Data analytics to target student support and address problems early

By Miranda Prynne, 22 December, 2022
View
How higher education institutions can incorporate data analytics into student support strategies to pinpoint risk factors that could disrupt a student’s progress and target assistance
Article type
Article
Main text

Higher education institutions have a responsibility to support students throughout their academic journeys. With proper intervention, we can help students address problems early. Many non-traditional learners are working adults who balance the pursuit of a degree with external commitments that can take priority over their education. After 19 years in the sector, I’ve seen how quickly students can lag in courses, possibly choosing to withdraw as they navigate work-life balance or have trouble grasping course content. For these individuals, the level of support received from their university is often the difference between them completing their degree and dropping out.

It’s important that institutions meet students where they are, supporting them with resources throughout their academic journey. In addition to personal touch points between students and their advisers and professors, advanced analytics give today’s educators an opportunity to identify student needs and inform proactive interventions to improve student outcomes.

Resource collection: Catching students before they fall
How to communicate your online teaching structure to students – and why it matters
Developing students’ confidence and sense of belonging online

Analytics provide valuable information to advisers

Hundreds of student data points can be used to inform advisers of potential risk indicators, enabling them to intervene and improve a student’s academic journey. Data points include:

  • readiness indicators such as grade point average (GPA) or credit hours taken before university enrolment
  • live course intelligence such as attendance frequency, grade progression or rate of adoption of support resources like tutoring
  • student-to-student interactions, which are gauged by assessing the percentage of enrolled students who engage in a university’s online community
  • term-to-term persistence and history of dropouts.

Activating analytics for student success

We monitor all students enrolled in classes through attendance management and student success intelligence systems. This monitoring allows us to pinpoint students in need of additional support throughout their academic journey. For instance, of all students enrolled between July 2021 and May 2022, we identified and personally engaged with 79 per cent of them in order to intervene with additional support, where necessary.

In addition, course inspection helped advisers isolate students with adequate engagement rates who were performing below expectations to identify the root causes of their low performance and tailor support accordingly. This included coaching around using available resources to get back on track, be it one-to-one support with their professor, using the 24/7 tutoring resources provided to all students or working with a licensed counsellor or social worker through our student-assistance programme. Success was measured through eligible students’ persistence, from term to term. Of those students who received targeted support, 84 per cent continued their education into the next term.

Live course intelligence technology such as Inspire for Advisors can equip faculty and student support teams with statistical tools to assess the probability of low student performance and persistence. The available data include previous attempts made by a student in the course that were unsuccessful, lack of assignment completion or poor grades. The real-time data allow educators to provide focused tutoring and resource options to ensure all students are prepared for success.

Supporting all stages of a student’s journey

Three key actions using advance analytics can be taken by institutions to support students at all stages.

Focus on incoming students 

Don’t wait until students get into classes to see if issues arise. Analyse their incoming attributes in advance, such as incoming high school or college grade point average (GPA), number of completed credit hours taken and historical enrolment patterns at prior institutions. These attributes, coupled with course sequencing or close mentorship, provide baseline insights to create “learner personas” that are derived from collected data to help pinpoint students in need of support and in what areas before they even enter their first course.

Identify where to leverage your advisers 

With labour and talent shortages, you need to strike the right balance between leveraging your institutional staff versus technology and automation. Adopt a test-and-learn approach to rolling out tech-based advice and support services and be willing to adapt and refine according to student responses and feedback.

It is worth noting that when seeking to test and learn, you must consider how your envisioned strategy aligns to an experiment framework in which you can apply A/B testing and isolate measurements to the test period and groups. For instance, take into consideration which aspects of data collection you will need to verify. Are you accounting for outside influences that could skew data? Does your data structure provide the necessary granularity or time to measure true differences? How do these compare with benchmark expectations that will help pinpoint the best strategy moving forward? Finally, it’s critical to capture quantitative and qualitative inputs to develop a holistic understanding of your experimented strategy.

Be proactive about student support 

Don’t wait for students to ask for help. There is a multitude of data points, such as attendance records, grade progression and engagement levels, to anchor personalised support for students. Identify the most important metrics for your institution to analyse as indicators of likely student success or failure, then test these through the experiences of your student population. For example, would a student benefit from being placed in a standard-level course or would a transitional level have greater impact? As the insights gathered from this data mature, institutions can add more points for performance analysis, creating a holistic approach to student support.

Many students have limited time to complete their courses alongside external responsibilities. Predictive analytics enables advisers to identify at-risk students, working with them one on one to address challenges. This allows them more time to complete their programme, ensuring their education is working for them, not against them.

Putting analytics into action can be challenging if a student is hesitant to accept assistance, but starting with initial touch points such as a preliminary adviser meeting can go a long way in making them feel supported. These relationships with tutors and advisers can change the course of an education experience and future career for the better.

Educational institutions should examine their student data capabilities to incorporate analytics into support strategies. Advisers and professors can then leverage analytics to pinpoint potential risk factors that could disrupt a student’s education pathway to address a problem before it is too late.

Elise Awwad is chief operating officer at DeVry University.

If you found this interesting and want advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the THE Campus newsletter.

 

Standfirst
How higher education institutions can incorporate data analytics into student support strategies to pinpoint risk factors that could disrupt a student’s progress and target assistance

comment