A Generational Model of Political Learning

I propose a mathematical framework for modeling opinion change using large-scale longitudinal data sets. The framework encompasses two varieties of Bayesian learning theory as well as Mannheim's theory of generational responses to political events. The basic assumptions underlying the model are (1) that historical periods are characterized by shocks to existing political opinions, and (2) that individuals of different ages may attach different weights to those political shocks. Political generations emerge endogenously from these basic assumptions: the political views of identifiable birth cohorts differ from each other, and evolve distinctively through time, due to the interaction of age-specific weights with period-specific shocks. I employ this model to examine generational changes in party identification using data from the 1952-1996 American National Election Studies. My estimates of the age-specific weights characterizing various points in the life-span are generally consistent (at least between the ages of 15 and 60) with a simple Bayesian model in which an individual's opinion at any given time is a simple average or "running tally" of past political experiences. I find no support for the hypothesis that more recent events receive disproportional weight, and only slight (and statistically uncertain) support for the notion that events experienced during a crucial period encompassing late adolescence and early adulthood have more powerful effects than those experienced later in life.