As was noted in the acknowledgments, this volume originally arose as a way to kill two birds with one stone: to help infuse additional energy into the quantitative analysis of national election policy and to help provide guidance to the Pew Charitable Trusts, which in 2012 was considering whether to launch what eventually emerged as the Elections Performance Index (EPI). At the time the papers for this volume were commissioned, Pew was considering more than twenty indicators for inclusion in the index. Although the authors were given freedom to approach the topics of their chapters how they wished, and to report on any conclusions they reached in the process of doing their analyses, each author (or set of authors) was asked to provide some assessment of the indicators that were most relevant to the topic they addressed.

Readers of this volume will recognize the independence with which the authors pursued their tasks. Some provided explicit discussions of the reliability and validity of a set of possible election index indicators, while other authors were more implicit in their assessments. It is to the credit of the authors that their analyses led to the abandonment of some of the indicators that that been proposed for the EPI. (The clearest example was a proposed indicator concerning the confidence voters had that their votes had been counted as cast. Paul Gronke’s analysis in Chapter 10 provided a good argument that measures of voter confidence are important for understanding how voters think about the elections they participate in, but that survey responses to standard voter confidence questions are too influenced by partisan attitudes to be considered useful for assessing how well state and local officials run elections.) In other cases, the analysis required Pew and its advisory committee to rethink how indicators were measured, or how they were conceptualized.