A method for assessing computational thinking in students using source code analysis
Steven Pacheco-Portuguez, Antonio González-Torres, Lilliana Sancho-Chavarría, Ignacio Trejos-Zelaya, Jorge Monge-Fallas, José Navas-Sú, Alberto J. Cañas, Andres Rodriguez, Carol Angulo Chinchilla · 2022 International Conference on Advanced Learning Technologies (ICALT) · 2022
This paper summarizes research in progress, at a national scale, that analyzes large volumes of elementary school and high school projects to assess the development of skills, attitudes, and practices that students develop when solving problems via computer programming, grounded on their under-standing of fundamental concepts in computing. The approach is independent of programming languages and uses a generic abstract syntax tree and projects’ metadata to calculate metrics, and establish relationships between measures, school regions, educational centers, and student groups. The research relates source code analysis using abstract syntax trees to some of the main computational thinking concepts. Preliminary results obtained using the proposed method are presented and discussed.