Master's thesis: "Personalising motivational strategies and gamification techniques with recommender systems"

In my Applied Informatics master's thesis, I incorporated my love for mathematics and education. You can read a Dutch article about my research on de Vlaamse ScriptieBank. The article underneath is an abridged English adaptation. Apart from a summary, my thesis text is written in Dutch. Feel free to contact me if you would like to discuss specific parts in English.

Few teenagers consider mathematics as cool, that is an open secret. The required abstract thinking often puts their motivation on the back burner. Therefore, teachers and educational publishers are looking for flexible tools that motivate students to practice mathematics. One possible approach to achieve that is gamification, whereby elements characteristic of game design are applied in non-gaming contexts. Research in human-computer interaction has shown that gamification can be successful in many domains, including education. However, when I was starting my master's thesis, none of the existing online mathematics platforms aligned with the curriculum for secondary schools in Belgium were gamified (that is, applying gamification).

Practising mathematics with personalised gamification

Therefore, I decided to develop Wiski (https://wiski.be), a free online mathematics platform that tries to motivate students with personalised gamification. On Wiski students can freely and unlimitedly practise the mathematics materials from secondary school. Covering 79 topics, the website contains over 2000 qualitative multiple choice questions, which were supplied by die Keure, the publisher of mathematics textbooks commonly used in Flanders. Exercise sessions typically consist of five multiple-choice questions about a specific topic and give students immediate feedback on their answers.

Wiski contains eight motivational strategies which I carefully picked based on the gamification literature, the mathematical-educational context and the target group. Specifically, students can:

1. earn experience points by solving exercises;
2. mutually compare their progress in a points leader board;
3. mutually compare their progress in an exercises leader board;
4. help others with difficult exercises;
5. collect hidden mathematical cartoons;
6. challenge themselves with mixed exercises;
7. receive feedback with mathematical jokes when answering questions;
8. build a day streak.

For my master's thesis, I developed a simple recommender system that suggested motivational strategies deemed fitting to the student, based on their personal preferences, their Big Five personality traits, and their Hexad gamification user type. However, since the latter two require students to complete quite extensive questionnaires, I adjusted this functionality after finishing my master's thesis: students now have full control over all available motivational techniques and can enable those they like.

Is personalising gamification more engaging?

To investigate whether personalising gamification techniques led to more motivation, I conducted a randomised controlled experiment: half of the students only saw three arbitrarily chosen gamification techniques, while the other half was presented a personalised selection. At the same time, I collected data on how students interact with the gamification techniques. Ultimately, 401 students from five different schools got cracking with Wiski for at least one hour in class.

The experiment revealed almost no significant differences in motivation between both groups, which suggests that teenagers are not necessarily motivated more in the short term when gamification is personalised in educative applications such as Wiski. However, some students in the control group did indicate that they would have prefered gamification techniques tailored to their preferences. This observation could indicate that personalising gamification may impact motivation in the long term.