How many Republicans are enough?
That’s the question that’s emerged from the Los Angeles Times‘ poll of June 5-8, which showed Kerry leading Bush, 51 percent to 44 percent. However, 38 percent of those polled identified themselves as Democrats, with just 25 percent stating they were Republicans. Today, the paper’s James Rainey gives us a balanced and detailed look at a story that has largely gone under the radar this campaign season: if, and how, the party affiliations of respondents can affect the results of a poll.
The debate first erupted on ABC’s The Note, a morning summary of political stories, in an exchange of statements between Bush campaign strategist Matthew Dowd and Los Angeles Times polling director Susan Pinkus. Dowd charged that the poll was a “mess” because of the imbalance of Democrats and Republicans. Pinkus responded that her organization used “sound statistical methods” to conduct the poll.
This back-and-forth then appeared in the Washington Times and the New York Times on June 11, as mostly a play of “he-said, she said.”
Today the Los Angeles Times‘ James Rainey gives us a breakdown of the debate over the paper’s poll. As Rainey describes it, independent pollsters such as the Los Angeles Times believe party allegiances fluctuate and therefore do not “adjust results to achieve partisan balance.” On the other hand, political operatives believe party affiliation is more static, and thus generally feel polls should reflect the partisan affiliations of voters in past elections (typically 35 percent Republican and 38 percent or 39 percent Democratic).
Rainey, through the voice of ABC polling director Gary Langer, notes that independent pollsters and political operatives have separate interests. News directors are interested in where voters stand now, and political operatives want to know with how people will vote on Election Day — thus their concern that polls mirror the partisan split of previous elections.
Not surprisingly, Rainey lends more space to defending independent pollsters’ (and the Los Angeles Times‘) methods. Regardless, Rainey and the Times deserve credit for taking a critical look at the inner workings of such polls — since, as we’ve already noted, the media’s obsession with numbers can drive how it covers the election.
–Thomas Lang
Thomas Lang was a writer at CJR Daily.