Using social network analysis to measure the effect of learning analytics in computing education
- ,
- Rafael Ferreira Leite De Melloa(Author),
- Olusola Adesopea(Author),
- Vitor Rolimb(Author),
- Dragan Gasevicc(Author),
- Christopher Hundhausena(Author)
- aWashington State University Pullman,
- bUniversidade Federal Rural de Pernambuco,
- cMonash University
Abstract
Student retention and learning in STEM disciplines is a growing problem. The 2012 report by the US President's Council of Advisors on Science and Technology (PCAST) predicts a future deficit in science, engineering, and mathematics (STEM) in the following decade and emphasizes the importance of addressing this issue. With this as a motivating factor, the OSBLE+ Social Programming Environment (SPE) was used to leverage social and programming data for the basis of automatically generated prompts inserted into the SPE. These prompts were designed to stimulate help-seeking, help-giving, and social interaction in the learning environment. A social network analysis was performed in order to determine whether exposure to the automated interventions would positively affect the relationship among students over time. Results of this study suggest that students in the experimental treatment who were presented with automated prompts developed more connected and social networks than those in the control treatment.