Improving retention by identifying and supporting ‘at-risk’ students

Type: Evidence | Proposition: C: Uptake | Polarity: | Sector: | Country:

This study provides promising results that may, in future, help to retain students. However, the paper does not provide evidence of increasing retention. It does point to uptake of learning analytics at scale at The Open University.

The paper states that future plans will include integrating with the data warehouse as it becomes available, so that the output can be used to inform lecturers using live data. Further development and refinement of the models, for example integrating the module data, will lead to additional requirements for the data warehouse and additional risk categories that must be reflected through the dashboard. With more data being used as input for the predictive models, new visualizations are being investigated that will allow lecturers to view the risk factors that the models have identified as having the most impact in predicting risk.

Key take-aways identified in the paper

  • Without the feedback from face-to-face interactions, lecturers using virtual learning environments may find it difficult to identify and focus on students who are struggling in class.
  • The Open University uses data from its virtual learning environment to pinpoint students who have an increased risk of dropping out of a class, as well as to study class structure and content with the goal of tailoring the learning experience to each student's unique profile.
  • Predictive data can help instructors not only identify "at-risk" students but also use this enhanced feedback to improve the virtual learning experience.
Citation: Wolff, Annika, & Zdrahal, Zdenek. (2012). Improving retention by identifying and supporting 'at-risk' students, Educause Review Online. | Url: http://www.educause.edu/ero/article/improving-retention-identifying-and-supporting-risk-students