### Monday, February 27, 2006

## Re: st: repeated anova

Ricardo Ovaldia wrote:

I am struggling with this seemly simple ANOVA, but I can't get it set correctly. I have 40 students randomly allocated to 4 experimental stress groups. The diastolic blood pressure was measured for each student at baseline and at five additional times. The lay out is something like this:

[redacted]

Note that there are 10 students in each -group-, but each student was measured at 6 different times. Could someone please tell me what is the correct Stata command for this repeated ANOVA?

--------------------------------------------------------------------------------

Using the baseline blood-pressure value as a covariate in repeated-measures ANCOVA might be a better option. I believe that the ANCOVA set-up that's shown in Winer is the first model fit procedure in the illustration below. See www.stata.com/statalist/archive/2004-01/msg00032.html for the references.

Joseph Coveney

clear set more off set seed `=date("2006-02-28", "ymd")' set obs 6 forvalues i = 1/6 { generate float a`i' = 0.7 + 0.3 * (`i' == _n) } mkmat a*, matrix(A) drawnorm bp0 bp1 bp2 bp3 bp4 bp5, corr(A) n(40) clear generate byte student = _n generate byte group = mod(_n, 4) + 1 * Begin here rename bp0 baseline reshape long bp, i(student) j(time) anova bp baseline group / student|group /// time group*time, continuous(baseline) repeated(time) exit

* * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

Tag: statalist