Description power twoway computes sample size, power, or effect size for two-way analysis of variance (ANOVA). By default, it computes sample size for given power and effect size. Alternatively, it can compute power for given sample size and effect size or compute effect size for given sample size, power, and number of cells. You can choose between testing for main row or column effect or their interaction. Also see [PSS] power for a general introduction to the power command using hypothesis

more »

... ts. Quick start Sample size for the main effect of the row factor for a 2 × 3 design specified using cell means with a within-cell variance of 27 and default power of 0.8 and significance level α = 0.05 power twoway 19 18 32 \ 23 25 26, varerror (27) Same as above, but specify cell means in the matrix cm matrix cm = (19, 18, 32 \ 23, 25, 26) power twoway cm, varerror (27) Same as above power twoway cm, varerror(27) factor(row) As above, but calculate sample size for the main effect of the column factor power twoway cm, varerror(27) factor(column) As above, but calculate sample size for the interaction of the row and column factors power twoway cm, varerror (27) factor(rowcol) As above, but for within-cell variances of 20, 25, 30, and 35 power twoway cm, varerror(20(5)35) factor(rowcol) Sample size for the row factor with power of 0.85 and α = 0.01 power twoway cm, varerror(27) power(.85) alpha(.01) Specify that the groups in the second row have twice the sample size as those in the first row power twoway cm, varerror(27) cellweights(1 1 1\2 2 2) showcellsizes Sample size when variance of the main effect of the row factor equals 1.2 in a 2 × 3 design power twoway, varerror(27) factor(row) vareffect(1.2) /// nrows(2) ncols(3) Power for the main effect of the row factor with a sample size of 180 for cell means stored in matrix cm power twoway cm, varerror(27) factor(row) n(180) 1 2 power twoway -Power analysis for two-way analysis of variance Same as above, but specify that the sample size per cell is 40 power twoway cm, varerror(27) factor(row) npercell (40) As above, but specify cell sample sizes of 40, 45, 50, and 55 power twoway cm, varerror(27) factor(row) npercell(40(5)55) As above, but display results as a graph of power versus sample size power twoway cm, varerror(27) factor(row) npercell(40 (5)55) graph Effect size and target between-group variance for three groups, sample size of 150, and power of 0.8 power twoway, varerror (27) nrows (2) ncols(3) n(150) power(.8) Menu Statistics > Power and sample size Syntax Compute sample size power twoway meanspec , power(numlist) options Compute power power twoway meanspec, n(numlist) options Compute effect size and target effect variance power twoway, n(numlist) power(numlist) nrows(#) ncols(#) options where meanspec is either a matrix matname containing cell means or individual cell means in a matrix form