In April 2014, Tim Groseclose published his book "Cheating An Insider's Report on the Use of Race in Admissions at UCLA". Admissions to elite universities is highly valued. Who should be admitted and what criteria should be used for judging files? To his credit, Tim provides all of the admissions data he was able to access through a California Public Records Act request. You can download the data here. For those applied econometrics teachers consider having your students extract a random subset of these records and estimating some "production functions" in which the key outcome variable is whether a UCLA applicant is admitted.
Your students can test for the role of the student's;
2. household income
3. standardized tests
4. grades in high school
as inputs in determining the probability of being admitted.
Using a linear probability model, the students can estimate multivariate regressions and the regression coefficients will have the interpretation of index weights and students will learn a good lesson in "isoquants". Such an analysis represents a revealed preference test of the UCLA bureaucracy's formula for being admitted to this prestigious institution.
This is clearly a highly controversial subject but Tim has performed an important service by bringing this subject into "the light". Admissions officers have great power and discretion. Big data methods allow us to take a close look at what they actually do rather than on relying on what they say they do. Such cross-checks on power create good incentives and guarantee the continued strength of our key institutions. We all need to face competition and cross-validation (by arms-length evaluators) that we are doing our jobs. I believe this statement holds for professors, administrators, deans and admissions officers.