This study examines the methodological and practical challenges of using Education Management Information System (EMIS) data for higher education research within the framework of the StdPerson - Student Persistence and Degree Completion in Higher Education: Theoretical and Empirical Models project. The study draws on large-scale microdata, notably the RAIDES database provided by the Portuguese Directorate-General for Education and Science Statistics (DGEEC), to move beyond traditional descriptive analyses based on aggregated indicators (Ferrão, 2023; Ferrão & Almeida, 2018, 2019a). A central methodological challenge lies in the hierarchical nature of educational data, in which students are nested within courses, departments, and institutions. This structure violates the independence assumption of standard regression models and requires the application of multilevel logistic and random coefficient models to properly capture variability across programs and institutional contexts. Longitudinal data integration constitutes another major challenge. Analysing student persistence and degree completion demands the linkage of pseudonymized enrollment records with graduation files across multiple academic years using unique identifiers. This process is technically demanding and must comply with strict data protection and GDPR requirements (Fazendeiro et al., 2025). Moreover, defining “dropout” is conceptually complex, as researchers must differentiate between students who permanently exit higher education and those who temporarily interrupt their studies, transfer between programs, or move across institutions—each reflecting distinct trajectories and determinants (Ferrão & Almeida, 2019b, 2021). Additional challenges include handling missing data, which often arise from alternative access pathways or incomplete academic records, potentially undermining model robustness. Finally, the StdPerson project underscores the limitations of single-institution analyses and highlights the value of integrating nationwide EMIS microdata to produce findings that are generalizable at the national level. By addressing these methodological constraints, the project establishes a robust empirical framework for identifying long-term predictors of student success—such as prior school failure and first-choice admission—thereby informing institutional policies aimed at promoting equity and the democratization of success in higher education.

Fazendeiro, P., Prata, P., & Ferrão, M. E. (2025). Subtle biases introduced in equity studies through data anonymization. PLoS ONE, October 8,. https://doi.org/https://doi.org/10.1371/journal.pone.0332441 Ferrão, M. E. (2023). Differential effect of university entrance scores on graduates’ performance: the case of degree completion on time in Portugal. Assessment & Evaluation in Higher Education, 48(1), 95–106. https://doi.org/10.1080/02602938.2022.2052799 Ferrão, M. E., & Almeida, L. S. (2018). Multilevel modeling of persistence in higher education. Ensaio: Avaliação e Políticas Públicas Em Educação, 26(100), 664–683. https://doi.org/10.1590/s0104-40362018002601610 Ferrão, M. E., & Almeida, L. S. (2019a). Differential effect of university entrance score on first-year students’ academic performance in Portugal. Assessment & Evaluation in Higher Education, 44(4), 610–622. https://doi.org/10.1080/02602938.2018.1525602 Ferrão, M. E., & Almeida, L. S. (2019b). Student’s access and performance in the Portuguese higher education: Issues of gender, age, socio-cultural background, expectations, and program choice. Avaliação: Revista Da Avaliação Da Educação Superior (Campinas), 24(2), 434–450. https://doi.org/10.1590/s1414-40772019000200006 Ferrão, M. E., & Almeida, L. S. (2021). Persistence and academic expectations in higher education students. Psicothema, 33(4), 587–594. https://doi.org/https://doi.org/10.7334/psicothema2020.68