University of New Hampshire’s Survey Center released its latest survey on Thursday and something seems extremely fish with the numbers for Representative Carol Shea-Porter (D – Congressional District 1). The DailyKos even point this out. When you take a look at the numbers from January’s poll and compare them to the latest poll taken in April, there are a few stark examples of numbers that make absolutely no sense.
The first numbers that jump off the page are the fact that Registered Democrats’ favorability of Shea-Porter actually dropped by 9%. On the opposite spectrum her favorability with Republicans suddenly rose by 17%. Really? It may be believable that some Democrats have lost favorability but Republicans’ favorability jumps by that high of a percentage? This is not realistic. Nothing major has happened in the campaign season so far, as DailyKos points out. Shea-Porter hasn’t even held a town hall in her district! How these numbers could swing like this is definitely fishy.
Other numbers that are completely out-of-whack are when you compare the districts they used. North Country’s favorability for Shea-Porter dropped by a 23% while the Central/Lakes district favorability rose by 21%. The Mass Border district’s favorability for Shea-Porter jumped a whopping 28%! These numbers are not credible. There is typically movement up and down over a period of time in all polls but not huge swings like the ones being reported by the latest UNH poll.
Is it possible the changes in the number of people they polled weighed more heavily? They polled more women this time than previously and more women than compared to men – 35% of those polled were men while 56% of those polled were women. In the previous poll that number was 36% men to 43% women. They interviewed more Registered Democrats in this poll over the last but fewer Registered Republicans and Independents. The number of Democrats polled increased by 11 while the number of Republicans and Independents decreased by 47. In January they polled 300 people while in April they only polled 256 people.
Clearly there are issues with this poll. It seems as though someone royally screwed up when posting the numbers. That could be the only reason other than outright devious polling which doesn’t bode well for the UNH Survey Center. It is unclear if they will re-check their numbers but if the same numbers once again appear in their next poll, Granite Staters will have the answer. It’s worth watching and comparing again because something is definitely ‘fishy’ with Shea-Porter’s number in UNH’s latest poll.