Hello.
I am testing for normality on a number of repeated measures variables in a sample of 29 people.
Both the Kolmogorov-Smirnov test and the Shapiro-Wilk test statistics say that all of these variables are significantly non-normal. However, observations of skew and kurtosis (as well as normality plots on histograms) suggests that the data are normally distributed.
In the end, I am looking to do repeated measures ANOVAs if this helps with any advice.
Anyway, I am wondering why the apparent contradiction? And what may be the solution?
I am very grateful for any help. I did find similar forums but they always dealt with very large samples hence why I am posting a new thread.
Thanks,
capn_kangaroo
I am testing for normality on a number of repeated measures variables in a sample of 29 people.
Both the Kolmogorov-Smirnov test and the Shapiro-Wilk test statistics say that all of these variables are significantly non-normal. However, observations of skew and kurtosis (as well as normality plots on histograms) suggests that the data are normally distributed.
In the end, I am looking to do repeated measures ANOVAs if this helps with any advice.
Anyway, I am wondering why the apparent contradiction? And what may be the solution?
I am very grateful for any help. I did find similar forums but they always dealt with very large samples hence why I am posting a new thread.
Thanks,
capn_kangaroo