Over the last decades there has been a trend towards the adoption of more ''personalized'' forms of electoral systems (Pilet and Renwick 2011). The adoption of electoral systems that allow for a greater role of a ''personal vote'' calls for a better understanding of under what conditions partisanship begins to give way to a personal vote. It also raises the question whether the move to a more personalized electoral competition has significant effects on who gets elected and, if so, which types of candidates are favored. Electoral systems that rely on preferential voting offer certain advantages when it comes to study these questions, if only for the reason that they offer more detailed data about voter behavior. Existing studies of preferential voting have, however, failed to fully exploit the data. This paper proposes a method for analyzing data on preferential voting that makes a more efficient use of the data and provides the tools that allow us to examine the questions identified above using unique datasets from elections in Ireland and Iceland held under preferential vote systems.
