Why is this necessary? I see "LION", upper case and delineated with " ". Thanks, Kelly

-----Original Message----- From: epidata-list@lists.umanitoba.ca [mailto:epidata-list@lists.umanitoba.ca] Sent: Sunday, May 01, 2005 2:16 PM To: epidata-list@lists.umanitoba.ca Subject: [EpiData-list] Re: selecting on string variables

1. How to select "string variables" in Epidata Analysis ?

select outcome="death"

An example should clarify: Try this: generate 3 // make 3 artificial records gen integer id = recnumber // create id variable define animal _________ // make a string variable * add value to the variable: if recnumber = 3 then animal = "monkey" if recnumber = 1 then animal = lower("LION") if recnumber = 2 then animal = "LION" . list ID ANIMAL 1 lion 2 LION 3 MONKEY * so we see that we have three values for the variable animal.

* and we have two variables: . var Name Type Length Decimals ID Integer 12 0 ANIMAL String 9 0 %s9

* now we try various selects:

select if animal = "LION" count No records available select // to select if upper(animal) = "LION" count No records available select select if upper(trim(animal)) = "LION" count Count : 2

The essense is therefore that to do a select on a string variable, you have to use as well sometimes the function upper() and the function trim() to get a proper selection.

regards

Jens Lauritsen Initiator of EpiData Coordinator EpiData Association