Learning analytics: student characteristics as predictors of study success

By Gijs de Groot

In the previous blogs, we explained what learning analytics is. We also used a dashboard to show how scale in higher vocational education affects study success and what interventions are possible based on this information. However, scale of an educational institution is not the only variable that affects study success. After all, students differ from one another. Are there certain types, or profiles, of students who do better than others? In this third and final blog of this series, we will look for an answer to this question, again using the dashboard.


Prior education of student has major impact on study success

Not everyone can take an HBO education. Only with certain previous education a student is admitted to HBO. As a rule, a student is admitted to a HBO program if he or she has a Havo, Vwo or MBO-4 diploma, possibly supplemented with additional requirements (for example, certain subjects). So although you can't go to HBO "just like that," there is a lot of difference in prior education. A student with a Havo diploma is often slightly younger than a student with Vwo as a previous education. Moreover, the level of Vwo is higher and the education is more theoretical in nature. A student with a MBO-4 diploma has already completed a three- or four-year advanced education, is a bit older and has more experience studying. The dashboard shows that there are large differences in study success between students with different prior education. Students with a Vwo diploma graduate the fastest: of students who started a HBO program in 2008, 46% graduated after four years. This is more than one and a half times as many as students with a Havo diploma (28%) and also substantially more than students with a MBO-4 diploma (35%). What is further striking is that after 8 years, the percentage of students with a Havo diploma who graduated (67%) is higher than the percentage of graduates with an MBO-4 diploma (60%). So the Havists are catching up with the MBO graduates, so to speak

Female students do better than male at HBO

Besides prior education, gender is an important characteristic that influences study success. 40% of female students who started an HBO program in 2008 graduated after 4 years, compared to 27% of male students. So there are one and a half times more female students who graduate nominal (that is, in 4 years) than male students: a big difference. Men also do not make up this deficit in the following years.

Knowledge about the influence of student characteristics on study success enables targeted support

Knowledge about the effects that certain characteristics or collections of characteristics (so-called profiles) of students have on study success is of great importance to educational institutions. The aforementioned characteristics in relation to study success are generally well known in higher education. However, if we go a little deeper and relate more detailed information about students to study success, surprising insights often emerge. Insights that are not based on gut and gut feeling, but on hard data. With this knowledge, an institution can tailor guidance and support for students as much as possible, such as with specific support packages and individual interventions. Recruitment, information and guidance can be adapted based on characteristics of the student population, so that new students have a realistic picture of programs and the workload involved.

The promise of learning analytics is great

In three blogs we have shown that learning analytics is about using insights obtained from analysis of (student) data at various levels. To get started successfully with learning analytics, it is important to approach it not as an ICT project but as a change task. Not only technology is important, but also strategy, processes, people and issues such as privacy, ethics and data protection. Ultimately, for a successful application of learning analytics, the teacher is the key to success. The use of data and analytics must find a place in the "professional image" of the teacher. It requires a different way of working that teachers and students must feel sufficiently senang with. 

We used analyses of open education data and a dashboard to show the relationship between scale of educational institutions and student characteristics and study success. Based on these relatively easily obtained insights, we outlined several possible generic interventions for HBOs. More detailed information and detailed analyses will lead to more detailed insights, allowing for customization and appropriate individual interventions ahead. In short: the promise of learning analytics is great.