Ten years of Computer-Supported Collaborative Learning: A meta-analysis of CSCL in STEM education during 2005–2014

The goal of this paper is to report on a meta-analysis about the effects of Computer-Supported Collaborative Learning (CSCL) in STEM education. The analysis is based on 316 outcomes from 143 studies that examined the effects of CSCL published between 2005 and 2014. Our analysis showed that the overall effect size of STEM CSCL was 0.51, a moderate but notable effect size in educational research. The effect was greatest on process outcomes, followed by knowledge outcomes, and affective outcomes. The sizes of the effects were moderated by types of technology and pedagogy, educational levels of learners, and learning domains. Moderators further interacted so that effects of technology and pedagogy varied depending on the modes of collaboration, learners' educational levels, and domains of learning. The current study demonstrates the overall advantage of CSCL in STEM education and highlights a need to understand how these variables may interact to contribute to overall CSCL effectiveness. •A comprehensive meta-analysis of the effects of Computer-Supported Collaborative Learning (CSCL) in STEM education.•The overall effect size of STEM CSCL was 0.51, a moderate effect size notable in educational research.•CSCL effectiveness was moderated by the educational levels of learners, domains of learning, technology and pedagogies.•The effect of CSCL technology and pedagogy was moderated by collaboration types, educational levels, and learning domains.•There is no one-size-fits-all solution in CSCL.