Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

simple question. Is it possible to find the average of the entries in a column using SPSS?

4,516 views
Skip to first unread message

Bogdan Lataianu

unread,
Dec 28, 2008, 8:06:44 AM12/28/08
to
Is it possible to find the average of the entries in a column using
SPSS?
More or lees as it possible in Excel. Does SPSS allow functions like
in Excel for column data?

Allen

unread,
Dec 28, 2008, 9:46:05 AM12/28/08
to

The easiest approach is to use the "Frequencies" function. As I
recall
you pick statistics and you can choose "Mean" under "Central
Tendency".

martijnv...@gmail.com

unread,
Dec 28, 2008, 4:20:41 PM12/28/08
to

If you need the average value in your dataset for computation you can
use data --> aggregate. If you want to find the average for your
entire dataset you have to add a variable with a constant value (e.g.
value 1 for all cases) and specify that constant as a break variable.
The dialog lets you choose between several options for what you want
to do with the result: add to the active dataset, or to a new dataset
with a specified name etc. Besides computing averages aggregate lets
you compute frequencies, maximum and minimum values etc over columns
in your dataset.

hollan...@gmail.com

unread,
Apr 15, 2016, 9:04:34 AM4/15/16
to
Unfortunately no.


IBM would do itself a ton of favors for SPSS if it would introduce a COMPUTE GroupMean = COL.MEAN(variableA).

Instead of

AGGREGATE
/BREAK Group
/mode=addvariables
/GroupMean = MEAN(variableA).
execute.

Bruce Weaver

unread,
Apr 15, 2016, 1:39:58 PM4/15/16
to
If you're that put out by having to use AGGREGATE, stick it in a macro.

DEFINE !ColumnMean
( V = !CHAREND('/') /
BreakVar = !DEFAULT('') !CHAREND('/') /
Result = !CMDEND ).

AGGREGATE
/OUTFILE=* MODE=ADDVARIABLES OVERWRITE=YES
/BREAK !BreakVar
/!Result = MEAN(!V).

!ENDDEFINE.

* COMPUTE GroupMean = COL.MEAN(variableA).
!ColumnMean V = variableA / BreakVar = Group /Result = GroupMean.
!ColumnMean V = variableA / Result = GrandMean.

Rich Ulrich

unread,
Apr 15, 2016, 2:27:16 PM4/15/16
to
On Fri, 15 Apr 2016 06:04:30 -0700 (PDT), hollan...@gmail.com
wrote:
You could write a macro that effectively does that.


On the other hand, I am intrigued by the thought from the
original poster, that there could conceivably be a
"spreadsheet version" of SPSS -- where there could
be a built-in function COL.MEAN (var) . You might use
this for smaller files until we reach the era of infinite, free
computing.

--
Rich Ulrich


0 new messages