SPSS v. 21 STAR JOIN
I have just started with version 21 and I have run into a major issue.
Merge is no longer the MATCH CASES command. It is now STAR JOIN. There is something different that this command is doing. When merging, the command just stops when it runs into a keyed variable that has no match in the other data set. It should just show up as missing since there is no match in the other file but it stops the process and doesn't match anything else.
Does anyone know what STAR JOIN is actually doing and why it does not complete the command?
Thanks.
Software/Hardware used:
SPSS version 21
Software/Hardware used:
SPSS version 21
ASKED:
September 11, 2012 5:25 PM
UPDATED:
September 11, 2012 5:54 PM
Answer Wiki:
Match Files is alive and well, but it requires pre-sorted data. If you
use the dialog interface to perform your merges, make sure you select
"Cases are sorted in order of key variables in both datasets", and the
dialog will generate Match Cases syntax (make sure your data file are
sorted by the key variables).
Star Join will only include cases that are present in the first file.
Any cases in the second file that are not also present in the first file
(based on the key variable values) will not be included. This is like a
SQL outer join, and unlike Match Files, which will include all cases
from both files. But it shouldn't just "stop" when it encounters the first unmatched case.
To see all answers submitted to the Answer Wiki: View Answer History.
Thanks, this is very helpful!Do you know what STAR JOIN is actually doing? I tried reading the IBM explanations but they are not helping much. What is the difference between match files and star join besides that fact that match files needs presorted data?