Abstract
This note uncovers a serious flaw in Abramson et al.’s (1992) influential application of expected utility theory to the study of strategic voting in U.S. presidential primaries. Due to this flaw, it is not clear what theory their positive empirical results actually support. Rather than dismissing their findings as anomalous or simply as disconfirming expected utility theory, this note uses conditional expected utility theory, a basic revision of the standard approach to voting, to explain these and other results related to primaries.
Get full access to this article
View all access options for this article.
