Tuesday, February 22, 2005

How To Lie About Math Aptitudes With Statistics

The FLUBA Committee on Bizarre Attempts of Academic Self-Parody notes this morning:

The Math Test Score Upper Tail: Is There Reason to Believe That Sociology Swamps Biology?

Yes:

In 1992, 2.8% of Asian-American women who took the Math SAT scored 750 or above.
In 1992, 2.1% of white men who took the Math SAT scored 750 or above.
In 1992, 0.4% of white women who took the Math SAT scored 750 or above.
In 1992, 0.2% of African-American men who took the Math SAT scored 750 or above.


Yet, when we visited the cited source for the above, we found:

Scores by Sex

In 1992, more females than males took the SAT; however, the mean score for males was 43 points higher than that for females. (See figure 1-7.) See also appendix table 1-18.

In addition, females are underrepresented among the highest scorers. ....

Which can be seen on this graph. The actual numbers for 1992, of those scoring between 750-800, being:

Total number of students: 14,132
Males......................................11,027 (2.2% of all tested)
Females...................................3,105 (0.6% of all tested)

So, just who is J. Bradford trying to kid? The usual suspects in the comments section?

We'd of thought that's so easy to pull off, that there's no sport to it. (Hint: Asians are only about 3% of the U.S. population, and there is a ceiling of 800 that conceals the real variances.)

No comments: