University of Twente Student Theses

Login

Learning analytics towards identifying the processes of higher order thinking in online discussions of MOOCs

Smalbergher, I. (2017) Learning analytics towards identifying the processes of higher order thinking in online discussions of MOOCs.

[img]
Preview
PDF
2MB
Abstract:The current evaluation systems in MOOCs lack in knowledge about students’ learning processes. This requires the development of an evaluation system that can automatically track learners’ progress, and constantly inform of what the students need for improvement. The first is to explore how higher order thinking processes in online discussions can be analyzed and to what extent it can be automatized. Ways to assess students’ thinking process in the online discussions were investigated. Then, a new framework was constructed, which helped in designed a coding schema for higher order thinking identification. The coding schema was then used to classify the data manually and further used for teaching a machine learning to identify higher order thinking indicators. The results show that a Supervised Multiclass Classification Model can recognize the indicators of higher order thinking processes and classify the comments of students from the online discussions of a MOOC in three levels of thinking in proportion of 67%, and can make a distinction between lower and higher order thinking in proportion of 85% by using a coding schema designed specifically for the identification of higher order thinking in online discussions.
Item Type:Essay (Master)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:70 social sciences in general
Programme:Educational Science and Technology MSc (60023)
Link to this item:http://purl.utwente.nl/essays/74246
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page