1-way repeated measures Anova and post hoc test

[Laurentseitz]

Hi all,

So I am trying to perform a 1-way repeated measures Anova and a post hoc test on my data. In the data editor, I have:
- variable 1 (labelled as 'id'): the participants id number (1 to 14),
- variable 2 (labelled as 'time'): the time the measurements were made (pre= 1, post 1 min= 2, post 2 min= 3 and post 3 min= 4),
- variable 3 (labelled as 'sbj'): a jump test in centimeters

In the command window I typed 'anova sbj id time, repeated (time)' and it worked.

Now when I want to perform a post hoc test to see where the differences occur I type 'pwcompare time subj time#sbj' BUT it gives me 'variable sbj not found - r(111);'

Does anyone have an idea where the issue may come from?

Thank you all for your precious help

[Joseph Coveney]

Are you sure the error message wasn't "variable subj not found - r(111)"?

It's usually best to show exactly what you typed and what resulted, as in copy-and-paste, or (better) attach a log file. See Section 12 of the forum's FAQ.

[Laurentseitz]

Hello,

Ok so I found my mistake. I can't include the outcome in the command, so it worked with:


pwcompare time, effects

Pairwise comparisons of marginal linear predictions

Margins : asbalanced

------------------------------------------------------------------------------
| Unadjusted Unadjusted
| Contrast Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
time |
2 vs 1 | .0914286 .0234124 3.91 0.000 .0444482 .138409
3 vs 1 | .0864286 .0234124 3.69 0.001 .0394482 .133409
4 vs 1 | .1092857 .0234124 4.67 0.000 .0623053 .1562661
5 vs 1 | .0921429 .0234124 3.94 0.000 .0451625 .1391233
3 vs 2 | -.005 .0234124 -0.21 0.832 -.0519804 .0419804
4 vs 2 | .0178571 .0234124 0.76 0.449 -.0291233 .0648375
5 vs 2 | .0007143 .0234124 0.03 0.976 -.0462661 .0476947
4 vs 3 | .0228571 .0234124 0.98 0.333 -.0241233 .0698375
5 vs 3 | .0057143 .0234124 0.24 0.808 -.0412661 .0526947
5 vs 4 | -.0171428 .0234124 -0.73 0.467 -.0641232 .0298376


However, do you know how to get the exact p values?

[Joseph Coveney]

What do you mean by exact p-values? Do you want something like this?

pwcompare time, pveffects mcompare(bonferroni)

[Laurentseitz]

Yes that is exactly what I want!!! Thank you so much.