Hi,
I am doing a 2x2 test in SPSS and one of the cells has a number less than 5. I have gone into descriptive statistics and ticked exact test as directed but the output does not show a number for the Fisher's exact test. It only has the same results as when exact is not ticked. Does anyone know why this may be?
The Fisher exact test is rarely appropriate as it assumes fixed marginal totals (which is unlikely in most research contexts). Generally one can safely use chi-square with most expected cells > 5 (the observed counts in each cell are not an issue). If it falls below this I'd suggest the Egon Pearson correction.
https://seriousstats.wordpress.com/2019/09/05/chi-square-and-the-egon-pearson-correction/
Hello Marie Quinn . Please share the 4 cell counts and your command syntax so that others can see the problem you are describing. Thanks.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS