Still Free

Yeah, Mr. Smiley. Made it through the entire Trump presidency without being enslaved. Imagine that.

Thursday, January 31, 2013

Specious Correlation

I'm not sure why I keep volunteering to receive mail from Diversity Inc. but I do. Today's award for "specious correlation" goes to the linked report that Segregated Blacks More Likely to Die of Lung Cancer
Blacks more likely to die of lung cancer than whites. Researchers have found that Black patients living in segregated counties have a lung cancer mortality rate about 10 percentage points higher than those living in diverse neighborhoods. When looking at specific counties, the researchers found that the mortality rate of white lung-cancer patients remained steady between diverse and predominantly white counties—between about 50 percent and 53 percent. Rates were comparable for Black patients living in diverse counties. But Black patients living in highly segregated counties had a mortality rate of about 63 percent. Black patients living in moderately segregated areas had a mortality rate of 57 percent.
It must be the "segregation". Oh wait, Segregation, as in legally enforced "you must live here or else" has been illegal for many decades now. And yes, I do strenuously object to referring to any neighborhood in which folks choose to live with other folks who look like them (or whatever other category one wishes to use) as "segregated".

"Separated"? Fine? "Segregated"? Please pass me with that. I live where I please. And if I choose to live among other black folks that's my choice, my business and not "segregation".

Even the summary shows this bias. "Predominately white neighborhoods' verses "highly segregated" black neighborhoods. Why not call the "Predominately white" neighborhoods "highly segregated"? Hmmm?? Why not? Why have "segregated" only attached to blackness? Anyway, first back to the report.

Now the reason I said all of that is because the report makes one believe that lung cancer is somehow "caused" by segregation without actually coming out and saying it. There are a lot of reports like that.

If one wants to see how silly it is to even mention "segregation" one has only to look at Beijing China. Millions and Billions of Chinese suffering under air pollution and the directly associated effects and "segregation" has nothing at all to do with it. We wouldn't even bother to ask whether these Chinese are "segregated" because it has NOTHING to do with what causes the lung problems they have. Instead we would ask about factories, vehicle density and other things that are direct contributors to disease. Clearly discussing who each Chinese person lives next to has nothing to do with lung disease mortality.

Going to the Reuters report we find:

He told Reuters Health that by comparing different counties, a person would find one has resources the other does not, such as hospitals and doctors.
So it's not segregation it's where hospital and other healthcare resources are located and whether residents can reach those resources. Clearly one can have a so called "segregated" neighborhood full of hospitals and doctors no? of course, because the "segregated" white neighborhoods apparently have lower incidences of lung cancer mortality. So since they are just as "segregated" why even bother pointing to so called "segregation"?

Politics, that's why.

StateHealthFacts.org has a nice chart of Percent of Adults who Smoke by Race/Ethnicity, 2011

Blacks are shown to have a higher percentage of smokers than whites nearly across the board and Native Americans beat out everyone else wherever the data is available.

Guess what happens when you belong to a group that smokes the most? Higher lung cancer incidences and deaths.
Shocking! I know.

o_0

Let's take the data for Washington DC, a pretty "segregated" city. Blacks are reported to have a smoker rate of 31%. The white rate is 9%. Guess who's going to have a higher incidence of lung cancer deaths? Even my cat can figure this one out. I don't need to even discuss "segregation" when such blatant differences in smoking rates exist.

This is the same old 1954 "if they aren't in proximity to white folks then they are damaged goods" argument that Diversity Inc. is trying to push (again).

What does the JAMA article conclude:

Lung cancer mortality is higher in blacks and highest in blacks living in the most segregated counties, regardless of socioeconomic status.
Oh really? I think I just posted that. Oh so the issue isn't actually "segregation" but a combination of blacks smoking more and a "lack of resources". Well, DUH.

Look, This is about culture. It is about generally black folks avoiding doctors and hospitals and "sucking it up" rather than getting medical attention.

A couple of things need to be done immediately for studies like this. If a researcher want to make "segregation" claims they should first have to "prove" segregation. They should need to prove that folks in the area were forced to live there and prevented by law (or by accepted illegal business practice) to live anyplace else.

Secondly, they should need to study other so called "segregated" groups. For example they need to go to "Chinatowns" and "Little Italies' and the like and perform the same studies. This is the only way to prove show that "segregation" is the issue. If they do not then any claim of "segregation" as a factor should immediately void the research project from publication.

Thirdly, if researchers want to study blacks in various social situations then they should do so and state so. For example this study should have said blacks who live in neighborhoods with "ready access to medical facilities" and those who do not. It should include data such as health consciousness and activities since these issues will confound any results.

It should simply no longer be acceptable to simply point "segregation" at black folks. Scientific publications should use terms such as "highly homogenous neighborhoods" or "heterogenous neighborhoods" to describe areas that are predominated by one population or a plurality of populations respectively.

Black folk need to stop using the term "segregated" to describe neighborhoods and schools in which they predominate. It is negative. No wonder so many black neighborhoods, particularly those that are poor.

But so long as certain groups and organizations can get paid and get recognition by misusing the term "segregation" it will continue to be misused. I'm certain I'll get another mailing from Diversity Inc. about "segregation" in the near future. I just love it when non-black people tell me about "segregation".