Got a table attached here (test.jmp). After I used split function like below, I got duplicate items and the N number is also not correct. Anybody knows what is wrong? Thanks a lot!
Table:
Split setting:
Result I got, which contains duplicate items:
You need to put supplier_wafer_id into the Group role.
That will create one row per supplier_wafer_id.
That's not what I get. First of all, I would not select all 3 columns to keep - the one you are splitting and the one you split by will already be used - just select "Keep all." When I do that I get 9 rows and the ID column shows no duplicates. But another strange thing is that the ID column contains a string of around 13 characters - in your image it looks like you have a single letter. So, I'm not sure what is going on since it looks like you get 9 rows like I do. But I would not select all 3 columns to keep when two of them are being used in the dialog to split and split by.
Thanks. I think I uploaded the wrong file, but it basically the same. If you look carefully, it has duplicate items for 13 characters item too.
I re-uploaded the file. I tried keep all, it is the same.
If you look at the distribution of defect_class and select a couple and subset them, you will see that the duplicate ID rows show up for some defect classes and not others. I'm not entirely sure what is happening, but I think what you might want to use is Analyze - Tabulate and put defect_class in the columns and ID in the rows and then put the N_rows int the table portion (selecting either mean, N, max, or some other statistic, as appropriate).
