Explore Harvard's Nieman network Nieman Fellowships Nieman Lab Nieman Reports Nieman Storyboard

OK, what are we expecting for black turnout?

ASK THIS | October 16, 2008

Reporters and editors shouldn’t be waiting for Election Day to report Republican efforts to suppress the black vote or the Democrats’ drive for a high turnout. They should also ask the pollsters a question or two.

By Barry Sussman

John McCain has a daunting job in states with large black populations; the arithmetic works against him. So it follows that he’s making an issue of voter fraud, trying for a chilling effect on black voters on the one hand and a gridlock at largely black precincts on the other. We’ve seen this picture before.

On the Democratic side, starting with the Iowa caucuses, the Obama camp has been unusually keen and alert when it comes to process and is no doubt getting ready for the assault. A first step was to go all-out to increase black voter registration. A second is to have as many people as possible vote before Nov. 4th. And a third  is to have in place plenty of observers and lawyers in key places.

The tug of war has already started; reporters and editors should be working on it now in all the battleground states. What are the opponents doing at this moment? Are elections officials prepared? How?

Virginia, North Carolina and Florida are three of the states I have in mind. Reliably Republican in recent presidential contests, these states don’t look at all reliable in 2008. The last time I looked, opinion polls in two of them have Obama ahead and tied in the third. But there’s a key factor about the polls, often overlooked: It’s not the raw numbers alone that figure in the computations—there’s the matter of turnout expectations.

More than science is at work in opinion surveys. News media pollsters and others often adjust their samples to match census figures for characteristics such as age, education, gender and sometimes race. In drawing a likely voter model, what weighting will they assign to blacks this time around, if any? Something less than their proportion of the electorate? The same amount? More?

It’s likely that turnout by blacks will be higher than ever. Will pollsters take that into account? If yes, how much higher? Reporters should ask them. And ask them to explain their reasoning.

It’s also expected by some (me included) that a good number of GOP voters across the country, upset with the party’s leadership in the Bush years, will stay home – the way some did after Watergate when Gerald Ford lost to Jimmy Carter.

That could be true in Virginia, where a well-liked former governor, Mark Warner, is running miles ahead of his Republican opponent, James Gilmore, for an open U.S. Senate seat.

Census figures show Virginia to be 19.9 percent African American. Logically, blacks may be expected to account for more than 19.9 percent of the electorate, perhaps 22 percent or even more. Let’s say they account for only 20 percent. Let’s also say they vote 90 percent for Obama, 10 percent for McCain.

That means that before any other votes are counted, Obama has 18 percent of the total and McCain 2 percent. To break exactly even, McCain would have to get 60 percent of the non-black vote. That’s a very tall order in a state that has been moving more Democratic in recent years.

In North Carolina, blacks make up 22 percent of the population. If there too they turn out only in proportion to their numbers and vote 90 percent for Obama, then McCain would need 61.5 percent of all other voters just to get over the 50-percent mark. As in Virginia, chances are good that blacks and other Democrats are more up for this election than are Republicans.

What if blacks make up 25 percent of the electorate in North Carolina? Well, then, McCain would need 63 percent of everyone else just to tie.

Florida has fewer blacks—15.8 percent, according to recent figures—but the principle is the same: For McCain to win the state, he’ll need landslide support among non-black voters.

The state figures I've used are for all blacks, not just those of voting age or those registered to vote. I think those must be somewhat lower. Regardless, blacks are a big chunk of the electorate in these states. Black turnout may well determine the outcome in Virginia, North Carolina and Florida and other states also. That's threatening for the Republicans and explains why some of them make extravagant charges about voter fraud even when it is almost nonexistent.

McCain himself seems to be calling the attack signals. On Oct. 15th, in his third debate with Obama, he said that Acorn, a voter-registration group, "is now on the verge of maybe perpetrating one of the greatest frauds in voter history in this country, maybe destroying the fabric of democracy."

That sounds overblown. You wonder why he’d say such a thing. Maybe he’s setting the stage?

The NiemanWatchdog.org website is no longer being updated. Watchdog stories have a new home in Nieman Reports.