Post Hoc Tests - Agricultural Statistical Data Analysis Using Stata - page 155

Analysis of Variance
Source SS df MS F Prob > F
---------------------------------------------------------------
Between groups 1.22886109 4 .307215272 185.11 0.0000
Within groups .082981547 50 .001659631
-------------------------------------------------------------
Total 1.31184263 54 .024293382

Bartlett's test for equal variances: chi2 (4) = 6.7369
Prob>chi2 = 0.150

Comparison of ELISA Absorbance value
by 1-5: Different virus inoculated watermelon germplasm
(Bonferroni)
Row Mean-|
Col Mean | PI 025 PI 026 PI 261-1 PI 528
---------+--------------------------------------------
PI 026 | -0.014
| 1.000
|
PI 261-1 | -0.245 -0.231
| 0.000 0.000
|
PI 528 | -0.283 -0.269 -0.038
| 0.000 0.000 0.340
|
Egun | -0.372 -0.359 -0.127 -0.090
| 0.000 0.000 0.000 0.000

he oneway command offers three different multiple compari-
son tests as options. This includes Bonferroni, Scheffé, and Šidák. All
three are presented in the Results window as a triangular matrix of
the differences between the treatment means and a probability that
these differences are significant. All three use different approaches to
control the familywise Type I error rate. The oneway command is
limited to a single factor model, which is not applicable except for a
CRD. In agriculture, other designs, particularly RCBD (randomized
complete block design) for field experiments, are much more common.
For example, the Bonferroni adjustment divides the selected
probability by the number of comparisons and declares only those
significant at this new probability level. So, for example, with the
absorbance data above, there are 10 possible comparisons; therefore,
the Bonferroni adjustment at the 5% level of significance would be

005
10

.

= 0 005 .