Análise visual do percurso acadêmico de estudantes ao longo do ensino superior

• Ano de defesa: 2022
• Autor(a) principal: Lunardelli, Fernando
• Orientador(a): Manssour, Isabel Harb
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Pontifícia Universidade Católica do Rio Grande do Sul
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação
• Departamento: Escola Politécnica
• País: Brasil
• Palavras-chave em Português: Análise Visual; Ensino Superior; Percurso Acadêmico; Diagrama de Sankey; Evasão; Modelo de Predição
• Palavras-chave em Inglês: Visual Analytics; Higher Education; Academic Path; Sankey Diagram; Dropout; Prediction Model
• Área do conhecimento CNPq: CIENCIA DA COMPUTACAO::TEORIA DA COMPUTACAO
• Link de acesso: https://tede2.pucrs.br/tede2/handle/tede/10327

According to the Census of Higher Education in Brazil, dropout in undergraduate courses is a problem getting worse every year in higher education institutions. However, constant and unified analysis of the student’s path can help improve this scenario, enabling understanding or predicting when these courses will not be completed. Nonetheless, analytical tools are needed to facilitate these follow-ups and make decision-making feasible. In this context, the present work proposes creating a data visualization model that allows the analysis of the academic path of one or more students during higher education. Through the exploration of data and statistical analysis, this model aims to identify individuals, or groups of individuals, with a tendency to not complete their courses successfully, in addition to allowing a “view of the whole” concerning their academic career and key indicators. In this way, it seeks to help decision-makers of educational institutions (administrators, educators, technical managers, etc.), in conducting guidelines, applying policies, and other actions, which minimize the conditions that lead these students to drop out. The proposed model, centered on a visualization that uses a Sankey diagram connected to an evasion prediction model, and its implementation, were based on the requirements identified from a systematic literature review, the implementation of a prototype, and interviews with four domain experts. The implementation of the model was also validated through interviews with four domain experts, who considered it adequate to contribute to the improvement of student progress monitoring.