Modeling MOOC Student Behavior With Two-Layer Hidden Markov Models

Geigle, Chase; Zhai, ChengXiang

doi:10.1145/3051457.3053986

Cited by 46 publications

(35 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Our review of the literature finds a distinct common thread connecting learning design and learner behavior studies, namely, that of abstraction and complexity reduction. In addition, many of the methods in our work are inspired by research in the area of learner-behavior pattern mining [4,7,8,25]; we find that many methodologies in this field have have potential applications to pattern mining of course structure and pedagogy.…”

Section: Related Workmentioning

confidence: 99%

“…The research presented in [4,7,8,25] characterizes MOOC learners through their clickstream data tracking their transition between activities. Reference [25] first identified common 2-gram event transitions; [4] next extended these to 8-gram event sequences and labeled the sequences as various motifs representing a study pattern; and [8] extended this by connecting these event transitions to self-regulated learning strategies using learner self-reported survey results as well.…”

Section: Learner Behavior Patternsmentioning

confidence: 99%

“…In another area of research, learner behavior modeling has taken off for MOOCs [4,7,8,25]. However, there has yet to be any large-scale or automated evaluation of the effectiveness of various learning design patterns using the tools from the learner-behavior community.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Toward large-scale learning design

Davis

Seaton

Hauff

et al. 2018

Proceedings of the Fifth Annual ACM Conference on Learning at Scale

View full text Add to dashboard Cite

This paper applies theory and methodology from the learning design literature to large-scale learning environments through quantitative modeling of the structure and design of Massive Open Online Courses. For two institutions of higher education, we automate the task of encoding pedagogy and learning design principles for 177 courses (which accounted for for nearly 4 million enrollments). Course materials from these MOOCs are parsed and abstracted into sequences of components, such as videos and problems. Our key contributions are (i) describing the parsing and abstraction of courses for quantitative analyses, (ii) the automated categorization of similar course designs, and (iii) the identification of key structural components that show relationships between categories and learning design principles. We employ two methods to categorize similar course designs-one aimed at clustering courses using transition probabilities and another using trajectory mining. We then proceed with an exploratory analysis of relationships between our categorization and learning outcomes. * The author's research is supported by the Leiden-Delft-Erasmus Centre for Education and Learning.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Learner Behavior Patternsmentioning

confidence: 99%

See 1 more Smart Citation

Toward large-scale learning design

Davis

Seaton

Hauff

et al. 2018

Proceedings of the Fifth Annual ACM Conference on Learning at Scale

View full text Add to dashboard Cite

show abstract

“…The availability of rich data sources has accelerated the development of learning analytics and educational data mining tools. For example, some of the work has been focusing on pedagogical recommendation [13,26], measurements of student behaviors in ITSs [18,8,11], and the detection of learning behaviors in a physical environment, such as wandering detection [22]. Some other work has concentrated on the student performance prediction, such as dropout prediction [23,7], grade prediction [1,19], and knowledge tracing [5,16,17,25].…”

Section: Introductionmentioning

confidence: 99%

Incorporating Features Learned by an Enhanced Deep Knowledge Tracing Model for STEM/Non-STEM Job Prediction

Yeung¹,

Yeung

2019

Int J Artif Intell Educ

View full text Add to dashboard Cite

The 2017 ASSISTments Data Mining competition 1 aims to use data from a longitudinal study for predicting a brandnew outcome of students which had never been studied before by the educational data mining research community. Specifically, it facilitates research in developing predictive models that predict whether the first job of a student out of college belongs to a STEM (the acronym for science, technology, engineering, and mathematics) field. This is based on the student's learning history on the ASSISTments blended learning platform in the form of extensive clickstream data gathered during the middle school years. To tackle this challenge, we first estimate the expected knowledge state of students with respect to different mathematical skills using a deep knowledge tracing (DKT) model and an enhanced DKT (DKT+) model. We then combine the features corresponding to the DKT/DKT+ expected knowledge state with other features extracted directly from the student profile in the dataset to train several machine learning models for the STEM/non-STEM job prediction. Our experiments show that models trained with the combined features generally perform better than the models trained with the student profile alone. Detailed analysis on the student's knowledge state reveals that, when compared with non-STEM students, STEM students generally show a higher mastery level and a higher learning gain in mathematics. 2

show abstract

“…Section 2), we test the extent to which the testing effect can be leveraged in one of today's most popular digital learning settings: Massive Open Online Courses (MOOCs). Research into both MOOC platforms and MOOC learner behaviour has found that learners take a distinctly linear trajectory Wen & Rose´, 2014;Geigle & Zhai, 2017) through course content. Many learners take the path of least resistance towards earning a passing grade (Zhao et al, 2017), which does not involve any backtracking or revisiting previous course units -counter to a regularly spaced retrieval practice routine.…”

Section: Introductionmentioning

confidence: 99%

Scaling Effective Learning Strategies: Retrieval Practice and Long-Term Knowledge Retention in MOOCs

et al. 2018

View full text Add to dashboard Cite

Large-scale online learning environments such as MOOCs provide an opportunity to evaluate the efficacy of learning strategies in an informal learning context with a diverse learner population. Here, we evaluate the extent to which retrieval practice -recognized as one of the most effective learning strategies -facilitates long-term knowledge retention (and thus learning outcomes) among MOOC learners using an instructional intervention. We observed no effect on learning outcomes and high levels of treatment non-compliance. In addition, we conducted a series of exploratory studies into long-term recall of knowledge acquired in MOOCs. Surprisingly, both passing and non-passing learners scored similarly on a knowledge post-test, retaining approximately two-thirds of what they learned over the long term. Notes for Practice• We developed and tested a system that prompts students with previous questions to encourage retrieval practice in a MOOC.• Retrieval practice showed no effect on learning outcomes or engagement.• Learners who passed the course scored similarly on a knowledge post-test as did learners who didn't pass.• Both passing and non-passing learners retained approximately two-thirds of the knowledge from the course after a two-week testing delay. RQ4To what extent is long-term knowledge retention facilitated in a MOOC? RQ5 What is the learner experience using ARPS to engage in retrieval practice?The remainder of the article is organized as follows: In Section 2 we review the literature on retrieval practice, spacing, and knowledge retention. In Section 3 we introduce ARPS in detail. Section 4 discusses the methodology and our study design. Results are presented in Section 5, and Section 6 offers a discussion of the main findings, limitations to consider, and implications for future research.

show abstract

Modeling MOOC Student Behavior With Two-Layer Hidden Markov Models

Cited by 46 publications

References 5 publications

Toward large-scale learning design

Toward large-scale learning design

Incorporating Features Learned by an Enhanced Deep Knowledge Tracing Model for STEM/Non-STEM Job Prediction

Scaling Effective Learning Strategies: Retrieval Practice and Long-Term Knowledge Retention in MOOCs

Contact Info

Product

Resources

About