Strategy ’08

Obama vs. the other guy, 2008

Confronting Press Corps Myths, Vol. 1

If there’s been one truly disheartening development this election season (and perhaps it developed before this year), it’s the degree to which even the most highly intelligent members of the press corps have incorporated and repeated factually incorrect myths as part of their work. On the subway this morning I was reading John Heilemann’s mostly strong New York Magazine piece on Obama and the race factor. For the record, I think Heilemann is a supremely intelligent, sharp journalist – and his emergence this year has been a net positive for political reporting (or reportage as the old folks would say). But then I caught this section:

Even now, there is intense dispute about whether the so-called Bradley effect—named after the African-American former mayor of Los Angeles, Tom Bradley, whose race for governor of California became synonymous with white voters misleading survey-takers about their intentions—helps explain Clinton’s shock-the-world comeback victory in New Hampshire. And in primaries too numerous to list, exit polls overpredicted Obama’s performance, leading cable commentators to hint that blowouts were at hand, only to watch the results roll in and prove tighter than anyone expected.

I can’t tell you how disillusioning it was to read this statement (and another one I highlight below).

There are a number of what I call “Press Corps Myths” that have infected political reporting this year. I’m not talking about the myths we’re all aware of out there: Obama’s a secret Muslim! His wife yelled at whitey! Obama swore his oath on the Koran! No, those are just myth myths. “Press Corps Myths” operate at a different level: they are factually incorrect statements repeated so frequently that somehow even extremely astute reporters incorporate them and repeat them. I mentioned in an earlier post how one of the biggest “Press Corps Myths” is the notion that John McCain is a “known quantity.” That will be covered in a subsequent post, but for now, let’s take down this big one that Heilemann used:

In the passage above, Mr. Heilemann wrote: “in primaries too numerous to list, exit polls overpredicted Obama’s performance, leading cable commentators to hint that blowouts were at hand, only to watch the results roll in and prove tighter than anyone expected.” Where to begin?

One: the Bradley Effect, as far as I can tell, has to do with actual polls, not exit polls. More importantly, a poll before an election and an exit poll are vastly different items, apples and oranges, and equating the two with the ease that Heilemann does is highly disturbing. The article itself is about Obama’s standing in the polls, NOT the exit polls, so throwing in that nugget is a dubious intellectual maneuver.

But more importantly, how often did this really occur? Yes, exit polls sometimes (but NOT always) showed Obama doing better than he actually did, but the same thing happened to John Kerry in 2004 and countless other candidates as well.  Bradley Effect?  Or cause and effect reporting?

The real issue is this concept that’s taken hold in the minds of the press that polls before elections overstated Obama’s support – Aha! See? Race is a factor!  If only the concept of the over-stated support were true.  The funny thing is, when I think back to the primaries, with the notable exception of New Hampshire, if there’s one thing I remember well it’s how strongly Obama OVER-performed what polls leading up to the events suggested.  In Wisconsin, Virginia, and numerous other states, Obama’s actual performance was WAY over what every poll actually suggested it would be.

The complete picture, as is often the case, is given to us by Nate at In his analysis he finds that averages across all states show that polls UNDERSTATED Obama’s support by about 3.3%. Any attempt to claim otherwise is dishonest:

The notion of the Bradley Effect gained a lot of currency after the New Hampshire primary, when Hillary Clinton did much better than anyone expected and won the state. However, the 8.9-point gap separating the pre-election polls and the actual results in New Hampshire represented only the seventh-largest error in the primaries. There were bigger discrepancies in Iowa, South Carolina, Alabama, Georgia, Wisconsin and Mississippi, all of which favored Barack Obama. These discrepancies did not receive as much attention as New Hampshire because they did not change the outcome of the election. But mathematically speaking, they were just as important.

And yet this notion has taken hold so strongly in the media that rarely does a day pass that you don’t hear some variation of “we saw the Bradley Effect proven in the primaries.” Last week I saw Clarence Page on a panel on TV and he said the following (paraphrasing from memory):

I had a rule of thumb in the primaries, anytime a poll showed Obama with a slight lead or close to Clinton, I added four or five points knowing that Hillary Clinton would end up ahead.

The other panelists nodded in agreement. No one corrected him. He’s not a dumb guy. Neither is John Heilemann.

And yet these intelligent journalists continue to perpetuate not just something that’s factually incorrect, but something that’s quite easy to determine whether or not it’s factually incorrect. Other than NH and a couple less dramatic incidents, there was NO widespread case of Obama polling highly in advance of an election and ultimately performing significantly worse.  It just didn’t happen.  And these primaries weren’t that long ago.

How do we disinfect the press and get them to purge themselves of this myth? Other than emails, letter-to-the-editors, etc., I’m not really sure. But it’s a start.

This was volume 1. There are many more “Press Corp Myths” to confront. Heilemann, unfortunately, gave me another one to hit when he wrote in the same article:

Would it really seem strange from that vantage point if the first black major-party nominee—a guy with a thin résumé, no foreign-policy credentials in an era scarred by terrorism, a background alien to much of Wonder Bread America, and the full name Barack Hussein Obama—lost?

Stay tuned for the next edition…


August 12, 2008 - Posted by | Media Strategy, Uncategorized | , , ,


  1. Good piece, Dansac. The real trouble is that the big, baldfaced lies (like “Obama is a Muslim”) make the little lies seem somehow less dangerous. But, of course, they’re not, and when the press adopts falsehoods as facts, we’re in a bad state.

    Comment by zenbowl | August 12, 2008 | Reply

  2. and why the hell do smiley faces keep popping up in everything I write today?

    Comment by zenbowl | August 12, 2008 | Reply

  3. This is great, thanks so much.

    Are you going to do another installment on Obama’s “problem closing the deal with… [women, hispanics, working class people, etc.]”? If I hear that one more time my head is going to explode!

    Comment by Ruth | August 13, 2008 | Reply

  4. Good food for thought. I always found it suspicious that these exit polls showed Obama over performing, which made me nervous as a supporter, because I saw the true danger in this tactic. This would be used to create the perception that Obama cannot close the deal, leading to another false assumption that there are certain groups that he is not relating to (white working class voters, older-white women, hispanics). Perception is reality and the press did a good job in creating a false reality about Barack’s ability to connect with certain voting blocks.


    Comment by Shawn | August 13, 2008 | Reply

  5. The original Bradley effect was evident in the polls leading up to the election and the exit polls. Furthermore, real political scientists (as opposed to amateur hacks with a blog) don’t mind comparing exit polls of one election to current polls of another election. They do it all the time.

    Something else to remember is that in the primaries we were just looking at a slice of Democrats, not the entire voting population. I don’t think we have any evidence to say whether we’ll see the Bradley effect in Novemeber.

    Comment by Andy | August 13, 2008 | Reply

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: