I am probably wrong about this (corrections welcome!), but there seems to be a non-zero chance that Amherst is misreporting its SAT data and that, therefore, Williams should be alone at #1 in the US News rankings. Consider the data for the class of 2011 on what percentages of the class took the SAT and the ACT.

Williams: SAT: 95% and ACT: 18%. Total: 113%.
Amherst: SAT: 76% and ACT: 23%. Total: 99%.

The Williams numbers are not unusual for this year. For the class of 2002, we have SAT: 99% and ACT: 15% for a total of 114%. And that makes sense. The SAT is now truly a national (even international) exam for those high school students thinking of elite schools. Although many/most of those schools will also accept the ACT, it is now common for academically gifted students in the South and Midwest (ACT strongholds) to take the SAT as well. That pattern explains why almost every single member of the Williams class of 2011 took the SAT. (Why many students also take the ACT is a topic for another day.)

Is it really plausible that 111 of the 474 students in the Amherst class of 2011 did not take the SAT? No. (The only plausible explanation that I can think of is that Amherst is now drawing from a signficantly different student population that Williams. Yet that seems highly unlikely and I have never heard of such a tendency. Has anyone?) If (many of) those students did take the SAT, is it plausible that Amherst does not have access to their scores? No. (I think that Amherst requires SAT subject tests and that any student who took such subject tests and reported them to Amherst would have to allow Amherst to see her SAT scores as well.) If Amherst does have access to their SAT scores, then why doesn’t it report those scores in its Common Dataset? And, even more interesting, are the SAT scores for those students particularly low?

Comments welcome. (I especially hope that data-maven HWC will chime in.) Are other elite schools more like Williams or Amherst? (At Swarthmore (pdf), 95% of the members of the class of 2011 took the SAT.) Without looking at the data yet, one hypothesis would be that Amherst reports only ACT scores (and not SAT scores) for its weaker students. That would cause its ACT scores to be lower (relative to its SAT scores) than those for other elite schools. Do we see that?

Print  •  Email