Race, Race, Race

Man. Now everybody's calling everybody a racist.

Wait a minute. Before it slips beneath the waves forever, I would like to take a quick second look at the recent Dallas Morning News series, "Striking Differences," in which the newspaper accused the Dallas County prosecutor of racism.

Race is a tough issue in town. Everything in Dallas is about racism. Again. I don't know how it could be otherwise. The entire conservative wing of black leadership in the city is under FBI scrutiny in a massive corruption probe. No white elected official has been named yet. How can you have a deal like that going on and people not talk about race?

That's why I thought the Morning News series on jury selection in Dallas County criminal district courts was potentially volatile. In a racially charged atmosphere, the city's only daily newspaper publishes a huge two-years-long study accusing Dallas County District Attorney Bill Hill of running a racist department.

Dallas County District Attorney Bill Hill thinks The 
Dallas Morning News did two years of digging, only 
to find that he and his assistants are not using racist jury 
selection techniques.
Dallas County District Attorney Bill Hill thinks The Dallas Morning News did two years of digging, only to find that he and his assistants are not using racist jury selection techniques.

And nothing. No volatility. As far as I can tell, the whole series--26 copyrighted stories, by my count--sank beneath the surface without making a ripple. It was like watching the Titanic disappear into a goldfish pond.

No other media picked it up. There was no brouhaha. I've been calling lawyers and law professors about it all week: The most common reaction was that they'd have to go back and read it before they could comment.

How can that be, in a town where almost everybody black is shouting racism and almost everybody white is shouting not-racism? Here's a case of alleged racism in a very important public office. Everybody shrugs.

I think I know why. This is really a story about journalism more than racism. This story didn't get any traction because the Morning News failed to make its case. And that happened because of the way they went after it.

By the way, I don't normally earn my rent money defending prosecutors. I figure most prosecutors can damn well defend themselves. But under these circumstances in the city, this issue seems important to me for what it says about accusations of racism and how careful we all have to be about making them.

The main ammunition the newspaper offered against Hill was an arcane statistical study that they said found racism in the techniques used by assistant district attorneys to strike black people from juries. The Morning News has been drifting toward this sort of thing in the last couple years, using social science techniques rather than old-fashioned reporting and, in some instances ("Dallas at the Tipping Point"), even hiring commercial consultants to do their reporting for them.

In this case, the newspaper offered the results of a "logistic regression analysis model" to prove Hill and his staff are racists. The paper said the model showed that all of the explanations offered by prosecutors for their peremptory strikes keeping members of the jury pool off juries were smokescreens. The real aim, the paper said, is to get blacks off juries.

The model works like this: You take all the documented factors that could explain a strike--age, race, income, prior experience with law enforcement and so on--and pour them in. The model whirs and chunks along for a while like an electric martini mixer. You open the lid, and what do you find?


In fact, logistic regression is probably only arcane to me. It's used commonly in all kinds of social research and has been around for a long time. It can be tweaked and tuned to do a pretty good job of analyzing social outcomes.

Or a pretty bad job. Depending. Kind of like a hunting rifle with a high-powered scope. You still have to know cows from deer.

Or, as the social scientists say, "Garbage in, garbage out." The accuracy of the model's predictions depends entirely on the design of the model, the variables fed into it and the values assigned to those variables. You could have a model that gets it wrong most of the time.

In 1986--the year the Supreme Court decided Batson v. Kentucky, ruling against purely racial striking of jurors--the Morning News published an important series of stories exposing flagrant racism in jury selection by Dallas County prosecutors. The point was that hardly any black members of a jury pool ever got picked for a jury.

The strangest thing about this recent series of Morning News stories is that when you read down, the facts presented by the News' reporters seem to contradict the sensational headlines. The first story starts out, "Key Findings--Dallas County prosecutors excluded black jurors at more than twice the rate they rejected whites."

But when you read on, you find that the racial makeup of juries today in Dallas County almost exactly matches the makeup of the jury pools. The percentage of the original pool that is black is the same as the black percentage of the picked juries.

So how does the Morning News figure blacks are being unfairly knocked off juries during the picking process?

Bill Hill told me: "I think what happened is that it took them two years to gather all this information and develop these models and crunch all these numbers. I think they were probably disappointed when they came to the conclusion that the number of minorities on the panel ended up that same percentage on the jury.

Next Page »
My Voice Nation Help