Pollster.com

Articles and Analysis

 

The Times, Obama and Race

Topics: Barack Obama , Blogosphere , CBS/New York Times , Measurement , NY Times

While I was busy column-writing this morning, the blogosphere was busy discussing today's New York Times story on the just released CBS/New York Times poll. The story, by Adam Nagourney and Megan Thee, had two central themes. First, it confirms the results of decades of prior surveys showing that "Americans are sharply divided by race," and the second theme comes from the story's headline: "Poll Finds Obama Isn't Closing Divide on Race:"

The survey suggests that even as the nation crosses a racial threshold when it comes to politics — Mr. Obama, a Democrat, is the son of a black father from Kenya and a white mother from Kansas — many of the racial patterns in society remain unchanged in recent years.

Indeed, the poll showed markedly little change in the racial components of people's daily lives since 2000, when The Times examined race relations in an extensive series of articles called "How Race Is Lived in America."

My first reaction was similar to that of Ben Smith, who wrote this morning:

Whether or not it's notable in a broader social context, it doesn't really seem like big news on the campaign front, though I guess you could argue that it means Obama has not personally transformed American race relations, at least not so far. I'm not sure I'd seen anyone claim he had.

Over at Swampland, Karen Tumulty passed along an email from Jackson Dykman (Tumulty describes as Time's "datameister") who put it more strongly:

[T]he premise of the story is, well, utter nonsense.

Are we really supposed to think that because a black man has become the Democratic nominee in recent weeks that he somehow should have cured (or markedly improved) race relations in this country? This is just a silly premise, yet the story thrust of the story seems to be shock and surprise that the mere fact of Obama's candidacy hasn't reversed—or obliterated—the slight increase in racial tensions in this country over the past 8 years.

Dykman's critique is worth reading in full.

Later, my colleague Marc Ambinder received a response from Nagourney to the criticism (which also included push-back from the Obama campaign) including this partial concession:

One last point: I do think there is room for discussion about the headline – "Poll Finds Obama Candidacy Isn't Closing Divide on Race". The point of the story is that black respondents apparently do not see the fact of Mr. Obama's candidacy as evidence of significant improvement in race relations. The story does not suggest that there is some onus on Mr. Obama himself to be closing this divide. I also, on a smaller matter – and the one matter the Obama campaign did raise with me – should have included, in saying that 20 percent of white voters had a favorable view of Michelle Obama, the fact that 72 percent either have no opinion about Mrs. Obama or hadn't heard enough about her, to avoid any suggestion that 80 percent had an unfavorable view of her.

Which brings me to the poll question lesson for the day. The standard CBS/NYT favorable rating question differs from other pollsters by offering respondents an explicit option to say they are unfamiliar with the person in question, e.g. "is your opinion of Michelle Obama favorable, not favorable, undecided, or haven't you heard enough about Michelle Obama yet to have an opinion? (emphasis added)."

As a result the "favorable" percentage produced by the CBS/NYT poll typically lower than that yielded by other surveys, but my sense has always been that their wording provides a more accurate read on the real perceptions held by voters. Force respondents to choose between "favorable" and "unfavorable," and those just vaguely familiar with the name of a public figure (but little else about them) will typically say they are "favorable."

According the the CBS/NYT survey, 20% of white voters have a favorable opinion of Michelle Obama, while 19% are unfavorable, but 17% are "undecided" and 38% have not heard enough to say. In other words, most do not know her yet.

 

Comments
hobetoo:

I'm a little sympathetic to Nagourney in one sense. From my own experience, headline writers can make fools of us all. At least in academic writing we generally get to write our own headlines.

That said, I wish the NYT only assigned people to do polling analysis who had a strong academic or scientific sensibility about "data." Instead of writing a "story," they should try to prove a point by rejecting plausible alternatives. This can be done in a stylish way, but the burden of proof should fall on the writer-analyst to assure that the data interpretation is accurate.

____________________

jsh1120:

Mark,

Thanks for noting the difference between the CBS/NYT wording of their "fav/unfav" measure and that of other polling organizations. I find this particular indicator an interesting one since it both encompasses a broad range of factors and isn't necessarily affected by comparisons among particular figures.

At the same time, I've been struck by the apparently wide range of results pollsters get on such a measure. Your comments about an explicit "no opinion" option are well taken. In addition, I suspect the measure is particularly susceptible to recent news stories.

I haven't seen an analysis of the relationship between favorability and candidate choice (other than the recent USA Today/Gallup cluster analysis.) It would be interesting to see a more detailed analysis of the power of the variable.

____________________

PatrickM:

I'm looking forward to the next story in this series: "Poll Finds McCain Candidacy Isn't Closing Divide on Age".

Perhaps they'll even extend it to spouses: "Poll Finds Cindy McCain Isn't Closing Divide on Tastes Great vs. Less Filling".

____________________

David Moore:

The problem with the NYT article on the racial divide was that it misrepresented what results the poll questions provided. See my post on skepticalpollster.com for a more extensive critique.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR