Solidarity and Self-Interest: Using Mixture Modeling to Learn about Social Policy Preferences

José Alemán, Dwayne Woods


This article addresses the problem of measuring social policy preferences in a valid and reliable way. Scholars have faced a number of challenges in measuring these preferences. First, it is not clear how exactly we should conceive of this domain. Second, the literature presents contradictory findings regarding the effect of contextual factors on policy preferences. Third, abstract preferences regarding the welfare state and information about its performance can affect each other, complicating the attempt to distinguish between the two. Finally, latent manifestations of these preferences might not be equivalent across countries. We develop an approach that validly and reliably measures attitudes about the role of government in addressing inequalities in the market distribution of resources. Mixture modeling and in particular latent class analysis enables us to take advantage of information for multiple countries and survey questions while doing justice to the characteristics of the survey data. Using three waves of the International Social Survey Programme’s module on social inequality, we find that preferences towards the market and the role of government in the economy form four distinct clusters of individuals that we refer to as “moderate altruists”, “moderate egoists”, “extreme altruists”, and “extreme egoists”. These clusters tend to be homogenous with respect to both abstract notions of the role the government should play in the economy as well as about evaluations of actual performance. The exceptions are the last two survey waves, for which we find that one class exhibits a mixed profile of individuals: solidaristic with respect to some indicators, but self-interested with respect to others.


solidarism, self interest, social policy preferences, latent class analysis, mixture modelling

Full Text:




  • There are currently no refbacks.

Copyright (c) 2019 José Alemán, Dwayne Woods

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.