Date: Thu, 9 Jan 1997 10:48:13 EST
Reply-To: Gloria Edwards <firstname.lastname@example.org>
Sender: "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From: peter homel <HOMEL@SACC.HSCBKLYN.EDU>
Subject: renaming multiple variables
This may not be feasible for Gloria Edward's problem, but I have found
a quick and not so dirty way of renaming variables from multiple files
where there are duplicate variable names. The trick is to merge the
files first in a program that does allow for duplicate "variable names".
Thie can be done, for example, in Excel (using copy and paste, say).
When you bring the Excel file into SPSS, SPSS will automatically rename
duplicate variable names by adding a numeral to the duplicate
variable names. For example, say you bring in an Excel file which contains the
following column names: X1 X2 X11 X1 X2 X11 X1 X2 X11. SPSS will rename these
as X1 X2 X11 X12 X21 X111 X13 X22 X112. It can be seen that SPSS simply adds
a numeral that will make the renamed variable unique (e.g., X2 -> X21 in
the first duplicate, X2 -> X22 in the second duplicate). SPSS always checks
to see whether renaming the variable will result in another duplication.
Since X11 already exists, X1 -> X12 in the first duplicate while X1 ->
X13 in the second duplicate.
This tends to work fairly well with variable names that do not end with
numerals. SPSS will truncate variable names before adding numerals in
order to keep within the eight character limit for variables names.
Working with variable names that already end in numerals may make it
confusing you to tell which ones come from which file (although that
can be done based on the order of the variables in the file).
Another problem that precludes using this method for Gloria Edwards data
is that (as far as I know) Excel is limited to 255 (256?0 columns which would
not be enough for what Gloria is doing. Anyway, this method works fine for
small files with not too many variables.
PETER HOMEL PHD
HEALTH SCIENCE CENTER BROOKLYN
STATE UNIVERSITY OF NEW YORK
450 CLARKSON AVENUE BOX 7
BROOKLYN, NY 11203-2098
TEL: (718) 270-7424
FAX: (718) 270-7461
MOTTO: STATISTICS DON'T LIE!(PEOPLE DO!)
I have a client who wants to merge three datasets. The problem is that
the datasets have duplicate variable names (e.g., v1-v225) that represent
different things in each dataset. Therefore, the variables need to be
renamed before merging.
Is there a way to rename multiple variables in SPSS without specifying
each variable individually? The datasets are so large that individual
renames would be extremely cumbersome, but the menu options in SPSS Win
7.0 didn't appear to provide a work around. Can we do this with syntax
Any help would be appreciated.
Instructional and Research Computing
235 Bryan Building