The problem of estimating the parameters of the Burr distribution is considered using logarithms or some order statistics. One of the parameters is eliminated by finding the ratio of the logarithm of some ordered statistics and estimate the second parameter. Asymptotic variances and covariances of the estimators are obtained and an efficiency of the estimator of the parameter relative to its maximum likelihood estimate when one of the parameters is known, is obtained. A table is constructed for computational purposes. An example is given to illustrate the application of the estimating equation.