Mark Mellman: Of leaners and swingers

My column last week on independents who lean toward a party ignited a (tiny) controversy.

Some of those who questioned my argument asked questions like “Are independents independent?” and “Are independents a myth?”

{mosads}Conducting a nuanced conversation in 140 characters, or even in 600 words, can be difficult, but in my view the more appropriate question is “How independent are independents?” My answer is that most are not as independent as their initial claim suggests, though some clearly are.

I noted that those independents who, upon further questioning, express a closer identification with a party vote like partisans.

Between 1992 and 2012, on average, 79 percent of independents who leaned toward Democrats voted for that party’s presidential candidates, while more than 80 percent of Republican leaners did likewise, rivaling the support offered by those who initially claimed to be partisan.

Two scholars who originally alerted us to the partisanship of leaners — Professors David Magleby and Candice Nelson — recently found that the issue positions of these leaners also closely track those of partisans. For example, in nearly two dozen polls on ObamaCare between 2010 and 2012, on average only 12 percent of Republicans expressed favorable views of the reform law, as did 14 percent of Republican leaners. By the same token, 67 percent of Democrats and 62 percent of Democratic leaners had favorable impressions. Thus, leaners were barely distinguishable from partisans and dramatically different from each other.

On issues from the stimulus bill to abortion and gay marriage, the story is similar: independent leaners look very much like partisans and quite different from pure independents.

Another stream of evidence comes from social psychologists administering the implicit association test. Without rehearsing the complex experimental paradigm, the conclusion is that even at a subconscious level, independent leaners identify with their respective parties, while pure independents do not.

My friends at Third Way, and some scholars, argue that independent leaners are not partisans, based mainly on a 2000-2004 panel study that indicates these leaners were more likely to change party than partisans. By 2004, about 28 percent of independent leaners in 2000 had switched sides in the partisan divide. Of course that means most didn’t switch, but leaners were more likely to do so than others.

Summarizing all the evidence then, while independent leaners might be more likely to switch parties than partisans, at any particular point in time, leaners act as if they are partisans.

That may be an accurate summary, but it is not the whole story. Part of the rest might arise from conflating concepts.

Though ascribing motives to others is fraught with danger, I cannot help but believe that part of the reason groups like Third Way feel so strongly about counting leaners as independents is out of fear that Democrats will abandon efforts to move persuadables and instead focus only on mobilizing the base.

Here they are right — that would be a grave error.

But independents and swing voters are not the same thing. Independents are defined by their answer to a question about partisanship. Swing voters are defined by being open to persuasion. Not all, not even most, swing voters are independents.

Professor William Mayer authored one technique for identifying swing voters. While all such methods are open to critique, his findings are instructive. Mayer establishes that, on average, between 1972 and 2004, swing voters made up 23 percent of the electorate. However, just 13 percent of those swing voters were pure independents, 28 percent were independent leaners and 60 percent were ostensible partisans. Other scholars and practitioners, employing different analytic strategies, arrive at the same conclusions: There are a significant number of swing voters and they are mostly not independents.

Arguing, as I have, that independent leaners are closet partisans is not to diminish the importance of a wholly different segment: persuadable or swing voters. In otherwise competitive contests, winning requires both the base and the swing.
 
Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982. Current clients include the majority leader of the Senate and the Democratic whip in the House.