The statistical assumption of sphericity comes into play when conducting repeated-measures ANOVA. In essence, sphericity accounts for the random variation and error associated with measurement in inferential statistics. Multiple observations of an outcome can artifically inflate or deflate the underlying covariance matrices associated with the algebra used when calculating the degrees of freedom and F-value for repeated-measures ANOVA. The assumption of sphericity is often not met when conducting repeated-measures ANOVA analyses, strictly due to the aforementioned variation and error that exists when collecting more than two observations of a continuous outcome.
The Greenhouse-Geisser correction is used to correct for sphericity. Click on the button below to learn more about interpreting the Greenhouse-Geisser correction and sphericity in SPSS.
The Greenhouse-Geisser correction is used to correct for sphericity. Click on the button below to learn more about interpreting the Greenhouse-Geisser correction and sphericity in SPSS.