CCES-2010-Partisan-Breakdown.jpg

“Independent” Voter Poll Analysis Is Mostly Useless

I can’t take it any more. This drives me absolutely up a wall (and I’m certainly not alone). 

From the IBD/TIPP poll write-up:

A key swing group making up nearly a third of voters, independents favor Romney over Obama 52%-34%.

Independent swing voters make up around a third of voters? No, they don’t. 

From Ed Morrissey at HotAir:

[Polling on “Independents”] suggests that some of the assumptions built into the pollster models are still leaning too far to 2008, and that Romney is actually in better shape than those toplines suggest.

The vote split among “Independents,” as broken out by most polls tells us a lot about the state of the race? No, it doesn’t.

Why? Because most of the “Independent” voters in polls lean toward one or the other party. 

More from Ed Morrissey at HotAir:

Some argue, though, that these independents are in some significant part Republicans who aren’t affiliating themselves with the GOP. That, however, doesn’t make much sense. 

But it does make a lot of sense. And this isn’t really an argument. It’s just a matter of looking at some facts, some data. These Leaners are, at the very least, as partisan as “Not-Strong” Republicans and Democrats. In fact, in 2010 Leaners of both parties were substantially more partisan than “Not-Strong” identifiers, and Dem Leaners were more partisan in 2008. 

Here are the numbers from the vote-validated CCES 2010 data, with vote for House candidates:

So, why don’t these partisan voters, who vote 85-95 percent for one party or the other, self-identify with one of the two major parties? It could be for a host of reasons . . . 

Leaners in both parties are more ideological than those who weakly identify with a party. Some of them vote 100 percent for one party, but consider it ideologically wanting, corrupt, compromised. They aren’t “Party” people. Think Tea Party folks, libertarians, hard-core Progressives and Greens. The Democratic Party to a Progressive? A bunch of corporate sellouts and compromisers. But most of them will pull the lever 9 times out of 10 for the Democratic candidate. Same goes for most Tea Partiers and libertarians.

So Mitt Romney is winning “Indies” going away? Great . . . better than not doing so. But in order to say anything at all about what it means for the race, we need to know the split in Leaners and the split among Pure Independents, who really do throw in about 50/50 for each party.

But analyzing the muddled lump of “Independent” voters that most polls break out for public consumption is likely to be a frustrating and pointless use of time.