Estimates of population, or unrestricted, variance are generally required to correct correlations for restriction in range. Several articles have been concerned with the advantages of being able to correct correlations for variables that are restricted in range but for which the population variance is not known. Methods previously suggested use a tabled function, as per Cohen (1959), to estimate the population variance. This article presents an empirically derived formula that permits an accurate determination of corrections without the use of a table.
Get full access to this article
View all access options for this article.
References
1.
Alexander, R. A. , Alliger, G. M., and Hanges, P. J. (1984). Correcting for range restriction when the population variance is unknown. Applied Psychological Measurement, 8, 431-437.
2.
Alexander, R. A. , Hanges, P. J., Kollar, L., and Alliger, G. M. (1986). Estimating parameters under range restriction when the unrestricted population variances are unknown. Manuscript submitted for publication.
3.
Cohen, A. C. (1959). Simplified estimators for the normal distribution when samples are singly censored or truncated. Technometrics, 1, 217-237.
4.
Thorndike, R. L. (1949). Personnel selection: Test and measurement techniques. New York: Wiley.
