This study measured the impact of homework and student homework preferences on the learning of 9th grade geography students of average and below average reading levels. Quiz averages were compared between two student groups that had alternating periods of exogenous manipulation, where one group was assigned homework while the alternate group was not assigned homework. Surveys determined the students' perceived gaps between preferred homework conditions and actual homework conditions. These gaps were then compared to quiz averages and homework completion rates. Data showed that quiz averages remained similar whether students were assigned homework or not, and that gaps between actual and preferred homework conditions had minimal and inconsistent effects on quiz scores. Homework seemed to have little relationship to learning. (Contains 1 figure and 1 table.)