Statistics Education

Ask anyone who deals with mathematics and the general
population, and they will tell you the general level of
understanding is low. But I was still surprised to read an
editorial
in this morning’s New York Times, which said:

Whereas native-born children’s language skills follow a bell curve, immigrants’ children were crowded in the lower ranks: More than three-quarters of the sample scored below the 85th percentile in English proficiency.

I believe that the general point is true, but the statement as
written makes the opposite point. In the general population, by definition 85
percent are below the 85th percentile. So if only three
quarters (75%) of immigrants’ children are below the 85th
percentile, that means that 25% of them are above the 85th
percentile, which is better than in the general population.

I suspect that it’s a typo of some kind, and maybe it’s that
three quarters of the immigrants’ children are below the 65th
percentile or some such. But the Times proofreader would have
caught it as fast as I did if he or she really understood the meaning of “percentile”. Or maybe it was the translation of three quarters into 75% that the proofreader was weak on.