GydEd started out as a research project looking into the issue of student attrition. While examining retention tools currently used by higher ed institutions, we found a major limiting factor; they rely solely on measuring quantitative data points such as high school GPA, test scores, and demographics. However, when looking at the research it became clear that these solutions were missing a major aspect of the problem: the students’ well-being.
In the United States 40% of undergraduate students are estimated to drop out of university before completing their studies, and 1/3 of that dropout rate is first-year students. The top reasons for dropping out include financial pressure, academic disqualification (although 40% of college dropouts have a 3.0 GPA or higher), poor social fit, and lack of mental health support.
Most universities have resources to support their students, yet many students disappear from their courses before they can be identified for supportive intervention. Even more so now, with the demands of the current pandemic, students are struggling with their academic studies.
After many conversations with schools from all over the world, we learned that universities lacked a way to understand how their students were feeling, and why their students were dropping out. We realized that the best way to solve this problem was by providing a real-time insight tool to gauge the student experience.
Before we could start building the tool, we had to understand what we needed to measure. An immense body of literature has emerged reviewing factors that contribute to undergraduate dropout. A meta-analysis by Robbins, Oh, Lee, and Button, showed that three categories of psycho-social factors (PSFs) were good predictors of higher education attrition rates: motivation, emotion, and social control. Following Robbins et al.’s research, we decided to focus on these PSFs to gauge the student experience and understand why students are dropping out.
To measure the scales above, we built our questionnaire model using a total of 12 questions. For each of the three scales, we ask one main question and three correlating follow-up questions. If a student answers with a low score on a main question, they will be asked its three follow-up questions to identify the source of the low score. The follow-up questions are a mixture of modified items from the above sources, as well as additional items requested by the partnered university.
According to Baik, Larcombt, and Brooker, student well-being is underserved in higher education despite universities’ emphasis to add supportive resources. Therefore, we also connect students to relevant resources, based on their answers. The resources are obtained from the partnered university and supplemented with external resources.
Our interactive chatbot “Stella” is integrated directly into the LMS using the Learning Tools Interoperability (LTI) Advantage framework developed by IMS Global. The chatbot appears once a week to ask the questions, where each interaction takes between 30 seconds to a couple of minutes to complete, depending on the students answers and interest. Students are completely anonymous and may opt out at any time if they so choose.
The information gathered by the chatbot is aggregated and presented in a live dashboard format to the schools. Giving schools access to this real-time look into the well-being of their students allows them to understand how their students are feeling, and why they feel that way. The dashboard also shows individual student scores, anonymously, enabling targeted interventions through the chatbot. This allows administrators to take proactive measures to ensure higher retention rates and better student well-being.
Coupling our research-based methodology with our interactive chatbot, we want to help more students finish their studies by helping their schools understand their needs.
If you would like to learn more, signup with your details below!