BI-based Methodology for Analyzing Higher Education: A Case Study of Dropout Phenomenon in Information Systems Courses

• André Andrade Menolli UENP
• Flávio Horita UFABC
• José Jorge Dias UFPB
• Ricardo Coelho UENP

DOI: https://doi.org/10.5753/sbsi.2020.13771

Resumo

Through a Design Science Research method, this paper presents a BI-based methodology to analyze the Information System (IS) courses, in particular the dropout phenomenon and their enrolled students, using open public data from the Anísio Teixeira National Institute for Educational Studies and Research. Data related to Higher Education in Brazil were analyzed and understood, to create a dimensional model. Based on this, a DataWarehouse was created, and the original data was then extracted, transformed and loaded, and a Business Intelligence (BI) environment was deployed and set. This BI environment was utilized for carrying out several descriptive statistics analyses about the dropout phenomenon. Results showed differences in dropout rates in the dimensions of race, type of city where the course is located, type of institution (public or private), teaching modality (on-site or e-learning), year of entry into course, and course period. Thus, this paper increase the knowledge about IS courses and encourages discussion about dropout phenomenon.

Palavras-chave: Information System Courses, Higher Education, Business Intelligence, Dropout, Data Model, OLAP, ETL process.