Under the assumption of normality, a formula is derived for criterion-related validity of a partial composite score based on the highest (or lowest) of several equivalent observations. This partial composite score has higher validity than each single observation. However, it has less validity than the composite score based on all observations.
Get full access to this article
View all access options for this article.
References
1.
Cronbach, L. J. (1970). Essential of psychological testing. (3rd ed.). New York: Harper and Row.
2.
Curnow, R. N. and Dunnett, C. W. (1962). The numerical evaluation of certain multivariate normal integrals. Annals of mathematical statistics, 33, 571-579.
3.
David, H. A. (1981). Order statistics (2nd ed.). New York: Wiley.
4.
Dunnett, C. W. and Sobel, M. (1955). Approximation to the probability integral and certain percentage points of a multivariate analog of Student's t-distribution. Biometrika, 42, 258-260.
5.
Feldt, L. S. (1965). The approximate sampling distribution of the Kuder-Richardson reliability coefficient twenty. Psychometrika, 30, 357-370.
6.
Huynh, H. (1986a). On the reliability of the extreme score. Psychometrika, 1986, 51, 475-478.
7.
Huynh, H. (1986b). Reliability of composite measurements based on the m highest of n equivalent components. Journal of educational statistics, 11, 225-238.
8.
Lord, F. M. and Novick, M. R. (1968). Statistical theories of mental test scores. Reading, MA: Addison-Wesley.
