I have database which has data about 72 plants, that got 6 different treatments. I have measured the weight of those plants every day and every hour for the past two weeks.so basically there is column for:
treatment that was given
I would like to see for every day if there is a significant difference between the different treatment groups based on the measured weight.
My data is not normally distributed and repeated so I thought to use Friedman test. I have grouped all the plants according to the treatment they got by calculate the mean value of all the hours each group got, so this is how my table loooks like after the groupby (this is not all the days):
I want to check for each day if there is a significant difference between the different groups. I thought that the best way to check it is by using Friedman test (my data is not normal distributed) but I am not sure I can because in all the examples there is no different between the observations (or the rows) but here, we have given different treatments to the observations and that might affect the results.
so here, not only the days are different from each other, but also the treatment that were given.
My question is, is it still valid to use the Friedman test in this case to see if there is significant different between the groups every day? and if not, do you know any test that can fit this situation?
(each empty cell represent the mean value of all the plants that got the same treatment on that date)