Article Summary
Introduction
Research on polarization typically focuses on the people who are becoming polarized—extreme Democrats and Republicans. Yet as U.S. politics becomes increasingly polarized, more and more people choose to identify themselves as “independents” rather than as Democrats or Republicans. They do so in large part because they are so uncomfortable with intense partisan polarization. In many ways, however, these independents are really undercover partisans. Although about 40% of the U.S. public describes themself as independents, most of those people admit to leaning toward either the Democratic or Republican party. But they do so only after initially attempting to disguise their partisanship.
This article investigates these ‘partisan leaners’ or ‘undercover partisans,’ asking why people might attempt to distance themselves from the party they (eventually) admit to affiliating with. In particular, these authors aim to identify how “independent leaners” (people who genuinely identify as independents but somewhat prefer one party) differ from “disgruntled leaners” (people who view themselves primarily as partisans but use the independent label to express displeasure or separation from their party).
Understanding the differences between these groups helps us to understand the vote choices of political independents. Disgruntled leaners’ dissatisfaction with their party may make them less likely to vote for candidates from their party. Independent leaners, however, are likely to have relatively stable electoral preferences because they are not expressing dissatisfaction with a party; rather, they simply identify as independent because it is important to them personally. In other words, disgruntled leaners may vote less consistently like partisans.
Analytical Approach
To explore differences between political independents, the authors conducted a survey in June 2019 in which they asked 1,685 people in the U.S. different questions about their political affiliation. First, they asked a relatively standard question to reveal participants’ partisanship: “Generally speaking, are you a Democrat, a Republican, or an Independent?” (Response options: Strong Democrat, Weak Democrat, Independent leaning to Democrat, Independent, Independent Leaning to Republican, Weak Republican, Strong Republican).
Then, later in the survey, they ask “Thinking again about your political identity, which best describes you?” (Response options: Strongest Democrat, Democrat, Weakest Democrat, Independent, Weakest Republican, Republican, Strongest Republican). By replacing the “Independent leaning” options from the first question with “Weakest partisan” options in the second question, the authors gain insight into why people identify as independent leaners. Both disgruntled leaners and independent leaners select the “leaning” options on the first question. On the second question, however, disgruntled leaners will select “weakest partisan” to continue signaling displeasure with their party, whereas independent leaners will select “independent” to signal the value they place on being an independent. In this sense, the study is essentially a within-subjects experiment, where all participants are given two different questions to explore how their responses change with different measurement strategies.
To explore the practical implications of the motivations for identifying as a leaner, the authors also examine how ideological strength (that is, how liberal or conservative participants are) differs between disgruntled and independent leaners. They do so by comparing average ideological strength between the two groups of leaners, controlling for education, race, gender, and the importance participants place on their partisan identity. Next, the authors examine how the two groups’ vote choices differed. They do so by asking participants to think about the 2020 election and indicate whether they would vote for Trump, the Democratic nominee, someone else, or not vote at all. They then compare participants’ likelihood of saying they would vote for the nominee from their own party to the likelihood of selecting any other option, controlling this time for ideology in addition to the other variables (i.e., education, race, gender, and partisan identity importance). Finally, the authors examine whether there are any differences across the two parties in how likely people are to identify as disgruntled or independent leaners, controlling for education, race, gender, and partisan identity importance.
Main Findings
The “leaners” group is not a unified block of people who feel only weekly attached to the parties. Of those who identified as leaners on the first partisanship question, about 40% chose “independent” in the second question (and are thus categorized as independent leaners), 35% chose “weakest partisan” (and are categorized as disgruntled leaners), and the remainder chose a different category (i.e., a stronger partisan option). Thus, the “leaners” group masks differences between people who inherently value the independent label and people who are simply dissatisfied with their party.
These differences are substantively important. Disgruntled leaners have lower ideological strength than independent leaners. That is, people who identify as leaners due to dissatisfaction with their own party are more moderate than leaners with more sincere independent identity. In fact, independent leaners’ ideological strength is also greater than weak partisans—those who selected “weak Democrat” or “weak Republican” on the first partisanship question, rather than a leaner option. Similarly, disgruntled leaners are more likely than independent leaners to indicate that they would defect from their party and vote for a different candidate.
There also appear to be differences between Democratic and Republican leaners. Democrats are more likely to identify as independent leaners; Republicans are more likely to identify as disgruntled leaners. The authors argue that these differences are driven by Republicans’ distaste for an increasingly extreme Republican party and their preference for more moderate candidates.
