st: how to do Bonferroni test after MANOVA

[Lindsey Yang <lindsey.yang@gmail.com>]

I did a MANOVA analysis:

manova y1 y2 y3 y4 y5 y6 y7 y8 y9 = pcp6 age health2 medpay insusf exc
hinfoc enabm pcputil pcpsatisf
xi3: mvreg y1 y2 y3 y4 y5 y6 y7 y8 y9 = i.pcp6 age health2 i.medpay
i.insusf i.exc i.hinfoc i.enabm i.pcputil i.pcpsatisf

y1-y9 are means of 9 subscales of the health care quality measurement
instrument. pcp6 has 5 levels indicating 5 types of health care
facilities. The main purpose here is to compare the adjusted means in
each of 9 subscales across 5 types of health care facilities to see
whether each of the 9 adjusted means differ significantly by health
care facility types. Finally I hope to come up with a table below:

          Facility 1     Facility 2    Facility 3     Facility 4
Facility 5   P-values for group level comparisons
y1      adjusted
          mean (AM)    AM             AM             AM            AM
              ??????
..
y8           -               -                    -                  -
          -                    -----
y9           AM           AM             AM             AM
AM                ??????

Is the method here Bonferroni? Does the Bonferroni here mean
accounting for analysis of 9 dependent variables or for analysis of 5
facilities?

How to do Bonferroni in stata for my case? If involving matrix, what
should the matrix be in my case? The matrix is killing me!

Thanks so much!

Lindsey
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/