Student Dropout Research Articles

Reproducing Predictive Learning Analytics in CS1

Predictive learning analytics has been widely explored in educational research to improve student retention and academic success in an introductory programming course in computer science (CS1). General-purpose and interpretable dropout predictions still pose a challenge. Our study aims to reproduce and extend the data analysis of a privacy-first student pass–fail prediction approach proposed by Van Petegem and colleagues (2022) in a different CS1 course. Using student submission and self-report data, we investigated the reproducibility of the original approach, the effect of adding self-reports to the model, and the interpretability of the model features. The results showed that the original approach for student dropout prediction could be successfully reproduced in a different course context and that adding self-report data to the prediction model improved accuracy for the first four weeks. We also identified relevant features associated with dropout in the CS1 course, such as timely submission of tasks and iterative problem solving. When analyzing student behaviour, submission data and self-report data were found to complement each other. The results highlight the importance of transparency and generalizability in learning analytics and the need for future research to identify other factors beyond self-reported aptitude measures and student behaviour that can enhance dropout prediction.

Towards Sustainable Education: A Machine Learning Model for Early Student Dropout Prediction in Higher Education Institutions

Sustaining learners through an education cycle is a challenge for institutions at all levels. For higher education institutions, learners are presumed to be mature enough to complete their study courses. However, the challenge of student dropouts is prevalent. This paper seeks to address the key question of why students continue to drop out of learning institutions despite interventions undertaken by stakeholders. The attrition rates are a major concern that requires immediate attention if sustainable education is to be achieved. Dropping out of school is attributed to both individual factors and external factors. However, both require mitigation to save the future of education. This paper presents an analysis of challenges leading to student dropouts sampled from five institutions within the central region of Uganda (532 respondents). In addition, we leveraged the power of artificial intelligence (AI) to design and present a machine learning model for early student dropout prediction so that early interventions can be undertaken. The study adopted the design science methodology to scientifically support the design and validation of the machine learning student dropout prediction model. The early warning model presents key performance indicators to signal whether a student is predisposed to drop out or on course to completion. This way corrective intervention can be undertaken early enough for likely dropout. The validation experiment was conducted on a sample of 523 from the five institutions predicated a dropout of 10%. This proved the concept and the capacity of the model to predict learner dropout from university.

Student Dropout Analysis in Higher Education and Retention by Artificial Intelligence and Machine Learning

Student Dropout Analysis in Higher Education and Retention by Artificial Intelligence and Machine Learning

Debiasing Education Algorithms

AbstractThis systematic literature review investigates the fairness of machine learning algorithms in educational settings, focusing on recent studies and their proposed solutions to address biases. Applications analyzed include student dropout prediction, performance prediction, forum post classification, and recommender systems. We identify common strategies, such as adjusting sample weights, bias attenuation methods, fairness through un/awareness, and adversarial learning. Commonly used metrics for fairness assessment include ABROCA, group difference in performance, and disparity metrics. The review underscores the need for context-specific approaches to ensure equitable treatment and reveals that most studies found no strict tradeoff between fairness and accuracy. We recommend evaluating fairness of data and features before algorithmic fairness to prevent algorithms from receiving discriminatory inputs, expanding the scope of education fairness studies beyond gender and race to include other demographic attributes, and assessing the impact of fair algorithms on end users, as human perceptions may not align with algorithmic fairness measures.

Supervised machine learning algorithms for predicting student dropout and academic success: a comparative study

Utilizing a dataset sourced from a higher education institution, this study aims to assess the efficacy of diverse machine learning algorithms in predicting student dropout and academic success. Our focus was on algorithms capable of effectively handling imbalanced data. To tackle class imbalance, we employed the SMOTE resampling technique. We applied a range of algorithms, including Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), as well as boosting algorithms such as Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost), CatBoost (CB), and Light Gradient Boosting Machine (LB). To enhance the models' performance, we conducted hyperparameter tuning using Optuna. Additionally, we employed the Isolation Forest (IF) method to identify outliers or anomalies within the dataset. Notably, our findings indicate that boosting algorithms, particularly LightGBM and CatBoost with Optuna, outperformed traditional classification methods. Our study's generalizability to other contexts is constrained due to its reliance on a single dataset, with inherent limitations. Nevertheless, this research provides valuable insights into the effectiveness of various machine learning algorithms for predicting student dropout and academic success. By benchmarking these algorithms, our project offers guidance to both researchers and practitioners in their choice of suitable approaches for similar predictive tasks.

Early prediction of MOOC dropout in self-paced students using deep learning

ABSTRACT To address three issues identified in previous research this study proposes a clustering-based MOOC dropout identification method and an early prediction model based on deep learning. The MOOC learning behavior of self-paced students was analyzed, and two well-known MOOC datasets were used for analysis and validation. The findings are as follows: Firstly, the dropout rate among self-paced students in MOOCs exceeds 90%, with over 50% of students participating in online learning activities for only one day. Furthermore, the starting dates for students in the same course differ significantly. Secondly, leveraging early learning behavior and relevant background features, the proposed early prediction model accurately predicts over 98% of dropout cases and identifies over 50% of engaged students. Through training, the model's convolutional kernels capture meaningful weights for different days and activities. Lastly, background features related to students and courses have a more significant impact on dropout rates. The utilization of resources such as videos and active participation in learning activities, like asking questions, demonstrate a particularly significant influence on dropout rates. Notably, there is no fixed period that consistently affects dropout rates. These method and findings provide effective strategies for decreasing dropout rates and improving student engagement.

