Critical Insights' "Spring 2008" survey is out and the numbers are interesting.
The poll, conducted June 1-27, found the following:
Tom Allen (D): 40 (34)Numbers in parenthesis are from Critical Insights' last poll, conducted Oct. 12-30, 2007.
Susan Collins (R-inc): 50 (54)
Curiously, several Maine news outlets are reporting a 14 point spread, which refers--as far as I can tell--to an unscreened pool of Maine residents.
But the 50%-40% spread represents the result among likely voters, which is obviously the more important number. It's the number we focused on when the last Critical Insights' poll was released. And the "likely voters" number is the one that was widely reported at that time.
Of course, screening for "likely voters"--or at least "registered voters"--is pretty standard among pollsters.
So why are outlets going with the 14-point number, and ignoring the 10 point spread? Your guess is as good as mine.
In any case, looking at the Critical Insights result in the context of the last three independent polls leaves us with the following:
Critical Insights (July) (June) Collins +10It now seems more likely that the Pan Atlantic poll is an outlier. And it seems pretty clear that we've got ourselves a race.
Rasmussen (July) Collins +7
Pan Atlantic (June) +25
Rasmussen (June) +7
UPDATE: The data that Critical Insights includes with the results doesn't appear to be restricted to likely voters, so it's not as helpful as it might be.
But it is notable that there's only a slight discrepancy in outcomes between the first and second congressional districts: Collins does just three points better in the second than in the first.
Also of note: Allen gets only 65% of Democratic votes, according to Critical Insights. That number is sure to rise--and perhaps substantially--as voters begin to educate themselves about Collins' performance in office over the last six years.