Mixed-Measure ANOVAs in Stata
While it is possible to run mixed-measures ANOVAs using the ANOVA command it is often a lot simpler to use the wsanova
command. To learn how to use the wsanova
command, click here. If your design involves continuous measures or more than one repeated measure, you will have to use anova
. When dealing with complicated designs, the anova
gets correspondingly complicated. The main concept is using /
to make sure Stata tests each term with the proper error term. In Stata syntax, whatever follows a /
is considered to be the error term for whatever precedes the /
.
Past this basic concept, no explicit specific instructions exist because it is statistical theory, not software knowledge that will determine what the command looks like. That said, Stata offes a very nice internet helpfile with examples of relatively simple designs, which might be a good starting point. Click here to read Stata's help on repeated measure ANOVAs, including complex mixed-measure designs.
Back to ANOVA