z-scores across columns

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

z-scores across columns

la09
I was hoping someone could tell me how to calculate z-scores across colums in spss;

Normally i generate z-scores using the check box in the descriptives window, but this provides z-scores based on the rows within a column. My data set is setup with subjects in rows and conditions across columns; i want to look at the effect of condition, independent of indivudal differences in overall reponse times, and therfore need to generate z-scores per individual across condition. I imagine a simple syntax would be adequate and was hoping someone would know how to do this.

(p.s i have thought of rearranging my data set to my coditions are in rows but then for the analysis i would nee to re-arrange then to colums so this isnt an adeqaute solution)

Any advise on how to dothis would be greatly appreciated, thanks.

Reply | Threaded
Open this post in threaded view
|

Re: z-scores across columns

Art Kendall
It sounds like maybe you want to do what is called ipsatizing the data.
Something like this untested syntax should do it.

descriptives var=raw1 to raw22/save.
compute case_mean= mean(zvar1 to zvar22).
compute case_sd =    sd(zvar1 to zvar22).
do repeat zz = zz1 to zz22/zv = zvar1 to zvar22.
compute z=(zv-case_mean)/case_sd.
end repeat.

Art Kendall
Social Research Consultants


la09 wrote:

> I was hoping someone could tell me how to calculate z-scores across colums in
> spss;
>
> Normally i generate z-scores using case_ check box in case_ descriptives window,
> but this provides z-scores based on case_ rows within a column. My data set is
> setup with subjects in rows and conditions across columns; i want to look at
> case_ effect of condition, independent of indivudal differences in overall
> reponse times, and case_rfore need to generate z-scores per individual across
> condition. I imagine a simple syntax would be adequate and was hoping
> someone would know how to do this.
>
> (p.s i have thought of rearranging my data set to my coditions are in rows
> but case_n for case_ analysis i would nee to re-arrange case_n to colums so this
> isnt an adeqaute solution)
>
> Any advise on how to dothis would be greatly appreciated, thanks.
>
>
> --
> View this message in context: http://www.nabble.com/z-scores-across-columns-tp21244968p21244968.html
> Sent from case_ SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except case_
> command. To leave case_ list, send case_ command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send case_ command
> INFO REFCARD
>
>
>

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Art Kendall
Social Research Consultants