Objective: To analyze longitudinal binary data by using generalized linear models. with consideration of the correlation between repeated measures. Methods: Generalized estimating equations (GEE) proposed by Zeger and Liang was used. For several covariance structures, one method was given for estimating regression and correlation parameters. Results: Regression and correlation parameters were estimated simultaneously. A set of program was finished and an example was illustrated. Conclusion: Longitudinal data often occur in medical researches and clinical trials. To solve the problem of correlation between repeated measures, it is necessary to use some special methods to cope with this kind of data.