Blog post
Transforming curriculum planning with Sankey diagrams
Have you ever wondered how to gain deeper insights into the complexities of curriculum planning and to support data-driven decision-making? As an educator and researcher at ETH Zurich, I’ve witnessed firsthand the challenges we face in curriculum planning. With students increasingly seeking personalised learning experiences, we need tools that can help us visualise their pathways through complex academic programmes. This is where Sankey diagrams (figure 1) come into play, transforming how we analyse and improve our curriculum.
The power of Sankey diagrams
Sankey diagrams, a type of flow diagram, are increasingly recognised in educational research for their ability to visualise student trajectories, track course progression, and identify areas of improvement (Morse, 2014; Loder, 2024; Horváth et al., 2018; Raji et al., 2021). By visualising student pathways, we can uncover new opportunities for optimising course sequencing and ensuring that the curriculum meets both student needs and institutional goals.
Visualising student pathways at ETH Zurich
‘Sankey diagrams allow us to visualise the courses students take, their sequences and how effectively they navigate their educational journeys.’
In our project (Templ et al., 2024), we applied Sankey diagrams to map student flows in the Environmental Sciences programmes at ETH Zurich (figure 1). These diagrams allowed us to visualise the courses students take, their sequences and how effectively they navigate their educational journeys. What struck me was how much clarity these visualisations provided compared to traditional data analysis methods.
Figure 1. The Sankey plot illustrates courses completed in the semester by Environmental Systems majors before and after a selected data sciences course (e.g. ESDS)
Note: Data spans from 2013–2023 and includes courses with at least two participants.
Imagine trying to discern trends from a spreadsheet filled with numbers. Now picture that same data beautifully represented in a flow diagram that highlights critical pathways and bottlenecks. This approach is not just a visual enhancement; it’s a game changer for curriculum planners.
Key insights and applications for curriculum improvement
Our analysis revealed that students often complete foundational courses like ‘Multivariate Statistical Analysis’ before advancing to specialised courses such as Environmental Systems Data Science (ESDS). This insight is invaluable, helping us to refine course prerequisites and ensure students are adequately prepared for more advanced topics.
The interactive nature of the Sankey diagrams allowed us to dig deeper. We could filter data by various parameters – such as course type, ECTS credits (European Credit Transfer and Accumulation System is a standard for comparing academic credits: that is, the volume of learning based on the defined learning outcomes and their associated workload), or student demographics – enabling a nuanced understanding of how different student groups engage with their education. For instance, we learned that students with prior experience in data cleaning could progress directly into advanced topics, while others benefited from a tailored approach that built their foundational skills first.
Streamlining curriculum analysis and adjustments
Historically, we relied on manual reviews of course completion data, which was both time-consuming and often lacked clarity. The transition to using Sankey diagrams has streamlined this process and highlighted areas where our curriculum can be improved. For example, our findings prompted us to restructure the ESDS course into two half-semester offerings: ‘Data Crunching’ and ‘Machine Learning’. This adjustment aligns the course content with varying levels of student experience, ultimately enhancing the learning experience.
Fostering data-driven decision-making
‘Incorporating Sankey diagrams into our curriculum planning has not only improved the academic experience for our students but has also fostered a culture of data-driven decision-making within our institution.’
Incorporating Sankey diagrams into our curriculum planning has not only improved the academic experience for our students but has also fostered a culture of data-driven decision-making within our institution. As educators, we have a responsibility to leverage data effectively, and Sankey diagrams provide a compelling way to do just that.
I encourage other educational institutions to explore the potential of Sankey diagrams for their own curriculum planning needs. The ability to visualise complex pathways can lead to significant improvements in how we design and deliver academic programmes.
Conclusion
Sankey diagrams are more than just a visualisation tool; they are a pathway to enhancing the effectiveness of our educational programmes. By embracing such innovative approaches, we can ensure that our curriculum not only meets the demands of today’s learners but also prepares them for the challenges of tomorrow.
This blog post is based on the article ‘Leveraging Sankey diagrams for enhanced curriculum planning in higher education’ by Barbara Templ, Matthias Templ, Anouk N’Guyen van Chinh and Urs Brändle, published in the Curriculum Journal.
References
Horváth, D. M., Molontay, R., & Szabó, M. (2018). Visualizing student flows to track retention and graduation rates. In Proceedings of the 22nd International Conference on Information Visualization (IV).
Loder, A. K. F. (2024). Student Flow Visualization. European Journal of Education, 59(2), e12619. https://doi.org/10.1111/ejed.12619
Morse, C. (2014). Visualization of student cohort data with Sankey Diagrams via web-centric technologies. University of New Mexico.
Raji, M., Duggan, J., DeCotes, B., Huang, J., & Zanden, B.V. (2021). Modeling and Visualizing Student Flow. IEE Transactions on Big Data, 7(3), 510-523. https://doi.org/10.1109/TBDATA.2018.2840986
Templ, B., Templ, M., N’Guyen van Chinh, A., & Brändle, U. (2024). Leveraging Sankey diagrams for enhanced curriculum planning in higher education. Curriculum Journal, 36(1), 202–206. https://doi.org/10.1002/curj.299