We’re all pretty sick of the relentless drumbeat of the MSM over the mid-term meme that the Dems are going to take a shellacking in November, right? How much of the doom and gloom is driven by poll numbers? Oh, I’d say about 90 percent, with a margin of error of plus or minus four points.
But what if the polls were wrong? What if, because of the way telephone polls are conducted, they were inherently biased toward more conservative results?
According to Pew Research, there’s a pile of stinking bias behind almost all of the telephone polls, and it’s even higher and stinkier behind the automated polls. The reason is simple — people who only have cell phones tend to be more liberal while people who have land lines tend to be more conservative. Pollsters conduct polls using land lines, and automated polls are strictly land lines; very few polls reach a significant percentage of cell phone users.
Pew conducts its polls using both land lines and cell phones, and has identified bias in other polls it says could be worse than in 2008, when it was first detected:
In three of four election polls conducted since the spring of this year, estimates from the landline samples alone produced slightly more support for Republican candidates and less support for Democratic candidates, resulting in differences of four to six points in the margin. One poll showed no difference between the landline and combined samples.
In the Pew Research Center’s latest poll, conducted Aug. 25 to Sept. 6 among 2,816 registered voters, including 786 reached by cell phone, 44 percent said that if the election were held today that they would vote for the Republican candidate for Congress in their district or leaned Republican, while 47 percent would vote for the Democratic candidate or leaned Democratic. Among the landline respondents, 46 percent preferred the GOP candidate and 45 percent the Democratic candidate, a four-point shift in the margin. In this survey, both estimates would have shown a close race between Republicans and Democrats.
Limiting the analysis in the survey to those considered most likely to vote in this year’s elections, a similar bias is evident. The combined landline and cell estimate produced a seven-point Republican advantage: 50 percent supported the GOP candidate for Congress in their district while 43 percent backed the Democratic candidate. The Republican lead would have been 12 points if only the landline sample had been interviewed, a significant difference from the combined sample of five points in the margin.
Pew plans to release a comprehensive study of the affect of phones on polls after the Nov. 2 election, which could turn out somewhat differently than the pollsters predict. We’ll definitely cover it here, whatever the results.
“Pew plans to release a comprehensive study of the affect of phones on polls ..”
You need to learn V.A.N.E, “Verb effect, noun affect.” (Once you learn this rule, then you may be ready for the rare exceptions).
Sorry, it’s late… “verb affect, noun effect,” of course. Perhaps you were typing late also!