Is there a way to increase the number of characters printing (displayed) for a PRO FREQ table? It appears that the column width is defined by the number of characters in "frequency." My labels are over the 9-spaces provided and I want all of the VAR labels to be printed. Thanks. ... View more

I am trying to generate the reports described below and not sure if I should use PROC REPORT, TABULATE or another procedure. (I seem to go down one path only to see than another might have advantages). Of course I would like to complete it in as few "steps" as possible. Any assistance would be highly appreciated. Please let me know if any additional info is required. Also, are there any hard and fast rules to determine what procedure to use in such reporting? Or is the selection primarily based on programmer choice? I would like to find a process than I can use/standardize across any organization. Thanks, beforehand... date - Var1 - Var2 - Var3 04-04 5 6 6 04-07 3 2 2 04-09 5 6 1 04-11 4 4 6 04-13 2 6 4 04-15 1 6 3 04-19 1 6 6 (note: only a single month of data is being shown.) 3 Reports: * Total # Responses (greater than or equal to 3); * % of Responses (greater than or equal to 3) of the Total; * Month-to-month % change of Responses over 3. (note: I want it to be flexible enough to handle additional reporting, when needed.) ------Jan - Feb - Mar - Apr -> Dec Q1 Q2 Q3 ... View more

In doing some ETL development I am finding that SAS isn't allowing me to use some special characters (e.g., parentheses and single quotes) in some places (e.g., file descriptions, file names). Is there a complete list of non-usable special characters and what/if any escapes there are for their use. (Please inform me if this is posted in the wrong forum.) Please let me know if more info is required. Thanks Thanks, but I have answered this. Message was edited by: veblen ... View more

I have the table below being generated and I'm looking to exchange the null in var1 for obs 1 to 1. Basically I need to move the data for var1 up one observation and then drop obs 4. any ideas on the simplest way to complete this task? obs var1 var2 var3 1 null 22 33 2 1 22 33 3 2 22 33 4 3 0 0 Thanks Message was edited by: veblen Message was edited by: veblen Message was edited by: veblen ... View more

I'm looking for the best method to convert a character string to a numeric value. For example: 'EAST' should be converted to the number 1 'WEST' should be converted to the number 2 'NORTH' should be converted to the number 3 'SOUTH' should be converted to the number 4 Please let me know if this is not the correct forum. Please let me know if more info is required. Thanks ... View more

What is the simplest way to change a percentage of a vars observations in a data set based on a condition on that single var? For example: I want to change 50% of the observations in a data set where var A = 1 to var A = 2? (50 current obs where A =1 are to be changed to 25 obs where A = 1 and 25 where A = 2.) Even the ability to change x-number of obs where var A = 1 to A = 2 would work. For example: 100 obs where A = 1; A is changed to 2 Normally I have the luxury of a second variable to condition against but in this case I don't have anything to use. Thanks. ... View more

I am working with Integrity constraints and want to be able to use a second var in the IC condition. ic create X_chk = check (where= (( Y = 1 and X ~=999) or (Y =2 and X = 999))) Can I use two vars in an simple integrity constraint or do I have to check it against a second table? Or is there some other method that I'm missing. Thanks ... View more