Predicting National School Performance for Policy Making in South Africa

Wandera, Henry; Marivate, Vukosi; Sengeh, Moinina David

doi:10.1109/iscmi47871.2019.9004323

Cited by 4 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…There was an effort to predict students’ grades in order to provide support and personalization ( Blasi, 2017 ). For instance, Wandera et al (2019) proposed several models to predict school pass rate in order to support higher-level decision making. Similarly, Aguiar et al (2015) developed performance prediction models to help schools more efficiently allocate limited resources by prioritizing students who are most in need of help and target intervention programs to match those students’ particular needs.…”

Section: Resultsmentioning

confidence: 99%

“…The classifiers reached an accuracy higher than 95%. In the same direction, Wandera et al (2019) and Blasi (2017) applied deep neural network architectures to predict the final grades of high school students. Both papers reached accuracy of approximately 90%.…”

Section: Resultsmentioning

confidence: 99%

“…Both papers reached accuracy of approximately 90%. However, Wandera et al (2019) also included the SHAP (SHapley Additive exPlanations) analysis ( Lundberg and Lee, 2017 ) to provide insights into the most relevant features for the problem.…”

Section: Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Applications of Learning Analytics in High Schools: A Systematic Literature Review

Sousa

Alexandre

Mello

et al. 2021

Front. Artif. Intell.

View full text Add to dashboard Cite

Learning analytics aims to analyze data from students and learning environments to support learning at different levels. Although learning analytics is a recent field, it reached a high level of maturity, especially in its applications for higher education. However, little of the research in learning analytics targets other educational levels, such as high school. This paper reports the results of a systematic literature review (SLR) focused on the adoption of learning analytics in high schools. More specifically, the SLR followed four steps: the search, selection of relevant studies, critical assessment, and the extraction of the relevant field, which included the main goals, approaches, techniques, and challenges of adopting learning analytics in high school. The results show that, in this context, learning analytics applications are focused on small-scale initiatives rather than institutional adoption. Based on the findings of this study, in combination with the literature, this paper proposes future directions of research and development in order to scale up learning analytics applications in high schools.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Applications of Learning Analytics in High Schools: A Systematic Literature Review

Sousa

Alexandre

Mello

et al. 2021

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…Los árboles de decisión, las redes neuronales y los modelos multinivel, se constituyen en las terceras técnicas estadísticas de preferencia usadas para la comprensión o explicación del LA. Los árboles de decisión han sido utilizados con un enfoque principal de comprensión en el LA, así como articulado con otros modelamientos estadísticos para la explicación del LA (Cornell-Farrow & Garrard, 2020;Rebai et al, 2019;Wandera et al, 2019;Xu et al, 2019;Yang et al, 2020).…”

Section: Las Variables a Prioriunclassified

“…La presencia de recursos en el contexto cercano o de vivienda de los estudiantes, puede favorecer el desempeño de estos y en consecuencia, de las escuelas (Wandera et al, 2019). Los estudiantes que disponen de solo computador logran menores puntuaciones en diversas áreas, por el contrario, los que tienen mayor interacción computador-internet, incrementan sus puntuaciones (Ariza et al, 2021;Chacón-Vargas & Roldán-Villalobos, 2021).…”

Section: Las Variables Emergentes En El Logro De Aprendizajeunclassified

Técnicas estadísticas y logro de aprendizaje: revisión bibliográfica

Suárez Riveros,

Pineda Ríos,

Mendivelso Ramírez

2021

ECOMATEMATICO

View full text Add to dashboard Cite

El objetivo de este escrito fue describir las diferentes técnicas estadísticas que han sido empleados para comprender o explicar el logro de aprendizaje, en estudiantes en diferentes niveles educativos. Desde el punto de vista teórico se consolidaron las categorías a priori, provenientes de las técnicas estadísticas (Modelos Multinivel, Modelos geoespaciales, Regresión, Clustering, Análisis Descriptivo, Redes Neuronales, Árboles de decisión, Bosques aleatorios, NaiveBayes y Support Vector Machine), así como la conceptualización de Logro de Aprendizaje. El enfoque metodológico para la revisión se hizo a partir del mapeamiento informacional bibliográfico. Entre los resultados se encontraron 50 documentos de diferentes bases de datos (Elsevier (1), Google Scholar (6), IEEE (4), Scielo (2), ScienceDirect (5), Scopus (31), y Springer (1)), que estudian diferentes regiones del mundo (Asia (17), América del sur (13), América del norte (8), Europa (6), África (5), Oceanía (4), Centro América (3), junto con la orientación a explicar (17), comprender (31) o comprender y explicar (2).Adicionalmente, se identificó un conjunto de variables emergentes en los diferentes reportes, entre las que se encuentra, con mayor relevancia, el nivel socioeconómico, género, afectividad, antecedentes y características y posibilidades de los padres.

show abstract