[A method for describing monotonic dose-effect curves and areas in yes/no reactions (author's transl)].

A method combining a polynomial statistical model and multiple probit regression, is useful in describing nonmonotonic dose-effects in yes/no-reaction variables. An example shows the application of this method to the combination of two substances. The effects can be described with a minimum of polynomial coefficients. The application of a simple X2-test allows to test for additivity of substance effects by testing whether interaction coefficients improve the fit of the model significantly or not.