« Max Cleland and Jeff Merkley - YouTube style | Main | 140K strong Humane Society endorses Merkley! »

May 01, 2008

New Poll - surging Merkley ties Novick

SurveyUSA just released the latest poll commissioned by KATU. Merkley and Novick are statistically tied, according to SurveyUSA's analysis. Novick got 30% and Merkley got 28% with the margin of error being 3.9%. Here's SurveyUSA's own analysis:

Oregon U.S. Senate Primary: Merkley Support Nearly Triples, Now Tied with Novick -- In a Democratic Primary in Oregon for United States Senate today, 05/01/2008, three weeks till votes are counted, attorney Steve Novick and state House Speaker Jeff Merkley tie, according to a SurveyUSA poll conducted exclusively for KATU-TV Portland. Novick today gets 30%, Merkley 28%, within the survey's 3.9 percentage point margin of sampling error. 4 other candidates in single digits. 26% of voters say they are undecided, or will vote for some other unnamed candidate. Compared to an identical SurveyUSA poll released 04/07/08, Novick is up 7 points, Merkley is up 16 points. Among women, Merkley's support has tripled and he now leads Novick 30% to 25%. Among men, Novick is up 7, Merkley is up 8. Among Pro-Life voters, Merkely's support has quadrupled. Among voters age 50+, Merkley had trailed Novick by 12, now tied. As evidence of how closely matched and fiercely fought the contest now is: Among Conservatives, the two are tied. Among Moderates, the two are tied. Among Liberals, the two are tied.

I don't know what to make of the quadrupled support among pro-life voters except that it seems like it must be a reflection of Merkley's strong veteran support.

The tripled support among women makes sense. The Merkley campaign specifically targeted women with their first TV ad, and did it by highlighting his strong support for abortion rights. Which is why I'm left to speculate about the pro-life voter surge too.

Nits to pick are easy to find. Perhaps the biggest one is that just 650 of the 1800 people surveyed are deemed likely to vote according to SurveyUSA's own analysis. Also, 194 of those surveyed weren't registered to vote when they were surveyed and may have missed the deadline to register which fell smack in the middle of the three-day survey.

Posted by Kevin at May 1, 2008 06:46 PM