This paper extends the study of the behavior of the coefficient of determination (r) when two data sets are pooled. The manner and extent that the pooled r2 can vary due to the subgroups' relative location is graphically demonstrated by developing equations for "iso-r2" lines. These results are used to correct misleading statements about pooling effects which appeared in a previous article in this journal.
Get full access to this article
View all access options for this article.
References
1.
Adams, A. J. (1984). A cautionary note on the reliability of advertising test-retest scores. Journal of Advertising, 13, 41-45.
2.
Anscombe, F. J. (1973). Graphs in statistical analysis. American Statistician, 27, 17-21.
3.
Belsley, D. A. , Kuh, E., and Welsch, R. E. (1980). Regression diagnostics: identifying influential data and sources of collinearity. New York: Wiley.
4.
Brier, S. S. and Fienberg, S. E. (1980). Recent econometric modelling of crime and punishment: support for the deterrence hypothesis? in S. E. Fienberg and A. J. Reiss, eds., Indicators of crime and criminal justice: Quantitative studies, Washington, D.C., p. 89.
5.
Gatignon, H. and Reibstein, D. J. (1986). Pooling logit models. Journal of Marketing Research, 23, 281-285.
6.
Krasker, W. S. and Welsch, R. E. (1982). Efficient bounded-influence regression estimation. Journal of the American Statistical Association, 77, 595-604.
7.
Mahajan, V. , Sharma, S., and Wind, Y. (1984). Parameter estimation in marketing models in the presence of influential response data: robust regression and applications. Journal of Marketing Research, 21, 268-277.
8.
Sockloff, A. J. (1975). Behavior of the product-moment correlation coefficient when two heterogeneous subgroups are pooled. ED-UCATIONAL AND PSYCHOLOGICAL MEASUREMENT, 35, 267-276.
