|
Christer, Assuming that you have a reasonable size data try this procedure ( I tried it with 4000 ID's, and up to 25 Colors) , ; 1. Open your table, then select : Tables/Split: // from menu bar Add: Color // the name of the text column you want to concatenate Split Label Column: Sample // this is the name of the column with ID Output table name: Split or any name you want, select Drop ALL This will create table called "split"; 2. open the table "split", then select Tables/transpose Output table Name: transposed This will create table called "transposed" : your ID "SAMPLE" will be under column Label, your Colors will be under columns Row 1 , Row 2, ......... 3. create a new column in the transposed table that has the following formula: :Row 1 || ";" || :Row 2 || ";" || :Row 3 || ";" || :Row 4 ; complete it up to the number text columns you have 4. Finally you can Join this new table with the table that has the Sample and mean , by Id :Sample/Label good Luck Bassam Finally
Hi Emil! A specific example would be a table like this: Sample Value Color Sample 1 6 Green Sample 1 4 Red Sample 2 3 Blue Sample 3 8 Green Sample 3 10 Blue Sample 3 9 Black Where I would want to use Summary on the table to generate: Sample N Rows Mean Value Color Sample 1 2 5 Green;Red Sample 2 1 3 Blue Sample 3 3 9 Green;Blue;Black Where it is the last text column that I would like to get aggregated in this or a similar fashion. Thanks for the help! Best wishes Christer -----Original Message----- From: Emil M Friedman [mailto:emilfrie@alumni.princeton.edu] Sent: söndag, maj 15, 2005 05:02 To: jmp-l@lists.biostat.wustl.edu Subject: Re: [jmp-l] Summary with text columns Could you give us a specific example? That will make it easier to answer your question. ------------------------------------------- Emil M Friedman, PhD 2304 Richmond Road Beachwood, OH 44122 emil.friedman@alum.mit.edu emilfrie@alumni.princeton.edu 216-591-1750 http://www.statisticalconsulting.org ----- Original Message ----- From: <Christer.Westerlund@astrazeneca.com> To: <jmp-l@lists.biostat.wustl.edu> Sent: Friday, May 13, 2005 1:49 AM Subject: [jmp-l] Summary with text columns > Hi! > > New to JMP and find it a fantastic program! Have been trying to use > Summary > on tables with text columns included. I wonder if there is any way of > getting cells containing the different text strings aggregated and > separated > by say , or ; or something. I know that you can do a lot of statistical > treatmetns on duplicate rows if you have numeric columns which is great, > but > how about text columns, where you want to get the differnt text strings > for > duplicates into the same cell.... > > Hope this is understandable and grateful for any help! > > Christer >
|
