Keywords: Affective polarization, survey experiments, conjoint

Studies on affective polarization have traditionally discussed which causal mechanisms have the greatest influence on the phenomenon. In particular, it has been debated whether partisan identity is more important than ideological and issue positions or the other way around. Without an experimental approach, the evidence is contradictory and sometimes too dependent on the national context. However, in this paper we argue that the conjoint design is the best possible alternative to resolve a far-reaching debate and we use for this purpose an example of a conjoint experiment implemented in the framework of an opinion survey applied to a representative sample of the Spanish population. In my conjoint experiment, participants were asked to show their preference for a potential partner between two profiles. Each profile was composed by combining different attributes including: the party the person voted for, their ideology, some social identities (religiosity, economic class and territorial feelings) and their position on different issues. The results of the experiment show that partisan identity is the attribute that most influences the choice of partner, well above ideology.

In addition, the paper also discusses why it is preferable to measure the results of the experiment using a Bayesian hierarchical (HB) model. Unlike the aggregate multinomial logit (MNL) model, the HB model allows us to calculate utilities at the individual level and, at the same time, the utilities form a distribution between that is governed by meta-parameters (the mean and variance of the utility distribution). The analyses carried out, using the 'ChoiceModelR' package (a trademark of Decsion Analyst), which applies Markov chain Monte Carlo (MCMC) algorithms to estimate a hierarchical multinomial Logit model with a normal heterogeneity distribution, address some of the weaknesses of classical regression and AMCE (Average Marginal Component Effects) analyses, especially the effect of preference variability on the validity of the estimators.

In short, this paper addresses three main issues: why conjoints can solve one of the major gaps in the literature on affective polarization; how this has been demonstrated in a survey experiment specifically implemented in Spain; how we can advance in the analysis of data derived from conjoint experiments by implementing Bayesian hierarchical models and MCMC algorithms as opposed to the currently dominant approach based on MCE. I hope that these results will help to extend the use of conjoints in surveys to solve similar problems.