Comparing absentee and precinct voters: Voting on direct legislation

This paper addresses issues related to how absentee voters actually cast their ballots on propositions. If the liberalization of absentee laws changed either the composition or behavior of the electorate, then the outcome of the election may be affected. This paper tests whether the electoral behavior of absentee and precinct voters differs in regards to voting on propositions. The analysis is based on a sample of actual absentee and precinct voter ballots drawn from the approximately three million ballots cast in Los Angeles county for the 1992 general election. The analysis uses a nested model of voter participation and is estimated using the weighted exogenous sampling maximum likelihood method. We find that precinct and absentee voters do differ on both the propositions on which they cast votes and in their propensity to vote Yes for a proposition. For example, absentees appear to vote on fewer bonds and initiatives than do precinct voters. They also vote on fewer propositions dealing with state taxes, food taxes, and property taxes. In addition, given that a voter casts a valid vote, the propensity for absentee voters to vote Yes is higher on initiatives and propositions related to education, welfare, and health care than it is for precinct voters.