We share results from a quasi-experimental study in which we compared achievement between traditional lecture-based and computer-based sections of college algebra on a common multiple choice exam as well as performance on problem solving items. Students in the computer-based group performed better on the final exam and were also more likely to complete the problem solving tasks correctly than students in the traditional courses. However, the computer-based group showed limited ability to interpret an equation and relate it to a contextual situation. [For the complete proceedings, see ED584443.]