2014 Cooperative Election Study

Harvard University, Cambridge MA

Investigators

Abstract

The Cooperative Election Study is a data infrastructure project involving the collaboration of research teams from over 70 different academic institutions. Collectively, these research teams have fielded national surveys of 35,000 adults in 2006, 37,000 in 2008, and, with partial funding from NSF, 54,000 in 2010 and 55,000 in 2012. Each research team that wishes to be involved in the project purchases a 1,000 person sample survey from the same firm. Each individual team determines half of the questions on its survey. The other half of the content, Common Content, is created by a design committee, drawn from the participating teams. Common Content consists of questions that every team would like to measure or questions that are of broad interest and require a very large sample. The project, thus, fields as many as the there are teams and also produces a single large sample survey that consists of the Common Content. With 35 teams in 2006, there were 35 individual team surveys each with a sample of 1,000 and a single common survey of 35,000 cases. The survey, through the use of modules, encompasses a vast range of research topics that individual teams seek to answer. Researchers have used their team modules to conduct standard survey research on particular themes and questions for which appropriate public use data are not available. Others have used their modules to conduct experiments with political advertisements, pictures of candidates, information about the economy, and other treatments. Since its inception, the survey has produced studies that have appeared in over 100 journal articles and dozens of books. These include over a dozen articles that appeared in the top journals in the discipline as well as articles in law, sociology, statistics, and economics. Studies utilizing this data have focused on a wide variety of topics including elections, voting behavior, public opinion, political parties and polarization, the opinion on wars and other military interventions. The survey has helped to create and sustain a network of researchers interested in state and national elections, survey design, and public opinion. Participants have included graduate students and undergraduates who not normally have access to surveys of this size. These participants develop research skills especially in the area of questionnaire design, and generate further research ideas that can be pursued through this or other vehicles. Additionally, the scholars involved in this project will continue to use the data to develop educational material relating to the survey that can be adopted widely as part of Political Science undergraduate and graduate curricula.