As a spate of surveys show undecided voters breaking for Obama, Republicans are at it again, "unskewing" polls and complaining about poll samples that allegedly show Democrats outnumbering Republicans. This is, of course, when they are not attacking Nate Silver for spreading facts, figures, and statistics (the horror!).
All of this shows that right-wingers do not understand what party identification is all about. Republicans like Dan McLaughlin at RedState and Jay Cost at National Review are rallying the troops, arguing. that Romney will win because he is allegedly winning "independent" voters. It is unfortunate that these veteran analysts are giving unsophisticated readers false hope by allowing them to believe that independent voters and undecided voters are the same thing. They're not.
Party ID is self-reported (in credible polls)
Poll respondents SELF-identify and "independents" are not really independent: most lean strongly towards one of the parties. This self-identification is fluid; that is, the party whose partisans are less enthusiastic at any given time more likely to identify as "independent" rather than as a party-line voter.
Party ID is a measure of enthusiasm -- or the lack thereof
When a woman who typically votes Democrat is less enthusiastic -- when she is embarrassed about her party's policies, platform, political candidate, or prospects -- she is more likely to identify as independent. When a dude who almost always votes Republican is less enthusiastic, he is more likely to identify as independent.
Thus, a poll with a D+9 or R+5 sample does not necessarily mean the other party is underrepresented in the poll: it more likely means that for whatever reason one party's poll respondents were more likely to identify as "independent."
For example, If a poll's same is D-37, R-31, and I-32, it does NOT show definitively that the poll is predicting more Democratic voters than Republican voters. If most of the poll's "independent" respondents are actually Republicans for all practical purposes, then there might actually MORE Republicans in this D+6 sample.
Independent indeschmendent
The independence of "independent" voters is overstated and overrated, because most of them are independent like Bill O'Reilly and Bill Maher claim to be independent. Which is to say they are partisans in independent clothing. They are not undecided.
"Unskewing" polls and complaining about party identification in poll samples is a fool's errand. Conservatives who have pinned their hopes on the erroneous belief that pollsters are 'oversampling' Democrats are setting themselves up for a letdown.
Comments are closed on this story.