Learning trajectories and policy simulations: A new online tool and data explorer

• In Event: Tools for analysing education systems

Proposal

While most children now spend many years in school, they often emerge without even the most basic reading and math skills. Learning trajectories show how much children learn as they progress through school. They can help us understand the process that leads to low learning outcomes, when and how to intervene during that process, and what the simulated impact of different policies might be on the urgent challenge of low learning.

While a growing academic literature exists around learning trajectories (Kaffenberger, Savage and Crouch, 2021), these insights have not yet been disseminated and made widely available to a policy and practitioner audience. To facilitate this, a team of researchers from the Global Education Monitoring Report (GEMR) and Research on Improving Systems of Education (RISE) programme have designed a new online data explorer tool for learning trajectories. This interactive tool allows users to build, compare, and download bespoke learning trajectories and related policy simulations using data relevant to them. The current version of the tool uses MICS6 data, a large international household survey with recent data on foundational reading and mathematics skills that is nationally representative for children ages 7-14 in 32 countries (including 26 low and lower-middle income countries). The tool is embedded in the SCOPE website, an online monitoring tool telling the story behind SDG 4 data that is maintained by the GEMR.

We will showcase the tools’ capabilities by applying it to two critical questions:
• How quickly do children acquire foundational skills as they progress through school?
• What policy priorities for improving foundational learning can be most effective?

In answer to the first question, descriptive learning trajectories reveal that the pace of foundational learning is far too slow in many countries. Flat learning trajectories during the first years of school result in only 1 out of 4 children in the average country in the dataset acquiring basic reading skills by Grade 3. Since low learning has its roots in the first years of school, a key policy takeaway is for countries to measure and prioritize foundational skills from the early grades.

To explore which policies might be most effective at improving learning, the tool allows users to run two sets of policy simulations using learning trajectories. The first set of simulations show that even a tremendously successful set of access policies would have a modest impact on learning, but that achieving the learning per grade of higher-performing peer countries has significant estimated impacts. The second set of simulations estimate the impact of reducing inequality between advantaged and disadvantaged groups (i.e. rich and poor) within a country. The most striking finding is that even achieving complete equality between rich and poor – or boys and girls, or urban and rural areas - would have a limited impact in many countries. Learning is low for everyone, both the advantaged and disadvantaged alike.

Authors

• Michelle Kaffenberger, RISE Programme, University of Oxford

• Rastee Chaudhry, RISE Programme, University of Oxford

• Jason Silberstein, RISE Programme, University of Oxford