A carregar...
Projeto de investigação
Understanding the drivers of academic achievement: Evidence for Portugal’s high school system
Financiador
Autores
Publicações
Deep Learning in Predicting High School Grades
Publication . Costa-Mendes, Ricardo; Cruz-jesus, Frederico; Oliveira, Tiago; Castelli, Mauro; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Societa Italiana di Istochimica / PAGEPress Publications
This paper applies deep learning to the prediction of Portuguese high school grades. A deep multilayer perceptron and a multiple linear regression implementation are undertaken. The objective is to demonstrate the adequacy of deep learning as a quantitative explanatory paradigm when compared with the classical econometrics approach. The results encompass point predictions, prediction intervals, variable gradients, and the impact of an increase in the class size on grades. Deep learning’s generalization error is lower in the student grade prediction, and its prediction intervals are more accurate. The deep multilayer perceptron gradient empirical distributions largely align with the regression coefficient estimates, indicating a satisfactory regression fit. Based on gradient discrepancies, a student’s mother being an employer does not seem to be a positive factor. A benign paradigm shift concerning the balance between home and career affairs for both genders should be reinforced. The deep multilayer perceptron broadens the spectrum of possibilities, providing a quantum solution hinged on a universal approximator. In the case of an academic achievement-critical factor such as class size, where the literature is neither unanimous on its importance nor its direction, the multilayer perceptron formed three distinct clusters per the individual gradient signals.
Using artificial intelligence methods to assess academic achievement in public high schools of a European Union country
Publication . Cruz-Jesus, Frederico; Castelli, Mauro; Oliveira, Tiago; Mendes, Ricardo; Nunes, Catarina; Sa-Velho, Mafalda; Rosa-Louro, Ana; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Elsevier
Understanding academic achievement (AA) is one of the most global challenges, as there is evidence that it is deeply intertwined with economic development, employment, and countries’ wellbeing. However, the research conducted on this topic grounds in traditional (statistical) methods employed in survey (sample) data. This paper presents a novel approach, using state-of-the-art artificial intelligence (AI) techniques to predict the academic achievement of virtually every public high school student in Portugal, i.e., 110,627 students in the academic year of 2014/2015. Different AI and non-AI methods are developed and compared in terms of performance. Moreover, important insights to policymakers are addressed.
Determinants of academic achievement
Publication . Nunes, Catarina; Oliveira, Tiago; Castelli, Mauro; Cruz-Jesus, Frederico; Information Management Research Center (MagIC) - NOVA Information Management School; NOVA Information Management School (NOVA IMS); Elsevier
This study explores the contribution of various drivers of attainment in secondary education in Portugal. We propose a model explaining the influence of students, teachers, and parents' traits on high school achievement, measured by the self-reported Math and Portuguese final grades of 220 students. Using PLS-SEM, we show that previous achievement predicts current achievement in both subjects; however, noteworthy differences were found. Portuguese grades are significantly better for students whose parents have post-secondary education and communicate higher expectations about their offspring's school careers. At the same time, Math achievement is influenced by students' perception of teachers' involvement but not by parents' expectations or education. Previous retention and receiving educational allowance impair Math achievement, but not Portuguese. Results and implications are discussed.
Influence of computers in students’ academic achievement
Publication . Simões, Sofia; Oliveira, Tiago; Nunes, Catarina; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Elsevier
With fast-growing technology, schools have to adapt and use technology constantly as a tool to grow. This study aims to understand the influence of computer factors on students' academic achievement. We propose a model on the influence of computer attitudes, computer learning environments, computer learning motivations, computer confidence, computer use, computer self-efficacy, loneliness, mothers' education, parents' marital status and family size on academic achievement (AA). To validate the conceptual model, 286 students aged 16–18 years old answered an online questionnaire. The most important drivers that positively affect AA are computer use, employment motivations, and mothers' education. While enjoyment attitudes, school environment, interest motivations, and loneliness influence AA negatively. Also, family size and computer self-efficacy work as moderators, and computer use works as a mediator between computer learning environments and academic achievement.
A machine learning approximation of the 2015 Portuguese high school student grades
Publication . Costa-Mendes, Ricardo; Oliveira, Tiago; Castelli, Mauro; Cruz-Jesus, Frederico; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Springer
This article uses an anonymous 2014–15 school year dataset from the Directorate-General for Statistics of Education and Science (DGEEC) of the Portuguese Ministry of Education as a means to carry out a predictive power comparison between the classic multilinear regression model and a chosen set of machine learning algorithms. A multilinear regression model is used in parallel with random forest, support vector machine, artificial neural network and extreme gradient boosting machine stacking ensemble implementations. Designing a hybrid analysis is intended where classical statistical analysis and artificial intelligence algorithms are blended to augment the ability to retain valuable conclusions and well-supported results. The machine learning algorithms attain a higher level of predictive ability. In addition, the stacking appropriateness increases as the base learner output correlation matrix determinant increases and the random forest feature importance empirical distributions are correlated with the structure of p-values and the statistical significance test ascertains of the multiple linear model. An information system that supports the nationwide education system should be designed and further structured to collect meaningful and precise data about the full range of academic achievement antecedents. The article concludes that no evidence is found in favour of smaller classes.
Unidades organizacionais
Descrição
Palavras-chave
Contribuidores
Financiadores
Entidade financiadora
Fundação para a Ciência e a Tecnologia
Programa de financiamento
3599-PPCDT
Número da atribuição
DSAIPA/DS/0032/2018