After yesterday’s release of a KPIX-TV poll showing 11th District GOP nominee David Harmer with a 6-point lead, Democratic Rep. Jerry McNerney followed today with a statement about an internal poll that gives him a 10-point advantage.
What is a poor voter supposed to believe?
So, here’s the deal.
McNerney’s poll was done on Sept. 21-25, nearly a month ago. That makes a difference. His campaign did not tell us what question was asked, in what order it was ask or provide crosstabs that would break down the age, party or gender of the 500 people surveyed.
That’s typical. Campaigns commission internal polling all the time and they rarely release the details; it’s like asking a football team to publish their playbook. But it also means that we cannot judge the veracity of such polls with any great confidence.
In contrast, all the information about the KPIX-TV poll conducted by SurveyUSA is online. More significant, the television station doesn’t work for either of the candidates. (My personal thanks to the station for spending the money on an independent poll!)
For review purposes, the pollster surveyed 624 likely and actual voters on Oct. 8-11 and found that, if the election were held that day, Harmer would beat McNerney 48 percent to 42 percent. (Click here to read the initial post.)
Which poll is more accurate? Polls are a snapshot in time so the more recent survey is probably a better reflection of current opinions.
But there are some caveats to the SurveyUSA poll. The firm uses an automated telephone polling system. The individual who picks up the phone enters his or her answer via the telephone keypad. No live pollster asks the questions nor can the automated system confirm that an actual voter or likely voter has answered the telephone.
That said, KPIX commissioned similar polls in the 2009 10th Congressional District special election and the results were remarkably close to the final outcome.
The crosstabs show that among the 624 participants, 41 percent identified themselves as Republicans; 39 percent as Democrats and 18 percent as nonpartisan. The most telling result may be among the independents, which prefer Harmer over McNerney 50 percent to 32 percent.
Actual party registration in the district is much closer between the two major parties. But keep in mind that turnout won’t necessarily match registration. Candidates win with a plurality of the votes cast by the people who actually show up.
Read on for McNerney’s full news release. Read the rest of this entry »