Review, Meta-Analysis, and Re-analysis of Conjoint Experiments on Immigration

• In Event: Advances in Public Opinion Research Methods

Abstract

Over the last decade, a popular experimental method—the conjoint design—has been used widely by political scientists to understand how immigrant characteristics shape individual attitudes toward them. We collect replication files for all studies consisting of conjoint experiments where respondents are asked to evaluate hypothetical immigrant profiles with attributes varying along economic (e.g., educational attainment and occupation) and cultural (e.g., nationality and religion) dimensions. We meta-analyze their estimates to assess the relative effect of each set of attributes on evaluations of immigrants. Our results confirm prior findings suggesting that preferences are driven by both sociotropic economic considerations and cultural or norms-based concerns. However, our re-analysis of replication data files demonstrates the effects of attributes conveying sociotropic benefits are weaker for immigrants perceived as culturally distant. Furthermore, among individuals predisposed against immigration, cultural threats outweigh economic reasoning, with the opposite being true for individuals predisposed in favor of immigration.

Authors

• Marco M. Aviña, Harvard University

• Taeku Lee, Harvard University

• Marcel Roman, University of Texas, Austin

• Priyanka Sethy, Harvard University

• Mashail Aman Malik, Harvard University

• Reed Rasband, Harvard University