“Learn to Argue” and “Argue to Learn”: meta-analysis of effective instructional design for online scientific argumentation activities

Online scientific argumentation activities have significant value for students’ cognitive ability. The purpose of this study was to solve the issues of online scientific argumentation. Firstly, the influences of online scientific argumentation were classified into “Learn to Argue” and “Argue to Learn” abilities. And then the technological empowerment scientific argumentation teaching model was constructed. Second, a total of 5415 students’ data from 46 studies were captured. It is proved that online argumentation activities have an upper-middle positive effect on students’ “Learn to Argue” (Hedges’ g = 0.603, P = 0.000) and “Argue to Learn” (Hedges’ g = 0.596, P = 0.000) abilities. Finally, the results of moderator analysis showed that the natural science issues have the greatest effect. Compared with higher education, online scientific argumentation is of greater value in K-12 classroom. Moreover, the CRE framework has been proved to have greater application potential. Virtual reality game platform and game based learning can more effectively promote students’ “Learn to Argue” and “Argue to Learn” abilities. Besides, intelligent evaluation based on machine learning has been proved to be a significant way to evaluate students’ argumentation abilities. Based on this, this study made discussion to provide valuable reference for practitioners and researchers.