syntax for scoring repeated measures

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

syntax for scoring repeated measures

joseph bianco
Hi All:
I'm not syntax savvy so I hope someone can assist.

I have a dataset containing responses from several questionnaires that were
administered 12 times (once per week during an intervention).  One
questionnaire, the GDS, has 30 items and I have the scoring syntax for it.
In the spreadsheet, each administration of the GDS consists of 30 variables
named by session number and item number: S1GDS01 for session 1, item 1--and
so on.

My question is: is there a command or syntax that I can use to score all of
the 12 administrations of this measure without having to retype 30 new
variable names for each session?
Thanks in advance,
Joe

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: syntax for scoring repeated measures

Maguin, Eugene
Joe,

So, your data are in wide format with 360 variables per person for the GDS.
If there were no missing data, I think I can tell you how to compute a set
of total sums or total means using a vector and loop-end loop structure.
But, if you have missing data I would just write out 12 statements. The
mean.x function has more flexibility and that flexibility would be more
involved to replicate using a loop-end loop structure. I think this
statement form will work.

Compute s1mean=mean.25(S1GDS01 to S1GDS30).

Gene Maguin


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
joseph bianco
Sent: Thursday, March 11, 2010 11:38 AM
To: [hidden email]
Subject: syntax for scoring repeated measures

Hi All:
I'm not syntax savvy so I hope someone can assist.

I have a dataset containing responses from several questionnaires that were
administered 12 times (once per week during an intervention).  One
questionnaire, the GDS, has 30 items and I have the scoring syntax for it.
In the spreadsheet, each administration of the GDS consists of 30 variables
named by session number and item number: S1GDS01 for session 1, item 1--and
so on.

My question is: is there a command or syntax that I can use to score all of
the 12 administrations of this measure without having to retype 30 new
variable names for each session?
Thanks in advance,
Joe

=====================
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

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: syntax for scoring repeated measures

Richard Ristow
At 01:37 PM 3/11/2010, Gene Maguin wrote:

So, your data are in wide format with 360 variables per person for the GDS. If you have missing data I would just write out 12 statements. The mean.x function has more flexibility and would be more involved to replicate using a loop-end loop structure. I think this statement form will work.

Compute s1mean=mean.25(S1GDS01 to S1GDS30).

I'm not sure this is any neater, but you could use DO REPEAT with the 'mean.25'  function. Not tested:

DO REPEAT  Mean =s1mean TO s12mean
          /First=S1GDS01, S1GDS31, S1GDS61, S1GDS91,
                 S1GDS121,S1GDS151,S1GDS181,S1GDS211,
                 S1GDS241,S1GDS271,S1GDS301,S1GDS331
          /Last =S1GDS30, S1GDS60, S1GDS90, S1GDS120,
                 S1GDS150,S1GDS180,S1GDS210,S1GDS240,
                 S1GDS270,S1GDS300,S1GDS330,S1GDS360.
.  Compute mean =mean.25(First to Last).
END REPEAT.

===================== 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