# Learning Analytics in Massive Open Online Courses

**Authors:** Mohammad Khalil  
**arXiv:** [1802.09344](https://arxiv.org/abs/1802.09344)  
**Date:** 2018-02-17  
**Summary page:** https://www.summarizepaper.com/en/paper/1802.09344v1/

## Key points

- Educational technology has grown significantly in the past 15 years
- Two emerging topics are learning analytics and Massive Open Online Courses (MOOCs)
- MOOCs have attracted numerous participants due to their open accessibility and no-cost structure
- Large datasets generated by MOOC platforms require advanced tools and methodologies for further examination, highlighting the key importance of learning analytics
- This thesis focuses on integrating learning analytics into MOOCs to investigate further steps in the capabilities of both fields
- The research was conducted on the Austrian iMooX xMOOC platform using prototyping and case studies research methodology
- Main contributions include designing a general learning analytics framework, developing a learning analytics prototype, recording students' behavior across various MOOC variables such as discussion forums, interactions in videos, self-assessment quizzes, login frequency, and creating a cluster of student engagement.
- A systematic review was conducted on 1,315 articles published between 2011-2016 in peer-reviewed journals or conference proceedings written in English with an abstract available.
- Results showed that many publications described their research as use cases while older publications focused more on this approach than recent ones.
- A brief text analysis was also done on article abstracts to examine robust trends in higher education's prominent field of learning analytics resulting in a word cloud representation tool depicting prominent terms such as academic performance prediction models, distance learning development, personalized feedback systems and intelligent tutoring systems.
- Integrating learning analytics into MOOCs can improve educational outcomes while reducing costs and expanding accessibility for students.
- -The systematic review provides insights into the current trends and challenges in the learning analytics research field emphasizing the need for clear research questions and a focus on practical applications.

## Summary

The field of educational technology has grown significantly in the past fifteen years, with a wide range of engaging online environments and fields falling under its umbrella. Two emerging topics in this domain are learning analytics and Massive Open Online Courses (MOOCs). MOOCs have attracted numerous participants due to their open accessibility and no-cost structure, making them an effective tool for scaling education in disparate areas. However, the large datasets generated by MOOC platforms require advanced tools and methodologies for further examination, highlighting the key importance of learning analytics. This thesis focuses on integrating learning analytics into MOOCs to investigate further steps in the capabilities of both fields. The research was conducted on the Austrian iMooX xMOOC platform using prototyping and case studies research methodology. The main contributions include designing a general learning analytics framework, developing a learning analytics prototype, recording students' behavior across various MOOC variables such as discussion forums, interactions in videos, self-assessment quizzes, login frequency, and creating a cluster of student engagement. To gain more context on the challenges emerging from the learning analytics research field, a systematic review was conducted on 1,315 articles published between 2011-2016 in peer-reviewed journals or conference proceedings written in English with an abstract available. The analysis process used Latent Dirichlet Allocation (LDA), a popular probabilistic topic modeling technique that identifies latent themes based on word co-occurrences. The results showed that many publications described their research as use cases while older publications focused more on this approach than recent ones. A brief text analysis was also done on article abstracts to examine robust trends in higher education's prominent field of learning analytics resulting in a word cloud representation tool depicting prominent terms such as academic performance prediction models, distance learning development, personalized feedback systems and intelligent tutoring systems. Overall these findings highlight how integrating learning analytics into MOOCs can improve educational outcomes while reducing costs and expanding accessibility for students. The systematic review also provides insights into the current trends and challenges in the learning analytics research field, emphasizing the need for clear research questions and a focus on practical applications.