Automatic question generation about introductory programming code

• Autor(a) principal: Soares, Tiago Filipe Martinho
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/23611
• topic: Introductory programming; Pedagogical tool; Intelligent tutoring systems; Introdução à programação; Ferramenta pedagógica; Sistemas de tutoria inteligentes

Many students who learn to program end up writing code they do not understand. Most of the available code evaluation systems evaluate the submitted solution functionally and not the knowledge of the person who submitted it. This dissertation proposes a system that generates questions about the code submitted by the student, analyses their answers and returns the correct answers. In this way, students reflect about the code they have written and the teachers of the programming courses can better pinpoint their difficulties. We carried out an experiment with undergraduate and master's students in Computer Science degrees in order to understand their difficulties and test the prototype's robustness. We concluded that most students, although understanding simple details of the code they write, do not understand the behaviour of the program entirely, especially with respect to program state. Improvements to the prototype and how to conduct future experiments are also suggested.