Tuesday, December 15, 2009

On Scientific Illiteracy

In discussing the recent recommendation against yearly mammograms for women between 40 and 50, NBC reporter/analyst Dr. Nancy Snyderman opined that "we are on the verge of becoming scientifically illiterate". Dr. Snyderman is much too generous in her observation. We have actually been scientifically illiterate as long as I can remember.

The agency involved balanced the negatives and the positives in making the new recommendation. The negative of course is that of every 1900 women, one will actually get breast cancer. The positives are an avoidance of anxiety based on the many false positives in those 1900 cases, and of course the reduction in expense and trouble.

The media perpetuates this scientific illiteracy by prefacing every statement with "Of course if you are that one person, then it is important." This is akin to the statement often heard from folks when you try to tell them that the odds of winning the lottery are one in ten million. The response one often gets is, "But what if you are that one person". Or, upon telling my ex-wife the odds on something, she said "But there's a 50-50 chance the odds are wrong!"

Recently I heard it said that a single person who wants to work and can't find a job is unacceptable. This represents not only scientific illiteracy, but a sort of pandering to the least common denominator of intelligence which is completely unacceptable, whether it comes from the media or from the administration. It is the same sort of mathematical illiteracy demonstrated by the acceptance of Dan Quayle's justification for getting into the Indiana National Guard during the Vietnam War, which was that "there were a hundred openings at the time", ignoring the fact that in a Guard of 10,000, there are always going to be at least 100 openings based on normal comings and goings in personnel. In other words, the Guard was full and the stats show that Quayle *did* get special treatment; but of course the media, in its ignorance, let this slide.

Now today we have a report that radiation causes 29,000 deaths per year. Part of the problem is said to be the insistence of many patients that they be given CT scans. It will be interesting to see if this information is processed by the media and the public with any kind of scientific literacy.


Laura said...

I agree, the level of scientific literacy is pretty bad. It's especially disheartening to then hear of all these climate-change deniers deny science they don't even understand!

Just as an aside, while I indeed don't buy the line that "if that one is you, it matters" ... the 1 in 1900 does matter a bit more if we look at it in terms of real numbers instead of percentages. Lakoff wrote a great article on how many more women would actually risk getting breast cancer if the age was raised to 50: 47,000. Also, we must remember to take other "costs" into account than just the monetary cost of mammograms and anxiety ... what about the extra cost of health care for those 47,000 women who only discover their cancer at a late stage?
As far as I know, Kathleen Sebelius still opposes the new guidelines.

chessart said...

I don't necessarily agree with the new guidelines, the point is to analyze the issue correctly (i.e., with scientific literacy). I would trust my former governor to get it right.

Laura said...

I saw this comic today and it reminded me of your post:


chessart said...

Interesting cartoon!