Double Negative

All attack ads aren’t created equal

Both the Obama and McCain campaigns, their coffers flush with cash, have been taking their messages to the airwaves, and the folks at the Wisconsin Advertising Project have been paying attention. In a comprehensive analysis, the group at the University of Wisconsin, Madison, categorized all the campaign ads as either positive, negative, or contrast ads, and concluded that both campaigns have been equally negative.

According to a press release (pdf) from the project, “if one allocates the contrast ads as half positive and half negative or considers the contrast ads as negative—as the Adverting Project does—the tone of the McCain and Obama campaigns has been absolutely identical.” These conclusions were cited during Wednesday’s debates and, later, all over the media.

“To date 73 percent of McCain’s ads and 61 percent of Obama’s have been negative, the report said,” was a fact mentioned by the Associated Press, the Chicago Tribune, The Hill and many others.

But while the study’s findings are consistent with its own methodology, its tidy conclusions don’t accurately represent the totality of issues associated with campaign ads, be they negative or positive.

For example, Media Matters pointed out that “the project made no effort to assess the veracity or fairness of the ads in question.” And, as project deputy director Jacob Neiheisel told me, the group doesn’t distinguish between negative policy ads and negative character ads.

So in the eyes of the project, the following two ads are equally negative:

Barack Obama: Keating Economics

John McCain: Celebrity

The criticism here isn’t aimed at the Wisconsin research group. As academics, they are obliged to reach only the conclusions that their study design allows, and the figures they’ve compiled reflect their methodology.

But the press has to go a step beyond and evaluate the conditions and conclusions of the study against the realities on the ground.

In the last weeks, the media have cited the group’s statistics, creating a false equivalency between McCain and Obama’s ads.

From CNN:

Though Sen. Barack Obama’s campaign circulated a University of Wisconsin Advertising Project study earlier this week indicating that nearly all of Sen. John McCain’s ads are negative compared to just 34 percent of Obama’s, both campaigns are spending about equal amounts on attack ads.

From The Boston Globe:

While most polls show that voters believe John McCain is running a more negative campaign than Barack Obama, a new count out today suggests they are running nearly equal numbers of negative TV ads in local markets.


Although it has often been said in recent days that John McCain has run a far more negative campaign than his main rival for the presidency Barack Obama, research shows that both campaigns spent approximately the same amount on negative ads.

One appeal of the Wisconsin statistic is that it’s counterintuitive. The consensus seems to be that McCain’s ads are more negative than Obama’s, yet the numbers appear to contradict that. But in this case, the collective intuition is correct, because McCain’s ads have been more consistently vicious, and in some instances, empty, throughout this election (see lipstick on a pig). Political journalists who have been closely watching this race are qualified, in an objective way, to make that assertion. Instead, the press has relied on a study that lets them avoid drawing their own conclusions. What’s more, they’re letting the study become the proxy for their consensus, without explaining or examining how its conclusions were reached.

In the case of the Wisconsin project, the effect of its coding system was to reduce the amount of variation among the ads. To consider ads as not only positive, negative, and contrast, but also as true-false, and personal-policy, would have introduced many more variables and, likely, made it difficult to reach easily digestible conclusions.

Some studies accommodate nuance, and some don’t.

Too bad. Reality is messy, and when the press relies on compact factoids derived from reductive research, everyone loses.

Has America ever needed a media watchdog more than now? Help us by joining CJR today.

Katia Bachko is on staff at The New Yorker.