Modelling the dropout patterns of MOOC learners

Xie, Zheng

doi:10.26599/tst.2019.9010011

Cited by 20 publications

(14 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, if a student achieved greater scores on assignments lower the probability of being a dropout. A similar finding can be seen in many dropout predictions model [7,43]. This is because the cost of failure is low.…”

Section: Discussionsupporting

confidence: 82%

“…Because of that, we developed two predictive models. One, where dropout is defined as a student who fails to pass an exam [36] and another, where a dropout is defined as a student who will withdraw from or fail to pass the exam [43,40]. Besides using two definitions of the dropout, we created and evaluate logistic regression models in eight different time frames, ranging from the beginning of the course up to the mid-term of the course.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting dropout in online learning environments

2021

View full text Add to dashboard Cite

Online learning environments became popular in recent years. Due to high attrition rates, the problem of student dropouts became of immense importance for course designers, and course makers. In this paper, we utilized lasso and ridge logistic regression to create a prediction model for dropout on the Open University database. We investigated how early dropout can be predicted, and why dropouts occur. To answer the first question, we created models for eight different time frames, ranging from the beginning of the course to the mid-term. There are two results based on two definitions of dropout. Results show that at the beginning AUC of the prediction model is 0.549 and 0.661 and rises to 0.681 and 0.869 at mid-term. By analyzing logistic regression coefficients, we showed that at the beginning of the course demographic features of the student and course description features are the most important variables for dropout prediction, while later student activity gains more importance.

show abstract

Section: Discussionsupporting

confidence: 82%

Section: Introductionmentioning

confidence: 99%

Predicting dropout in online learning environments

2021

View full text Add to dashboard Cite

show abstract

“…This phenomenon also emerges in many empirical citation distributions of papers. The mechanisms underlying this phenomenon in both cases are the same, namely cumulative advantage, but known as the Matthew effect and the Lindy effect respectively [47].…”

Section: Discussionmentioning

confidence: 98%

Bridging MOOC Education and Information Sciences: Empirical Studies

Xie

2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

In the rise of massive open online courses (MOOCs), massive data of online learning behaviors are generated, which are valuable for the research on distance education in the aspects of making courses more attractive, learning more effective, and platforms wiser. Two empirical studies, here, utilized basic technologies of information sciences to analyze the log data of viewing behavior provided by the MOOC platform iCourse, illustrating the potential values of scientometrics and network sciences in the assessment of course attractions and in the improvement of teaching quality. One study extended information entropy to describe the diminishing marginal utility of repeated viewing and the increasing information of viewing new videos and further showed the possible applications of scientometrics to course assessment by integrating the viewing time on course videos invested by learners and the number of videos viewed by learners. The other study derived a network to represent learning paths adopted by the crowds of learners, which sheds light on the applications of network navigation and link prediction to the MOOC education in the aspects of teaching optimization and learning navigation. This paper adumbrates not only the practicability but also the limitations of the provided methods. INDEX TERMS Data science applications in education, distance education and online learning, evaluation methodologies.

show abstract

“…Since viewing is a prominent behavior of MOOC learners, how are the viewing time lengths utilized to assess courses' attraction? Fat-tail appears in the length distributions of learners' time spent on viewing specific MOOCs [22]. In other words, those length distributions exhibit a right skewness, relative to either normal or exponential distributions.…”

Section: Introductionmentioning

confidence: 99%

Assessing the Attractions of MOOCs From the Perspective of Scientometrics

Xie

2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

A range of empirical observations show that the right-skewed phenomenon emerges in the citation distributions of papers. The phenomenon also exists in the length distributions of learners' time spent on viewing the videos of a massive open online course. The mechanisms underlying this phenomenon in both cases are the same, namely cumulative advantage, known as the Matthew effect and the Lindy effect respectively. These similarities make it possible to apply the theories and methods in scientometrics to attraction assessments of massive open online courses. Massive data of learning behaviors are recorded on the MOOC platforms with high measurability and wide coverage, setting the precondition for this study. Based on the data derived from the log records of viewing behavior, we showed how to utilize the ideas of typical scientometric indicators, such as the impact factor and h-index, to assess courses' attraction. Our work adumbrates not only the practicability but also the limitation of the provided indicators.INDEX TERMS Data science applications in education, distance education and online learning, evaluation methodologies.

show abstract

Modelling the dropout patterns of MOOC learners

Cited by 20 publications

References 28 publications

Predicting dropout in online learning environments

Predicting dropout in online learning environments

Bridging MOOC Education and Information Sciences: Empirical Studies

Assessing the Attractions of MOOCs From the Perspective of Scientometrics

Contact Info

Product

Resources

About