Abstract
The past decade has witnessed profound changes in the tenor of American party politics. These changes, in tandem with growing affective polarization and residential segregation by party, raise the question of whether party identification is itself changing. Using three multi-wave panel surveys that stretch from the first Obama Administration through the Trump Administration, this paper takes a fresh look at the stability of party identification, using several different statistical approaches to differentiate true partisan change from response error. Perhaps surprisingly, the pace of partisan change observed between 2011 and 2020 is quite similar to the apparent rates of change in panel surveys dating back to the 1950s. Few respondents experience appreciable change in party identification in the short run, but the pace at which partisanship changes implies that substantial changes are relatively common over a voter's lifespan. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11109-022-09825-y.