Using learning analytics to enhance your math course

Teaching mathematics in higher education is challenging due to factors like high student numbers, different student profiles and proficiency levels, and lack of time to provide students with insightful feedback. Learning analytics can be a vital ally to help math educators overcome these obstacles.

In the digital era, learning analytics involves collecting, analyzing, and interpreting data on students’ learning performance and behavior on digital traces of learning activities. We can use this information to optimize instruction, improve student engagement, and enhance student learning outcomes and satisfaction.

In this blog post, we will discuss the following topics:

Our goals are to explore the benefits of learning analytics in math courses and give you practical tips you can integrate into your class.

Benefits of using learning analytics in your math classroom

Monitor students’ progress

Math teachers often struggle with large student numbers and a heavy workload. Learning analytics can provide you with a quick overview of your course. This feature can facilitate you to be on top of your class, detecting and solving problems promptly. Moreover, tracking students’ progress over time will provide you with practical insights for your following courses.


Upgrade your teaching

Learning analytics can help you improve your curriculum and teaching. For example, data on how students are performing with the exercises can let you know if learners already master a topic and if they are particularly struggling with an element of the coursework.

Provide students with personalized support

No two students are the same. They usually have different proficiency levels and needs. Many learners might suffer from math anxiety and struggle to stay on track. Often, some students don’t ask for assistance before it’s too late or avoid doing it completely.

Thanks to learning analytics, you can identify on time who is underperforming individually. This enables you to offer them additional resources, studying strategies or one-on-one instruction to aid them in improving their understanding of the material.

These conversations can also let you know if someone needs special adjustments (e.g., giving extra time to a student with dyscalculia). A personalized approach can lead to better engagement and higher student satisfaction. Additionally, it can help you close the gaps between underperforming students and their counterparts.

teacher talking with a student

Dirk Tempelaar teaches math with SOWISO at Maastricht University. He has used SOWISO’s learning analytics module in his research on modelling student learning and achievements in learning from an individual difference perspective. That includes “big data” approaches to learning processes and predictive models for generating learning feedback based on computer-generated learning data.

Tempelaar (2020) conducted a study to explore how to help students who need more support when taking a challenging math or stats course in the first year of their bachelor’s. The courses in his study use a problem-based learning approach, so students need to take charge of their learning. However, not all students are ready for this format.

So Tempelaar contrasted the learning experiences of two student groups with different learning approaches:

  1. Students who emphasized deep learning: they scored much higher on critical processing, relating and structuring knowledge than on analyzing and memorizing.
  2. Learners who focused on stepwise learning: this group scored much higher on analyzing, memorizing and recalling procedures.

Tempelaar’s research suggests that the former type are self-regulated learners. These students can reflect on their own learning to select, plan and control appropriate learning strategies by themselves. The latter type has a low level of self-regulation and requires more external regulation. This means that stepwise learners are highly dependent on the teacher’s guidance to supervise their learning processes.

Tempelaar analyzed SOWISO’s data and the information gathered with surveys measuring student dispositions. He found that using a blended module design with a digital environment with learning analytics and actionable learning feedback can be especially useful for first-year students, particularly when they are having trouble adapting to university courses that require a more autonomous learning approach.

His study provides a strategy showcasing how using learning analytics contributes to bridging the gap among students with different knowledge backgrounds, past teaching experiences, and learning dispositions.

Inform students and help them succeed

Learning analytics can also benefit students directly by giving them a realistic overview of their performance and progress. This digested data can stimulate them to reflect, detect and fill their knowledge gaps.

Following Tempelaar et al. (2013), learning analytics can also become an ally in providing students with insightful information to adapt their learning to their own strengths and weaknesses. This can enable them better understand themselves and build better study habits. Learning analytics can empower students to succeed in their future academic and professional pursuits.

However, this requires that students take an active role in their learning, and they might differ in the information they need to advance.

Prevent student dropout

First-year math courses in higher education can feel intimidating for a large part of the student population, leading to adverse outcomes like avoidance and high failure rates.

Teachers from the Swinburne University of Technology addressed in a study how learning analytics can ease improving retention in first-year mathematics (Faridhan et al., 2013). As they discuss, a lot of student data is being collected nowadays. However, these data should be further analyzed to predict students’ success and improve retention interventions by targeting and supporting at-risk students.

Photograph of a student doing math homework

How to use learning analytics in your math course with SOWISO

Before looking at some tips, let’s first have a look at all the data that can be retrieved in SOWISO’s learning analytics module, which can be found in the Report tab.

Fine-grained data

Teachers can look at the data from all the different available tasks (practice material in content, tests, assessments) and in various granularities.

