Crime,time, and punishment: An exploration of selection bias in sentencing research

The sentencing decision reflects the culmination of a long series of processing and, thus, selection decisions, with cases leaving the system at each decision point. Accordingly, the substantive implications of bias due to sample selection are of particular concern for sentencing research. In an effort to assess the existence and manifestations of selection bias, the sentencing decision is modeled for three samples, each of which was selected from different stages of the justice process. Event-history data on felony arrests in the State of California over a 3-year period are used, along with a relatively simple analytic technique which reduces such bias. Results indicate that biasis introduced when censored observations are excluded from the analyses. Also, the effects of certain exogenous variables on sentence length differ, depending upon the selection criteria. Of these, the influence of pleading guilty rather than going to trial is especially interesting. Overall, our findings are consistent with the possibility that selectivity bias has concealed effects of sentence bargaining in some earlier studies.The data utilized in this study were collected and made available by the State of California Department of Justice, Bureau of Criminal Statistics. The Department of Justice bears no responsibility for the analyses or interpretations presented here.