The effects of recruitment are often overlooked in studies of validity. This paper demonstrates the effects of recruitment on corrections for restriction in range in the univariate case. If Company A were to recruit very specifically for a particular position and Company B recruited for the same position very generally, the unrestricted variances in the applicant groups would be considerably different. This difference would result in a larger corrected correlation for Company B. This difference is demonstrated in the paper for different selection ratios and across 100 values of the unrestricted correlation.
Get full access to this article
View all access options for this article.
References
1.
BrewerJ. K.HillsJ. R.Univariate selection: The effects of size of correlation, degree of skew, and degree of restriction. Psychometrika, 1969, 34, 347–361.
2.
GulliksenH.Theory of mental tests. New York: Wiley, 1950.
3.
LandyF. J.Adventures in implied psychology: On the value of true negatives. American Psychologist, 1978, 33, 756–760.
4.
LordF. N.NovickM. R.Statistical theories of mental test scores. Reading, Mass.: Addison-Wesley, 1968.
5.
RydbergS.Bias in prediction: On correction methods. Stockholm: Almquist & Wiksells, 1963.
6.
ThorndikeR. L.Personnel selection. New York: Wiley, 1949.
7.
Uniform Guidelines on Employee Selection Procedures. Federal Register, 1978, 43 (August 25), 38290–38315.
