SKEWED DATA

James Joyner · Tuesday, May 6, 2003 · 1 comment

Power Line has an interesting analysis of Save the Children’s country rankings, where the US again did poorly. Why?

The problem with rankings of this sort is that advocacy groups nearly always put their thumbs on the scale by including factors that are mostly political. Thus, the data show the U.S. ranking 4th on the “Children’s Index,” but only 13th on the “Women’s Index.” (The two are combined to form the “Mothers’ Index.”) Why this difference?

It doesn’t take long to figure out. In the “Women’s Index,” along with such logical items as “percent of pregnant women with anemia” and “adult female literacy rate,” Save the Children includes a “political status” column, measured by the percentage of seats in the national legislature held by women. This accounts for most of the difference between the U.S. and European countries like Sweden, Norway and Switzerland, which rank at the top of the index and have more women in their legislatures.

There is also a sneakier reason. The “Women’s Index” includes infant mortality as a factor–as it should, given the purported purpose of the study. But the infant mortality rate is measured as the “lifetime risk of maternal mortality.” That language puzzled me for a moment, until I realized what was going on. The birth rate in the U.S. is much higher than in European countries like Sweden and Switzerland. By adding up the total “lifetime” infant mortality risk, instead of assessing the risk on a per-baby basis, as would be logical, Save the Children is penalizing American mothers for having more children. This can hardly be unintentional.

It is always a terrific practice, and one seldom engaged in by journalists, to check the study methodology.

(Hat Tip: InstaPundit)