Hi, I'm trying to merge files from a data base of 20 files, each containing about 4000 subjects. Unfortunately the ID numbers are actually a combination of numbers and letters (e.g., b1458219). SPSS appears (?) to only allow merging of files if the files are strictly numerical / do not contain letters, even as string variables. What's the best way to merge these files, either in SPSS or by converting to Excel, which I'm not terribly familiar with?
Thanks!
Personally I'd consider using a combination of converting or creating a new variable so that the id columns to be common across files first.
I'd then use vlookup in excel. But if using this be conscious that it will put a load on the memory so you may want to copy and paste the values once you've merged a dataset before starting on anew one. There's many ways to skin a cat and without more info this is just one way that *may* work but others may have better solutions.
For instance I've used R and MATLAB to create index vectors which then can be used in merging. Again getting each file to have the correct I'd format is prob going to be quicker and easier
I believe each "key" variable (unique identifier) needs to be the same type (e.g., string) and in exactly the same format (e.g., A24). After converting key variables in each file to a common type and format, merging should work as expected.
- Badges
- Science topic
- Similar topics
- Filing