This article describes a simulation activity that can be used to help students see that the estimator "S" is a biased estimator of [sigma]. The activity can be implemented using either a statistical package such as R, Minitab, or a Web applet. In the activity, the students investigate and compare the bias of "S" when sampling from different populations. In the case of sampling from a Normal population, the exact bias of "S" is derived. For advanced students, a nonlinear regression method can be used in the activity to obtain an approximation to the bias of "S" when sampling from non-Normal populations (see the R code on the web). The activity can be easily modified to investigate the bias and possibly other properties (such as consistency) of other statistics, providing teachers a flexible pedagogical tool. (Contains 4 figures and 1 table.)