Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
In this paper college admissions are based on test scores and students can exert two types of effort: real learning and exam preparation. The former improves skills but the latter is more effective in raising test scores. In this setting the students with the lowest skills are no longer the ones with the lowest aptitude, but instead are the ones closest to the borderline for college admission. Increased access to college leads to greater income inequality between college graduates and non-graduates. Overall, the ability to study for the test leads to higher expected test scores but lower skills.