The Problem With Public Policy Polling–Is Obama Really Up By 5 In Ohio?

The headline reads, “Obama leads by 5 in Ohio”. However, few details as to the methodology and sampling are given in the article itself. You have to download a pdf file to discover that the respondents self-identified 41% Democrat, 37% Republican, and 22% independent. Is it really all that unusual that Obama would lead by 5 in a poll that oversamples Democrats by 4? Then again, Obama won Ohio in 2008 by 4.6%. Given the degree of disenchantment with Obama and the state of the economy, is he really going to do as well in Ohio as he did in 2008? Is it really plausible that given the state of the race at the moment, the split of actual voters in November will be 41% Democrat and 37% Republican?

Then there is this:

Romney actually leads 46-44 with independents but Obama has the overall advantage thanks to a more unified party base. He leads 86/11 with Democrats, compared to Romney’s 83/13 advantage with Republicans.

All things being equal, this would indicate that once sampling errors are corrected, Romney is effectively tied in Ohio, which indeed corresponds to about every other measure one can find as to the state of the race.

Of course, all things are not equal. While Obama got 9% of the Republican vote in 2008, it is simply impossible to believe that 13% of Republicans in Ohio plan on voting for Obama in 2012. Obama did not get that high of a percentage in 2008, why should he do so now? 13% of Republicans may stay home in 2012, but 13% aren’t going to vote for Obama.

The poll was also conducted though automated phone interviews. Many, many people refuse to answer such polls, because they are just too darned busy, or because they are too tired and stressed to put up with this kind of nonsense, especially when they have had a hard day at work, and especially when they find out that it is a liberal polling organization such as PPP. Such polls tend to be answered by people who are idle and have a lot of free time on their hands. Thus, white working males–a large group that in the past has overwhelmingly voted Republican–will be undersampled by such polls. In the case of this poll, 63% of respondents were over the age of 46, and women outnumbered men by 54% to 46%. While the poll purports to be of likely voters, the sampling of the poll does not really resemble the cross-section of voters in 2008, or in any national election we have ever had.

None of this is to say that Romney is actually ahead in Ohio or anywhere else. The point is that we do not know how well Romney is doing in Ohio, and neither does PPP.

Enhanced by Zemanta
This entry was posted in politics and tagged , , , , , , , . Bookmark the permalink.

2 Responses to The Problem With Public Policy Polling–Is Obama Really Up By 5 In Ohio?

  1. Public polling should always be taken with a skeptical view depending on who is doing the polling. The agenda of whom is doing the poll can certainly make you wonder and the honesty of them not disclosing who they polled. I usually myself just ignore them all!

  2. Pingback: The GOP’s Advantage In 2012–Why The Polls Are Bunk | The Tree of Mamre

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s