This report investigates an area of uncertainty in using the Angoff method for setting standards, namely whether or not a judge's conceptualizations of borderline group performance are realistic. Ratings are usually made with reference to the performance of this hypothetical group, therefore the Angoff method's success is dependent on this point. A technique whereby the conceptualization of borderline performance can be defined in reference to the performance of the total group is demonstrated. If total group performance is more familiar to judges, either through past experience or training provided during the standard setting activity, use of total group performance to provide a frame of reference within which to define group performance may lead to more realistic item ratings. (Author/DWH)