This study evaluates an interactive web-based homework designed to increase student understanding in a principles of microeconomics course. Employing concepts from Action Research, the preliminary investigation was undertaken based on assumptions about the efficacy of online educational resources. Do students who do well on online homework improve performance on exams? Which characteristics of students explain achievement differences? The authors address these questions in relation to the use of a web-based, interactive online homework service which accompanied the text. An econometric model using exam performance as the dependent variable and various course component and demographic characteristics as independent variables. The study found no statistically significant impact of online homework on the mean exam score. This preliminary study indicated no gender, ethnicity, class standing, or declared major impact on exam performance though the limited degrees of freedom make generalization questionable. The only statistically significant predictor was shown to be GPA.