Tuesday, October 24, 2006
By Steven L. Taylor

The NYT has a piece in today’s paper that seems to be a requisite discussion every election season (along with, for example, the “October Surprise” pieces that have been circulating that last week or so): In Southwest, a Shifting Away From Party Ties. The piece mainly focuses on anecdotal evidence from Arizona and Colorado about the alleged rising tide of independent voters.

I am always skeptical of such “analyses” for several reasons.

First, what does it really mean to be an “independent” voter? The way it is treated in these types of articles tend to make it sound as if these are voters who could, at the drop of a hat, vote for either party and shift between the two with great fluidity. I find this to be unlikely. Indeed, the numbers tend to indicate that while there are a number of self-identified independents out there in the electorate, many of them heavily lean towards the Ds or the Rs. In other words: if a given independent voter votes mainly with one party or the other, are they truly independent? Is the proper operationalization of the concept “independent” simply someone who thinks of themselves are independent, or should it be understood as a voter who has a significant variation in their voting patterns in respect to party? Clearly the right answer in the latter, although journalists tend to prefer the former.

In short: the issue should ultimately be how people vote regularly, not how they think about themselves (or how they register).

Second, when it comes right down to it, most people don’t want to be thought of as ideologues or partisan zombies, so they like to say things like “I vote for the person, not the party” or somesuch. I think this plays into the numbers in the graphic associated with the story which states that 58% of the population would prefer a system with no party labels v. 35% which prefer the current system. Ot strikes me that at the moment I would expect a great deal of negativity to be directed at the current parties. Also, I wonder what the results would have been if one of the questions had asked about a multiparty system?

Back to the “voting for the person” issue: the real question is when we look at which type of person that the “independent” votes for is there a pattern? Does it just so happen that the “independent” votes mostly for Democratic persons or Republican persons?

Third, given our two-party system, if one wishes to vote for a candidate who actually has a chance to win office, one has to channel one’s political support to the Ds or the Rs. Granted not all do–some consciously vote third party because of their independent-mindedness. However, the numbers clearly indicate that this is a small percentage of the population.

Further, as the article notes, if one is counting on voter registration as a measure (which appears to be what the article it doing), it isn’t a very good one:

Some election experts are skeptical about voter statistics and say party registration as counted by most states is a poor measure of the electorate. Massachusetts, for example, has rules that encourage strategic registration, since independents can vote in any party’s primary. Most national surveys have shown little change in the numbers of self-described independents since the late 1970’s, but the polls tend to ask about ideological preference rather than actual party registration.

Not to mention that the majority of states do not require partisan identification on voter registration.

And the surveys cited above are the real key: ideological preferences rather than partisan ones.

More from the article:

In Colorado’s closely fought Seventh Congressional District outside Denver, lawn signs for the candidates — Ed Perlmutter, a Democrat, and Rick O’Donnell, a Republican — mostly carry no party identification.

That is just analysis via anecdote. Campaigns in the US are candidate-centric, not party-centric. As such, it is hardly unusual, or indicative of a trend, to see a bunch of signs with only the candidates’ names on them. To know if a given set of signs means anything one would have to know what the norm was in terms of party identifiers on signs and then figure out whether one is seeing significant deviation from that norm to say if anything of significance was taking place. That sentence is just the author of the piece trying to confirm the hypothesis of his piece.

Further, we get back to the question of definition:

About one in three voters in Colorado are independents — in the Seventh it is even higher, 37 percent.

How does the author know this? Voter registration? Self-identification? I find it rather difficult to accept that 37% of Colorado voters could, in any given election, vote for either party–which is the impression given in the piece.

Indeed, after copious research (I used “The Google”) I found the registration form for Colorado and I see that, in fact, one registers one’s party affiliation on the form (more copious researach reveals that one indicates party affiliation in Arizona as well). Hence, the author of the piece is using voter registration stats to represent voting behavior. However, the two are not the same thing–not by a longshot.

Update: Andrew Sullivan notes the piece as well (posting the graphic referenced above). He hits on the emotional aspect of the piece: a lot of people are thinking themselves as independent because of their discontent with the GOP, although I am not sure that that is what the graph really shows. The right-hand side tracks registrations in AZ, but that really doesn’t tell us much (as I note above). The piechart is as much a reflection of the desire for more or different kinds of choices than anything else.

Filed under: Uncategorized | Comments/Trackbacks (2)|
The views expressed in the comments are the sole responsibility of the person leaving those comments. They do not reflect the opinion of the author of PoliBlog, nor have they been vetted by the author.

2 Responses to “Speaking of Independents…”

  1. LaurenceB Says:

    Case In Point:

    I always call myself an “Independent”, but I definitely plan on voting a straight Demo ticket this time.

    So I guess I’m “Independent” only in the same sense that Bill O’Reilly claims to be.

  2. Scott Says:

    > the issue should ultimately be how people vote regularly, not how they think about themselves

blog advertising is good for you

Visitors Since 2/15/03

Wikio - Top of the Blogs - Politics



Powered by WordPress