Quality Development Using Text Analysis Tools on Course Reflections

Quality development and quality assurance in higher education is an important topic. In this process, collecting information about students’ experiences and perceptions of courses and programmes is an important part, but getting such data is problematic. Few students answer course questionnaires, and then often answer very briefly, which decreases the reliability of the results. There is a risk that the answers received are not representative, and that changes are made to courses and programmes that does not align with the perceptions of the majority of the students.

In this project we plan to use reflective texts on courses the students provide in a meta-course in their educational programme. In that course, the students write a reflective text on the courses they have taken, in order to take a step back from the micro perspective of the course, and see the course’s place in the bigger picture. Even though these reflections were initially not meant to be course evaluation, the students have approached the student management in using these texts as material for course development, and for the last couple of years they have also been used as such. However, we now face the opposite problem of before. Instead of receiving too little feedback we now get too much feedback, with close to 100 reflective texts for courses, and if aggregated to the programme management there are several thousands of reflections per year.

In this project we will investigate how these texts can be automatically analysed, using sentiment analysis and text analysis tools in order to both detect problems in couses, and see if students’ perceptions of courses change when courses are modified. In the project, students’ comments on other student texts will also be evaluated. 

The project is using the text analysis tools supplied by Gavagai .

Team