Data per task

The systems track the performance and progress in all the different available learning tasks. This information is displayed in the following Report subtabs:

  • Content: Data from self-practice tasks by students.
  • Tests: Data from formative or summative tasks created by the teacher and computer-generated diagnostic tests.
  • Assessments: Data from practice tasks selected by the teacher.

Data granularities:

  • Data per exercise: The system tracks three scores in each exercise per student: the mastery score, the time spent if the solution was requested from the system, and the number of attempts required to achieve the mastery level.

    • Per class: results of an exercise from all the members in a class in one view.
    • Per student: detailed data of every exercise attempt for each student.
  • Data per topic/subchapter/chapter: The system aggregates the data of all exercises contained in the relevant topic/subchapter/chapter.

    • Per class: results from all the members in a class.
    • Per student: results of each student.
  • Data per student with time filter: Besides progress and performance in each exercise and topic, the system provides a student’s dashboard, which includes a general overview of the students’ activities and learning experiences, monitoring the overall average score, the activity rate, activities by date and the most recent exercises entertained. The student’s dashboard includes a time filter to look at the activities in a specific period.

  • Data per class with time filter (Dashboard tab): an analytical overview of the activities and results of your entire class, such as the overall average score, the overall average progress, the activity rate, the activities by date, the most recent packages entertained, and the average score per chapter.

Tips (what, how, and why): Teachers can use all these types of data to…

1. Regular use of formative tests to prevent cramming before the final exam

Frequent formative tests will increase the data collection for a more insightful learning analytics report. It will also help students improve their knowledge and study habits. You can learn more about the benefits of formative tests in this blog post.

Screenshot of SOWISO learning analytics

2. Identify underperforming students and provide timely feedback based on early performance

Teachers can use the topic and students’ data to provide timely feedback on early performance per topic. Then, students can have extra practice and ask for support or new study tactics before a summative exam or evaluation.

Moreover, at any time and independently of their teacher, students can monitor their own performance and progress and obtain detailed insights into their learning activities on the platform.

GIF of SOWISO learning analytics

3. Monitor students’ engagement level

Various forms of students’ disengagement have increased considerably in recent years, possibly as an after-effect of the covid pandemic. So, having ways to measure this can assist teachers in finding solutions.

With the reporting environment, teachers can measure the students’ engagement level with the learning materials by monitoring the time spent on a topic or exercise, by looking at the number of attempts in assignments/practice material and by using the time filter to look into their recent activities in the selected time frame.

Screenshot of SOWISO learning analytics

This action helps teachers acknowledge the topics and concrete learning activities that call for more attention. Exploring students’ learning patterns can also enable you to connect with them.

4. Identify challenging or neglected topics for course redesign

The topics’ data provide insights for teachers to update course modules that may need certain amendments. This information can simplify spot course elements that could be better explained or require extra attention during your classes. For instance, pinpointing overly-challenging exercises or super easy ones (relative to a group of students) can help you redesign assignments or tests, include other parts of the course or modify the teaching style.

Using learning analytics facilitates you to try new things, get quick feedback about your teaching and keep improving your methods.

5. Track long-term students’ performance

Teachers can keep track and visualize students’ long-term progress and activities by setting the time filter to 12 or 6 months in the student or class dashboard. This helps teachers, course coordinators and educational researchers interested in comparing and analyzing the activities and results of different classes or groups over a long period.

Do you need more information? All the buttons in our learning environment are explained here.

Do you want to use our learning analytics for a research project?

SOWISO’s system tracks each student’s interactions with the platform’s features and learning materials while complying with GDPR.

Our tool enables the collection of a wide range of learning metrics that can be used by their teachers and for research purposes when anonymizing the data. Do you have any research ideas on using learning analytics in math classrooms? Would you like to collaborate with SOWISO? Don’t hesitate to contact Dr Ana-Lucia Vargas Sandoval, our Product Specialist.


  • Faridhan, Y. E., Loch, B., & Walker, L. (2013). Improving retention in first-year mathematics using learning analytics. In ASCILITE-Australian Society for Computers in Learning in Tertiary Education Annual Conference (pp. 278-282). Australasian Society for Computers in Learning in Tertiary Education.
  • Tempelaar, D. (2020). Supporting the less-adaptive student: The role of learning analytics, formative assessment and blended learning. Assessment & Evaluation in Higher Education, 45(4), 579-593.
  • Tempelaar, D. T., Heck, A., Cuypers, H., van der Kooij, H., & van de Vrie, E. (2013, April). Formative assessment and learning analytics. In Proceedings of the third international conference on learning analytics and knowledge (pp. 205-209).

Do you want to be on top of the STEM Education world?