thank you so much for the guideline, im having trouble with this now and find this very helpful for me. However, I have a problem with data normality, which is pretty low than the p value. What can I do in this case?
Hi! thanks for the tutorial. I have one question tho: isn't the normality assumption of linear models supposed to be checked on the residuals of the model? i.e. on the pooled deviation between the values as predicted by the model and the actual data?
Thank you for your guide. However I have some issues: When I run the anova_test I get this error: Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) : 0 (non-NA) cases What should I do? (obviously there are not NA cases in these variables)
@@nguyenthithuycntynguyenthi8113 Hello there fellow companion, have you found out how to asses such problem? In the case you have, do you mind to share the solution you optted for? Thanks anyways.
Hi all, I also have encountered the same problem when adding a second factor. The test worked when just looking at changes among time, but not a second, or in my case, third factor either. If any of you find out how to do this and can share, that would be greatly appreciated!
Hello, what is the alternative to using shapiro-wilk when you have a large dataset (>5000 datapoints) ? Also, what do you do if you don't have normality ?
