## Thursday, June 11, 2015

### Texas, Abortion, and Math

While reading this New York Times article about a federal appellate court upholding pretty strict limits on abortion in Texas, I came to the following passage:
The Fifth Circuit panel found that the percentage of affected women who would face travel distances of 150 miles or more amounted to 17 percent, a figure that it said was not a “large fraction.” An abortion regulation cannot be invalidated unless it imposes an undue burden on what the Supreme Court has termed “a large fraction of relevant cases.”
And I got to thinking...

So, what does constitute a "large fraction"?

Well, I found this Slate article discussing the difference between outbreaks and epidemics. In it, they state that, in terms of the flu:
If the number of flu-caused deaths exceeds 7.7 percent of the total, then the United States officially has an epidemic on its hands.
Then I got to thinking about autism, because, well, you know, we're an autism family, and all the statistics that get thrown around regarding that issue, like 1 in 88 kids has autism. You've heard the stories and statistics and the desire to determine what's been causing this autism epidemic so I won't bore you with all that. Instead I'll just say that there does seem to be a very big concern regarding the supposed increase in the number of children diagnosed with autism. Some statistics even say that 1 in 50 children have autism.

That sounds scary, and I admit it is concerning, but 1 in 50 amounts to 2% of the children. So my question is: If 7.7% of the total number of deaths are attributed to a certain ailment and that constitutes an epidemic, or if there is so much concern over the well being of 2% of a given population, then how can it be said that 17% is not a "large fraction" of the given population in Texas?

It seems very large.

Seventeen percent of just about anything would be a cause of great concern.

Except for abortion rights in Texas.

#### 1 comment:

1. I think it is a matter of vagueness of language, not one of math. I think it also has to do with the perceived severity of the problem, and where one stands (pro or opposed).

Suppose that in Texas all smokers could smoke for free. Then suppose that for some reason, a law was passed that smokers in the more rural areas had to pay for their cigarettes, and that this affected 17% of Texan smokers. If one were a smoker in Texas, this might seem like a huge number... almost a fifth of the population would go from having free cigarettes to having to pay for them. This would be outrageous. However, if you are working to eliminate smoking, you would take the opposite view. You would think it was an ok start, but a whopping 83% of smokers are completely unaffected.

Or suppose they passed a law that said that 17% of kids of families falling below a certain income would get free meals at schools. I may think that it was a good step in the right direction, but I'd find it hard to call 17% a large fraction of cases.

Or consider the example with the flu. I'd agree that 7.7% of cases resulting in death would seem like a significant number to call it an epidemic, but what if we changed it to say that 7.7% of cases resulted in hospitalization, with some/most cases recovering? Or 7.7% resulted in people missing work? I think we can all agree that the most drastic case, death, is cause for alarm. But once we back off of that extreme, our perception changes. I wouldn't blink if someone told me 7.7% of the population had to use sick days last year because of the flu. I'd have a different reaction if someone told me it was 75% of the population.