Loading...
19 results
Search Results
Now showing 1 - 10 of 19
- Estudo Longitudinal sobre Eficácia Diferencial e Equidade Social no BrasilPublication . Ferrão, Maria EugéniaO estudo de eficácia diferencial e equidade social entre o 5º e 9º ano do ensino funda-mental segue orientações teórico-metodológicas de eficácia educacional. Modelos de coeficientes aleatórios são aplicados aos dados longitudinais da Prova Brasil 2011 em-parelhados com os da Prova Brasil 2015, envolvendo mais de 1.2 milhões de alunos. A proficiência aferida no 9ºano e conhecimento prévio aferido no 5ºano, na escala SAEB, foram padronizados para o propósito do artigo. Foram testadas ainda variáveis socio-demográficas do aluno, tais como sexo, etnia/raça autodeclarada, nível socioeconómi-co, situação face ao trabalho e se a mãe é ou não alfabetizada. Os resultados sugerem que a relação linear entre conhecimento prévio e proficiência no 9º ano apresenta va-riação ténue entre escolas, municípios ou entre UFs e o mesmo acontece com a relação entre nível socioeconómico e proficiência do 9º ano, apesar de alguns dos parâmetros aleatórios serem estatisticamente significativos. Os indicadores de equidade social e eficácia diferencial por conhecimento prévio variam pouco. Os resultados reforçam a relevância da investigação institucional, especialmente quanto aos fatores intra-es-colares, porque a razão entre a variância do nível Aluno e o somatório de variâncias de nível Escola-Município-UF é maior de 10 em Português e quase 8 em Matemática.
- Student’s access and performance in the Portuguese Higher EducationPublication . Ferrão, Maria Eugénia; Almeida, Leandro S.The purpose of this article is to characterize and contribute to the debate on the democratization of Portuguese higher education, both in terms of access and the performance of students enrolled in a public university. The analyses concern the sociodemographic characteristics and schooling trajectory of the 2,697 students enrolled for the first time in the University of Minho in the academic year 2015/16. The relationships between such characteristics and the choice of program, expectations regarding higher education, the criteria of admission, and the association with their permanence and performance in the first year of studies are explored as well. Several statistical tests were applied, such as those based on multivariate analysis of variance, chi-squared test for the independence between variables, or the tStudent test for the comparison of means of two independent samples. Results suggest that student’s gender, socio-cultural background and schooling trajectory are related to the choice of the programe, university entrance score and the entrance option. The multivariate analysis of variance of student’s grade point average at the end of the first year suggests the influence of the interaction between the fixed term of scientific-disciplinary area of the program attended and the program option of access to higher education. We did not find any statistically significant association between socio-cultural background and permanence in higher education; i.e, the socio-cultural origin of the students does not seem to influence the decision to abandon, suspend or transfer program, at least during their first year of studies. Our findings suggest student’s resilience and/or institutional action meaning a step further on the path for social equity in the Portuguese higher education.
- Imputação de Valores Omissos em Análise Descritiva de Dados, em RPublication . Salambiaku, Luzizila; Prata, Paula; Ferrão, Maria EugéniaOs valores omissos representam um problema frequente no processo de análise de dados. Neste artigo foram comparados seis métodos distintos de imputação, disponíveis no software R e avaliado o seu desempenho em conjuntos de dados relacionados com a área da educação. Foi estudada uma amostra de 20408 estudantes para testar os seis algoritmos em quatro conjuntos de dados gerados por simulação com diferentes percentagens de valores omissos, considerando 5%, 10%, 15% e 20% nas variáveis de interesse. Foram explorados métodos de imputação simples (Média, Mediana e Moda), métodos baseados em aprendizagem automática (kNN e bPCA) e um método de imputação múltipla (MICE). Foi avaliado o desempenho de cada método calculando os respetivos erros de imputação através as métricas RMSE e MAE. Os resultados obtidos mostram que a imputação pela Moda forneceu quase de forma constante menores valores de erro.
- Anonymized Data Assessment via Analysis of Variance: An Application to Higher Education EvaluationPublication . Ferrão, Maria Eugénia; Prata, Paula; Fazendeiro, PauloThe assessment of the utility of an anonymized data set can be operational-ized by the determination of the amount of information loss. To investigate the possible degradation of the relationship between variables after anony-mization, hence measuring the loss, we perform an a posteriori analysis of variance. Several anonymized scenarios are compared with the original data. Differential privacy is applied as data anonymization process. We assess data utility based on the agreement between the original data structure and the anonymized structures. Data quality and utility are quantified by standard metrics, characteristics of the groups obtained. In addition, we use analysis of variance to show how estimates change. For illustration, we apply this ap-proach to Brazilian Higher Education data with focus on the main effects of interaction terms involving gender differentiation. The findings indicate that blindly using anonymized data for scientific purposes could potentially un-dermine the validity of the conclusions.
