Prometheus 6

The central question though is, how much more well-off were middle and upper-income African Americans? If THEY were much better off during Clinton than they were during his predecessors...then it makes sense for even working class and poor African Americans to think that they were better off. Because the way it stands now, as the middle and upper income blacks go, we all go. Melissa's paper made it seem as if blacks made these mistakes because they were in love with Clinton. 

I think it more likely that they drank the bougie koolaid. 

The article strikes me like she's saying Black folks love the Clintons because they made the mistakes. But yeah, folks fraggin' DROWNED in the kool-aid. Confusing wealth with the ability to make the minimum payments...

You know, I really don't mind that some folks don't like the Clintons and don't want Ms. Clinton to be nominated or elected.

On the other hand, I DO mind when somebody tries to play their readers for rubes. Ms. Harris-Lacewell seems to be doing exactly that.

Specifically:

Her article uses vague insinuation: "The data show that nearly a third of black Americans held false understandings of black economic conditions during the Clinton years."  Did Mr. Clinton do anything to foster these false understandings? Maybe he did, but the writer doesn't tell us. The author insinuates that the economic optimism reflected in the survey is a) false, and therefore b) the result of being misled by Mr. Clinton. She doens't maker her case.

She improves on the facts: "By 2000, nearly 30 percent of African-American respondents believed that blacks were doing better economically than whites." Nearly 30 percent? How near? 29? 28? She doesn't say. But earlier in the very same paragraph, she stretches "nearly 30" to "nearly 33 1/3": "The data show that nearly a third of black Americans held false understandings of black economic conditions during the Clinton years."

She uses hidden sources--she refers to a survey in which a sizable number of African-American respondents say they belive "blacks were doing better than whites," but she does not cite the source. The result she quotes is not just surprising, it's bizarre, and the author even says so: "But it strikes me as bizarre that nearly a third of blacks perceived a reversal in the deeply historically entrenched economic position of the races." I'd love to know more about that survey, but the author won't tell me where to look.

She outright misrepresents the facts: "The poorest African-Americans experienced an absolute decline in income, and they also became poorer relative to the poorest whites." That's just not true according to the United States Census Bureau. (If you click the link, check out table F-3) In 2000, the mean income for the poorest 20 percent of black families was $7,833. That number grew steadily throughout the 1990s. As a percent of the bottom quintile of white families, black families earned just 47 percent. As bad as that is, it's also the end result of steady improvement through the 1990s. In 1993, the difference was much greater with poor black families earning just 38 percent of white family income.

She uses emotionally loaded language: "The hypnotic racial dance of cultural authenticity that Bill Clinton performed in office lulled many blacks into perceptual fog." This is just insulting to everybody. Bill Clinton is a charlatan. Black people got fooled. He blinded them with his Jedi mind tricks. The author owes her readers a few details, don't you think?

There's a lot to discuss and understand about how the American economy works--and doesn't work--for different segments of the population. Ms. Harris-Lacewell isn't helping.

I don't believe

"By 2000, nearly 30 percent of African-American respondents believed that blacks were doing better economically than whites."

and

"The data show that nearly a third of black Americans held false understandings of black economic conditions during the Clinton years."

refer to the same questions. On the other hand, I need more details on that first answer myself. Since it's referred to in the abstract for the paper.

I'd love to know more about that survey, but the author won't tell me where to look.

The paper this is drawn from is here. And seriously, Sage Online Journals is offering free access to its sociology journals intil the end of February . The Journal of Black Studies is on the list.

The source for the survey is the 1994 National Black Politics Study I think. It could be the 1996 National Black Election Study but I don't think so. The article itself appeared in the Journal of Black Studies in 2005.

To be fair, Melissa Harris-Lacewell, who is a professor at an Ivy League university, should have been identified as a supporter of Barack Obama. It is only fair.

you're right, but what does where she works have to do with how she was identified by slate?

she uses the 84 and 88 national black election study the 94 national black poilitics study, the 96 national black election study, and the 2000 dubois center for the study of race, politics, and culture study for her data about black attitudes towards the economy.

Now that I see the data, I'm even more convinced that Ms. Harris-Lacewell is stacking the deck.

