Two-way repeated-measures ANOVA: two within-subject factors

Eight subjects each tested at two intensities (Low, High) and three times (T1, T2, T3) — six measurements per subject. Both factors are within-subject: every subject appears in every cell. MinutePlot tests each effect against its own error term (A against A×Subject, B against B×Subject, A×B against A×B×Subject), the standard univariate formulation.

The data

One row per subject, the A×B combinations across the columns:

Two-within layout: subjects down the rows; factor A as column blocks, factor B within each block. Every cell holds exactly one value per subject.

In MinutePlot

Choose Two-way repeated-measures ANOVA, paste the table, run. Mauchly’s test is reported for every effect with more than two levels; when sphericity is rejected the Greenhouse–Geisser correction is applied and both p-values are shown.

The result

On the validation dataset behind this guide: intensity F(1, 7) = 371.9, p < 0.001; time F(2, 14) = 94.5, p < 0.001 (ε = 0.76, corrected p still < 0.001); the interaction just short of significance (F(2, 14) = 3.60, p = 0.055).

The figure MinutePlot produces: intensity along the axis, one bar per time, error bars, and the compound letters (capitals across intensity, lower-case across time).

Styling and exporting this figure: see Figure styling and Exporting & downloading.

Report it like this

“Intensity (F(1, 7) = 371.9, p < 0.001) and time (F(2, 14) = 94.5, Greenhouse–Geisser p < 0.001) affected the response; their interaction did not reach significance (F(2, 14) = 3.60, p = 0.055).”