I would greatly appreciate it if some-one would point me in the right direction.
I have to critique the statistical methods used in a few academic journals. I am aware that it is important to test the assumptions of parametric data... normally distribution etc. However when reviewing the journals do I just assume that these assumptions have been tested, should the authors state that they have been tested or is there a way that I can check them myself from the data in the journal.
Sometimes they mention it, but that is most of the time not really the case. Or they say something like, we assume that the data is normally distributed, because our sample size is very large. If the data is normally distributed, etc then there should be a parametric test. However, sometimes you see in the tables something like (Mann-Whitney test P-value or KW test). Then you know, even that they not really mention it, that they did not meet the assumptions and therefore had to execute a non-parametric test.
Reply With Quote
