UM: publishes 2014-2015 Common Data Set

Colleges and Universities A-Z University of Michigan
1

http://obp.umich.edu/wp-content/uploads/pubdata/cds/cds_2014-2015_umaa.pdf

A few “highlights” (as I construe them):

  1. Applicants: 49,776;
  2. The percent offered admission: 32%; (7,895M and 8,152W) = 16,047
  3. Enrolled (3229M and 3246W) or yield of around 40%
  4. 67% have a composite ACT about 30 (roughly the Ivy League 25th percentile);
  5. Interquartile ACT: 29-33; roughly 1 point under the Ivy numbers at each end of the range;
  6. Interquartile SAT: 1920-2210;
  7. Average GPA 3.81;
  8. Freshman retention rate: 97.2%;
  9. Undergraduates: 28,395; graduates: 15,230;
  10. Six year graduation rate 91%
  11. Percent take off of wait list only around 2% of list;
2

It seems the final enrolled number of students is around 200-300 less than the number who had paid the deposit. The original calculation of yield rate based on deposit was 42%.
@blue85 The summation of 25th percentile section scores does not equal the 25th percentile of composite score. The same for the 75% scores. One that has a lower score in one section would need a higher score in other section. If one has only 25% in each of the 3 section scores, his/her composite score would be slightly lower than the 25th percentile for composite score. You can use the admission statistics as an example. The summation of the 3 sections’ 25th percentile score is 2010 while the composite 25th percentile score is 2040.
http://admissions.umich.edu/apply/freshmen-applicants/admitted-student-profile

3

“@blue85 The summation of 25th percentile section scores does not equal the 25th percentile of composite score. The same for the 75% scores.”

I think that what you are saying is that the summation (c.d.f.) under the curve (p.d.f.) is inherently non-linear and the non-linearity precludes direct summation. While true, that principle is frequently violated in quotational practice (from having seen a number of such direct summations where the full distribution/all-quantiles is not available).

4

In 2009, Michigan’s mid 50% ACT was 27-31 and its mid 50% SAT was 1220-1420. Michigan’s private peers had mid 50% ACTs in the 30-34 range and mid 50% SAT in the 1350-1550 range. Today, Michigan’s mid 50% ACT range is 29-33 and its mid 50% SAT range is 1280-1480. Michigan’s private peers have not experienced much change at all. Most are still at 30-34 and 1350-1550. I think the gap will be completely closed in the next 4-5 years, which is what I had predicted would happen several years ago.

5

It has nothing to do with non-linearity. It is just the distribution of scores. If one has score at 25% of each section, he would be below the 25th percentile for composite score. When you are weak in one section, you need to have stronger score in the other section to get in.

6

How does this compare to previous years?