The results of STAR JOIN matching may be inaccurate when non-sequential numerical keys with large gaps are present in the primary file.
The STAR JOIN merge functionality may produce inaccurate results when non-sequential numerical keys with large gaps are present in the primary file. The resulting dataset may not contain all the merged data that would be expected if compared to the results of the alternate MATCH FILES procedure. The original files that are used in the STAR JOIN process are not corrupted. Please see PM85344 for additional details on test scenarios.
IBM SPSS Statistics 21.0 FP1
Until a fix for this issue is applied, customers should not use the STAR JOIN feature and should use the older Merge Files procedure as an alternative. The Merge Files procedure is described in the product Help and the Syntax Guide.
SPSS Statistics 21.0 FP1 IF003 has been released to address this problem. Please see the SPSS Statistics 21.0 FP1 (Interim Fix 3) for more details
Note: The content of this Alert is based on the best available information as of the date of its publication.