Green Room

Marist poll in Ohio has D+9 sample

posted at 10:23 am on November 3, 2012 by

I’d spend a half-hour analyzing the latest NBC/WSJ poll conducted by Marist in Ohio shown Obama with a 6-point lead over Romney, 51/45, but all you need to know is this: the D/R/I is 38/29/32.  In 2008, the exit polls showed a split of 39/31/30, and in 2010 36/37/28.

I’ll stipulate that if Republican turnout in Ohio is lower as a percentage of the electorate than in 2008, then Obama’s going to win the state.  But consider that even the similarly-skewed Quinnipiac poll conducted for CBS/NYT showed Republican enthusiasm outstripping Democratic enthusiasm by fourteen points:

This poll isn’t worth spending any more time analyzing.

Recently in the Green Room:



Trackback URL


You can skew a poll easily enough, but it’s hard to match Romney’s 30 – 35,000 crowd last night in Ohio to Obama’s crowd of 4000, held in the very same gym as McCain’s late rally in 2008, except McCain at least packed the place.

Adjoran on November 3, 2012 at 3:27 PM

Animal60 on November 3, 2012 at 11:02 AM

Yep. THIS is the reason for the topline number. Jay Cost has been all over this. The likely voter screen is waaaaaaaaaaaay too loose. so this is basically a poll of REGISTERED not LIKELY voters.

This has been true for the Marist polls all this season. Perhaps others as well.

Polling is expensive, response rate is low (and in Ohio it’s probably even lower because people are sick of it) so the easier it is to fill your ‘quota’ of 1000 likely voter responses the better. Haven’t a clue if this is on purpose or laziness, boredom or what on the part of individuals at the call center — but it takes thousands and thousands of calls to get a sample of likely voters these days.

MaggiePoo on November 3, 2012 at 3:31 PM

Marist Marxist poll in Ohio has D+9 sample

(Corrected obvious spelling error in headline)

landlines on November 3, 2012 at 3:46 PM