<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />I have a table with columns named A, B, C, D for instance. These columns contain numbers and are set to Numeric, Continuous. I am trying to find the second highest value in these four columns using the Sort List function within a JMP Column formula. If I do something like Sort List( { 22, 44, -54, 33} )[3] it returns 33, which is correct. But if I replace this with the named columns such as Sort List( {:Name("A"), :Name("B"), :Name("C"), :Name("D")}[3] It gives me an error "Expression does not fit available column types 1 times At rows: {1} Operation: Bad Argument.
So if I do a combination of this like, Sort List( {:Name("A"), 44, -54, 33})[3] it returns 44 independent of what is in A. My theory is that it is ignoring the named field. To test this theory, I change the subscript from 3 to 4 and now it returns the same error as above.
I have tried wrapping the named column with a Num() function call. I have tried replacing the Sort List function call and the subscript with N Items function call it always says that this is a list of 4 items.
Any hints? Is there a better way to approach this?
Great!!! The trick was adding the Eval List function call as you mentioned. Thanks for your help.
If your data table has only one row, use the following:
Sort List(Eval List(List(:A[1],:B[1],:C[1],:D[1])))[3]
If you have multiple rows, you'd have to decide which row you want the value from, sort the rows first, or use a For Each Row loop.
