To test for an association between the marker alleles and the disease, we consider the model: E(Y) = μ = (μ1, ⋯, μN)T with μi = p + r if i is a case and μi = p if i is a control. The null hypothesis is H0 : r = 0. The proposed corrected χ2 test is one that extends the χ2 test by taking into account the correlation structure among individuals. The test statistic of the corrected χ2 test is [Bourgain et al., 2003] :
