Biased headlines

I didn't hear much hubbub in copy-editing cirlces when the American Enterprise Institute released a report on bias in newspaper coverage earlier this week.

The conclusion of the conservative think tank was not surprising: Economic reporters slant the news to favor Democrats.

But the method the two economists used to come to this conclusion did catch me off guard (emphasis mine):

The two economists combed through 389 newspapers and A.P. reports contained in the LexisNexis database from January 1991 through May 2004, during the administrations of George H. W. Bush, Bill Clinton and George W. Bush. They picked out headlines about gross domestic product growth, unemployment, retail sales and orders of durable goods and classified the headlines' depiction of the economy as either positive, negative, neutral or mixed. Then they crunched some numbers.

They looked only at headlines.

Is this fair? Of course, not. "A headline is not coverage," as Slate's Jack Shafer put it.

But does it matter that it's fair? Not so much. The whole job of a headline is to sum up the stories, to draw readers in. I'd say it's even worse for headlines to be found biased than for stories.

So, write off the study for its partisan overtones if you'd like. But don't write it off because it looked at headlines instead of articles. Readers don't see the difference.

>Do Newspapers Make Good News Look Bad? [The New York Times]
>To analyze articles, you must read them [Baltimore Sun]
>Headline bust [Testy Copy Editors]