You can define user-missing values for short string variables (8 bytes or less), but not for long string variables (more than 8 bytes).

You can autorecode long strings into consecutive integers, preserving the original string values as value labels), and then define missing values for the new integer values.

data list free /longstr (a20). begin data "a string" "another string" " " end data. autorecode longstr /into numvar. *the blank comes first in sort order and so gets assigned the integer value 1. missing values numvar (1).

*in SPSS 13, you can automatically assign string blanks as user-missing in the autorecoded variable. data list free /longstr (a20). begin data "a string" "another string" " " end data. autorecode longstr /into numvar /blank=missing.

-----Original Message----- From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Stewart, Suzanne Sent: Monday, January 24, 2005 12:15 PM To: SPSSX-L@LISTSERV.UGA.EDU Subject: Missing values for string variables

> I would like to define missing values for string variables and can't > seem to do that. The Help Section says this: > > All string values, including null or blank values, are considered to > be valid unless you explicitly define them as missing. To define null > or blank values as missing for a string variable, enter a single space > in one of the fields under the Discrete missing values selection. > This doesn't seem to work though as I am not able to change anything > (or even access) under Missing for strings variables like you can for > numeric variables. > I hope this makes sense. I am really stuck here and need to find the > quickest and easiest way to define missing values for string variables > without recoding them all! > > Any help you can provide would be MUCH appreciated. > Thanks, > Julie > > > >