Students at a large urban community college enrolled in seven classes of an experimental remedial algebra programme, which integrated study skills instruction and collaborative problem solving. A control group of seven classes was taught in a traditional lecture format without study skills instruction. Student performance in the course was measured by a common 25-question multiple-choice exit exam. After controlling for cognitive and affective differences among students (N = 233), when all other variables were held constant, students in the experimental classes answered two more questions out of 25 correctly than students in the control group classes. The results suggest problem solving and study skills are potential areas for further improvement in learning.