<p>Made a mistake on the title, it's actually a 40 year difference, 1966-2006 scores, not 1966-1996. </p>
<p>The 1966 scores have been re-centered to make an apples to apples comparison with 2006 scores:</p>
<p>As I believe the SAT to be a great indicator of IQ (maybe not in every case, but in aggregate case, one can most certainly derive an accurate IQ of student body),these numbers pretty much to me shoot down the theory of grade inflation being due to students getting so much smarter.</p>
<p>[National</a> Trends in Grade Inflation, American Colleges and Universities](<a href=“http://www.gradeinflation.com%5DNational”>http://www.gradeinflation.com)</p>
<p>Decline of the LAC and increased extremism.
Top 6 in 1966 (4 LACs):
- Stanford - 1488
- Harvard - 1460
- Pomona - 1410
- Smith - 1410
- Reed - 1400
- Carleton - 1390</p>
<p>Top 6 in 2006 (1 LAC):
- Harvard - 1484
- Princeton - 1475
- Duke - 1451
- Dartmouth - 1442
- Pomona - 1437
- WUSTL - 1434</p>
<p>More interesting for many other reasons.</p>
<p>there were no uniform reporting standards in 1966. as a result, id be very hesitant to reach ANY conclusion based on published scores from that era.</p>
<p>“Decline of the LAC and increased extremism.”</p>
<p>lolwut.</p>
<p>And history has ended, mirite?</p>
<p>Amherst, Williams, Wesleyan, Swarthmore–you’ll find none of them on this list. A lack of data does not imply a “decline.”</p>
<p>FWIW I have data for over 100 schools, for circa 1970, on a spreadsheet.
I got out one of the old college guides when D1 was looking at colleges, because she said everything I “knew” was obsolete. So I went back and looked at the ordinal relative selectivity levels from when I was looking vs. when she was looking. Though much was similar there were indeed some changes.</p>
<p>The data that appeared in the guide books from that era were actually more complete than what I see now. They broke out colleges of multi-college universities separately, and also reported scores separately by gender.
that data was really interesting to see, the ranges for M and F were often quite different.</p>
<p>Huh? Where is a lack of data implying anything? This (very significant) sample of LACs has declined relative to the large universities. There is no reason to assume that the sample is biased towards LACs which have declined. Even if the this sample is not at all predictive of the rest of the population, and the other LACs have remained constant, the population as a whole will have declined.</p>
<p>tomslawsky-
Very interesting. Thanks for posting this. From where did you get the 1966 SATs? How did you recenter? Did you use the conversion table on the College Board website? Thanks again.</p>
<p>I didn’t convert, they did:</p>
<p>"Here we compare average 2006 SAT scores (Math + Verbal) with re-centered 1966 SAT scores for the schools for which we could find 1966 data. The average difference is 16 points, a negligible amount. Breaking down differences by selectivity indicates that highly selective schools, on average, have seen modest improvements in student test scores (55 points) and non-selective schools have seen modest drops (33 points) for the schools examined here. Note that these are estimates of SAT scores based on Department of Education data. "</p>
<p>The only caveat I would have to this is that in 1966, there were entire regions of the country where very few people took SAT’s. Therefore, schools in those regions have SAT data that is likely drawn from a small # of people.</p>
<p>but wouldn’t all of the applicants have taken the SAt though? Was it required then?</p>
<p>I give about 1% probability that the data on Stanford for 1966 are correct. </p>
<p>However, let’s assume they are. What does this mean?</p>
<p>One conclusion possible is that Stanford has decided to select its class with less weight on SAT in its decision criteria in 2006 than in 1966. Stanford has never been known as the “1600 SAT school”. I has always been known as the “Student Body Presidents’ School”. Different selection criteria. </p>
<p>If someone could verify these data, then we could have some interesting discussions about why Stanford has changed its selection criteria away from SAT, and Princeton, WashU, and a few others have done the opposite.</p>
<p>if it helps you any, the mean SAT data I have for Stanford circa 1970, per the Cass & Birnbaum college guide of the day, was 651 verbal, 680 math. It was the 6th most selective college, by my weighting, that year.
Princeton was 645, 674.
Wash U was 600, 619.</p>
<p>monydad, 1331 sounds about right to me. When I attended in 1975, my 1310 (1370 corrected) was just below the median. Like most at the time, I studied for about 4 hours in an SAT prep book the week prior, took the test once, and never thought about it again.</p>
<p>Sixth at the time? That seems high in rank to me… Stanford has never given the impression that beyond a certain score, it cares much how high the SAT is.</p>
<p>6th according to the monydad selectivity index, involving a weighting of admit % and SATs. Not necessarily 6th in mean SATs. I am too lazy to look at the spreadsheet now.</p>
<p>^ Oh, OK… I’ll bet the SAT rank was probably somewhere from 10-15.</p>