[aprentic]

This was actually the very first statistics project I ever did. We had the basic high school level "Probability and Statistics class". We learned some simple combinatorics, some formulas for the normal distribution and the basics of hypothesis testing. The we were told to do some project (it was junior or senior year so they gave use fairly wide latitude).

I wrote up a survey of people and distributed to classmates who had either taken the SAT more than once or had taken the PSAT and then the SAT. I asked them if they took a prep class and what kind.

The results strongly supported the meme.

There was a big gap between people who did and did not take prep classes. There was a small but statistically significant difference between the Princeton Review and Stanley Kaplan (I don't remember which one won) and there was a large gap between people who took group classes and those who had private tutors.

I didn't collect any demographic information but it was at a fairly expensive private school in Manhattan. Mostly white kids. The minority kids were mostly from prep-for-prep.