@LisaRenee #5: most likely rounding error. My problem with the poll is their assumptions about party affiliation. Even with the open primary system we have in Ohio, there is no way either major party comes close to 50%. Even in a good year for the Rs, I doubt 48% of Ohioans would consider themselves Republicans, and this year the Republican brand is in shambles. In a Presidential year, when even the “sometimes or never” voters will turn out, those figures have to be closer to 30% for either party, and the “everyone loves a winner” factor gives the Dems the edge.

@Steve: the 4 million “Independents” cited by the Dispatch no doubt includes a substantial number of newly registered voters who either didn’t vote in the primary or registered after the primary (and have not been “flagged” D or R).

My takeaway is that the methodology is deeply flawed, and the results are not representative of the current electoral sentiment in Ohio.

Right on. I looked at the PD this morning and rolled my eyes, especially at the BIG BOLD HEADLINE on the front page: 48-46. I thought – I PAY for this kind of misinformation?

I have always thought the PD caters to its most right wing readers, thanks, of course, to the influence of Kevin O’Brien.

Your observation is part of what makes Ohio so swingy. Our semi-open primary system creates difficulty in labeling people as Dem or Republican. For example, it seems as though, while we do disagree on the number of voters who voted for Clinton, definitely some folks who usually pull a GOP ballot for the primary pulled a Dem one this year for Clinton. Now that Obama is the candidate, they may very likely pull a GOP ballot.

And then, when you look at the turnout for the primary, you can see why there are so many millions of unaffiliated – these are folks who haven’t pulled a primary ballot in either two or four years (I forget which it is) and can vote however they wish, crossing or not crossing lines, in the general.

I think it’s better for the voter – why should we care so much about who knows our party affiliation unless we want someone to know?

But for organizing and poll predictions? It’s a nightmate and keeps people guessing.

On August 22, 2008 the Columbus Dispatch reported party affiliations for Ohio’s registered voters as 2.4 million Dems, 1.5 million Repubs and 4.0 million without affiliation.

I am not a professional pollster, but I wonder how the accuracy of the Ohio Newspaper Poll’s results can be evaluated without weighting results based on self-identification of party affiliation, if I have correctly read its description of its methodology.

Trends for Ohio:

http://www.electoral-vote.com/evp2008/Pres/Graphs/ohio.html

Obama takes Ohio and rolls to a big electoral college landslide. Take it to the bank.