Analysis of the Impact of the Pandemic on Social Inequalities in Enem 2019 and 2020 using Machine Learning

Sep 5, 2026

Upload by Quinn

ENEM assesses the abilities and knowledge of students in secondary education or those who have already completed it, and its scores enable enrollment in SISU, a key pathway to public universities. During the pandemic, planning in schools—especially public ones—was disrupted, leading many students to stop taking ENEM in 2020. This research analyzes social inequalities among ENEM participants using ENEM 2019 and 2020 data with machine learning. Cluster analysis with K-Means and performance classification with Random Forest, K-Nearest Neighbors, and MultiLayer Perceptron are combined with Select K-Best feature selection. Two groups emerge and the MultiLayer Perceptron achieves 85.18% accuracy in 2019 and 83.63% in 2020.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Abstract

Methodology

Results and Findings

Keywords

COMPUTER SCIENCE 10. 5433/1679-0375.2023.v44.48234 Analysis of the Impact of the Pandemic on Social Inequalities in Enem 2019 and 2020 using Machine Learning Análise do Impacto da Pandemia nas Desigualdades Sociais no Enem 2019 e 2020 utilizando Aprendizadode Máquina Bruno da Silva Macedo1 ; Camila Martins Saporetti2 ABSTRACT ENEM measures the ability and knowledge of students who are in high school or have already completed it. With the scores obtained in the exam, the student can enroll in SISU, which is one way to enter public universities. During pandemic, the planning of schools, mainly public, was affected so that many students gave up taking the ENEM in 2020 . To identify the proﬁle of those enrolled in ENEM and verify which portion was most affected, this research analyze their social inequalities using data from ENEM 2019 and 2020 and machine learning methods. The methodology is based on cluster analysis where K-Means was applied and on performance classiﬁcation where Random Forest, K-Nearest Neighbors, and MultiLayer Perceptron were used, and Select K-Best was used to select features. The results of the grouping generated two groups, one composed of subscribers with lower ﬁnancial conditions and another with greater ones. In the classiﬁcation, the MultiLayer Perceptron obtained an accuracy of 85.18% for 2019 and 83.63% for 2020 . The results showed that the proposed methodology was able to identify the differences between the subscribers and classify their performance. keywords ENEM, machine learning, pandemic, social inequalities RESUMO O ENEM mede a capacidade e conhecimento dos estudantes que estão no ensino médio ou já concluíram. Com as notasobtidas no exame, o estudante pode se inscrever no SISU que é uma das formas de entrar nas universidades públicas. Durante a pandemia, o planejamento das escolas, principalmente as públicas, foi afetado de forma que muitos estudantes desistiram de realizar o ENEM em 2020 . Para identiﬁcar o perﬁl dos inscritos no ENEM e veriﬁcar qual parcela foi maisafetada, esta pesquisa analisa as desigualdades sociais usando dados do ENEM 2019 e 2020 e métodos de aprendizadode máquina. A metodologia se baseada na análise de agrupamento, onde K-Means foi aplicado, e na classiﬁcação dedesempenho, onde foram utilizados Random Forest, K-Nearest Neighbors e MultiLayer Perceptron, e o Select K-Best foi empregado para selecionar características. Os resultados do agrupamento geraram dois grupos, um composto por inscritos com menores condições ﬁnanceiras e outro com maiores condições. Na classiﬁcação, o MultiLayer Perceptronobteve uma precisão de 85,18% para o ano de 2019 e 83,63% para 2020 . Os resultados mostraram que a metodologiaproposta conseguiu identiﬁcar as diferenças entre os inscritos e classiﬁcar seu desempenho. palavras-chave ENEM, aprendizado de máquina, pandemia, desigualdades sociais Received: May 19, 2023 Accepted: October 3, 2023 Published: November 6, 2023 1 Student, Computer Engineering, UEMG, Divinópolis, Minas Gerais, Brazil. E-mail: bruno. 1694393@discente.uemg.br 2DSc., Dept. Computational Modeling, IPRJ-UERJ, Nova Friburgo, Rio de Janeiro, Brazil. E-mail: [camila.saporetti@iprj.uerj.br](camila.saporetti@iprj.uerj.br) Macedo, B. da S.; Saporetti, C. M. Introduction The National Secondary Education Examination (Exame Nacional do Ensino Médio-ENEM) was created by the Ministry of Education (MEC) in 1998, with the aim of measuring the ability and knowledge of students who have completed or are still in secondary education. The ENEM is submitted by MEC to help the school build student learning. Since its inception, ENEM has aimed to be more than just a diagnostic assessment of the portrait of Brazilian education, but also to lead individuals to make choices according to their abilities. In addition, it is used as a complementary or substitute exam for other exams to enter the job market and higher education (Santos, 2011) . Currently, ENEM has b

FAQs

What problem does the research address?

The study examines how the COVID-19 pandemic influenced social inequalities among ENEM examinees across 2019 and 2020.

Which machine learning methods are used?

Clustering uses K-Means, and performance classification compares Random Forest, K-Nearest Neighbors, and MultiLayer Perceptron, with Select K-Best used for feature selection.

What are the key results for classification accuracy?

MultiLayer Perceptron reaches 85.18% accuracy for 2019 and 83.63% for 2020, and the clustering separates participants into two groups based on financial conditions.

Sep 5, 2026

Upload by Quinn