In this article it is argued that actual selection can for several reasons be different from the selection as it is intended. A method is described to construct a composite variable that may be considered as the one that caused selection as it actually took place. This selector variable, rather than the one on which selection is supposed to be based, should be used in correcting for restriction of range. The application of the method is illustrated in a case study, in which actual selection appears to be remarkably different. Possible reasons for the discrepancies are discussed.
Get full access to this article
View all access options for this article.
References
1.
Bryant, N. D. and Gokhale, S.Correcting correlations for restriction in range due to selection on an unmeasured variable. EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT, 1972, 32, 305-310.
2.
Elshout, J. J.De selektieprocedure op het Instituut voor Sociale en Bedrijfspsychologie aan de Universiteit van Amsterdam (mimeographed. Amsterdam: ISBP, University of Amsterdam, 1965.
3.
Guilford, J. P.Fundamental statistics in psychology and education. New York: McGraw-Hill, 1965.
4.
Gulliksen, H.Theory of mental tests. New York: Wiley, 1950 .
5.
Modderaar, J.Over de validiteit van de selektie voor bedrijven. Nederlands Tijdschrift voor de Psychologie , 1966, 21, 573-589.
6.
Roe, R. A. and Boerkamp, R. G.De selektie voor funkties uit de groep Uitgebreid Algemeen. Een validatiestudie (mimeographed. Amsterdam : ISBP, University of Amsterdam, 1971.
7.
Roe, R. A. and Elshout, J. J.Some new formulas for the correction for restriction of range. Nederlands Tijdschrift voor de Psychologie, 1972, 27, 134-139.
8.
Vernon, P. E. and Parry, J. B.Personnel selection in the British forces. London: University of London Press, 1949.
