Why learning loss matters

Learning loss has become a defining theme of Covid. Last month we reported our staggering finding that almost half of students in the US and UK experienced some knowledge loss in mathematics during the first lockdown, with an average loss of eight months.

But not everyone is convinced by the importance of ‘learning loss’. A fixation with quantifying students’ knowledge, it is suggested, risks neglecting the other impacts of Covid that are not so easy to measure, such as their social and emotional wellbeing. Such arguments, which have previously taken aim at attempts to establish global learning metrics, ask legitimate questions around the purpose of education. They query whether we are measuring the right things and if, in light of Covid, we should instead rethink our very conceptions of education and schooling.

The educational impacts of Covid are of course many, with some easier to measure than others. But this is not a zero-sum game, and we should resist false choices. Short-term measures of learning loss (perhaps more aptly termed ‘knowledge loss’) carry profound implications for students’ – and thus society’s – long-term prosperity. To pit this against other (equally important) educational outcomes is a dereliction of our duty to students.

Figure 1: Estimates of learning loss by topic during 2020 lockdowns in US and UK, based on a sample of 883 students

A partial measure

Learning loss, as the term is usually invoked, gives some sense of how much knowledge has slipped away from students’ minds. Ever since Herman Ebbinghaus published his ‘forgetting curves’ over a century ago, educators have recognised the importance of sustained access to learning. Our findings simply put numbers to the inevitable. Switch off over a summer, or longer as is sadly the case during a pandemic, and that hard-earned knowledge erodes in a predictable manner. It can be won back, of course, but only if students receive targeted support that addresses their specific knowledge gaps. This isn’t about administering high-stakes exams and putting teachers and schools to the sword; it’s about rescuing learning potential, one student at a time. Measuring learning loss is simply the first step towards recovering it.

There is a case to be made that we were expecting too much of students (and parents and teachers) in the first place; that curricula worldwide are a mile wide and an inch deep, and that they would now benefit from a reduced emphasis on knowledge acquisition. It would be a welcome consequence of Covid if our governing assumptions around what students should know, and how much they should know, were challenged. Maths curricula have generally been slow to adapt to rapid changes in society. Technological development, and the very real threat of automation, demands that we scrutinise the very knowledge and skills students will need to thrive in tomorrow’s world. Covid may trigger a long-overdue debate around both the quantity and quality of school maths itself. But that is no reason to ignore learning loss in the meantime, especially when it is so pronounced.

A class issue

Above all, metrics like Whizz’s Maths Age can shine a light on inequality. We know that it’s the most vulnerable communities – those starved of resources and with limited or no access to online learning – that suffer the most. This is known for the so-called ‘summer slide’ and a study carried out by the Education Endowment Foundation affirms as much for the damaging effects of Covid. The argument that students are lifelong learners who acquire knowledge naturally relies on having a nurturing home environment with parents who are both available and informed enough to support their children’s development. This is not a given, and it is a cruel consequence of societal inequality that the safety net is weakest for the poorest communities. It is helpful to have an order-of-magnitude estimate of just how stark educational inequalities are during times of crisis. I would challenge those who reject the term on whether they could stomach an eight-month knowledge drop for their own children. It’s not something most educated, middle-class parents have to worry about – they have the resources and mechanisms to ensure their children can continue to learn and thrive even during school closures. The same simply isn’t true across all of society. Learning loss is nothing short of a class issue.

So yes, let’s rethink the purposes of schooling. Let’s redesign the curriculum and if we must, let’s strip away some of the presumptive foundations of the curriculum. Let’s also pay due attention to children’s holistic development and safeguard their mental wellbeing alongside learning itself. But let us not forget, even for a moment, that learning loss is a real phenomenon that widens the opportunity gap in society. We can discuss and debate what kinds of knowledge should fill a curriculum to our heart’s content, but we can surely all agree that every child, regardless of their background, is entitled to a basic level of education. It is, after all, enshrined in the universal declaration of Human Rights and SDG4. Global learning metrics like Maths Age are simply a reminder that we are far away from securing the safety net every child has a right to.

When we lurch to the extreme of removing all measurement from education, we create blind spots in the places that are most deserving of our attention.