This study assessed the capacity of a major Problem-Based Learning (PBL) Assignment to improve students' content knowledge surrounding relevant course concepts in two sections of "The Politics of Welfare Reform," a 200-level class offered at the University of Massachusetts-Dartmouth. The hypothesis was that performance on the PBL assignment, which occurred between the midterm and final exams, would raise students' scores from the multiple-choice midterm to the multiple-choice component of the final exam. The models suggest an independent PBL effect of improved exam performance among students with fewer preexisting academic resources in the larger section, and the lack of an independent PBL effect among students in the smaller, more academically prepared section. (Contains 5 tables and 3 notes.)