Abstract:
Most models of representation posit at least some relationship between
roll call votes and constituent opinions. But do constituents really
know how their elected representatives vote? This paper compares
public awareness of Senate roll calls on two highly controversial
issues, the 1991 Gulf War resolution and the October, 1991 vote to
confirm Clarence Thomas, using the 1992 Senate Election Study. The
authors show that the proportion of the sample that claim to remember
how their Senators voted is remarkably stable across these two cases,
as is the estimated impact of a series of individual and contextual
variables on probability of remembering. Variations emerge when the
authors compare the determinants of accuracy, in models estimated for
those who claimed to remember the vote and those who were willing to
guess. The authors close with some speculations about when and how
citizens know (or can accurately guess) roll call positions.
Paper presented at the 1996 Annual Meetings of
the American Political Science Association. The data used in this
paper was originally collected by the National Election Studies, and
was distributed by the Inter-University Consortium for Social and
Political Research. We would like to thank J. Matthew Wilson for
research assistance. Gronke would like to thank the Arts and Sciences
Council at Duke University for partial funding of this project. A
copy of this paper can be retrieved from the Political Methodology home
page: http://wizard.ucr.edu/polmeth/polmeth.html.
Paul Gronke
Sun Nov 24 22:06:23 EST 1996