I am in the process of aggregating different longitudinal samples from different surveys into one SPSS file. However, as this was not planned when the different surveys were set up, I have to rearrange quite a few of the variables, so the order between surveys match.
I would like to do that when exporting the respective survey to SPSS. What is the best way to approach that problem? I searched the forum, but did not find anything enlightening at all. I can't imagine I am the first one to tackle such an issue.
Hi Felix,
I had some similar problems with combining 2 different surveys with similar questions (I had 1 student survey and 1 other). What i did was to export them both from lime to spss, then i exported the data into excel and I combined them there. The benefit is that you get the sorted SPSS variables of the 1st survey and can work easily with the 2nd without making mistakes. When I was certain all is fine i just reimported the data back to the SPSS. There are several issues with that that you can live with...
1. you need all answers to be defined with same labels (if you have 4 labels defined and suddenly a 5th or 6th label appears from the 2nd data file)
2. Also had to recheck the variable names many times not to copy the wrong answers to the wrong rows... i had many 0-1 answers that was pretty hard to sort all of them with no mistakes...
thanks a lot for your input. I ended up changing the design of the analysis.
So I first made sure the variables between waves are the same per dataset. After that, I added a prefix to each variable which shows the waves the variable belongs to. After that, I merged the datasets. Going on from there, it's easy to compute new variables containing all relevant cases.
