I just read the help for -anovacontrast- which says,
"Note: anovacontrast gives incorrect and misleading results when
there are missing or empty cells."
I guess that explains it.
-Dave
On Oct 25, 2008, at 10:12 AM, David Airey wrote:
Phil Ender's -anovacontrast- commands gets the same contrasts
_except_ for the desired one:
anova nnn round size
Number of obs = 22 R-squared
= 0.7465
Root MSE = .082094 Adj R-squared
= 0.6673
Source | Partial SS df MS F
Prob > F
-----------
+----------------------------------------------------
Model | .317523495 5 .063504699
9.42 0.0002
|
round | .15988107 4 .039970267
5.93 0.0040
size | .15876267 1 .15876267
23.56 0.0002
|
Residual | .107829606 16 .00673935
-----------
+----------------------------------------------------
Total | .425353101 21 .02025491
anovacontrast round, values(-1 1 0 0 0)
Contrast variable: round (-1 1 0 0 0) Dep Var: nnn
source SS df MS Contrast = -0.01
---------+--------------------------------- N of obs = 22
contrast | .000060166 1 0.0001 F = 0.01
error | .107829606 16 0.0067 Prob > F = 0.9259
---------+--------------------------------- t = 0.09
anovacontrast round, values(-1 0 1 0 0)
Contrast variable: round (-1 0 1 0 0) Dep Var: nnn
source SS df MS Contrast = 0.49
---------+--------------------------------- N of obs = 22
contrast | .345744034 1 0.3457 F = 51.30
error | .107829606 16 0.0067 Prob > F = 0.0000
---------+--------------------------------- t = 7.16
anovacontrast round, values(-1 0 0 1 0)
Contrast variable: round (-1 0 0 1 0) Dep Var: nnn
source SS df MS Contrast = 0.04
---------+--------------------------------- N of obs = 22
contrast | .004915206 1 0.0049 F = 0.73
error | .107829606 16 0.0067 Prob > F = 0.4057
---------+--------------------------------- t = 0.85
anovacontrast round, values(-1 0 0 0 1)
Contrast variable: round (-1 0 0 0 1) Dep Var: nnn
source SS df MS Contrast = 0.18
---------+--------------------------------- N of obs = 22
contrast | .087318113 1 0.0873 F = 12.96
error | .107829606 16 0.0067 Prob > F = 0.0024
---------+--------------------------------- t = 3.60
*
* 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/
*
* 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/