Propaedeutic Programme to Prevent Student Dropout at the Specialized University of the Americas in Veraguas, Panama

The objective is to design and validate a student dropout prevention programme at the Specialized University of the Americas in Veraguas, Panama. The research has a quasi-experimental design, pre-test and post-test, with a descriptive and explanatory type of study, which allowed us to respond to the research problem, hypotheses and objectives. The population consisted of sixty-six (66) students. The students' need to know the level of development of multiple intelligences and learning styles was proven, and that, based on this knowledge, they learn to use the most appropriate learning strategies for them, in addition to structuring their study habits.

Analysis of poverty as a factor in the dropout rate of university students in Mexico City from 2000 to 2022

The analysis of the correlation between poverty and student dropout has been a phenomenon of deep interest that must be studied, a problem presents in a dynamic and diverse region such as Mexico City. This research covers the last two decades, from 2000 to 2022, taking into account the factors that influence the educational continuity of university students in an environment marked by socioeconomic variability. In this analysis, the complexity of poverty is evident as a determining factor in student dropout in the university context of Mexico City; using indicators developed by the National Council for the Evaluation of Social Development Policy (CONEVAL) and the National Council of Science and Technology (CONACYT); as well as data collected through censuses implemented by the National Institute of Statistics and Geography (INEGI). Seeking to effectively explain the correlation between student dropouts and poverty

Reasons for school dropouts in suburban areas near Villupuram district: A retrospective study

Abstract Background: When a student leaves school or a college before completing their academic program, they are referred to as dropouts. A school dropout is anyone who does not receive their high school diploma. It is a serious issue that impacts students all over and can have negative outcomes in their lives. In general, those who struggle to engage behaviorally and/or academically drop out of school. Hence, we have chosen this study to find various reasons behind student dropouts in sub-urban areas of schools covered under Koliyanur block, Villupuram district. The objectives of this study were to determine the proportion and reasons for dropouts among school students in Koliyanur block, Villupuram district. Materials and Methods: It was a record-based retrospective study done among school students (1st–12th standard) who dropped out of school in July 2022. The secondary data on school dropouts (government and private) in the Koliyanur block of Villupuram district were obtained from the Chief Educational Officer of Villupuram. Out of 148 government and private schools, 36 schools had dropouts which were included in the study. Results: The prevalence of school dropouts among 36 schools was found to be 64 students. Migration, lack of interest, poverty, and lack of awareness remained the most common causes of dropouts in government schools, and fees not paid being the reason in private schools. Conclusion: Evidence generated from this study can aid education planners in the early identification of various reasons for school dropouts and take appropriate measures for their reduction.

PRÁTICAS PEDAGÓGICAS E EVASÃO DISCENTE: UMA ANÁLISE NO CURSO TÉCNICO

ABSTRACT: In recent years, mainly from 2000 to 2016, Brazilian Professional and Technological Education has transformed due to public policies aiming to expand and restructure it. Therefore, vocational schools are challenged to rethink and resignify how they maintain a link with society. Several aspects need to be observed and questioned to contribute to students' permanence and success in professional schools. Student dropout is one of the problems faced in professional education, influenced by several factors that weigh on the student's decision to stay or not in the course. Among these elements, pedagogical practices and their relationship with students' learning and motivation stand out, providing greater involvement with the educational institution. This study aimed to analyze the influences of teaching pedagogical practices on students' evasion in the Electromechanics Technical Course of IFCE/Campus of Pecém. This qualitative, applied, and exploratory research was developed through bibliographic surveys, case studies, and experience reports. The results allowed a better understanding of evasion in this course, listing its possible motivations. We perceived the vital importance of pedagogical practices for students' permanence in the course, as they significantly contribute to the teaching and learning process.

Unlocking Success: Transforming Challenges into Solutions in the World of Remote Learning and Online Teaching for Distance Education

This research highlights the challenges that students faced when using different modes to participate in online classes. Encouraging the expansion of private educational institutions is seen as a way to promote competition throughout the entire higher education sector. This, in turn, is expected to result in a wider range of options, offering more innovative and higher quality products and services, all at a reduced cost(Boliver, 2023). The goals of this study were primarily centred on three characteristics of the learners: their personal, academic, and technical challenges. Because it depends on offline modes of instruction, this work has a research gap. Since the COVID era(Pareek & Soni, 2020), there have been very few studies on online instruction, and it is difficult to gauge how successful virtual education is and how well students are responding to it. While taking courses online using a mobile device or a computer, learners are wary of technological development and their frequent upgrades because of various groups like dropout students, working people, elderly people, and homemakers. In order to gather information from the 623 respondents who are enrolled in online education programmes, the researcher prepared a well-structured questionnaire. The SYSTAT software(Wilkinson, 1990) was used to utilise univariate and multivariate normality, and JAMOVI(Şahin & Aybek, 2019) was used for SEM(D. A. Jalaludeen, 2022). In order to analyse the data, we also formulated 4 null hypotheses (H0), of which 3 were disproved. The study's findings indicate that the majority of comments concern how taking online courses through ODL helps me advance professionally or pursue higher education. The analysis led the researcher to the final conclusion that, of the three attributes, personal problems had the greatest impact on how the online courses were run. The institution will expand and be able to strengthen its position if it adopts or incorporates all of the researcher's recommendations.

