INCORPORATING CO-OFFENDING IN SENTENCING MODELS - AN ANALYSIS OF FINES IMPOSED ON ANTITRUST OFFENDERS

Analyses of sentencing (and other criminal justice processes such as t he decision to prosecute, plea bargaining, and contact with the police ) often use the isolated individual as the unit of analysis. However, the criminal justice system often processes either offenses or court c ases rather than persons. If court cases always involved one individua l, this would have little impact. However, offenses involving co-offen ding-two or more persons acting together-comprise a substantial propor tion of criminal activity (Reiss, 1980, 1986). Depending on the preval ence of co-offending, it may be very likely that two or more individua ls involved in the same case will be selected as members of the same s ample of criminal justice or criminological data. Unless it can be sho wn that both the individual-level variables of co-offenders and their error terms are mutually independent, analyses based on methods such a s ordinary least-squares multiple regression would violate the underly ing assumptions of such models. However, alternatives to linear models assuming either type of independence are available. Among the most us eful of these are mixed models, specifically those assuming compound s ymmetry. This is illustrated with an analysis of fines imposed on crim inally convicted antitrust offenders. These models may yield results w hich are substantially different than those from models which ignore c o-offending. In a model of fines imposed on antitrust offenders, model s which ignore co-offending generally overstate both estimates and sta tistical significance of offense-level variables and understate those of offender-level variables.