When the trait is categorical and subjects are grouped as cases and controls, there are several association tests that one can use. General genotype association can be tested using the traditional χ2 test of independence in a 2 × 3 contingency table. More parsimonious procedures include the Armitage trend test in which genotypes are recoded to model a linear increase in the odds of the disease on the logarithmic scale for each different genotype, or associations of dominant or recessive models in which genotypes are aggregated in two groups [48]. The dominant model for the allele M, tests the associations of genotypes grouped as MM and Mm versus mm, whereas the recessive model for the allele M uses the grouping MM versus Mm and mm. Allelic association can be tested by recoding the data from genotypes into alleles. However, the association is difficult to interpret at the level of individuals. Some care is necessary when testing the associations of rare variants. In this case, the frequency of some genotypes may be too small (<5) for the χ2 test to be
