I am trying to find a solution for the next problem: I was trying to test my data using a factorial ANOVA but I just found that the data is not normally distributed then I cannot use an ANOVA anymore; this was my only way to test my hypotheses and test the size effect of a between subjects 2X2X3 experiment on my dependent variable. Do you have any idea of what type of test can I use that Substitute the factorial ANOVA for nonparametric data? Thanks for any contribution.
Thanks for your answers. What I did now is a transformation to apply the Kruskal Wallis test but I am not sure how to proceed because as you said is a one-way ANOVA and I have a factorial design... Other way is to make a normal transformation (log) but the Welch test is still significant.
Each variable represents each axis. You can only use non-parametric methods if you don't know the distribution. The variance represents the degree of deviation of the coordinate axis. Just make sure that the fairness between the variables is ok, and it doesn't matter if it's a factorial variance.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA