Researchers are often confronted with missing data for some subjects and variables. The usual method of handling this problem deletes subjects with missing data and thereby reduces the sample size. An alternative approach suggested by Cohen (1968) investigates the randomness of the missing data with respect to other variables and permits the researcher to use the information obtained to further understand the relationships among variables. This note compares three methods of treating missing data with respect to the results obtained, the informational value of each method, the population to which the results can be referred, and statistical power.
Get full access to this article
View all access options for this article.
References
1.
Cohen, J.Some statistical issues in psychological research. In B. B. Woleman (Ed.), Handbook of clinical psychology. New York: McGraw-Hill, 1965, 95-121.
2.
Cohen, J.Multiple regression as a general data-analytic system. Psychological Bulletin1968, 70, 426-443.
3.
Cohen, J., & Cohen, P.Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, N.J.: Lawrence Erlbaum Assoc., 1975.
4.
Timm, N.H.The estimation of variance-covariance and correlation matrices from incomplete data. Psychometrika1970, 35, 417-438.
5.
Winer, B.J. Statistical principles in experimental design (2nd ed.). New York: McGraw-Hill, 1971.
6.
Wolf, G., & Cartwright, B.Rules for coding dummy variables in multiple regression. Psychological Bulletin1974, 81, 173-179.
