Answers are grounded in this paper and cite their sources.
A FORTRAN program is presented which calculates power curves and determines sample sizes for a balanced ANOVA for either a one-way layout or a randomized complete block design. The user specifies the number of treatment groups, the Type I Error probability, and either an alternative hypothesis in a specific form and a desired power or a sample size. If an alternative hypothesis and desired power is specified by the user, the program determines the required sample size. If the user specifies a particular sample size, a power curve is presented in either tabulated or plotted form.
No takes yet. Share an insight, caveat, or question.
Thomas P. Turiel (1989) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: