Election Surveys That Screen Out 'Unlikely' Voters Might Be Outdated Al Hunt
Presidential elections are poll-driven. The candidate ahead in the surveys usually gets better coverage, and the results energize supporters. The one behind often comes across as doing little right, and campaigns and constituencies lose confidence.
But what if the polls are wrong, and we aren't surveying the real likely electorate?
This might be more than an academic issue. A number of polls this presidential race show a gap in the preferences of registered voters vs. likely voters. In these models, the president usually does better with likely voters, the figure most news organizations emphasize. To get to likely voters, all polling organizations use what is called a "screen," asking questions to determine who is likely to actually turn out on election day.
These screens differ greatly, as there is no consensus among experts on what works best. "This is an art, not a science," says Peter Hart, the prominent Democratic polltaker who has helped conduct The Wall Street Journal/NBC News survey for 15 years.
This controversy will be fueled by today's just-released Gallup poll that shows George Bush with a 13-point lead over John Kerry. That is at variance with other surveys this week, which suggest a tight race with a much smaller Bush tilt. But the likely voters margin also is considerably larger than the eight-point advantage in Gallup's registered voters in this survey. The likely voters match-up invariably gets more attention.
Gallup explains it has what it considers a time-tested formula for determining most likely voters. It asks eight questions, such as current intensity of interest, past voting behavior and interest, and whether you know where your voting place is.
"We've discovered that if we ask a set of more indirect questions, we can better predict who is or is not likely to vote," Frank Newport, editor in chief of the Gallup Poll, has said.
But there is reason to suspect those criteria are outdated, especially in an election where both sides say the intensity level is much higher than four years ago and get-out-the-vote organizations are considerably better than ever -- few people on Nov. 2 will be in the dark on where the voting polls are.
"A formula that made sense years ago may not recognize all the changes in society," notes Mr. Hart. "It gives more credence to past behavior and too little to current interest."
"For low-turnout elections those old models work well," suggests Bill McInturff, a Republican, and the other WSJ/NBC News pollster. "But in today's presidential election those models tend to [tilt to] a little older, a little more white, a little more affluent and a little more Republican voters. They may miss some of the extraordinary activity going on in African-American and Latino communities."
The registered-likely voters dichotomy also is evident in some of Gallup's state surveys including last week's Ohio results." Among registered voters in the Buckeye State, Bush-Cheney had a 48%-to- 47% edge, a dead heat. Among likely voters, however, this poll had the Republicans up 52%-44%; that garnered all the attention, followed by a spate of stories suggesting this key battleground state was moving to the president.
Curiously, the Gallup poll in the similar state of Pennsylvania at the same time showed a virtually even race among both registered and likely voters. And occasionally, the screen favors the Democrats; a Marist survey this week of New York state showed Sen. Kerry 11 points ahead among likely voters, but only seven points ahead among registered voters.
But most of the time the screen for likely voters tilts Republican. In 2000, Gallup's election eve survey showed George Bush ahead by two points among its likely voters; he trailed Al Gore by a point among registered voters, very close to the final outcome.
In 2000, the next to last WSJ/NBC poll before the election showed Republicans doing three points better among likely voters than registered voters. The election eve survey showed Bush up three points among likely voters, but failed to tally registered voters and didn't predict Al Gore's victory in the popular vote.
The Wall Street Journal and NBC News have settled on one question to screen likely turnout. Registered voters are asked their interest level in the election on a scale of 1-10, and those that respond 9 or 10 are considered likely voters.
Both camps expect an increase in the 105 million Americans who voted last time; the Bush camp looks for abut 111-112 million while the Kerry campaign projects 116-118 million; nobody can be sure exactly who those additional voters might be.
The probable outlook: Polls will vary and conflict if this race remains tight. Also, poll watchers must remember that the best survey has a three or four-point margin of error; that means if it shows the race even, one or the other candidate actually could be up by a half-dozen. Here's a final guide: if almost all the election eve polls show one candidate up four or five points or more, take it to the bank. But if most show the race within a couple of points, plan on staying up late election night.