I am trying to fit a 3 way ANOVA using R. Then i realized the sums of squares of the output has changed when the order of the variables in the model are changed.
Can anyone figure out the reason for that ?
There seems to be something wrong because in normal situation the sums of squares in R not affected by the order of variables in the model
Asked
Active
Viewed 80 times
0
Sam88
- 328
- 3
- 10
-
See [this question](https://stats.stackexchange.com/q/13241/28500). – EdM Mar 28 '18 at 01:47
-
Different "type" sums of squares--see https://stats.stackexchange.com/questions/20452/how-to-interpret-type-i-type-ii-and-type-iii-anova-and-manova – Richard Border Mar 28 '18 at 01:52
-
replace `aov` with `lm` and the order will not matter – Richard Border Mar 28 '18 at 01:52