A peek under the hood of the NYT/CBS poll

Allahpundit did a great job last night of dissecting the latest NYT/CBS poll based on its results from CBS’ reporting.  The survey found that a large majority of their respondents support keeping current collective bargaining rights in the public sector (60/33), and a slightly smaller majority opposed cutting pay or benefits of public employees to close deficits (56/37).  However, the New York Times published the actual sample, and it calls into question at least the magnitude of the results.

First, the partisan split in the sample gave a ten-point advantage to Democrats.  Their sample for this poll had a D/R/I split of 36/26/31, an absurd sample for political polling.  In December, Rasmussen’s general-population survey put Republicans ahead, 36.0% to 34.7% for Democrats.  A recent poll by Gallup shows erosion in Democratic affiliation all through 2010.  In 2008, Barack Obama won the popular vote by seven points nationwide, and the NYT/CBS poll assumes that the electorate has grown more Democratic in 2011.

Next, 20% of the poll’s respondents claim to come from union households.  However, only 11.9% of American workers belong to a union, according to a report published by the Bureau of Labor Statistics last month and noted by none other than the Times itself.  Union membership fell to a 70-year low as a percentage of the workforce, which in itself is a rather damning statement about the view of collective bargaining by the vast majority of American workers.  How exactly did the survey manage to comprise itself of almost twice as many union-household respondents for a poll on union policies as union members in the workforce?  Interesting.

Finally, 25% of respondents are either public employees or share a household with a public employee.  Federal employees comprise less than 2% of the workforce at around 2 million.  Overall, the US has 22.22 million government employees out of an employed workforce of 130.27 million, according to the Current Employment Statistics survey at the BLS.  Government employment accounts for 17% of all workers, so a sample consisting of 25% public-sector households for a survey of adults (not registered voters) seems a little off.

The gaps in the results are wide enough that these issues by themselves might not have swung them to the opposite.  However, at least the magnitude of the results can certainly be questioned in a poll this flawed.

Update: William Jacobson finds similar problems in a widely-quoted PPP poll claiming that Scott Walker would lose to Tom Barrett in a re-do.