Abstract
A two-state Markov model of individual voting transitions is developed which allows voters to differ with respect to their voting propensities and loyalties. Such a model allows the estimation of various process parameters without the specification errors common in other models. Additional parameters are introduced to allow for the effects of a third party, considerably extending the scope of application. The model is applied to the electoral voting histories for 1964, 1966, and 1970 of 671 individuals from sixty-seven English constituencies.
Get full access to this article
View all access options for this article.
