At 11:01 AM 3/28/2007,
[hidden email] wrote:
>I am trying to work with a file and each time I want to run some
>analyses (including a simple freq) I get the following error message.
>
>>Command name: FREQUENCIES
>>Input error when reading a case.
>>This command not executed.
>
>I didnt create the file. The dataset is a public access dataset from
>NCES (ECLS-K).
Maybe somebody can answer this offhand. If not, the following my help
fix or diagnose the problem.
. Is the file on disk, CD, or where? If it's on CD, DVD, or otherwise
not on your hard drive, copy it to your hard drive and try it from
there. If it's on your hard drive, copy it and paste it under another
name, and try it from there. (This may either bring out, or fix, errors
in the file, simply as a file.)
. Is the file
- Raw data (read with DATA LIST or GET DATA TYPE=TXT),
- SPSS saved file (GET FILE),
- SPSS portable (IMPORT),
- or other (GET {CAPTURE,SAS,TRANSLATE,...} or GET DATA, any type but
TEXT)?
These are all subject to different errors. In particular, if it's a
saved file, what system and version of SPSS wrote it?
. Try reading and saving the file, rather than running FREQUENCIES.
Occasionally SPSS can do this, when it can't run a procedure on the
file. If it works, run FREQUENCIES or whatever you like, on the new
saved file.
. Maybe, PRINT or WRITE can log how far SPSS got into the file. There
are very different reasons for problems
- With the very first case
- With the last case
- With a case somewhere in the middle, after reading some cases
successfully.
Running with, say,
N OF CASES 10
may help diagnose this, as well. If it works - you get the 10 cases -
it's pretty diagnostic of an internal error in the file.
Good luck!
Richard
Locked