PERMANÊNCIA DISCENTE EM CURSOS DE PEDAGOGIA A DISTÂNCIA: UM ESTUDO A PARTIR DA UNIVERSIDADE ABERTA DO BRASIL

ABSTRACT: Distance Education (DE) is growing rapidly in Brazil and nowadays it comprises more than a third of all enrollments in undergraduate programs (BRASIL, 2022a). However, the high dropout rates put pressure on Higher Education Institutions (HEIs) to define strategies to promote the permanence of their students until graduation. In this sense, this research was structured from the need to look at the institutions and to evaluate the way they are committed to the permanence of the students in distance learning. This research aims to identify and analyze how the HEIs affiliated with the Open University of Brazil (UAB) promote the permanence of students in the Pedagogy (teacher training) distance learning programs. For this purpose, eight semi-structured interviews were conducted with coordinators of six different HEIs, and institutional documents were analyzed. Both the understanding of the managers’ statements and the analysis of the documents emphasized that the institutions have different levels of commitment to the consolidation of distance education, which reflects directly on their performance to mitigate the dropout of students. Based on the identification of weaknesses of the programs, as well as on the good practices referred to in the literature, guidelines were presented that aim to promote the permanence of students in the contexts investigated.

Identifying and Understanding Student Dropouts Using Metaheuristic Optimized Classifiers and Explainable Artificial Intelligence Techniques

Identifying and Understanding Student Dropouts Using Metaheuristic Optimized Classifiers and Explainable Artificial Intelligence Techniques

What Works? Interventions Aimed at Reducing Student Dropout in Higher Education

What Works? Interventions Aimed at Reducing Student Dropout in Higher Education

Analysis of Deep Learning Development Platforms and Their Applications in Sustainable Development within the Education Sector

Educational institutions use information and communication technologies effectively to meet the innovation requirements that will increase their competitiveness. In this context, the rapid progression of deep learning has become a focal point for educational sustainability. Deep learning is increasingly integrated into education, driven by its advantages, including personalized learning experiences, elevated course material quality, student development enhancement, predictive analysis for student dropout prevention in massive open online courses, and streamlining instructional tasks. Notably, major corporations such as Amazon, Apache, Google, IBM, Microsoft, NVIDIA, and others actively contribute to the continuous development of deep learning tools and platforms. This section aims to provide a comprehensive understanding, starting with the definition of deep learning, its foundational principles, development tools, and platforms, followed by a discussion of its applications in education for sustainable development, illustrated with relevant examples.”

Toward Educational Sustainability: An AI System for Identifying and Preventing Student Dropout

Toward Educational Sustainability: An AI System for Identifying and Preventing Student Dropout

Enhancing the Early Student Dropout Prediction Model Through Clustering Analysis of Students’ Digital Traces

Enhancing the Early Student Dropout Prediction Model Through Clustering Analysis of Students’ Digital Traces

Predicting Student Dropout Risk Using LMS Logs

Predicting Student Dropout Risk Using LMS Logs

Log Linier Binomial Negatif dalam Memodelkan Data Siswa Putus Sekolah Tingkat SMA/SMK

Log linear negative binomial model is an extension of the linear regression model that can be used to analyze count data in the form of a contingency table when dealing with overdispersion where the variance value is greater than the mean value. Research in the field of education often produces contingency table data, one of which is data on students dropping out of school. The 2018 BPS survey showed that the number of students dropping out of school at the secondary school level in West Java province had the highest number of dropouts. Unfortunately, quantitative research on dropout students is still rarely done. Log linear negative binomial model is applied to determine the factors that influence the number of SMA/SMK dropout students in West Java province in 2021. The results show that gender, type of middle school, school status, interaction of gender and school status, and the interaction of middle school type and school status significantly affects the number of students dropping out of school. Furthermore, SMA/SMK dropouts in the province of West Java are dominated by male students, vocational students, and public-school students. Females are 1% more risky to drop out than males in public schools, while males are 1% more risky to drop out than females in private schools. SMA students are 59% more risky to drop out than SMK students in public schools, while SMK students are 71% more risky to drop out than SMA students in private school.

Effects of Infrastructure Impediments on Dropout of Female Students: A Cross-Sectional Based Study of District Bajaur Khyber Pakhtunkhwa

Effects of Infrastructure Impediments on Dropout of Female Students: A Cross-Sectional Based Study of District Bajaur Khyber Pakhtunkhwa