If a battery of tests is administered to the same individual, the question is how to combine the various resultant scores so as to obtain a composite score for prediction purposes. Even if preas-signed weights are used, they may not be the true weights, especially, when the test scores are intercorrelated. In the following the author briefly reviews the existing methods and proposes some new ones for obtaining weights that are best in some sense. Further, he illustrates the methods by using the data on certain city government employees.
Get full access to this article
View all access options for this article.
References
1.
Ary, D. , Jacobs, L. C., and Razavieh, A. (1985). Introduction to research in education (3rd ed.). New York: Holt, Rinehart and Winston.
2.
Downie, N. M. and Heath, R. W. (1983). Basic statistical methods. New York: Harper and Row.
3.
Johnson, R. A. and Wichern, D. W. (1982). Applied multi-variate statistical analysis. Englewood Cliffs, NJ: Prentice Hall, Chapter 8, 361-382.
4.
Lindquist, E. F. (1942). Afirst course in statistics (Revised Edition). Boston: Houghton Mifflin.
5.
McDonald, R. P. (1968). A unified treatment of the weighting problem. Psychometrika, 33, 351-381.
6.
Roscoe, J. T. (1975). Fundamental research statistics for the behavioral sciences (2nd ed.). New York: Holt, Rinehart and Winston.
7.
Sax, G. (1980). Principles of educational and psychological measurement and evaluation (2nd ed.). Belmont, CA: Wadsworth.
8.
Terwilliger, J. S. and Anderson, D. H. (1969). An empirical study of the effects of standardizing scores in the formation of linear composites. Journal of Educational Measurements, 6, 145-154.
9.
Wang, M. W. and Stanley, J. C. (1970). Differential weighting: A review of methods and empirical studies. Review of Educational Research, 40, 663-705.
10.
Wilks, S. S. (1938). Weighting systems for linear functions of correlated variables when there is no dependent variable. Psychometrika, 3, 23-40.
