Looks like he may be desperately needing her help again in the general election this Fall. His campaign is beginning to take on a lot of water and the candidate is beginning to look more like the failed Democratic Presidential candidates.

Interesting, just more information supporting his celebrity label which is going to stick....

Sounds right to me. Talk to any pollster and they will tell you that at least as many people judge candidates by their personal associations as by their positions on issues. Which is why I think Obama should reach out to Merle Haggard, who obviously doesn't want to vote Republican, and could influence exactly the sort of "disaffected white males" Obama is desperately trying to reach.

For more, please see: how-barack-can.html

Can someone help me understand why Oprah has not been MORE public about her support for Obama? Getting some frequency of the two of them together seems like a powerful weapon. Not every day, but get it out there.

The celebrity message is for angry older voters who lament the general decline for society. Lots and lots of people in this country actually think celebrity is a good thing - sometimes the IDEAL.

Oprah has reinvented what a good celebrity can be (entrepreneur, literacy advocate, movie star, media mogul, social conscience). Her audience is huge and spans across ethnic groups and cultural differences. I don't think that the GOP wants to go after Oprah, and Obama can benefit from that.

Oprah's presence in Obama's campaign has the power to take the GOP 'celebrity' arrow and redefine it as something smart, healthy, and popular.

So what you're saying is ... affluent blacks weren't sure if or how they were going to vote until Oprah endorsed Obama?

A likely story...

Zac: Don't be a dope. Oprah's fans are mostly white. When the authors say her magazine's readership is "disproportionately" black, that just means it's more than 10%.

Some folks just don't have enough to do.

Could someone remind me how many votes Obama got in total in the primary? It helps to put an extra million in context.

I can't stand Oprah Winfrey personally, but I like her magazine and I'm voting for Obama only by default. So where in this tortured demographic do I fit, exactly?

My other question about their methodology: Is lame-ass hyphenated?

Quite interesting. I presume those who consider this lame, a waste of time, etc. would prefer we just leave the election analysis in the hands of the competing narrative crews?

BUT - quoting their estimate with 7 sig figs is like a red flag to me.

O-bama is going to need all the help he can get.
Check this out:
http://www.huffingtonpost.com/alex-castellanos/the-molten-core-of-barack_b_116904.html.

Oprah 4 Veep!

Hey, if William Randolph Hearst could help get FDR in the White House (which he did), I don't mind Oprah Winfrey helping Obama get there.

So it WAS her endorsement that gave us Bush in 2000! Thanks for clearing that up for us.

No confidence interval, at least not in the abstract. That's a bad sign.

The first author, Craig Garthwaite, is apparently a right wing think tank type. While he holds various titles such as "director of research" and "chief economist, it looks like his highest degree is a masters in public policy (under wingnut welfare rules, you can be a chief economist without a degree in economics, whodda thunk it).

source: http://mediamatters.org/items/200503050001

Kevin, while I kinda agree with you on real vs. nominal economic numbers, I personally get more worked up by point estimates given without any indication of statistical confidence.

My narrative of the 2008 Democratic primary is now complete: You are all advised to adopt this, because it happens to be true.

1. Clinton's vote for the war authorization left an opening for an anti-war candidate.

2. Obama's Iowa win meant the African Americans would rally to his candidacy. With African Americans being 25% of the Democratic primary/caucus electorate, this meant that Clinton had to get almost 2/3rds of the non-African American vote.

3. Oprah's endorsement of Obama meant that Clinton would not be able to climb that 2/3rds hill.

Congratulations to President Obama.

This is Tim Moore, one of the authors of this paper. Thanks for the interesting comments.

I think it is important to provide some information about our affiliations and motivations. Both Craig and I are PhD students in Economics at the University of Maryland. We thought it would be interesting to see if Oprah's endorsement had any effect on the Democratic Primary, and tried to think of ways to do so. We do not really have a view on who it may be good or bad for - hopefully it provides us with more insights into the political process and voting behavior (although it is still a working paper and we are aware of the difficulties in inferring causation).

We have been full time graduate students for the past few years, with no current employment by think tanks or similar (although Craig did work for a think tank before entering grad school). We received no funding or support on this project.

Tim Moore, thanks for the info re your affiliations, etc. Did you estimate a confidence interval for your 1,000,000 figure (putting aside the issues with inferring causation as you say)?

OK, so I found the confidence interval in the body of the paper (p.38):

"The 95% confidence interval for this estimated effect is 423,123 to 1,596,995."

Fair enough. So with caveats about the model assumptions, that's a pretty serious effect.

Tim, if I may give you an editing suggestion for the benefit of lazy readers like me, please put that in the abstract. The conclusion too, while you're at it.

What was the impact of Oprah's endorsment of Obama on the Oprah brand?

Aw. Poor widdle Repubwicans. Defeated by Oprah. How sad for you.