Understanding preferences for income redistribution

Recent research suggests that income redistribution preferences vary across identity groups. We employ statistical learning methods that emphasize pattern recognition; classification and regression trees (CART™) and random forests (RandomForests™), to uncover what these groups are. Using data from the General Social Survey, we find that, out of a large set of identity markers, only race, gender, age, and socioeconomic class are important classifiers for income redistribution preferences. Further, the uncovered identity groupings are characterized by complex patterns of interaction amongst these salient classifiers. We explore the extent to which existing theories of income redistribution can explain our results, but conclude that current approaches do not fully explain the findings.