In her paper, in Figure 6, she presents the Assessment of Black Economic Position Relative to Whites as a trend line. In fact (and as kspence correctly notes above) she doesn't have sufficient data to support a trendline--she has the results of five studies taken over a period of 16 years by different organizations. These are discrete events. One makes as unsupportable leap to assume the trend line she depicts.

Given this data set: 1984 - 8, 1988 - 7, 1992 - 7, 1996 - 10, 2000 - 30, there are two theories that immediately suggest themselves. Either something very striking has happened or you have a bad data point. If you validate the data, then yes, something unusual has happened.

At the same time, even if the data is accurate, you can't use two measurements taken four years apart to assume a smooth, uninterrupted trend. Harris-Lacewell's figure 6 shows the percentage of blacks who believe they are economically better off than whites rising evenly from 1996 to 2000. In fact, no measurements were taken during that interval. 

Now even if you trust the data, Ms. Harris-Lacewell is still abusing her readers. In her paper, she builds a causal model that says black respondents who favor the Democrats are more likely to assess black economic position favorably versus whites. When a Democrat is in the White House, black Democrats are more likely to have a rosy outlook.

But in her Slate article, she goes much further--the change in perception is no longer the result of black Democratic optimism. Now it's the result of vile Bill Clinton's "hypnotic racial dance of cultural authenticity."

Is it because he played the saxophone?

The time series raises another question. We have data points for '84, '88, '92, '96, '00. Was there a survey done in '04?

The time series raises another question. We have data points for '84, '88, '92, '96, '00. Was there a survey done in '04?

Couldn't tell you. Two of the organizations only did one study. 

she has the results of five studies taken over a period of 16 years by different organizations. These are discrete events. One makes as unsupportable leap to assume the trend line she depicts.

You know a series of data points are always connected in a curve. That curve is always an assumption. For that matter I've wondered for years why a curve instead of straight lines, but there you are.

It's the emotionally loaded language that got you, isn't it?

It's the emotionally loaded language that got you, isn't it?

Partly, at least. (Especially "hypnotic."--"That Bill Clinton hypnotized us!") Loaded language is one of the familiar markers that a writer is trying to lead you to a conclusion. But it's more than that. It's just what I said at the top of my first post--it bugs me when I see a writer tilt the game.

As far as the trend line goes, four years is kind of a long stretch between data points. Even if you assume the data is correct, it's plausible that there were sudden rises or falls in sentiment during the intervals between surveys. Maybe attitudes stayed fairly constant from (just for example) 1996 to 1998, but then took a sudden jump to more than 30 percent and then fell back. Or that they fell to less than 10 percent from 1996 to 1999 and then shot up.

My point is, you can't tell from the data and the chart suggests that you can. This data should be shown as a histogram, not a trendline.

The reason I wonder about whether there's any '04 data is because that might build the case that the 2000 data is accurate or not.

...but what does where she works have to do with how she was identified by slate?

I'm missing something here, Spence. Are you objecting to me having identified her as being a professor? I'm not following you right along through here. I happen to admire her work.  

Even if you assume the data is correct, it's plausible...

But is it probable? 

Far, far less so than the smooth curve Prof. Harris-Lakewell projected. And every four years is one per presidential campaign...there's all manner of statistics collected on that schedule. 

I'll give yo the 30% figure needs checking. But truth is, I do believe Bill served up some trickeration

PTC you caught me with my professor hat on. I'm not "objecting" as much as I am wondering what one statement (her being a professor at princeton) have to do with the other statement (your on point belief that slate should've identified her allegiances).

there is a survey that was conducted in 2004 for a different purpose that she could have used. but they didn't ask any questions that would've helped her i don't think. but by then the paper was already in submission. 

when she presented the paper i remember a discussion she had with a brother who is now the chair of the economics depatrment at howard. the issue i have with this paper is actually NOT the fact that she used a few different surveys. she did the best she could here. the problem i have is with the conclusion. so if i recall correctly the prime independent variable is a bill clinton feeling thermometer, right?

It was worth following up this thread, just to read this sentence:

But truth is, I do believe Bill served up some trickeration

Haw!