How to work with merged data with different ID numbers indicating the time points

pvl

New Member
#1
Dear all,

I hope someone can help me with this (maybe easy?) question.

I am currently writing my thesis, and for that I received an existing SPSS data file. The SPSS file is merged and consists of all kinds of variables at 4 different time point.

I think the ID numbers of the patients in the 4 different files were the same, so the file is merged in a way that the variable names are still the same (for instance: CFQ1, CFQ2, CFQ3) and then new ID numbers are created for the same patients. So instead of having the variable names indicating the different time points (CFQ1_T1, CFQ1_T2, CFQ1_T3, CFQ1_T4; and having just 1 ID number for each patient), the ID numbers are indicating the different time points (ID001_T1, ID001_T2..).

Usually, to test for change over time I would run for instance a paired sample t-test by inserting: CFQ1_T1 and CFQ1_T2, but with this data file I have no clue on how to test for differences in these variables over time or how to make predictions over time.

For instance, I want to know whether quality of life (total score) at T4 is predicted by the CFQ (Total score) at T1. What actions should I take in order to examine this effect? With this merged data-file I have no clue on how SPSS can filter out the quality of life T4 data and the CFQ T1 data of all patients (as there is just 1 variable 'quality of life' and 1 variable 'cfq').

I find it very hard to explain the exact situation, so I hope you understand.

Thank you in anticipation