Premium
Sample Selection in Models of Academic Performance
Author(s) -
Cushing Matthew J.,
McGarvey Mary G.
Publication year - 2004
Publication title -
economic inquiry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.823
H-Index - 72
eISSN - 1465-7295
pISSN - 0095-2583
DOI - 10.1093/ei/cbh063
Subject(s) - test (biology) , psychology , sample (material) , correlation , population , affect (linguistics) , entrance exam , selection (genetic algorithm) , econometrics , mathematics education , statistics , actuarial science , economics , predictive validity , clinical psychology , demography , mathematics , computer science , sociology , machine learning , paleontology , chemistry , geometry , biology , communication , chromatography
This article shows how admission and enrollment processes affect the interpretation of simple validation studies of academic performance. In a competitive market for students, optimal behavior of admissions committees and applicants drives the simple correlation between test scores and performance toward zero, regardless of the relationship in the population of prospective students. Data from our university's MBA program support the prediction that applicants exhibit a higher correlation between test scores and undergraduate GPAs than do current students. This suggests that standard validation studies will understate the importance of GMAT scores in predicting performance of potential MBA students. (JEL C24 , I121 , C52 )