Newbie technical query - SPSS could not save a data file
Andries du Toit
Question from a SPSS newbie.
Apologies if this is not the correct forum.
I am trying to set up variables for a data sheet. A fair amount of
variables - 90 and counting.
Suddenly I got a message saying that "the attempt to save the data file has
failed because the disk was full, and i/o error has occurred or the task was
interrupted."
Then there came (from nowhere ) (I am a newbie - 3 hours old!) what appeared
to be some sort of output log, informing me that there had been an invalid
format code in the data being saved" and an error message saying DATA1211
And another error code sayin ERROR 1407, error when writing the dictionary
to a new data file.
Then I got prompted to save some output files.
And then SPPS crashed with an IPF (I think).
Any ideas?
Andries
pete...@my-deja.com
> I am trying to set up variables for a data sheet. A fair amount of
> variables - 90 and counting.
>
> Suddenly I got a message saying that "the attempt to save the data
file has
> failed because the disk was full, and i/o error has occurred or the
task was
> interrupted."
>
> Then there came (from nowhere ) (I am a newbie - 3 hours old!) what
appeared
> to be some sort of output log, informing me that there had been an
invalid
> format code in the data being saved" and an error message saying
DATA1211
>
> And another error code sayin ERROR 1407, error when writing the
dictionary
> to a new data file.
>
> Then I got prompted to save some output files.
>
> And then SPPS crashed with an IPF (I think).
>
The first thing I would do is to check if SPSS is telling the truth. Is
there any space left on Your working drive? But if You are working with
a "dry" datafile (no data) with 90+ variables, the file is so small
that this probably isn't the problem.
Or maybe there are some damaged sectors on Your disc. You can check for
damages with scandisc. But this is a rare problem.
Take a look at the working memory of SPSS. You can manipulate it by
opening a new syntax window and type:
set mxmem=100000 .
set workspace=100000 .
exe .
Run this and try again. The size of the parameters is depending on Your
computers RAM (i think)...
Finally: can You replicate the problem? If so, You should talk to SPSS.
good luck
/s
Sent via Deja.com
http://www.deja.com/
Andries du Toit
I solved my problem.
What I was doing was trying to copy variable attributes from one variable to
another.
This resulted in a compatibility problem - a "string" variable with decimal
points. For some reason. I thought the software would automatically detect
these inconsistencies.
What I realised is that I can copy an entire variable (i.e. right-clicking
in the grey tab left of the variable name) and then make adjustments.
> S: Welcome to the wonderful world of SPSS...
Thanks 8-)
> The first thing I would do is to check if SPSS is telling the truth. Is
> there any space left on Your working drive?
More than a 1.20 G. That was the first thing I checked
>But if You are working with
> a "dry" datafile (no data) with 90+ variables, the file is so small
> that this probably isn't the problem.
Thought so.
> Or maybe there are some damaged sectors on Your disc. You can check for
> damages with scandisc. But this is a rare problem.
Checked. No bad sectors.
> Take a look at the working memory of SPSS. You can manipulate it by
> opening a new syntax window and type:
It is set at 512 K
> set mxmem=100000 .
> set workspace=100000 .
> exe .
>
> Run this and try again. The size of the parameters is depending on Your
> computers RAM (i think)...
I have 160 Meg of RAM - it's good to have elbopw room sometimes.
> Finally: can You replicate the problem? If so, You should talk to SPSS.
Yes, it is replicable. Every time I copy a "string" characteristic onto a
"numerical" variable the decimal value (set at 2) does not go down to 0
but is just greyed out - and then when I save the data error occurs. This
may not be a problem with SPSS but with me own competence.
happyholidays
A