I am attempting to analyze data in a 2x5 contingency table and there are >20% of the boxes that have expected values that are less than 5. What test, using SPSS, would be appropriate in allowing me to accurately analyze this data? My version of SPSS does not seem to have an option to perform a Fisher's exact test on a contingency table greater than 2x2. All help is appreciated.
Unfortunately, they cannot - they are distinct categories that should not be combined for this analysis.
What is the contingency table? And is the model that you want to test that the marginals are fixed, and that there is no association?
The contingency table is 2X5 and I am simply testing to see if there is a significant association between 2 groups with regard to another variable that has 5 levels.
- Badges
- Science topic