Discovering Statistics Using IBM SPSS Statistics, 5th Ed. Practice Test

Fisher's exact test is an exact statistical method used for small samples to determine whether there is a nonrandom association between two categorical variables in a contingency table. Unlike the chi-square test, it computes the exact probability of the observed data under the null hypothesis, avoiding approximation errors that occur with small counts. This makes it ideal when sample sizes are limited or when expected frequencies are too small for reliable chi-square results. The other terms refer to different ideas: factorial ANOVA deals with experiments that have multiple factors, fit relates to how well a model or distribution matches the data, and familywise error rate concerns error control across multiple tests. None of these describe the exact-probability approach used in Fisher's exact test.