Maine’s conservative on-line “news” service, The Maine Wire (which is run by the Maine Heritage Policy Center, MHPC), responds to a poll showing Angus King ahead in the Senate race, with House incumbents Pingree and Michaud ahead as well, by saying “Consider the source.”
Now, there is no doubt it’s worth looking at who conducted a poll. This would be especially important if the group hasn’t released public polls. But that’s not so in this case.
Thus one can do what one normally does, which is to look at a polling operation’s track record, comparing their poll results to the actual vote.
So what do we see?
Both MPRC and MHPC did polls on election day registration, as did other Maine polling firms.
MPRC did the best of all polls on 2011′s Question 1, which restored election-day registration by a 60-40% landslide.
MPRC estimated a 54.7% yes vote and the actual vote was 60%
In contrast, the poll done by the MHPC was badly flawed in design and was the only poll in the state to predict No would prevail, finding only 47% voting yes, with 53% voting no. Again, the actual yes vote was 60%.
Admittedly, the MHPC was released weeks before the election so perhaps opinion shifted somewhat before election day. But the flaws in its instrument were quite evident and were not surprising, given that the owner of polling company used acknowledged:
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.
This blog discussed the MHPC poll and found it to be a case study in what not to do in polling. See:
1) Polling election day registration in Maine (on sampling)
And on flaws in the MHPC’s analysis of problems with Maine’s system of election -day registration, see Mis-Under-Over-Estimating Mainers.