Effects of peer interaction facilitated by computer-mediated conferencing on learning outcomes

Social interaction has always been a defining characteristic of education, training, and more generally of learning. Interaction with others offers the potential for collaboration, negotiation of meaning, mutually shared knowledge, and the joint construction of solutions that otherwise could not be produced alone. Computer-mediated conferencing (CMC) gives designers and instructors of distance education programs the potential to design courses incorporating interaction among class members. While many studies report the advantages and disadvantages of incorporating CMC into the design of distance education, few studies report whether achievement of effective learning outcomes were facilitated by peer interaction. Consequently, many researchers suggest our pervasive adoption of CMC has far outpaced our understanding of what makes peer interaction good and when peer interaction is good. This study addressed that concern and tested a model of five variables—learner characteristics (measured by self-construal), teaching style, task design, course requirements, and prior CMC experience—and their impact upon peer interaction and learning outcomes. The model was derived from an extensive review of the literature and tested with structural equation modeling. Adjustments were made to the original five-construct model based on theory and developments in the research process. A simplified three-construct model and two-construct model revealed the significant effect of peer interaction on learning outcomes. The two-construct model of peer interaction in an online course was a better fit to the data. The two-construct model showed a path coefficient of .66 from peer interaction to learning outcomes. This finding indicated that for every one standard deviation increase in peer interaction, there was a .66 standard deviation change in learning outcomes—a large, strong effect. The variables, teaching style and prior CMC experience, plus peer interaction accounted for 65% of the variance in learning outcomes. When peer interaction was temporarily dropped from the two-construct model, teaching style and prior CMC experience accounted for only 29% of the variance in learning outcomes. While the literature has pointed to and suggested this relationship, this study has been able to show (a) the relationship between peer interaction and learning outcomes and (b) its significance.