Approval voting is a system in which voters can vote for as many candidates as they like, or approve of, in a multicandidate election. Each candidate approved of by a voter receives one vote, and the candidate with the most votes wins. Approval voting has a number of advantages over plurality voting, and plurality voting with a runoff, as well as ranking systems such as the Borda count and the Hare system of single transferable vote. Approval voting has been used in universities, professional societies (e.g., the Econometric Society), internal elections of political parties in some states, and the United Nations Security Council to elect a Secretary General. Approval voting ballot data from the Institute of Management Science (TIMS), with about 6,000 members, were recently analyzed and provided important tests of several theoretical propositions. Data from the 1987 elections of the Mathematical Association of America (MAA), with about 26,000 members, and the 1988 elections of the American Statistical Association (ASA), with about 15,000 members, and the Institute of Electrical and Electronics Engineers (IEEE), with about 280,000 members, will be available for analysis and provide additional bases for analyzing patterns of support of candidates under approval voting. The analysis of these data should shed considerable light on how approval voting might work in public elections. For example, is it likely to elect so.called Condorcet candidates .. those who, if they exist, can defeat every other candidate in pairwise contests . . as it has consistently done in private elections. The use of approval voting in public elections has not gone without controversy and has been debated in numerous scholarly journals. Nevertheless, bills to enact approval voting have been introduced in several states; in 1987, an approval voting bill was passed by the North Dakota Senate by a vote of 34.16 but failed in the House. In this project the empirical analysis of the MAA, ASA, and IEEE approval voting ballot data will be vital in assessing the use and practicality of approval voting in public elections, as will data collected in the 1988 National Election Survey that will include an approval voting item. Further, an extension of approval voting to party.list systems using proportional representation (PR) in Parliamentary democracies will be analyzed. In particular, models to assess whether approval voting might encourage parties to split up in a PR system wil be investigated in order to evaluate the feasibility of marrying approval voting and PR using the combination in parliamentary elections. This research will contribute not only to the social choice literature but also indicate if there are any serious limitations to using approval voting to elect legislators in parliamentary democracies.
