SQL UNION operator
- Article History
- RSS Feed
- Mark as New
- Mark as Read
- Bookmark
- Subscribe
- Printer Friendly Page
- Report Inappropriate Content
The UNION operator in Proc SQL is used to append the rows of two or more SELECT statements having the same number of columns with similar data types. Similar columns in each SELECT statement should normally be in the same order, as the Union is done by position not by column name, except where appending dissimilar columns from a select statement is explicitly desired.
Usage notes
Note that, unlike using the SET statement with multiple datasets, the default behavior of the UNION operator is to remove duplicate rows, from the results of each SELECT statement as well as overall, resulting in only DISTINCT rows. If retention of duplicate rows is desired, either OUTER UNION must be specified, or else the ALL operator must be used on individual SELECT statements where duplicates are desired.
Syntax
SELECT column_name(s) FROM table_name1 <OUTER> UNION <ALL> SELECT column_name(s) FROM table_name2 ...
This article was originally posted by Gerald Zuckier on sasCommunity.org.