For independent categorical observations cross-classified into an by contingency table, let be observed counts and let be their row and column totals. Under the null hypothesis of independence, fitted expected counts are , and the Pearson statistic isWith fixed and strictly positive underlying cell probabilities, its null distribution tends to as . The unrestricted cell-probability model has dimensions; the independence model has , giving that difference.
For counts , all fitted counts are , so with one degree of freedom. Sparse counts or structural zeros require different calibration or an appropriately specified exact procedure; the large-sample approximation is not a guarantee for every table.
Articles by others on the same topic
There are currently no matching articles.