This research illustrates a new procedure of analyzing data from soil- or agar-block decay tests using non-linear regression techniques to estimate a toxic threshold retention point for wood preservative chemicals. The model can be used to provide an objective estimate of the level of retention that is fully adequate, and can be applied satisfactorily to decay data from laboratory decay tests. This so-called broken line model procedure analyzes the decay data simultaneously by separating it into two different parts according to the threshold retention point. Such a procedure is preferable to previous methods which suffer from the need of transforming decay data. Our analysis can be performed in any statistical analysis package that includes a non-linear regression procedure, regardless of having a small or large data set, under the assumption that residuals are normally distributed with a homogenous variance. The method of estimating weight loss due to decay and the one due to operational procedure are also described.