- Utility-driven assessment of anonymized data via clusteringPublication . Ferrão, Maria Eugénia; Prata, Paula; Fazendeiro, PauloIn this study, clustering is conceived as an auxiliary tool to identify groups of special interest. This approach was applied to a real dataset concerning an entire Portuguese cohort of higher education Law students. Several anonymized clustering scenarios were compared against the original cluster solution. The clustering techniques were explored as data utility models in the context of data anonymization, using k-anonymity and (ε, δ)-differential as privacy models. The purpose was to assess anonymized data utility by standard metrics, by the characteristics of the groups obtained, and the relative risk (a relevant metric in social sciences research). For a matter of self-containment, we present an overview of anonymization and clustering methods. We used a partitional clustering algorithm and analyzed several clustering validity indices to understand to what extent the data structure is preserved, or not, after data anonymization. The results suggest that for low dimensionality/cardinality datasets the anonymization procedure easily jeopardizes the clustering endeavor. In addition, there is evidence that relevant field-of-study estimates obtained from anonymized data are biased.
- Multilevel modeling of persistence in higher educationPublication . Ferrão, Maria Eugénia; Almeida, Leandro S.The dropout or evasion rates in higher education are now a social and institutional concern, justifying the implementation of public policies to prevent this phenomenon. These policies need studies on the most determinant variables of the risk of dropout. The main objective of this study is to analyze the student’s persistence in undergraduate courses, and the relationship with the student’s previous school trajectory and with the conditions of entrance into higher education, controlling for students’ sociodemographic characteristics, such as gender and age. We applied multilevel logistic regression models to data of 2.697 freshmen enrolled in a Portuguese public university in the academic year 2015/16. The results suggest that failure in basic education (ISCED 2) has a long-term effect. According to the estimates obtained, students who declare not having failed in basic education have odds ratio of persistence 2.7 times higher than students who declare having failed in basic education. The conditions of student’s admission to the course he/she attends are relevant variables to persistence in Higher Education, for example, whether s/he was admitted to her/his first option course and the student’s university entrance score. The results also show that older and male students have lower probability of persistence.
- Data Anonymization: K-anonymity Sensitivity AnalysisPublication . Santos, Wilson; Sousa, Gonçalo; Prata, Paula; Ferrão, Maria EugéniaThese days the digitization process is everywhere, spreading also across central governments and local authorities. It is hoped that, using open government data for scientific research purposes, the public good and social justice might be enhanced. Taking into account the European General Data Protection Regulation recently adopted, the big challenge in Portugal and other European countries, is how to provide the right balance between personal data privacy and data value for research. This work presents a sensitivity study of data anonymization procedure applied to a real open government data available from the Brazilian higher education evaluation system. The ARX k-anonymization algorithm, with and without generalization of some research value variables, was performed. The analysis of the amount of data / information lost and the risk of re-identification suggest that the anonymization process may lead to the under-representation of minorities and sociodemographic disadvantaged groups. It will enable scientists to improve the balance among risk, data usability, and contributions for the public good policies and practices.
- A Measure of Child Exposure to Household Material Deprivation: Empirical Evidence from the Portuguese EU-SILCPublication . Ferrão, Maria Eugénia; Bastos, Amélia; Alves, Maria Teresa G.Although monitoring and evaluating child poverty has been recognized as important, there is little statistical information focused on children. Because the annual EUStatistics on Income and Living Conditions (EU-SILC) survey does not include child-specific information on an annual basis, this study proposes a measure of child exposure to household material deprivation based on this dataset. The study considers four domains of deprivation that have a direct impact on child development: housing conditions, household financial capacity, household durable goods, and environmental living conditions. Although developing a child-centered measurement of child deprivation is important, the EU-SILC considers the household as the unit of measurement. Therefore, our proposal is household-based, allowing annual monitoring of children’s exposure to deprivation—an important insight for social policy purposes to tackle the problem of child poverty. Using the 2017 Portuguese sample, we applied graded response models to assess the psychometric properties of the EU-SILC items and fit separate indexes per domain and the composite index. Item selection was based on their characteristic curves and information functions. The results allow for the selection of more informative items for every domain to obtain the composite index. In general, the empirical analysis confirmed the theoretical approach for item selection. The methodology may be directly applied to the full EU dataset or to each country individually.
- Analysis of grade repetition through multilevel models: A study from PortugalPublication . Bastos, Amélia; Ferrão, Maria EugéniaThis paper investigates grade repetition in Portugal using microdata. Drawing on multilevel models, we analyse the number of times the student repeated a grade in compulsory education – our dependent variable – in association with children’s individual characteristics, household sociodemographic and economic background, and children’s living conditions – our covariates. Furthermore, we also attempt to shed light on the impact of schools on the endogenous variable. Our results confi rm the importance of individual, family, and neighbourhood characteristics on the rate of grade repetition. In terms of schools, the results obtained show that the student’s probability of failure vary across schools, demonstrating the importance of the impact of the school itself on grade repetition.
- Differential effect of university entrance score on first-year students’ academic performance in PortugalPublication . Ferrão, Maria Eugénia; Almeida, Leandro S.The main goal of this study is to show that the association between university entrance score and first-year students’ academic performance varies randomly across courses after controlling for students’ sociodemographic, schooling trajectory and motivational variables. The sample consists of 2697 first-year students who were enrolled in 54 courses at a Portuguese public university in 2015/16. Multilevel modelling of academic performance suggests that 34% of variability in grade point average is due to differences among courses and that 80% of such variability is explained by the field of study, whether the university is the student’s first choice, and the student’s gender, age and parents’ level of education. In addition, the results corroborate that the university entrance score is the strongest predictor of first-year academic performance.