Precise p values oneway scheffe

[Senor Massao]

Is there a way to obtain precise p values after one way scheffe test? I’m only getting 0.0000 which is of course enough to write p<0.001 but I need to report precise p values.

oneway continuous categorical, scheffe

[Felix Bittmann]

Apparently, return list does not even provide a p-value so you need do compute it manually.

oneway ...
di %20.18f chi2tail(`r(df_bart)', `r(chi2bart)')

[Senor Massao]

oneway ...
di %20.18f chi2tail(`r(df_bart)', `r(chi2bart)')

[/QUOTE]

Thanks for feedback. I think the code you gave is displaying p value for bartlett’s test for equal variances and not F test (I’m using binary dummy variable for each contrast)

[Fei Wang]

oneway ...
di %20.18f chi2tail(`r(df_bart)', `r(chi2bart)')

Thanks for feedback. I think the code you gave is displaying p value for bartlett’s test for equal variances and not F test (I’m using binary dummy variable for each contrast)

How about this:

di %20.18f Ftail(r(df_m), r(df_r), r(F))

[Felix Bittmann]

Sorry, I picked the wrong statistic to report. As you see, the principle is the same for the actual p-value of the test. The only difference is that you need to use the suitable distribution function of the statistic, as Fei Wang did here (Ftail instead of chi2tail).

[Senor Massao]

di %20.18f Ftail(r(df_m), r(df_r), r(F))

[/QUOTE]

Many thanks, yes, that works :-)