Implications
This study shows that, although leaners express less affiliation with parties than other kinds of partisans, they are not necessarily moderate. Independent leaners—those who identify as leaners because they prefer one party but are primarily independents—have strong, ideological, ‘partisan-like’ opinions. They are more liberal or conservative than both weak partisans and disgruntled leaners—who identify as leaners because they are dissatisfied with their preferred party. Independent leaners vote in basically partisan ways (i.e., consistently for their preferred party candidate), unlike disgruntled leaners, who are much more likely to defect and vote for other parties.
This study also shows that differentiating between these two groups of leaners is relatively simple empirically. Doing so may be important for scholars interested in dynamics related to partisan identity and ideological extremity. Why and when do people choose to distance themselves from parties? How does this affect the ways that they learn about and process political information? This study suggests, for instance, that independent leaners may be less susceptible to partisan motivated reasoning (a process in which people more easily accept information that supports their partisan identity) than other kinds of leaners, despite their relatively strong political opinions.
Questions left unanswered
The authors note that their findings regarding partisan differences in leaners’ vote choices are complicated by the electoral context in 2019, in which the Republican party had a clear nominee in Trump but the Democratic party had no clear frontrunner yet. Thus, participants who leaned to the Democratic party may have imagined different kinds of nominees, whereas participants who leaned Republican indicated their support for Trump specifically. Given this complexity, it would be worth confirming that the results apply in other electoral contexts and that the partisan differences in leaners’ motivations and vote choices are not just driven by Trump.
Methods and Analysis
Was the study and its analyses pre-registered?: No
Did the study rely on proxy variables to measure polarization?: No
Were standard p-value thresholds used (p<.05 or 95% Confidence Intervals that don’t overlap zero)?: Yes
- Largest p-value presented as significant: 0.05
Were correlational results interpreted with causal language?: No
Limitations / Weaknesses
The authors are centrally interested in people who identify as independents leaning toward one party. The traditional measure of partisanship in the U.S.—the one that gave rise to the “leaner” label—asks a two-step question, first asking whether people generally think of themselves as Democrat, Republican, independent, or something else, then following up with a second question to ask whether independents lean toward one party. The authors of this piece do not use this standard two-part measure, instead providing the option to identify as an independent leaner on the first question. That is not inherently flawed, but it means that conclusions about leaners as measured in this study do not necessarily generalize to conclusions about leaners as more typically measured.
In all of their regression analyses (i.e., the analyses that explore relationships between leaner type and ideological strength, vote choice, and partisanship), the authors control for the importance of partisan identity. They do so to account for differences in political engagement between people who value their partisan identity and those who do not. However, the authors’ argument about independent and disgruntled leaners implies that one of the central conceptual differences between the two groups is that they differ in partisan identity importance (i.e., independent leaners should attach little importance to their partisan identity because they are primarily independents, and disgruntled leaners should attach some importance to their identity because are primarily partisans despite their dissatisfaction). If partisan identity importance partially causes both leaner type and the outcome of interest (e.g., vote choice), including this control is appropriate. However, if leaner type causes reduced partisan identity importance (for instance, if disgruntled leaners partially discount the importance of their partisan identity), controlling for partisan identity importance will bias the estimated associations in the models. It would be useful for the authors to more thoroughly discuss their logic in including this control and/or to present alternative models excluding it to ensure that results are robust to model specification.
Open Data & Analyses
Does the article make the replication data publicly available?: No
Does the article make the replication analysis scripts publicly available?: No
Article Citation
Klar, S., Krupnikov, Y., & Ryan, J. B. (2022). Who are leaners? How true independents differ from the weakest partisans and why it matters. The Forum, 20(1), 155–167. https://doi.org/10.1515/for-2022-2048
Bibtex
1
2
3
4
5
6
7
8
9
10
11
12
@article{KlarKrupnikovRyan+2022+147+159,
url = {https://doi.org/10.1515/for-2022-2048},
title = {Who Are Leaners? How True Independents Differ from the Weakest Partisans and Why It Matters},
author = {Samara Klar and Yanna Krupnikov and John Barry Ryan},
pages = {147--159},
volume = {20},
number = {1},
journal = {The Forum},
doi = {doi:10.1515/for-2022-2048},
year = {2022}
}