In polling, you get what you pay for
To date, only one public poll has been released about the effort to restore Maine’s 38 year-old practice of election day registration.
The effort to place additional burdens on Maine voters is part of a national project which, in some places, has even targeted military voters.
As Steve Mistler reported, the Maine Heritage Policy Center has a “recently formalized effort to oppose the effort to reinstate the 38-year-old law.”
For a state-wide poll, this is very cheap. Now, one reason why the cost is so low is because it was a robopoll. This means that human beings did not conduct the polls. Rather, people at home picked up the phone and heard a recording. They responded not with words, but by pushing keys to indicate their answers. The person who did this and who answered might have been a 12 year old who thought it was cool to answer a survey. There would be no attempt, typical for the best polling firms, to randomize the sample by asking for certain types of people in the household, such as the oldest eligible female or youngest eligible male.
But there’s more to this — Pulse specializes in do-it-yourself polling. Rather than having questions written by someone who is trained to do this, the client writes the question.
This is really important because, as Shaw points out, A “major way that a poll can be bad is by using questions that lead respondents to answer in ways that don’t reflect their “true” opinion — in other words, which don’t predict their future voting behavior.” And, as her analysis demonstrated, the MHPC/Pulse Opinion poll had some problems with question wording.
Moreover, the owner of Pulse, Scott Rasmussen, acknowledges that, with Pulse’s do-it-yourself polling, question wording and thus results could be off.
All that Pulse does is take the questions, turn them around, and give them back to the client,” company president Scott Rasmussen said in a phone interview. “If you went and asked some off-the-wall question, Pulse would not vouch for your interpretation of the data or the reasonableness off the question.
When it comes to question-wording, these self-service polls are probably fine with simple questions on preferences for particular candidates. For more complex situations, not so much.
Does this mean we can say for sure that this poll’s findings are wrong? No. But there’s not a lot to give one confidence in its findings.