New research provides evidence that party identification in the United States remains remarkably stable over short periods, even in the face of political turbulence, such as the transition from the Obama to the Trump administration. The findings, published in Political Behavior, indicate that the pace at which people shift their party loyalties has remained surprisingly consistent over the past several decades.
The motivation behind the study was influenced by the dramatic changes within the major political parties in the United States. On the Republican side, the advent of Donald Trump marked a pivotal shift. His presidency brought about profound changes in the party’s platform, characterized by a more pugnacious and nativist approach, which contrasted sharply with traditional political conventions.
Simultaneously, the Democratic Party experienced its own evolution, particularly with the rise of its progressive wing. The presidential bids of figures like Bernie Sanders in 2016 and the emergence of Alexandria Ocasio-Cortez in 2018 highlighted this shift. Consequently, there was a growing perception, especially among white citizens, that the Democratic Party was moving progressively further to the left, especially on economic issues.
One key question was whether the recent tumult in party politics, marked by these transformations, was leading voters to reassess their alignments with these parties. Was the rapidly evolving political landscape causing a hastening in the pace of change in party identities?
“The stability of party identification (relative to what scholars of public opinion have long termed ‘short-term forces’ such as recessions or scandals) has long been a cornerstone of theories of public opinion,” said study author Donald P. Green, the J.W. Burgess Professor of Political Science at Columbia University.
“The central argument of my co-authored book ‘Partisan Hearts and Minds‘ is that party identification tends to change very gradually over the course of one’s adult life, despite the vicissitudes of party politics. In that sense, party attachments are more likely social identities (e.g., religious attachments or sports team affiliations).”
For their study, the researchers analyzed data from three multi-wave panel surveys: the Institute for the Study of Citizens and Politics survey, the Democracy Fund Voter Study Group survey, and The American Panel Survey. Each survey employed nationally representative sampling procedures to recruit thousands of U.S. participants. Together, these surveys encompassed a period from November 2011 to October 2020, providing a rich longitudinal perspective.
At the heart of the study was the measure of party identification, which was done using a seven-point scale. This scale, adapted from the American National Election Study (ANES), allowed the researchers to gauge the strength and nature of political affiliations among participants. It was not just a matter of identifying as a Democrat or Republican; the scale also captured nuances such as whether individuals considered themselves strong or not very strong affiliates of their chosen party.
The frequency with which party identification was measured varied across the surveys. The Institute for the Study of Citizens and Politics panel measured party identification on nine occasions, the Democracy Fund Voter Study Group six times, and The American Panel Survey twenty-four times.
One of the key findings was the remarkable stability of party identification in the short term. The data showed that, from one survey wave to the next, there was very little change in how individuals identified politically. The percentage of new variance in partisanship was found to be relatively small, ranging between 2% and 20%. This suggests that despite the dynamic nature of political events and the heated political climate, most Americans maintained consistent political affiliations.
Despite the substantial changes in the social, political, and technological landscape over the past several decades, the pace of change in partisan identification from 2011 to 2020 was found to be quite similar to that observed in panel surveys dating back to the 1950s.
“Much has been said in the popular press about the change that Donald Trump’s candidacy/presidency has wrought on the Republican Party and party politics more generally,” Green told PsyPost. “One might infer that nowadays Americans are thinking differently about the parties and their connection to them, which in turn would imply that party attachments are more unstable over time than they were in the pre-Trump era. Our paper says no: to our surprise, party attachments are about as stable as they have been in panel surveys dating back to the 1950s.”
Contrasting with the short-term stability, the study also highlighted a different picture over the longer term. When examining political identities over the span of several years, the data revealed more significant shifts. This longer-term fluidity suggests that while individuals may hold firm to their political beliefs and party affiliations in the short run, these affiliations are subject to change over a longer period. The study projected that these small, incremental changes in party identification could accumulate over a person’s lifetime, leading to more substantial shifts in their political allegiance.
“The fact that party identities tend to change gradually over time does not mean that they do not change at all,” Green explained. “Gradual changes can become meaningful over the span of decades. The point is not that party identities are unchangeable but rather that they rarely change abruptly in response to current conditions.”
The study, “Partisan Stability During Turbulent Times: Evidence from Three American Panel Surveys“, was authored by Donald P. Green and Paul Platzman.