This paper outlines key assumptions of the Tobit model, and then highlights the risk of violating these assumptions. The study reviews alternative flexible parametric and semiparametric modeling techniques, currently used sparingly in criminology, which it suggests researchers may find helpful when assumptions regarding the error terms are untenable. It is noted that many dependent variables of criminological interest have censored distributions, and that investigations that use such variables have increasingly turned to the Tobit model, a censored regression technique that is specified based on a latent dependent variable. It is indicated that, when used under suitable circumstances, this model provides appropriate estimates. By using an empirical example focused on sentencing outcomes and comparing estimates across analytic methods, this study sought to illustrate the potential utility of simultaneously estimating the Tobit model along with some alternatives. The data for this empirical work was from the Federal Justice Statistics Program U.S. Sentencing Commission File, which provided 42,107 cases. From this data file, a random sample of 25 percent was drawn and refined. Tables, figures, appendix, and references