We have investigated the effect of peer-review on the development of students' problem-solving abilities in an introductory physics course. Specifically, we report the results of a multiple-group pre/post-test quasiexperiment comparing two groups receiving different treatments with respect to activities completed after cooperative group problem-solving (CGPS) sessions. Both the treatment and control groups received identical instruction in lecture and CGPS discussion sections. Individual student problem solutions submitted on-line were evaluated via a validated problem-solving process rubric that was supplied to all participants at the beginning of the course. The treatment group completed rubric-based peer-review on-line for three randomly selected students, whereas the control group did not. Reliability-corrected analysis of covariance showed that the treatment group demonstrated greater improvement in problem-solving process compared to the control over the four-week study period.