I am trying to conduct a three-way robust ANOVA in R using the WRS2 package as:
t3way(Stem ~ Species*Temperature*Salinity, data = tutto)
and there is an error saying : Error in if (chkit$teststat > chkit$crit) break : missing value where is required TRUE/FALSE.
If for example I try changing it to t2way it functions but as a 2-way ANOVA... So I tried modifying the dataset to use only the t2way thinking that maybe there is some glitch with t3way as:
t2way(Stem ~ Temperature*Salinity, data = Specie)
And it gives me the same type of error as above, but it functions as t1way... Has anyone had a similar problem? Does anyone know a possible solution?
Thank you in advance
Thank you for answering Jos Feys, indeed I have found the discussion you pointed out. But there, it seems that the problem was in using the WRS package that was later resolved by using the WRS2 package that I was also using. In addition, if I understood correctly, one of the problems could have also been that the dataset used in that example was a wide type, while mine is a long type... And it is also strange that t2way works with 3 factors while t3way won't.
I will give a look at the other script you proposed, thank you very much for that.
Actually that is the main problem, I do have many outliers that are making it impossible to achive the normality and homogeneity needed for a normal ANOVA... Therefore I was looking for something more robust in order to be able to keep all the data.
Then you should use the WSR2. But, maybe you could test if you can use the parametric version. The important test is whther the residuals are normally distributed.
see:
https://www.datanovia.com/en/lessons/anova-in-r/
You can identify the outliers. Then:
Outliers can be due to:
1) data entry errors,
2) measurement errors or
3) unusual values.
Yo can include the outlier in the analysis anyway if you do not believe the result will be substantially affected. This can be evaluated by comparing the result of the ANOVA test with and without the outlier.
It’s also possible to keep the outliers in the data and perform robust ANOVA test using the WRS2 package.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA