Bartlett, C. J., & Green, C. G. (1966). Clinical prediction: Does one sometimes know too much? Journal of Counseling Psychology, 13(3), 267-270.
http://dx.doi.org/10.1037/h0023710
Based on the statistical theory of shrinkage, it was hypothesized that in clinical prediction a similar reduction in predictive efficiency may occur as a result of using too many predictors. Experienced counseling psychologists predicted grade-point average under 2 conditions, 4 and 22 predictors. All predicted better with 4 than with 22 predictors, supporting the hypothesis. The results are discussed in terms of their possible implications for clinical prediction. (PsycINFO Database Record (c) 2016 APA, all rights reserved)