Conditional Correlation Phenomena with Applications to University Admission Strategies

This paper considers the situation (as in college admissions) where one is given two attributes,X and Y, which one uses to predict a third attribute, Z, by some function Ẑ ofX andY. However, one only retains values ofX, Y, andZ for which Ẑ is large. A thorough discussion, under fairly general conditions on the distributions, is given of how the correlation coefficients of X, Y, andZ are affected by this restriction of the range of values. In the case of the normal distribution, where linear prediction is optimal, the role of suppressor variables is discussed.