An Exploration of Impact Factors Influencing Students’ Reading Literacy in Singapore with Machine Learning Approaches

Dong, Xin Qi; Hu, Jie

doi:10.5539/ijel.v9n5p52

Cited by 29 publications

(35 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The PISA 2018 assessment does not provide actual reading achievement scores for each student; instead, it measures proficiency in each domain using ten plausible values that represent ten random values drawn from the posterior distribution of the student's scores for reading [6]. We used the first plausible value for the overall reading proficiency; previous studies on the PISA dataset have used only one plausible value [8,9,45] based on the assumption that one plausible value is said to provide unbiased estimates of population parameters. The distribution of students based on their reading level and group is summarized in Figure 2, which also shows that 55% and 45% of the students belonged to the low and high performing groups, respectively.…”

Section: The Datasetmentioning

confidence: 99%

Using Machine Learning Approaches to Explore Non-Cognitive Variables Influencing Reading Proficiency in English among Filipino Learners

et al. 2021

View full text Add to dashboard Cite

Filipino students ranked last in reading proficiency among all countries/territories in the PISA 2018, with only 19% meeting the minimum (Level 2) standard. It is imperative to understand the range of factors that contribute to low reading proficiency, specifically variables that can be the target of interventions to help students with poor reading proficiency. We used machine learning approaches, specifically binary classification methods, to identify the variables that best predict low (Level 1b and lower) vs. higher (Level 1a or better) reading proficiency using the Philippine PISA data from a nationally representative sample of 15-year-old students. Several binary classification methods were applied, and the best classification model was derived using support vector machines (SVM), with 81.2% average test accuracy. The 20 variables with the highest impact in the model were identified and interpreted using a socioecological perspective of development and learning. These variables included students’ home-related resources and socioeconomic constraints, learning motivation and mindsets, classroom reading experiences with teachers, reading self-beliefs, attitudes, and experiences, and social experiences in the school environment. The results were discussed with reference to the need for a systems perspective to addresses poor proficiency, requiring interconnected interventions that go beyond students’ classroom reading.

show abstract

Section: The Datasetmentioning

confidence: 99%

Using Machine Learning Approaches to Explore Non-Cognitive Variables Influencing Reading Proficiency in English among Filipino Learners

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The performance of models with more than 20 is negatively influenced. Second, SVM-based studies in the social sciences have identified 20 to 30 features as a good number for an optimal feature set [ 108 ], and 20 features were selected for inclusion in the optimal feature set [ 95 ]. Therefore, in this study, the top 20 features were selected for subsequent analysis, as proposed in previous analyses that yielded accepted measurement rates.…”

Section: Resultsmentioning

confidence: 99%

“…The SVM contains the following two modules: one module is a general-purpose machine learning method, and the other module is a domain-specific kernel function. The SVM training algorithm is used to build a training model that is then used to predict the category to which a new sample instance belongs [ 95 ]. When a set of training samples is given, each sample is given the label of one of two categories.…”

Section: Methodsmentioning

confidence: 99%

Differentiating the learning styles of college students in different disciplines in a college English blended learning setting

Peng

Chen

et al. 2021

PLoS ONE

Self Cite

View full text Add to dashboard Cite

Learning styles are critical to educational psychology, especially when investigating various contextual factors that interact with individual learning styles. Drawing upon Biglan’s taxonomy of academic tribes, this study systematically analyzed the learning styles of 790 sophomores in a blended learning course with 46 specializations using a novel machine learning algorithm called the support vector machine (SVM). Moreover, an SVM-based recursive feature elimination (SVM-RFE) technique was integrated to identify the differential features among distinct disciplines. The findings of this study shed light on the optimal feature sets that collectively determined students’ discipline-specific learning styles in a college blended learning setting.

show abstract

“…Pese al eclecticismo de las comunidades mencionadas, la tendencia más común se vincula a que en el modelo tradicional, en el que el docente instruye y al alumnado adquiere los contenidos de forma pasiva, los resultados se caracterizaron por un lado, por la escasa influencia y relación entre las variables y, por otro, por una gran dispersión entre las mismas. Este hallazgo se contrapone a la mayoría de la literatura vinculada a la temática objeto de estudio en la que, en la mayoría de casuísticas, la utilización de modelos y métodos tradicionales se asocia a un mayor rendimiento académico del alumnado en PISA lectura (Dong & Hu, 2019;Giménez, Barrado & Arias, 2019;Gubbels et al, 2020;Perera & Asadullah, 2019;Torrpa, Eklund, Sulkunen, Niemi & Ahonen, 2017).…”

Section: Discusión Y Conclusionesunclassified

Uso metodológico docente y rendimiento lector del alumnado: análisis fundamentado en PISA lectura 2018

Rodríguez-García

Gago

2021

REIFOP

View full text Add to dashboard Cite

Las investigaciones en las que se vincula el rendimiento académico en lectura del alumnado y la utilización de modelos didácticos por parte del profesorado se caracterizan por el carácter exiguo y dispar de los resultados. Esta investigación analizó la influencia y relación entre la utilización metodológica que realizan 13119 docentes de España y el rendimiento académico del alumnado español en PISA lectura 2018. Para este propósito, se implementó un diseño de investigación descriptivo-correlacional en el que se relacionaron los resultados del alumnado español obtenido en la prueba de PISA lectura 2018 con los resultados extraídos tras la administración del cuestionario UMEPE a los docentes de esos alumnos. Los resultados arrojaron relaciones significativas y una influencia positiva y lineal entre el uso de un modelo activo y el rendimiento del alumnado en lectura. Por el contrario, cuando se implementa un modelo tradicional no se produce influencia ni relación entre las variables. La aplicación ecléctica de ambos modelos puede generar un mayor o menor rendimiento del alumnado en lectura en función del contexto. Estos hallazgos se anteponen a la mayoría de la literatura vinculada y se posicionan junto a los estudios que asocian métodos y estrategias activas con un mayor rendimiento. The studies that relate the use of teaching models and student´s achievement in reading are characterized by their recent and disparate nature. This research analyzed both the influence and relationship between methodological usage made by 13119 Spanish teachers and the achievement of Spanish students in PISA reading 2018. To this end, a descriptive-correlational research design was implemented in which the results of Spanish students obtained in PISA reading test are related to the results obtained after UMEPE questionnaire administration to the teachers of those students. The results showed significant relationships and a positive and linear influence between the use of active models by teachers and students’ achievement in reading. On the contrary, when a traditional model is implemented, there is no influence or relationship between the variables. The eclectic application of both models can generate more or less students´ achievement in reading depending on the context. These findings taker precedence over most of the connected literature and rank alongside studies that associate active methods and strategies with higher achievement.

show abstract

An Exploration of Impact Factors Influencing Students’ Reading Literacy in Singapore with Machine Learning Approaches

Cited by 29 publications

References 65 publications

Using Machine Learning Approaches to Explore Non-Cognitive Variables Influencing Reading Proficiency in English among Filipino Learners

Using Machine Learning Approaches to Explore Non-Cognitive Variables Influencing Reading Proficiency in English among Filipino Learners

Differentiating the learning styles of college students in different disciplines in a college English blended learning setting

Uso metodológico docente y rendimiento lector del alumnado: análisis fundamentado en PISA lectura 2018

Contact Info

Product

Resources

About