Hello,
I am having trouble merging files on SPSS. I've ordered (alphabetically and in ascending order) the key variable I am using to match the data sets (Email Addresses). I have a three part follow up study, which means that i have something like 8,000 participants on Day 1, 420 at Week 1, and 350 at the one month follow up (horrible attrition rate, I know). I've tried to "match cases on key variables" and I have also clicked "both files provide cases", but when I try to add them into the "key variable" box, a message pops up and says "variables are different type or length" and will not let me proceed. Has anyone encountered this issue before, or knows how to overcome it?
Thank you!
There's a good walkthrough in the link below:
http://www.ats.ucla.edu/stat/spss/modules/merge.htm
There's a good walkthrough in the link below:
http://www.ats.ucla.edu/stat/spss/modules/merge.htm
The main requirements are that BOTH files have been sorted on the key variable(s), which needs not only to have the same variable name(s), but also be of the same type, e.g. string, numeric, and--also very important--the same width. That's all I can think of.
As Christopher says, a key variable that is a string has to be formatted to the same length in both (or all) files. Here is one (untested) way to simplify things (as much as they can be simplified):
GET FILE = "C:\MyFolder\File1.sav".
DATASET NAME F1.
SORT CASES by StringKeyVar.
* The following line assumes no string will be longer than 100 characters.
ALTER TYPE StringKeyVar(A100).
GET FILE = "C:\MyFolder\File2.sav".
DATASET NAME F2.
SORT CASES by StringKeyVar.
* The following line assumes no string will be longer than 100 characters.
ALTER TYPE StringKeyVar(A100).
MATCH FILES
FILE = 'F1' /
FILE = 'F2' /
BY StringKeyVar.
EXECUTE.
DATASET NAME Merged.
DATASET ACTIVATE Merged.
DATASET CLOSE all.
* Reduce length of StringKeyVar to maximum length required.
ALTER TYPE StringKeyVar(AMIN).
* End of syntax.
HTH.
Please see this site
http://www.ats.ucla.edu/stat/spss/modules/merge.htm
But why match files? In a three part follow up study statistical procedures, except RM Anova, will need a long data structure.
Thank you! As soon as I adjusted the width to be the same in all three cases, it worked :)
Re André's question about "why match files", ADD FILES (for stack-merging) also requires any common string variables to be formatted to the same length.
I'm conducting a RM Anova for most of the data so that's why I was keen on merging them in this case :)
Dear,
Your question: How do you merge files on SPSS?
Kindly see the following links, then you will get some idea.
http://www.youtube.com/watch?v=2vVS4cH6OI8
http://www.youtube.com/watch?v=MFXPWT0ERoE
http://www.youtube.com/watch?v=vhv2tgbK_x0
http://www.youtube.com/watch?v=CN8EG_Nk6Mw
http://www.youtube.com/watch?v=REjFDGz-8TQ
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis