Table 3.

Computed sample size, estimated power, and simulated power for the ANOVA, approximate ANCOVA, and exact ANCOVA methods when $G=3$, ${\mathrm{\sigma }}_Y^2=10000$, $\mathrm{\rho }=0.9$, ${\mathrm{\sigma }}^2=1900$, $\{{\mathrm{\mu }}_1,{\mathrm{\mu }}_2,{\mathrm{\mu }}_3\}=\{400,450,500\}$, ${\mathrm{\delta }}^2=0.8772$, Type I error $\mathrm{\alpha }=0.05$, and nominal power $1-\mathrm{\beta }=0.80$

P	ANOVA				Approximate ANCOVA				Exact ANCOVA
	$N_{\mathit{TO}}$	Estimated power	Simulated power	Error	$N_{\mathit{TA}}$	Estimated power	Simulated power	Error	$N_{\mathit{TE}}$	Estimated power	Simulated power	Error
1	63	0.8148	1.0000	$-$ 0.1852	15	0.8108	0.7696	0.0412	18	0.8751	0.8812	$-$ 0.0061
2	63	0.8148	1.0000	$-$ 0.1852	18	0.8934	0.8509	0.0425	18	0.8409	0.8485	$-$ 0.0076
3	63	0.8148	1.0000	$-$ 0.1852	18	0.8867	0.8026	0.0841	18	0.8007	0.8086	$-$ 0.0079
4	63	0.8148	1.0000	$-$ 0.1852	18	0.8784	0.7702	0.1082	21	0.8614	0.8628	$-$ 0.0014
5	63	0.8148	1.0000	$-$ 0.1852	18	0.8681	0.7050	0.1631	21	0.8272	0.8368	$-$ 0.0096
6	63	0.8148	1.0000	$-$ 0.1852	18	0.8548	0.6520	0.2028	24	0.8807	0.8873	$-$ 0.0066
7	63	0.8148	1.0000	$-$ 0.1852	18	0.8373	0.5948	0.2425	24	0.8516	0.8578	$-$ 0.0062
8	63	0.8148	1.0000	$-$ 0.1852	18	0.8136	0.5257	0.2879	24	0.8173	0.8263	$-$ 0.0090
9	63	0.8148	1.0000	$-$ 0.1852	21	0.9048	0.6459	0.2589	27	0.8735	0.8786	$-$ 0.0051
10	63	0.8148	1.0000	$-$ 0.1852	21	0.8904	0.5855	0.3049	27	0.8442	0.8505	$-$ 0.0063