Re: PSPP or any othe free software: timeline-reserach/longitudinal-research possible?

[John Darrington]

On Tue, Apr 12, 2016 at 10:28:48AM +0200, "Bettina-Maria G??rdel /// BKJ" wrote:
     Hallo to all PSPP-experts,
     
     does there exist a free software for timeline- respectively 
longitudinal-research, i.e. to compare data-sets collected at several points in 
time from always the same sample?
     
     To explain in more detail the functions I need:
     
      
     
     -      a given sample is surveyed with always the same questionnaire at 
ten different points in time. 
     
     -      Then, I create a PSPP-file for every point in time I conduct the 
survey 
     
     -      in the long run, I would like to compare the different answers, 
i.e. I need to be able to compare certain items in the different files
     
     -      I could do that by putting those items I would like to compare into 
a new file. However, this would be quite time consuming
     
I think your question is about how to automate this last step.

Here is one way to do it:

Suppose you have 4 .sav files survey1.sav, survey2.sav, survey3.sav and 
survey4.sav  each of which have the
variables q1, q2 and q3.  The following syntax will combine them into one file, 
with twelve variables
q1_1, q2_1, q3_1, q1_2, q2_2, q3_2, q1_3, q2_3, q3_3, q1_4, q2_4, q3_4

MATCH FILES
 /FILE='survey1.sav' /rename=(q1=q1_1) (q2=q2_1) (q3=q3_1)
 /FILE='survey2.sav' /rename=(q1=q1_2) (q2=q2_2) (q3=q3_2)
 /FILE='survey3.sav' /rename=(q1=q1_3) (q2=q2_3) (q3=q3_3)
 /FILE='survey4.sav' /rename=(q1=q1_4) (q2=q2_4) (q3=q3_4)
 .

Section 10 of the user manual "Combining Data Files" tells you exactly what 
this is doing, and
how to adapt it to your specific needs.
     

You also talk about comparing different answers.  Obviously there are many 
different ways to compare things.

You might want to look at the ratio of the first and last answer to q2.  So you 
could run

COMPUTE q2_fl_ratio = q2_4 / q2_1.

You also might want to run a paired samples T-test on q1.  So you would do

T-TEST 
  PAIRS= q1_1 q1_2 q1_3 q1_4
  .


You asked about other free software.  I'm not terribly familiar with it, but I 
am sure that R will have similar 
capabilities.

I hope this helps.

J'
      
     

-- 
Avoid eavesdropping.  Send strong encryted email.
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.

signature.asc
Description: Digital signature