I have a data set from an experiment where one of my plots was lost because the plants I was growing were eaten by goats, which resulted in unrepresentative data. I had 4 replicates for the experiment
I usually deal with this by hiding and excluding the questionable data before I run an analysis (multifactor ANOVA with subsequent planned contrasts for linear trends). When I do the analysis, JMP generates a value for the missing data points, and I get good results. But if I generate a summary table, so as to make a bar graph of plant yield, JMP gives me only the mean of the 3 replicates. As a consequence, what I get visually does not at all match the results of the analysis I am running.
The question is this: Is there a way to get JMP to return a mean in the summary table that uses the value calculated for the excluded data point (N=4 rather than N=3)?