Some of my variables are skewed but most of them are parametric. In this case, do I have to log-transform all of the variables?

I would like to use ordinary one-way ANOVA. I have 4 groups with 4 samples and more than 10 variables. Most variables show nonskewed distribution based on the QQ plot but some do not fit well. I...

02 February 2020 1,459 3 View