Party identification has been studied extensively using both individual- and aggregate-level data. This paper attempts to formulate a statistical model that can account for the range of empirical generalizations that have emerged from aggregate time series and panel surveys. Using Monte Carlo simulation, we show that only certain types of data generation processes can account for these empirical regularities. Deciding which of the remaining types best explains the data means investigating the ways in which individual-level partisanship behaves over time. Partisanship at the aggregate-level tends to be highly autocorrelated, reequilibrating slowly in the wake of each perturbation. Working downward from the analysis of aggregate data, previous researchers argued that aggregate partisanship is fractionally integrated and contended that dynamics at the individual level are therefore heterogeneous. Using data from three panel surveys, we present the first direct assessment of individual-level dynamics. We also investigate the hypothesis that these dynamics vary among individuals, a claim that motivates much recent work on fractionally integrated time series. The model that best explains the observed characteristics of party identification is one in which individuals respond in similar ways to external shocks, reequilibrate rapidly thereafter, and seldom change their equilibrium level of partisan attachment.

Panel data are a very valuable resource for finding empirical solutions to political science puzzles. Yet numerous published studies in political science that use panel data to estimate models with dynamics have failed to take into account important estimation issues, which calls into question the inferences we can make from these analyses. The failure to account explicitly for unobserved individual effects in dynamic panel data induces bias and inconsistency in cross-sectional estimators. The purpose of this paper is to review dynamic panel data estimators that eliminate these problems. I first show how the problems with cross-sectional estimators arise in dynamic models for panel data. I then show how to correct for these problems using generalized method of moments estimators. Finally, I demonstrate the usefulness of these methods with replications of analyses in the debate over the dynamics of party identification.

This paper develops an estimator for models of election returns in multiparty elections. It shares the same functional formas the Katz—King estimator but is computationally simpler, can be used with any number of parties, and is based on more conventional distributional assumptions. Small sample properties of the estimator are derived, which makes it particularly useful in many of the applications where there are a relatively small number of voting districts. The distributional assumptions are contained in two elements. The first treats the observed votes as the outcomes resulting from sampling the voters in each district. The second stochastic element arises from the usual treatment of the stochastic term in a regression model, namely, the inability of the included variables and the linear form to match the underlying process perfectly. The model is then used to analyze the 1993 Polish parliamentary elections. The results from this analysis are used to develop Monte Carlo experiments comparing several different yet feasible estimators. The conclusion is that a number of accessible estimators, including the standard seemingly unrelated regression model and the Beck-Katz model with panel-corrected standard errors, are all good choices.

Katz and King have previously proposed a statistical model for multiparty election data. They argue that ordinary least-squares (OLS) regression is inappropriate when the dependent variable measures the share of the vote going to each party, and they recommend a superior technique. Regrettably, the Katz-King model requires a high level of statistical expertise and is computationally demanding for more than three political parties. We offer a sophisticated yet convenient alternative that involves seemingly unrelated regression (SUR). SUR is nearly as easy to use as OLS yet performs as well as the Katz-King model in predicting the distribution of votes and the composition of parliament. Moreover, it scales easily to an arbitrarily large number of parties. The model has been incorporated into Clarify, a statistical suite that is available free on the Internet.

Katz and King have previously developed a model for predicting or explaining aggregate electoral results in multiparty democracies. Their model is, in principle, analogous to what least-squares regression provides American political researchers in that two-party system. Katz and King applied their model to three-party elections in England and revealed a variety of new features of incumbency advantage and sources of party support. Although the mathematics of their statistical model covers any number of political parties, it is computationally demanding, and hence slow and numerically imprecise, with more than three parties. In this paper we produce an approximate method that works in practice with many parties without making too many theoretical compromises. Our approach is to treat the problem as one of missing data. This allows us to use a modification of the fast EMis algorithm of King, Honaker, Joseph, and Scheve and to provide easy-to-use software, while retaining the attractive features of the Katz and King model, such as the t distribution and explicit models for uncontested seats.