What Does Low Learning Really Feel Like?

What Does Low Learning Really Feel Like?

Source: Luis Crouch, using data from EGRA learning assessments and other publications, and taking a stylized average across sets of countries.


Using timed learning assessments that look at reading fluency, I created the animated chart shown above, also featured in a blog in Quartz. The unabashed aim is to try to shock more actors into action. True, this looks at fluency, not comprehension. And it is comprehension that is the goal -- and what gets measured in assessments such as PIRLS and PISA. Nonetheless, the two may be causally connected, at least partly. But completely aside from that (we don’t want to get into that debate here), even if fluency is not causally connected to understanding, it should be clear that if a child, later a youth, cannot read at a fair rate, she simply won’t get past a decent secondary school’s reading load: thus, a flat learning path. RISE leaders have argued (e.g., here) that to avoid a flat learning path, it is necessary to make sure that children master the basics as early in their schooling careers as possible, even if at first this may require an apparently simpler curriculum or apparently more basic ambition. RISE leaders have also shown evidence (e.g., here) that educational progress seems to happen best (at least in going from very low average performance to middling average performance) when countries take care of those who are learning the least and make sure they get off the lowest proficiency levels, again, by making sure that the basics are covered. It does not get more basic than actually being able to read, with fluency and comprehension.

Dramatizing the issue can help. Various ways of showing the problem have been put forth. In RISE vision documents (available here) RISE leaders have produced graphical illustrations of how flat the “learning path” is for children in most of the developing world. The UNESCO Institute for Statistics has put forward the sobering estimate that there are some 617 million children in the world not learning at a basic level, and that two-thirds of them are in school. The data numerically show the truth behind “schooling ain’t learning,” the subtitle to a book by Lant Pritchett, Research Director of RISE.

The RISE Programme is informed by the fact that education systems in much of the developing world are coherent around, or optimized for, getting children into school, but they are not coherent around learning. This makes sense. Systems respond to the incentives and signals they are given by their “principals”: political and technical leaders both within countries and in the global community. The MDGs were highly focused on access. Perhaps the single most important indicators that the global community focused on were numbers of children out of school or enrollment rates. NGOs typically advocated for the elimination of fees. Communities advocated and pressured so that schools would be built and teachers placed in them. Governments were rewarded by their own electorates and by the international community for creating policies such as free, universal primary education. Getting so many children into school, and even getting them to stay there for quite a few years, was a historical achievement. But in the last decade or so, governments and the international community have come to recognize that children sit in school for a lot of years without learning much, and that this is not just a waste of money, but, more importantly, a tragic waste of children’s lives and potential. Increasingly, countries and the international community are paying attention to learning levels. The SDGs, unlike the MDGs, call for tracking the percentage of children that reach at least minimum proficiency in reading and mathematics.

Another fact informing the RISE Programme is that part of the reason for the failure to improve average learning levels is that school systems tend to focus on the children who are easiest to teach and that’s not where the problem is. Urban children and children of wealthy parents tend to learn more. Urban and relatively wealthy parents are able to extract accountability and some standards, even from public schools. Rural and poor children tend to learn much less. For poor children, quality is not only lower, but also strongly random. Even inside classrooms, the children who are “destined to learn” are often seated at the front, and teachers work mostly with them. Curricula and testing are often aimed at selection, not education. 

In order to make progress, systems have to care about learning levels of all children. Only then can we succeed in preparing the next generation to achieve its full potential. And a good place to start is by making sure that all children master the basics quickly and efficiently, so that the readers on the left hand side of the illustration come to resemble the ones on the right hand side. This will require systems to be optimized around learning, not just access.


Luis Crouch is the Chief Technical Officer at RTI’s International Development Group and a member of the RISE Intellectual Leadership Team. He specialises in education policy, decentralised finance (e.g., funding formulas), political economy of reform, education statistics, planning, and projections. He has experience in all key areas of education data analysis, from the generation of primary data via surveys and citizen input, to statistical and econometric analysis, to evidence-based, Cabinet-level policy dialogue. He has previously worked at the World Bank and at the Global Partnership for Education. He has worked closely on South Africa’s education sector funding reforms, Egypt’s decentralisation experiments, and decentralisation and other policy reforms in Peru and Indonesia. His more recent work is in early grade reading and Early Childhood Development, as key entry-points to improving quality. He has worked in more than 25 countries in a 30-year career in development, and is the author of reports, technical papers, and books. 


RISE blog posts reflect the views of the authors and do not necessarily represent the views of the organisation or our funders.