LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (April 2006)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Fri, 21 Apr 2006 18:24:41 -0500
Reply-To:     "Oliver, Richard" <roliver@SPSS.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Oliver, Richard" <roliver@SPSS.COM>
Subject:      Re: Combining categories with Recode
Comments: To: "Peck, Jon" <peck@spss.com>
Content-Type: text/plain; charset="iso-8859-1"

Minor clarification (since the dreaded Appendix A might not be clear on this point):

The Value Labels command takes effect immediately. The command doesn't read the data, but it doesn't require some other command to read the data either, since all it does is alter the dictionary. But even if its behavior was similar to that of Recode, its results would still be available for the subsequent Crosstabs command. For example:

recode var1 (lo thru 5 = 1) (lo thru 10 = 2) (else=sysmis). recode var2 (lo thru 10 = 1) (lo thru 20 = 2) (else=sysmis). value labels var1 var2 1 "Low" 2 "High". crosstabs tables=var1 by var2.

This should produce a 2x2 table, and assuming value labels are displayed, both the rows and columns will be labeled "Low" and "High".

Another command that alters the dictionary and takes effect immediately is Missing Values, which can sometimes lead to results you perhaps did not expect.

-----Original Message----- From: SPSSX(r) Discussion on behalf of Peck, Jon Sent: Fri 4/21/2006 4:23 PM To: SPSSX-L@LISTSERV.UGA.EDU Subject: Re: Combining categories with Recode -----Original Message----- From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Bob Schacht Sent: Friday, April 21, 2006 2:34 PM To: SPSSX-L@LISTSERV.UGA.EDU Subject: Re: [SPSSX-L] Combining categories with Recode [snip]

[>>>Peck, Jon] Value Labels is not a procedure. It does not pass the data. This is spelled out in the dreaded Appendix A of the Command Syntax Reference.

Thanks. This is, in essence, what I wound up doing. BTW, is the "value labels" command considered as a procedure? In other words, can I slip in a value labels command between the Recode and the Crosstab so that the Crosstab procedure will be processed using both Recode and Value labels modifications? ------------ [>>>Peck, Jon] The power for this already exists in Ctables. You would use the Hiding Subtotal feature, which allows you to subtotal specified categories (and label it as you wish) and hide the subtotaled categories. Ctables also gives you full control over the order and inclusion of categories on an individual basis.

While I'm at it, can I dream a little and suggest that SPSS add an optional category list following the name of a variable in the CROSSTAB procedure? So, for example, instead of having to recode as above, one could simply stipulate

CROSSTABS ED_LEVL(2+3, 4+5, 6+7+8) BY ... /STATISTICS = CHISQ This would pool the categories as stipulated, and use labels consisting of the original labels strung together (e.g., the label for category 3 would simply be added on to the end of the label for category 2).

Bob

Robert M. Schacht, Ph.D. <schacht@hawaii.edu> Pacific Basin Rehabilitation Research & Training Center 1268 Young Street, Suite #204 Research Center, University of Hawaii Honolulu, HI 96814


Back to: Top of message | Previous page | Main SPSSX-L page