While the Ctables dialog uses value labels to show the table preview and to allow the user to specify which categories are in or out, subtotals etc, the syntax does not care about the labels, so removing a value label or assigning a null label will not affect whether the value is included or not. Of course it does affect how the category is displayed.

But you can use an explicit category list in syntax to specify what values should be tabulated (and in what order they should be displayed).

/categories variables=varlist [value, value, value ...] produces a table with only those values displayed. You can use special keywords in the category list, such as OTHERNM to refer to any other values not missing and MISSING to refer to missing values.

This can be combined with EMPTY=INCLUDE|EXCLUDE to determine whether the listed values are always displayed or not. (It might be a metaphysical question what the meaning is of OTHERNM combined with EMPTY=INCLUDE, but the effect of this in SPSS is to add any other labeled values to the table.)

HTH Jon Peck SPSS

-----Original Message----- From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Spousta Jan Sent: Thursday, February 09, 2006 1:25 AM To: SPSSX-L@LISTSERV.UGA.EDU Subject: Re: [SPSSX-L] Deleting a Value Label

Hi Georg,

Try something like

add val lab myvariable 97 ''.

(untested - I do not have Tables licensed here)

Greetings

Jan

-----Original Message----- From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Georg Maubach Sent: Wednesday, February 08, 2006 7:35 PM To: SPSSX-L@LISTSERV.UGA.EDU Subject: Deleting a Value Label

Dear Listers,

we need to delete a value label from a hole bunch of variables to properly do a CTABLES like this:

1 very good 10 2 indifferent 0 3 very bad 12 99 don*t know/no answer 4

The actual coding is:

1 very good 10 2 indifferent 0 3 very bad 12 97 don't know 1 98 no answer 3

We need to recode all variables like

RECODE ALL (97, 98 = 99) .

But: We need the zero as a value in CTABLES for all categories with no response. The solution to just recode is not sufficient cause the CTABLES would look like this

1 very good 10 2 indifferent 0 3 very bad 12 97 don't know 0 98 no answer 0 99 don't know/no answer 4

but we want to have

1 very good 10 2 indifferent 0 3 very bad 12 99 don't know/no answer 4

Therefore we would like to delete the value label from all the variables. Is there a way to delete the value label completely using SPSS 13.0 so that it does not show up in CTABLES?

Any hint is appreciated.

Best regards

Georg Maubach Market Analyst Deutsche Post World Net Business Consulting GmbH Market Research Service Center