Modeling the effects of victim behavior and moral character on prosecutors' charging decisions in sexual assault cases.

Prior research, modeling the effects of the victim's behavior and character on prosecutors' charging decisions, has used either a dichotomous variable that reflects the presence of any risky behavior or moral character issues or an additive index that captures the number of related items in a case file. We suggest that these measures do not adequately identify the specific issues that prosecutors take into consideration when making charging decisions. Using data on 666 sexual assault cases that resulted in arrest in three urban jurisdictions and a multivariate modeling strategy, we examine specific risk-taking behaviors and issues related to the victim's moral character in an effort to determine if certain behaviors and characteristics have a more substantial effect on charging decisions than others. We also examine the extent to which the effects of these blame and believability factors vary based on the nature of the cases. Our results reveal that although charging decisions in stranger cases are largely determined by legally relevant factors, these decisions in nonstranger cases are affected by several legally irrelevant victim characteristics: whether the victim had a prior criminal record, whether the victim had been drinking alcohol prior to the assault, and whether the victim invited the suspect to her residence. Further analysis, however, revealed that only the victim's prior record had a differential effect on charging decisions in cases involving strangers and nonstrangers and in aggravated and simple rape cases. Our results suggest that the focal concerns that guide prosecutors' charging decisions incorporate specific victim behaviors and background characteristics.