A curious CBS / NY Times poll found that, by a nearly 2-1 margin, Americans prefer to have taxes raised over a decrease in benefits to public employees. That would certainly be news if it were true. If that seems an unusual result, consider the sample. The poll is not a poll of voters, but of adults. About 40% of those polled do not vote. The poll sample is 36% Democrats versus only 26% Republicans. Only 27% liked the Tea Party. We’ve seen this all before in such polling.

But here’s a new twist. 21% of the poll’s respondents said that their household had at least one union member. 25% of the households had at least one public employee. These results are curious. Only 9% of people working in the US are employed in the public sector, so a 25% public employee figure seems extremely outsized. Moreover, the 21% union sample seems large compared to the US, where fewer than 7% of private sector employees belong to a union, and total union participation is in the low teens.

It would appear that the poll substantially oversampled both public employees and union members. Is this correct? Should the poll sample have been adjusted to reduce those figures to national averages? And how did they get such a sample in the first place?

This entry was posted on Tuesday, March 1st, 2011 at 1:08 pm and is filed under business, Democrats, General, MSM, Polling, Republicans. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.