ANN ARBOR— While the results of last week’s special U.S. Senate election in Oregon have a direct impact on the residents of that state, the way in which voters cast their ballots in that race may have a greater effect on the nation as a whole, say University of Michigan researchers.

The special election, in which Democrat Ron Wyden defeated Republican Gordon Smith to fill Bob Packwood’s vacant Senate seat, was the nation’s first congressional election to be conducted entirely by mail.

A team of researchers, led by Michael Traugott, a U-M professor of communication studies and program director of the U-M Center for Political Studies, will spend the next several months analyzing the impact of Oregon’s vote-by-mail procedures on the attitudes, composition and turnout of the electorate and outcome of both the Jan. 30 general election and the Dec. 5 primary race.

” A detailed examination of these elections is important because vote-by-mail, especially as it is being employed in this context, is but a step along a path that many election administrators see as a logical, technological progression to electronic voting from home by computer, cable television or phone,” Traugott says. ” A great deal of the attention that is being devoted to the Oregon experience is associated with the expectation that it is a natural laboratory for the voting system of the future.”

Traugott says that while the vote-by-mail process is expected to reduce costs and speed up the accurate tabulation of votes, it raises important questions about potential problems of fraud and undue influence; the impact of administrative procedures on various groups in the electorate, turnout and standing division of the vote (i.e., the number of Democratic votes vs. Republican votes); and the consequences for candidates and the campaign strategies they employ.

Specifically, the study will address the following questions: How satisfied are citizens with the election procedures and how they worked? Is there a relationship between vote-by-mail procedures and voter turnout, either in the aggregate or on subgroups in the electorate? Is the electorate in the vote-by-mail elections qualitatively different than the electorate that votes in person? What was the impact of election administration on turnout and voter characteristics in the election? Is there any evidence of fraud or the exercise of undue influence on voters?

Preliminary results of the study will be presented in San Francisco at the August conference of the American Political Science Association, with final results to be released in October in Washington, D.C.

The project will be funded by the Pew Charitable Trusts, Ford Foundation, Ralph L. Smith Foundation and Northwest Area Foundation. Other research team members include Erik Austin, director of data archives at the U-M Inter-university Consortium for Political and Social Research; Steven Rosenstone, U-M professor of political science and principal investigator for the American National Election Studies; and Robert Mason, professor of statistics at Oregon State University.

Ford FoundationNational Election Studies