Leiden Learning & Innovation Centre

Learning Analytics and Data Science

Exploring the value of data-driven approaches in education: collecting, analysing, and interpreting data from educational environments to improve teaching and learning outcomes.

By leveraging data-driven insights to stimulate informed decision making, our work in learning analytics aims to enhance instructional strategies, identify areas for improvement, develop targeted interventions, and personalise the learning experience.

We embrace a diverse range of methodologies, carefully selected to match the questions at hand. Ranging from the detailed monitoring of students’ learning experiences to applying data science methods for broader organisational contexts, the possibilities are many.

Why is learning analytics important?

Learning Analytics is relevant for teachers, students, administrators, and support staff across our university. It helps teachers interpret student engagement, supports students with personalised feedback, and enables faculties and services to make evidence-based decisions concerning courses and resources.

Dashboards to monitor progress, surveys to capture learning experiences, and predictive models to identify where support is needed - all of these have learning analytics at their core.

What kinds of questions can learning analytics solve?

As examples:

How are students engaging with my course materials?
Which students may need extra support to succeed?
At which point in my course do student often lose interest?
How effective are new tools, curricula, or interventions?
Where should we focus resources to have the greatest impact?
Where are gaps or overlap in our curricula?

Toolset

Our team applies a diverse set of industry standard analytics and visualisation tools – from Python and R for statistical analysis to Power BI, Looker, and Shiny for interactive dashboards. This mix ensures flexibility and reliability in transforming educational data into meaningful insights.

We select and apply the most appropriate tools based on the project’s goals and data characteristics, allowing partners to focus on organisation and educational goals without needing to determine technical details themselves.

Key terms: what they mean and offer

Data Science

At the core of learning analytics lies data science, which provides the foundational framework for collecting, processing, and analysing educational data. Data science techniques such as data mining, machine learning, and statistical analysis enable us to extract valuable insights from vast volumes of educational data.

Artificial Intelligence

Artificial intelligence plays a pivotal role in advancing learning analytics by automating tasks, personalising learning experiences, and generating actionable insights. AI adaptive learning systems can tailor educational content to individual student needs, help uncovering sentiment trends and predict student performance.

Visualisation

Visualisation bridges raw educational data and insights, making complex information interpretable. Interactive dashboards, charts, and graphs reveal student progress, engagement, and performance metrics, empowering stakeholders for informed decision-making and strategic interventions.

Contact us!

To see and hear more and discuss possibilities for potential projects and collaborations, you are welcome to get in touch with us at datascience@llinc.leidenuniv.nl.

Examples from our work

The images below provide a glimpse of the analyses and dashboards we develop and illustrate how we support data-informed decision-making in education.

Density plot comparing students’ first quiz grades with their final study advice. By visualising these patterns, we can explore how early assessments may indicate who could benefit from additional support.
Timeline of key student actions during a course — enrollment, first access, completion and leaving the course. Visualisations like this help identify how students progress through a course and when engagement typically begins or drops off.
Dashboard prototype for the Science Skills Platform providing insights into content usage across the transferable skills.