific practical concerns, detailed policy preferences or an inclination to vote for a specific political party (Burgoon et al., 2012; Cappelen et al., 2018). I test in the additional checks section the influence that the two indicators of mobility have on subjective job insecurity as it is a small piece of the puzzle. Again, following precedent, I use the question “How likely is it that during the next 12 months you will be unemployed and looking for work for at least four consecutive weeks?” (ESS, 2016). The respondents could choose between five options, 1= not at all likely; 2 = not very likely; 3 = likely; 4 = very likely; 5 = not working/not looking for work/never worked, which is recoded into a dummy variable for simplicity in order to indicate whether or not an individual expressed job insecurity. I also check for influence on welfare chauvinism, whether social benefits are considered a strain on the
RkJQdWJsaXNoZXIy MjY0ODMw