Hi All. This is my first post 
I want to check for the normality of my data using a Shapiro-Wilk test. It worked fine for all the other variables I have looked at, but suddenly it says "sample size must be between 3 and 5000". I am definitely in this range (n=21).
I have a column of data with 2 land management strategies (text), and a column containing a protein value (numbers). I want to see whether each management type's protein values are normally distributed. Below is the code. I would really appreciate your help with this!
Thanks
Daniel
#csv file name "Mean_Protein
attach(Mean_Protein)
RP.vnp <- Mean_Protein$Management == "VNP" #first management system
RP.konv <- Mean_Protein$Management == "KONV" #second management system
#for the konv management system:
shapiro.test(Mean_Protein$Protein_rel_TM_g_kg[RP.konv])
attach(Mean_Protein)
RP.vnp <- Mean_Protein$Management == "VNP"
RP.konv <- Mean_Protein$Management == "KONV"
shapiro.test(Mean_Protein$Protein_rel_TM_g_kg[RP.konv])