Earlier this evening I was robo-polled by an organization identifying itself as "DNC Services" -- which I assume is affiliated with the Democratic Party. (This may have been the same poll that David Bernstein received recently.) A few observations:
1.) I absolutely love answering political polls! This probably makes me a non-optimal responder, from a pollster's point of view.
2.) The questions centered around how likely I am to vote in the November 2 election, whether I'm likely to vote for Democrats or Republicans for Congress, and how I feel about the President. The only politician specifically mentioned was Barack Obama.
3.) One of the questions asked which national issue most concerned me. There was a list of five or six choices, none of which was "government spending" or "deficits." The closest choice to those was "the economy and jobs," but that's not really the same thing, is it?
By leaving spending off the list, this poll is going to miss -- or misidentify -- a major source of voter discontent. For the record, I chose health care, but I would have picked spending/deficits if that had been a choice.
4.) There were a bunch of demographic questions at the end -- age, gender, race -- that the recording kept stressing were "for statistical purposes only," So . . . the other questions have nothing to do with statistics?