Dispatch Polls Are STILL Within Their Track Record Margin of Error

This post will stay at or near the top for the rest of Sunday.

________________________________

Well, they’ve laid down their markers, haven’t they (Guv, Senate, All Other candidates and Issues 3, 4, and 5)?

So, once again, lest anyone think it’s somehow all over, let’s remind everyone of the Dispatch poll’s dismal track record in the six top-of-ticket statewide races from November 2005 and May 2006 that were ideologically contentious:

Note (*): The Dispatch predictions for the Reform Ohio Now issues at this post allocated Undecided answers proportionally to actual Yes and No responses. This differs from previous BizzyBlog posts, where all Undecideds were allocated to the No response to reduce the Dispatch’s embarrassment. No such break is appropriate today.

EVERY error is big, and is in the conservative direction. In general, the larger the ideological divide involved in the contest, the larger the Dispatch polls’ results have varied from actual (Flannery v Strickland was off by 14, and Blackwell-Petro was predicted accurately; neither race was particularly contentious ideologically).

No one can honestly look at the above and say that it’s over. Strickland, Brown, Dann, Brunner, Sykes, and Cordray supporters are, however, welcome to take their chances. :–>

_______________________________

UPDATE: When you look at the details, you just have to shake your head at the obvious slants in the design, and in the makeup of the respondents:

• The poll included a question “In deciding your votes, how important of a factor were the numerous reports of corruption and scandal in state government.” That may be a valid exit-poll question, but in a poll like this, it’s essentially a push-poll question. Nothing like being suggestive. How many people were influenced by it?
• The respondents in the two top races broke down as 45% Dem, 35%-36% Republican, and 19%-20% independent. That’s laughably absurd, even if you assume that a lot of Dems didn’t say they’re independent.
• The respondents “said” they voted for Kerry over Bush 53-47 in 2004; Bush won 52-48. That’s a 10-point swing all by itself.
• 11%-12% of the reponses came from Southwest Ohio; 36% came from Northeast Ohio. Puh-leeze.
• Aw, this is ridiculous — 27% of the responses came from people with at least one union member in the household.
• The responses were also very heavily skewed toward lower household incomes (from all appearances, the income question related to household and not individual income).

What odious junk. This is a mail-in poll dominated by liberal activists and unions; reporting the results serves no useful purpose other than to influence, instead of to report on, the election. I believe on Tuesday we will again learn that it’s “worse than worthless.”

______________________________

Selected Previous Posts:

• Oct. 31 — Well, Well, Well: The ‘Zogby’s Special Sauce’ Cries Will Come from the Left This Time (Blackwell down by 7.5%)
• Oct. 19 — Hugh Hewitt on Pre-Election Polls
• Oct. 17 — Why the Polls Get It Consistently Wrong
• Sept. 29 — The Polls Done by That Columbus Newspaper Need to Be Dispatched to the Trash
• Nov. 9. 2005: Worse Than Worthless: Ohio Polls