Hosmer-Lemeshow test is a goodness-of-fit test taken in use in logistic regression, particularly when there are regular covariates. Units are spitted into deciles based on predicted probabilities and a 2 × 10 contingency table of the expected and observed frequencies for the deciles attained. The Hosmer-Lemeshow test statistic is the Pearson chi-square statistic for this contingency table and has an estimated chi-squared distribution with 8 degrees of freedom if the model is correct. The test has restricted power.