Galluping Away with the Election?

The Left Coaster is the blog you should visit if you are easily discouraged by Gallup poll results on the 2004 presidential election.

Here's the Reuter's headline based on the latest Gallup Poll: "Bush Surges Eight Points Ahead of Kerry - Poll."

For your own sanity, check out Steve Soto's wonderful analysis of the problems with the Gallup poll:

According to the internals I just received from Gallup, they over sampled once again for Republicans in their latest poll. The LV sample they used for that hard-to-believe 8% lead for Bush sure enough had 3% more Republicans than Democrats, and the RV sample had 2% more Republicans than Democrats....

How likely is it that on Election Day the exit polls will reflect a 3% advantage for the GOP, given the registration gains that the Democrats have achieved, as well as the fact that such an advantage for the GOP has never existed in the last several decades? Again, the latest Gallup LV result is based on a sample that overstates GOP participation by 3% when compared to 2000 and understates Democratic participation by 4%.

Should you therefore pay a lot of attention to a poll that is swung 7% out of line with the 1996 and 2000 elections?

Soto thinks not, and I'd agree. Indeed, I've mentioned this argument before on the blog. Essentially, polls like this one demonstrate the rather unremarkable finding that Republicans plan to vote for Bush.

Just how much do these Gallup surveys vary? Soto provides a list of data and then summarizes:

Gallup wants you to believe that if the election was held at any time during the last four weeks, the GOP would constitute anywhere from a low of 35% up to 43% of those voting, and the Democrats would constitute anywhere from a low of 31% up to a high of 39% of those voting, again all of this happening in just one month. But if you factor out the one poll that resembled the 2000 exit polls, Gallup would have you believe that the GOP makes up between 38% and 43% of those voting, while the Democrats make up only between 31% and 35% of those voting.

And the punchline?

If Gallup had weighted this LV sample with the same methodology that John Zogby uses (the 2000 exit poll results), Kerry would have a 2% lead, rather than a 8% deficit.

Bottom line: It's a very close election, but there's hardly reason to dispair.

I'll return again to regular substantive issue blogging. After all, I believe this:

"When the focus so much is on the polls, that takes away from the specifics of public policy," says political scientist Linda Trimble of the University of Alberta.

"It's all about the game -- who is winning -- rather than the substantive issues that affect voters....Trimble says the media's obsession with polls leaves less room to debate the issues. An analysis she and PhD student Shannon Sampert conducted of the 2000 election coverage by the Globe and Mail and National Post found that each paper ran at least one poll story a day. They also found that between one-third and one-half of the front-page election stories were about poll results rather than issues.

This might also be true, though the analyst was talking about the 2000 Canadian national election:

[Doctoral student] Sampert, a former television reporter who specializes in studying the media, blames a misinterpretation of the polls for making the 2000 election look like a tighter race than it was.

"The polls didn't shift any more than three points from the beginning of the election campaign to the end," she says. "That was all within the margin of error, meaning there may have been no change at all.

"In other words, it was news about absolutely nothing."

Actually, the latest Gallup poll may be worse than nothing if if creates any kind of Bush bandwagon or discourages Dems worried about his invulnerability.


Updates: Still in dispair? Read about the "incumbent rule." You'll start ignoring the spread and looking at the point totals in the polls.

Also, anyone in need of any polling result can do some one-stop shopping at 2.0004.com. These are the daily tracking poll results, for instance.

I also recommend this "meta-analysis" of the election by Princeton Professor Sam Wang.