ERIC Number: ED412247
Record Type: Non-Journal
Publication Date: 1994-Apr
A New Sample Size Formula for Regression.
Brooks, Gordon P.; Barcikowski, Robert S.
The focus of this research was to determine the efficacy of a new method of selecting sample sizes for multiple linear regression. A Monte Carlo simulation was used to study both empirical predictive power rates and empirical statistical power rates of the new method and seven other methods: those of C. N. Park and A. L. Dudycha (1974); J. Cohen (1988); C. Gatsonis and A. R. Sampson (1989); S. B. Green (1991); E. J. Pedhazur and L. P. Schmelkin (1991); and J. Stevens (1992). The power rates of the new method were found to be superior, both relatively and absolutely, to other methods across most conditions examined. The results also demonstrate both the importance of using an effect size for determining regression sample sizes and the relative importance of predictive power over statistical power for regression. The new method of sample size selection developed in this paper provides a relatively simple means to account for both concerns. (Contains 9 tables and 87 references.) (Author/SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A