Using Network Meta-Analysis to Synthesize Research on Computer-Based Scaffolding in STEM Education

• In Event: Scaffolding STEM Learning With Technology

Abstract

This paper reports results of a network meta-analysis on computer-based scaffolding in problem-centered, STEM education. Network meta-analysis considers within subjects variance in addition to between subjects variance, which potentially allows for more experimental literature to be included in a particular analysis than in a traditional meta-analysis. We included 333 outcomes from 144 studies. Moderators include assessment level (concept, principles, and application) and scaffolding change (fading, adding, fading/adding, and none). Results indicated that the effect size of scaffolding was greater when measured at the principles assessment level than at either the application or concept level. When computer-based scaffolding is faded, the effects on learning outcomes are greatest as compared to other scaffolding customization (i.e. adding and fading & adding supports).

Authors

• Andrew Walker, Utah State University

• Brian R. Belland, Utah State University

• Nam Ju Kim, Utah State University

• Mason Lefler, Utah State University

• Brett M Whitney, Utah State University

• Lindi Andreasen, Utah State University