The Limits of Ecological Inference: The Case of Split-Ticket Voting

Authors


  • We would like to thank Bear Braumoeller, Bruce Cain, Michael Caldwell, Lawrence Cho, Christophe Crombez, Susan Jellissen, Masaru Kohno, Jim Kuklinski, Walter Mebane, Peter Nardulli, Brian Sala, Jasjeet Sekhon, Paul Sniderman, three anonymous AJPS reviewers, and reviewers 1–10 at the APSR for helpful comments. Cho also thanks the National Science Foundation (Grant No. SBR–9806448) for research support.

Wendy K. Tam Cho is Associate Professor of Political Science and Statistics at the University of Illinois at Urbana-Champaign, 361 Lincoln Hall, 702 S. Wright Street, Urbana, IL 61801. Brian J. Gaines is Associate Professor of Political Science and an affiliate of the Institute of Government and Public Affairs at the University of Illinois at Urbana-Champaign, 361 Lincoln Hall, 702 S. Wright Street, Urbana, IL 61801.

Abstract

We examine the limits of ecological inference methods by focusing on the case of split-ticket voting. Burden and Kimball (1998) report that, by using the King estimation procedure for inferring individual-level behavior from aggregate data, they are the first to produce accurate estimates of split-ticket voting rates in congressional districts. However, a closer examination of their data reveals that a satisfactory analysis of this problem is more complex than may initially appear. We show that the estimation technique is highly suspect in general and especially unhelpful with their particular data.

Ancillary