This article presents a direct comparison of three "new" parametric techniques for contingency table analysis--linear probability, logit, and log-linear modeling. The comparisons are rendered as nontechnical as possible, so that those familiar with ordinary regression will readily see how each of these techniques is analogous to regression in cases where one variable is deemed dependent on the others. The paper highlights the availability of linear probability modeling, a technique which has been largely ignored by sociological practitioners despite some attractive features. It also describes eight common errors in published applications of log-linear modeling.