Sorting and the Split-Ticket: Evidence from Presidential and Subpresidential Elections

Abstract

Although prior research demonstrates that strong partisans are less likely to cast a split-ticket, recent scholarly work hints that partisan-ideological sorting—the matching of an individual’s partisan and ideological identities—may play a comparatively stronger role in shaping this voting behavior. Simply, if a high degree of congruence between identities underscores psychological orientations that prevent association with an out-group, then highly-sorted voters should be less likely to cross-party lines within the voting booth. Using the 1972–2012 ANES Time-Series and 2010 CCES surveys, we provide evidence that demonstrates that a high degree of partisan-ideological sorting produces the strongest negative effect on split-ticket voting among a variety of alternative explanations at both the national and subnational levels. We then supplement these analyses with 1992–1996 ANES panel data to demonstrate how changes in sorting over time affect this voting behavior. Our results indicate that although an increase in partisan strength alone is insufficient to reduce an individual’s propensity to cast a split-ticket, an increase in identity sorting over time has a strong negative effect on split-ticket voting. We conclude with a brief discussion about the consequences of identity convergence; namely, that sorting fosters a unique form of electoral “polarization.”