SPSS Production Facility

[Craig Johnson]

I'm looking for some help using the SPSS Production Facility.

Background

Right now we are running through some different methodologies across
several data sets.  Each data set has different variables and has been
collected at different periods of time.  However we have went through
each data set and renamed a bunch of variables so each data file can
use the same generic syntax.  The trick now is to use the production
facility to automatically open each data file, run the syntax, and
save the output. By using the production facility we'll be able to
change the syntax in a single place, rerun the production facility,
and get an updated set of results across X studies.

Features

1)      It would be nice to have this all wrapped up in as few as possible
syntax files.  Ideally it would be our testing syntax and a syntax
file which opens the various data set.  (IE I realize I could make a
test syntax file for each data set we have and then add them all to
the production facility but that means I'd have to change the syntax
in N number of places for the number of data sets we have.  Not the
best solution….need something more generic for future use.)

2)      Ideally I'd be able to save an aggregate of the output.  IE if I
run the syntax on 100 studies the output is all saved in one output
file not 100 of them.  If this is possible I'd need to insert a title
or something which designates where the output starts and stops for
each file. If this isn't possible then I'd like to be able to save the
output file with the name of the data file.  Suggestions welcome as
I'm back and forth on the best way to do this….. =(

Solutions?

So……….oh guru's of the SPSS production facility how about some help? I
mean I'm pretty sure I could brute force this thing into working but
it will probably end up making me spend a lot more time on this then
needed.  =) Help!

[Natalie-24]

Hi,

I'm working on some problems that assessing the assumptions for the
regression model.  I attempted to plot the standard residual versus the
fitted or predicated values.  The graph is 90 degrees off, the pictorial in
the text.

Does anyone know why?

- nkc
----- Original Message -----
From: "Craig Johnson" <[hidden email]>
To: <[hidden email]>
Sent: Friday, August 17, 2007 7:00 PM
Subject: SPSS Production Facility

> I'm looking for some help using the SPSS Production Facility.
>
> Background
>
> Right now we are running through some different methodologies across
> several data sets.  Each data set has different variables and has been
> collected at different periods of time.  However we have went through
> each data set and renamed a bunch of variables so each data file can
> use the same generic syntax.  The trick now is to use the production
> facility to automatically open each data file, run the syntax, and
> save the output. By using the production facility we'll be able to
> change the syntax in a single place, rerun the production facility,
> and get an updated set of results across X studies.
>
> Features
>
> 1)      It would be nice to have this all wrapped up in as few as possible
> syntax files.  Ideally it would be our testing syntax and a syntax
> file which opens the various data set.  (IE I realize I could make a
> test syntax file for each data set we have and then add them all to
> the production facility but that means I'd have to change the syntax
> in N number of places for the number of data sets we have.  Not the
> best solution….need something more generic for future use.)
>
> 2)      Ideally I'd be able to save an aggregate of the output.  IE if I
> run the syntax on 100 studies the output is all saved in one output
> file not 100 of them.  If this is possible I'd need to insert a title
> or something which designates where the output starts and stops for
> each file. If this isn't possible then I'd like to be able to save the
> output file with the name of the data file.  Suggestions welcome as
> I'm back and forth on the best way to do this….. =(
>
> Solutions?
>
> So……….oh guru's of the SPSS production facility how about some help? I
> mean I'm pretty sure I could brute force this thing into working but
> it will probably end up making me spend a lot more time on this then
> needed.  =) Help!

[Albert-Jan Roskam]

Hi Craig,

I would create a master syntax in which all individual
syntaxes are 'called' using INSERT (or INLCLUDE)
commands. In the underlying syntaxes (let's call them
slave syntaxes) you'd have to specify the source file
location (i.e. it starts with GET FILE or GET DATA or
something similar). That's a handy habit anyway so you
know -for future reference- which source file the
syntax was originally written for. You could also
include a TITLE command there, as a header.

the master syntax would simply look like
* this is slave file 1.
INSERT FILE = 'D:\temp\file1.sav'.
* this is slave file 2.
INSERT FILE = 'D:\temp\file2.sav'.
* this is slave file 3.
INSERT FILE = 'D:\temp\file3.sav'.

This works with 'regular' SPSS, and probably also with
the production facility (PF). Note that the syntax
rules with batch processing (which occurs in PF) are
slightly less forgiving than those of a normal spss
session. Any line that doesn't end with a period (.)
should be indented.

Cheers!!
Albert-Jan

--- Craig Johnson <[hidden email]> wrote:

> I'm looking for some help using the SPSS Production
> Facility.
>
> Background
>
> Right now we are running through some different
> methodologies across
> several data sets.  Each data set has different
> variables and has been
> collected at different periods of time.  However we
> have went through
> each data set and renamed a bunch of variables so
> each data file can
> use the same generic syntax.  The trick now is to
> use the production
> facility to automatically open each data file, run
> the syntax, and
> save the output. By using the production facility
> we'll be able to
> change the syntax in a single place, rerun the
> production facility,
> and get an updated set of results across X studies.
>
> Features
>
> 1)      It would be nice to have this all wrapped up
> in as few as possible
> syntax files.  Ideally it would be our testing
> syntax and a syntax
> file which opens the various data set.  (IE I
> realize I could make a
> test syntax file for each data set we have and then
> add them all to
> the production facility but that means I'd have to
> change the syntax
> in N number of places for the number of data sets we
> have.  Not the
> best solution�.need something more generic for
> future use.)
>
> 2)      Ideally I'd be able to save an aggregate of
> the output.  IE if I
> run the syntax on 100 studies the output is all
> saved in one output
> file not 100 of them.  If this is possible I'd need
> to insert a title
> or something which designates where the output
> starts and stops for
> each file. If this isn't possible then I'd like to
> be able to save the
> output file with the name of the data file.
> Suggestions welcome as
> I'm back and forth on the best way to do this�..

=(
>
> Solutions?
>
> So���.oh guru's of the SPSS production
facility how
> about some help? I
> mean I'm pretty sure I could brute force this thing
> into working but
> it will probably end up making me spend a lot more
> time on this then
> needed.  =) Help!
>


Cheers!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Did you know that 87.166253% of all statistics claim a precision of results that is not justified by the method employed? [HELMUT RICHTER]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


      ____________________________________________________________________________________
Shape Yahoo! in your own image.  Join our Network Research Panel today!   http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7