SPSS Exact Fishers Message

[Heather Gilmartin]

Hi,

I am using SPSS 27 and running an Exact Fisher on data. I am receiving this message: Cannot be computed because unable to open temporary file.

The file is working for other calculations (descriptives, chi square, etc.) Any suggestions on how to fix?

Thanks

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

Googling <spss temporary file> took me to this tech support page:  

https://www.ibm.com/support/pages/cannot-be-computed-because-unable-open-temporary-file-when-attempting-run-ibm-spss-statistics-procedure

FWIW, when I open a DOS window and issue the command shown on that page, I
get a message saying "The system cannot find the path specified".  Either I
did something wrong, or the instructions given on that page need to be
updated.  (I am currently running SPSS 25 [64-bit] under Windows 10
Education, by the way.)

In any case, the info on that page suggests to me a couple of possibilities
about your problem:

1. You did not run as an Administrator when you installed SPSS; or
2. Your account on the computer where you are using SPSS is not an
Administrator account.  

If you are using Windows, you could try this advice from the tech support
website:

"Uninstall the product and reinstall the product by right-clicking on the
program installer executable file and selecting the 'Run as administrator'
option."

Maybe Jon or Rick have some other ideas?  

Good luck.  HTH.



Heather Gilmartin wrote

> Hi,
>
> I am using SPSS 27 and running an Exact Fisher on data. I am receiving
> this message: Cannot be computed because unable to open temporary file.
>
> The file is working for other calculations (descriptives, chi square,
> etc.) Any suggestions on how to fix?
>
> Thanks
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

Hello Heather.  In my first response (just a few minutes ago), I suggested
lack of Admin rights as a possible reason for the problem you're having.  In
this post, I'm asking for more detail about your data.  If you have a 2x2
table and are using Fisher's exact test because one or more of the expected
frequencies are low, it may be that you'd be better off using the N-1
Chi-square test.  See Ian Campbell's website (and Statistics in Medicine
article) for more details.

https://www.iancampbell.co.uk/twobytwo/twobytwo.htm

Here is Campbell's advice from the top of that page:

--- start of excerpt ---
The best policy in the analysis of two-by-two tables is:

    (1) Where all expected numbers are at least 1, analyse by the 'N - 1'
chi-squared test (the K. Pearson chi-squared test but with N replaced by N -
1).
    (2) Otherwise, analyse by the Fisher-Irwin test, with two-sided tests
carried out by Irwin's rule (taking tables from either tail as likely, or
less, as that observed).

These recommendations apply to data from either comparative trials or
cross-sectional studies, and update those of Cochran (1952, 1954).
--- end of excerpt ---

Regarding that last sentence, it is saying that the advice applies to study
designs where the marginal totals are not fixed in advance.  When the
marginal totals are fixed in advance, Campbell advises using Fisher's exact
test, or the Fisher-Irwin test as he calls it.  (He calls it that, because
J.O. Irwin also worked out the same test independently of Fisher, but rarely
gets remembered nowadays.)  

https://en.wikipedia.org/wiki/Joseph_Oscar_Irwin

Finally, if you do want to use the N-1 Chi-square test, notice that for 2x2
tables, the test of "linear-by-linear association" in the CROSSTABS output
is equivalent to the N-1 Chi-square.  

Busing FM, Weaver B, Dubois S. 2 × 2 Tables: a note on Campbell's
recommendation. Stat Med. 2016;35(8):1354-1358. doi:10.1002/sim.6808

HTH.






-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD