According to a recent Gallup Report the proportion of Americans classifying themselves as political independents averaged 42 percent in 2013, a 25 year high. Beyond disputing Gallup’s numbers, which are somewhat higher than those reported by other polls, several political scientists (eg, this Alan Abramowitz piece in Politico or this John Sides piece in the Washington Post's Monkey Cage) predictably downplayed the significance of independents, repeating the common claim that since the majority of independents admit that they “lean” toward one or the other of the parties, they actually are closet partisans, not independents. While Gallup's Managing Editor Jeffrey Jones exaggerates the significance of Gallup’s numbers, the counter-claim that leaning independents are really just partisans who like the independent label rests more on repeated assertion than on data.
The primary evidence cited to support the claim that leaning independents are closet partisans is that they vote like partisans. But this is a “chicken-and-egg” problem that has never been adequately addressed: does a Democratic leaning independent vote Democratic because she is a closet Democrat, or is she an independent who leans Democratic because she has decided to vote Democratic?
As shown in the accompanying table political scientists use a battery of survey questions to divide voters into seven categories of party identification. But these categories are not set in stone like racial, let along gender identification--people move around. Panel studies (wherein the same voters are interviewed in successive elections) show that from two-thirds to three-quarters of strong partisans put themselves in the same category from one election to the next, and 90 percent or more stay on the same side of the partisan divide. Less than 50 percent of weak partisans stay in the same category, but again, 85 percent at least stay on the same side of the partisan divide. Only 30 to 40 percent of independent leaners put themselves in the same category from election to election, however, and one- quarter to one-third move to pure independence or to the other side of the partisan spectrum. Independent leaners are much less consistent than partisans. Part of the reason they appear to vote like partisans is that they decide how they lean by considering how they are going to vote.
Other evidence also supports the conclusion that there are important differences between leaning independents and partisans. Leaning independents are more likely to support third party candidacies like George Wallace (1968), John Anderson (1980) and Ross Perot (1992-96). Leaning independents are much less likely than partisans to believe “that either party represents your views reasonably well.” And on various questions about specific issues, they differ consistently--although not greatly--from weak partisans.
All in all, the political science claim that leaning independents are closet partisans rests on a thin layer of evidence. But it is important to recognize there is equally little basis for the occasional claim that leaning independents are a mass of principled centrists. Yes, some independents are centrists, just as some are are closet partisans. But some are alienated from both parties and some just don’t care about politics. Still others are cross-pressured, agreeing with some Republican or Democratic positions and disagreeing with others. This is a subject that requires serious study rather than casual generalization. For however heterogeneous they are, independents are clearly important. Today’s electoral arena serves only hard partisans. Independents are a source of volatility in the electorate, and, as such, an agent of needed change.
About the Author: Morris P. Fiorina is the Wendt Family Professor of Political Science at Stanford University and a Senior Fellow of the Hoover Institution.