New York's 1978 Juvenile Offender Act created a new class of 13- 15-year-old defendants processed in the adult court system. This research isolated those factors which differentiated such offenders with one or more pretrial warrants from those warrant-free. These factors were used in a point scale to provide judges with specific pretrial release recommendations. The research population consisted of all 'juvenile offenders' (JO's) receiving prearraignment interviews between June 21 and December 31, 1982, and later arraigned in criminal court (501). In addition, a 50-percent random sampling of 16-year-olds arrested during the same period was studied so the JO's could be compared with the closest demographic group receiving adult recommendations (1,670). The prearraingment interview administered to all subjects included questions on family ties, home responsibilities, sources of financial support, and educational and employment history. These data were analyzed in conjunction with criminal and supreme court appearance history and warrant information. Warrant rates were consistently lower for JO's than for 16-year-olds. The warrant-occurrence variable correlated significantly with prior criminal record, school attendance, and defendants' arraignment expectations regarding the presence of a friend or relative at the hearing. The development and testing of the scale derived from these factors are described. 45 tables, interview form, and 15 references. ABI rcb