Over the most recent years, education and its use of technology has undergone many changes. At times fully online or partially online, the way we teach and the way students learn has shifted to incorporate more digital elements. With these changes come new challenges, one of which PhD researcher Gabrielle Martins Van Jaarsveld is choosing to tackle.
Text-based & rule-based
Gabrielle explains: "To start, a chatbot is a computer program that interacts with people who are using it, for a specific purpose. And in this case, this is a chatbot for academic goal setting. The chatbot we built is a text-based chatbot, which means input and output of the chatbot are written. It is also a rule-based chatbot, which means it relies on a set of pre-determined rules for responses. And these rules make the chatbot focus on its task: to help students through the goal setting process, which has clear phases."
In the context of goal setting, research has found that a personalized intervention carried out by a person is generally more effective than a standardized goal setting survey that a student is given to fill out. For one teacher to sit down with every student of their course is very difficult in large Bachelor or Master programmes. "However," Gabrielle explains, "the benefit of a chatbot is that it provides a solution to easily scale a more personalized goal setting procedure."
The development of the chatbot focused on the cycle of goal setting, which consist of three phases: the forethought phase, the performance phase, and the reflection phase. In the forethought phase, the chatbot guides the student through setting SMIP- smart, measurable, important, and multisource- goals (an adaptation of the SMART goal approach). Then, the student goes through the performance phase, where they work to realize their goals and keep track of their progress. Afterwards, in the reflection phase, the chatbot helps the student reflect and adapt their future goals accordingly. Gabrielle reflects on this: "I believe that setting up the chatbot with this structure, provides the most effective way for the student to set more important and achievable goals, and to learn from what didn’t go to plan, as well as celebrate their successes."
Gabrielle has conducted a pilot study with 16 Bachelor of Psychology students at ESSB. The feedback from this pilot study was positive, with all 16 students saying the chatbot was easy to use, and 14 of the 16 students saying they preferred the chatbot to doing a survey to set goals. She will now test the chatbot for the duration of a full block- five weeks- on a larger number of ESSB Bachelor students.
Technology driven support to students
When asked about the future, Gabrielle sees the chatbot as a very scalable and powerful tool for goal setting, but also for research as it can gather a large amount of data. "Because the more data the chatbot gathers, the more it will be able to improve. We aim to work on a machine learning module to implement in the chatbot, once enough data is gathered. This will allow the chatbot to assess whether a goal is of high quality by itself, rather than a moderator, such as a teacher, carrying out that process. My long-term goal is to use learning analytics to support the development of this chatbot, and to explore how technology can be leveraged to improve the support it can offer to students," she concludes.