I am trying to do a split plot design which has 3 hard to change factors (2,3,3 levels) and 1 easy to change factor (3 levels). Also there is going to be 4 replications per whole plot.

I have tried different software such as Minitab and I was so closed to get it done but since one of my HTC has 2 level and the others have 3 levels, I can't use Minitab since it only does a 2-level split plot design.

I think JMP might be the solution for me. I have seen the examples provided by JMP website such as Vinyl data.jmp and the other one. The issue is that they dont have a factor such as replication.

My question is should I type the number of whole plots as 2*3*3=18 and number of runs as 18*3(sub plot) *4 (rep) or whole plot as 2*3*3*4=72 and number of runs as 72*3=216 or maybe any other solution. Do I need to add a column representing replications? I know I have to add one representing the whole plot in my data.

Any help is highly appreciated.

I have tried different software such as Minitab and I was so closed to get it done but since one of my HTC has 2 level and the others have 3 levels, I can't use Minitab since it only does a 2-level split plot design.

I think JMP might be the solution for me. I have seen the examples provided by JMP website such as Vinyl data.jmp and the other one. The issue is that they dont have a factor such as replication.

My question is should I type the number of whole plots as 2*3*3=18 and number of runs as 18*3(sub plot) *4 (rep) or whole plot as 2*3*3*4=72 and number of runs as 72*3=216 or maybe any other solution. Do I need to add a column representing replications? I know I have to add one representing the whole plot in my data.

Any help is highly appreciated.

Last edited: