Hello! I appreciate your help in advance!
I am currently trying to figure out the proper way to analyze my data!
I had used the DOE feature of JMP to create this fraction factorial of 24 runs for m experiment. The goal is two screen which categories have the most effect on the success rate. I had initially use the response screening option to analyze my results, when I did that none of my values were within the threshold so I decided to use the robust option. however there doesn't seem to be any change in any of the data analysis comparing robust vs non robust.
I have seen some other posts where robust is typically used in the case of outliers. Can you assume that no change means that there are no outliers and that there is no need to bootstrap your data to make sure the original result is correct? I also tried using jmp to find any outliers However, the define outlier function will not work for me. I tried recoding my data from Pass/ Fails to 1/0 and it still didn't work. Is this due to the factors being categorical?
