Reconciling Individual and Aggregate Evidence Concerning Partisan Stability: Applying Time‐Series Models to Panel Survey Data

摘要: 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 have emerged from time series panel surveys. Using Monte Carlo simulation, we show only certain types data generation processes these regularities. Deciding which remaining best explains means investigating ways in individual partisanship behaves over time. Partisanship at tends be highly autocorrelated, reequilibrating slowly wake each perturbation. Working downward analysis data, previous researchers argued parti sanship is fractionally integrated contended dynamics are therefore heterogeneous. three surveys, present first direct assessment individual-level dynamics. We also investigate hypothesis vary among individuals, claim motivates much recent work on fraction ally series. The observed characteristics party one individuals respond similar external shocks, reequilibrate rapidly thereafter, seldom change their equilibrium partisan attachment.