Split Function overwritting values?

[#1]

I'm attempting to split this table to make it easier for analysis, but data is being overwritten.

Most rows in this source table (DT2) only have one set of "Markers" (the column in the picture below, not the JMP marking function). But, the CD3+ has 2 sets of 3, totaling 6 distinct data points, as shown in the picture below with yellow (with the same lot number and instruments). I would expect the split function (dt3 = dt2 << Split( Split( :Results ), Split By( :Marker ), Group( :Lot Number, :Instrument ));) to treat the additional set separately and add 6 lines in the new Table, but it overwrites the first set and only shows 3.

Is there a special function that would work better? A line of code I'm missing in the split? or a work around? Any help would be greatly appreciated!!!!

[#2]

Hi,
I think that you problem is that you have multiple rows with the same combination of Marker, Lot Number, and Instrument: JMP does not know that these are actually different items so it tends to take the first instance of the specific combination of columns to populate the Split Table.
Hence, you will need to create an identifier that distinguish between rows 13 - 15 and rows 22 - 24 in your example. Let me know if you need help with the creation of column formula that will create such identifiers.
Best,
TS

[#3]

That would make sense. So add an additional column like "Data Point #" or something. What formula did you have in mind TS?

[#4]

I think you also need to add the col "Marker" to the group role. Then you get more rows.

[#5]

That explodes the data table and doesn't contain the format I need. Most cells now are empty with that approach.

[#6 · Accepted solution]

I created a new column called "count" using the following formula

curmark = :Marker;
curlot = :Lot Number;
curins = :Instrument;
currow = Row();
number = N Rows(
	Current Data Table() << get rows where(
		:Marker == curMark & :Lot Number == curlot & :Instrument == curins & Row() < currow
	)
) + 1;

I then used it as an additional Grouping variable

 

and got the following results

[#7]

Thanks txnelson! That did the trick. Was able to drop that block into the existing script and it ran fine. Thanks!