Does the admissions test predict grades?
A highly selective graduate psychology programme in Minnesota admits only applicants who score in roughly the top 5% on its admission test. A faculty committee examined the 210 students admitted over five years and found that their admission test scores correlated only r = .08 with their first-year grade point average. Almost all of these students earned grades between B+ and A. The committee concluded that the test tells the programme nothing about who will do well and recommended dropping it from admissions, relying on interviews and personal statements instead.
Which one flaw undermines the conclusion?
This study is fictional, written for practice. No real study, institution or researcher is described.