Abstract To explore the possibility that robust estimators of location can be adapted to the problem of estimating the ED50 in binary-response bioassay, 10 estimators are compared. A Monte Carlo study is conducted to determine the mean squared errors (MSE) of the estimators. Taking into account the MSE for a variety of symmetric tolerance distributions, the sensitivity of the estimator to an anomalous response, and the possibility that the estimator is incalculable, the trimmed Spearman-Karber estimator is especially reliable. Adjusting the response proportions before calculating the ED50 significantly reduces the MSE of the minimum logit chi-squared estimator.