A newly released USA Today/Gallup Poll shows John McCain leading Barack Obama by four points, 49% to 45%, among likely voters. This on the heels of the latest Gallup tracking poll, which is based on a completely separate and much larger sample, that shows Obama leading McCain by 8 points among registered voters. In fact, the USA Today/Gallup Poll is the first national poll in almost three months to show McCain with a lead. Meanwhile, two other national polls released today show Obama maintaining his lead. A Democracy Corps Poll shows Obama ahead by 5 points among likely voters and a Research 2000 Poll shows Obama ahead by 12 points among likely voters.
So which poll should we believe? It's not unusual for polls of the same population at the same time to produce diverging results due to differences in sampling, interviewing, and weighting procedures along with differences in question order and wording. But there's something particularly suspicious about the results of the new USA Today/Gallup Poll. According to this poll, Obama is actually leading McCain by three points, 47% to 44%, among registered voters, but trailing by four points among likely voters.
How does Gallup get from a 47-44 Obama lead among registered voters to a 49-45 McCain lead among likely voters?
A few quick calculations show how. They have 900 registered voters and 791 likely voters which means that among their 109 unlikely voters, Obama leads McCain by a whopping 61% to 7%. Putting it another way, according to Gallup 16% of registered Obama supporters are unlikely to vote compared with only 2% of registered McCain supporters. This in an election in which numerous polls, including those done by Gallup, have shown Democrats more interested and enthusiastic about voting than Republicans.
Eight years ago the Gallup organization got in hot water for using a likely voter screen several weeks before Election Day that produced wild fluctuations in candidate preference. At one point, the Gallup tracking poll went from an 8 point Gore lead to an 11 point Bush lead in three days. Of course, this was nonsense. The wild swings in the tracking poll were almost entirely caused by the likely voter screen. Those results were not to be believed. And neither are these.