In this paper, I aim at methodologically defining and testing a new measure of opinion constraint, i.e. individuals’ capacity to connect their political ideas meaningfully and consistently. Opinion constraint has been originally defined by Converse as “the success we would have in predicting, given initial knowledge that an individual holds a specified attitude, that he holds certain further ideas and attitudes” (1964, 3). Methodologically speaking, this concept has been measured by means of several techniques, among which the most prominent is correlations between pairs of issues. However, correlations have the limit not to be suitable for individual-level analyses. To overcome this gap, this paper builds on previous studies (Sauer et al. 2011) by proposing a new measurement based on squared regression residuals. For the analyses, I focus on the European case, using data from the 2014 European Election Studies. First, I test what model of issue structure described in the literature best fits each country by means of CFA. Then, residuals are computed on the basis of OLS regressions, where one factor (i.e. issue dimension) resulting from the CFA is regressed on another.

Organisé par la Chaire de recherche en études électorales et la Chaire de recherche du Canada en démocratie électorale.

Contactez semra.sevi@umontreal.ca pour rejoindre la conférence sur Zoom.

DATE : mardi 14 juillet, 12h-13h