Date: Sat, 11 Feb 2006 07:11:44 -0800
Reply-To: Albert-jan Roskam <fomcl@yahoo.com>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Albert-jan Roskam <fomcl@yahoo.com>
Subject: Re: Identifying the syntax file in output and documentation
In-Reply-To: <F6C74A1471C6514393225D3CFF3323DBD7ADB5@natickexch03.natick.army.mil>
Content-Type: text/plain; charset=iso-8859-1
Hi Pat!
I only very occasionally save SPO-files. Sometimes I
do this to document 'final' runs. After all, the
syntax + data file can be used to reproduce analyses.
This saves a lot of files. The SPO/SPS files that I do
keep indeed have the same name, preferably a
meaningful (mnemonic) name. Often I number them to
indicate which one should be used first, and to
indicate how they're related (e.g.
01a-preprocessing.sps, 01b-first_checks.sps). When
using SPO files I use a SBS script I borrowed from
Ray's site (thanks a LOT, Ray!!!) to 'fill in' output
generated with the TITLE statement.
I also sometimes insert '* INCLUDE myfile+location' to
indicate which SPS preceded the current SPS. I have a
header in each file where the author, file creation
date, and the purpose of the file can be read.
Finally, within projects, often 'subprojects' can be
discerned. I save those in separate subdirs.
Another thing I consistently do, but which is less
related to your mail, is put the file attribute of the
source file to 'read-only', so it's impossible to
-somehow- overwrite the original file, after for
example a SELECT IF.
And still... it's a struggle to keep things organized!
=)
Good luck!
Albert-Jan
--- "Johnson, Wendy RDECOM (PKI)"
<Wendy.Kullen.Johnson@us.army.mil> wrote:
> I've learned to give them the same file names, so,
> for example, the output of file "ANOVA1.SPS" is
> "ANOVA1.SPO", that way I can match them up quickly.
>
> I also type the file name in as a comment at the top
> of the syntax file so when I print it I have a
> record of what file it is. Since I have spss set up
> to display the commands in the log, it shows in the
> output file, as well.
>
> I used to keep a seperate key file with a
> description of the syntax, too, but now that I can
> use long file names in Windows I put the details in
> the file name.
>
> Looking at my file list now I have:
>
> PreEval.sav
> PreEval 0 Labels.spo
> PreEval 0 Labels.sps
> PreEval 0 Labels.sav
> PreEval 1 Logic Checks.spo
> PreEval 1 Logic Checks.sps
> PreEval 2 Q1 thru Q12.spo
> PreEval 2 Q1 thru Q12.sps
> PreEval 3 Beverages.spo
> PreEval 3 Beverages.sps
> PreEval 4 Demographics.spo
> PreEval 4 Demographics.sps
>
> The names are not very detailed. But instead of one
> long syntax file I often use several short syntax
> (and output) files, anyway, to keep things orderly.
> It works for me.
>
> HTH,
>
> ~Wendy.
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion
> [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
> Patricia Cleland
> Sent: Friday, February 10, 2006 10:22 AM
> To: SPSSX-L@LISTSERV.UGA.EDU
> Subject: Identifying the syntax file in output and
> documentation
>
> Sometimes when I look at my output (.spo files), I
> can't remember what syntax file (.sps file) was used
> to create it. Does anyone have a method to
> automatically include the name of the syntax file
> that generated the output?
> Something similar to the script DataFooter.sbs from
> Ray Levesque's web site would be great.
>
> On a similar topic, does anyone have any suggestions
> for documenting analyses? I usually end up printing
> out the syntax file and highlighting and writing
> notes in the margin. However, if I need to re-run an
> analysis, I often need to read through all my
> syntax files, to figure out the one that I need to
> use. Sometimes I add comments to the syntax file
> that indicates its purpose.
>
> I have, very occasionally, developed an Excel or
> Word file that contains the name of the input data,
> syntax, output and output data files and a brief
> description of the purpose of the syntax file.
>
> For example,
> -syntax file ReadIn.sps, reads in the raw data file
> RawData.sav; -checks for valid data, recodes the
> variable age into deciles; -produces frequencies on
> recoded age, gender, SES; -produces output file
> ReadInandFreqs.spo -and the cleaned and re-coded
> data
> file RecodedData.sav.
>
> Needless to say, this is very labour intensive and I
> rarely do it. Does anyone know of any 'best
> practices' in documentation, preferably automated
> best practices?
>
>
> Pat
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
| 
