Disrupted learning during COVID-19: The impact of school closures on education outcomes in South Africa
In June 2020, after a two-month COVID-19 lockdown, South Africa’s schools started reopening in a phased approach. However, with the pandemic heading for a peak at the end of July, the government announced a further four-week postponement of most learners’ return. The learning losses resulting from this disrupted teaching will disproportionately affect the most disadvantaged learners, risking a deepening of inequality in the country. This is an updated version of an article by Vijay Reddy, Crain Soudien and Lolita Winnaar, which first appeared in The Conversation.
By mid-April about 1.725 billion students globally had been affected by the closure of schools and higher-education institutions in response to the coronavirus pandemic. According to the UNESCO Monitoring Report, 192 countries had implemented nationwide closures, affecting about 99% of the world’s student population. In the 1980s during apartheid many South Africans saw school boycotts bring schooling to sporadic halts — but the scale and complexity of what has been happening during the COVID-19 lockdown is entirely new territory.
In recent decades, crises such as natural disasters, armed conflicts and epidemics have disrupted education around the world. For example, in 2005, Hurricane Katrina in the US destroyed 110 of the 126 public schools in New Orleans. In the past decade at least 2.8 million Syrian children have been out of school for some period, and in 2013 the Ebola epidemic in West Africa saw 5 million children out of school.
School closures affect students, teachers and families and have far-reaching economic and social effects. This is especially so for fragile education systems and the negative effects will be more severe for disadvantaged learners and their families.
We see from previous health emergencies that the impact on education is likely to be most devastating in countries with low learning outcomes and high dropout rates. South Africa has low and socially graded learning outcomes. The 2016 Progress in International Reading Literacy Study showed that only one in four South African Grade 4 learners could read with meaning. The 2015 Trends in International Mathematics and Science Study (TIMSS) found that most Grade 9 learners were yet to achieve a minimum level of competency in mathematics and science.
In response to COVID-19 school closures and adherence to social distancing, UNESCO and many governments and agencies have recommended the use of distance learning, open educational applications and online learning to reduce disruption to education.
Richer households are better placed to sustain learning through online learning strategies, although with a great deal of effort and challenges for teachers and parents. In poorer households many children don’t have a desk, books, internet connectivity, a computer, or parents who can take the role of home schooling. The disparity in access to digital devices and connectivity between rich and poor is immense.
For learners in no-fee schools, only one out of five learners has a computer, and only half of them have an internet connection at home. For learners in fee-paying schools, around half have a computer and three-quarters an internet connection. Access to cell phones is higher, with close to 80% of learners having their own. One-third of learners in no-fee schools and just over half of those in fee-paying schools have parents with a post-secondary education. This differentiation in social capital and resources means a differentiated set of learning experiences at home.
While it is necessary to institute educational programmes during this period, these will not replace regular school. Despite the best efforts of the government, schools and parents there will be learning losses for almost everybody and worsened educational outcomes for the poor.
We applied the learning curve scenario methodology developed by the World Bank to the South African TIMSS data to illustrate the patterns of expected learning losses over the next few months due to school closures and disruptions.
Possible education outcomes
We plotted the kernel density learning curves for the 2015 study scores and a (hypothetical) reduced learning average across the distribution. We then plotted the kernel density curves for the fee-paying and no-fee schools and the reduced learning averages for each.
We expect higher learning losses in no-fee schools. While we don’t know the exact value of the learning losses, these graphs illustrate that learning loss patterns will be different for learners in different school types.
In graph A, the solid line shows South African learners’ mathematics achievement. The dotted line shows the impact of learning losses on the learner achievement profile. Graph B shows the expected learning graphs for current achievement and expected reduced achievement in no-fee and fee-paying schools.
We use the cut-off score of 300 to show the numbers of poorly achieving learners. The shaded portions in these graphs show the increased proportion of learners with very low achievement scores. The existing data show a larger proportion of learners in no-fee schools obtaining scores below the 300-point cut-off, compared to learners in fee-paying schools.
Our findings underscore the fact that disasters amplify existing structural inequalities in society and worsen inequalities through an unequal recovery process.
Parental and family support is important during this period. Parents and family must consciously and deliberately support children in completing a few hours of school work every day. An HSRC study on early educational environments found that close to one-third of parents reported that they read books to their children and played with them using the alphabet, number toys and word games.
Half of them reported that they wrote numbers, watched educational TV and sang songs with their children. The patterns are different for learners in fee and no-fee schools - but home-educational activities are happening and parents must be supported and encouraged to continue with them.
It is also important to start preparing for the recovery period when schools reopen. The curriculum must be simplified, targeting areas where learning loss will be most consequential for the following years. In the recovery phase, schools should arrange for additional lessons using the expertise of ex-teachers and university students from the community. While the short-term goal is to modify the curriculum, in the longer term the conceptual gaps have to be filled to ensure that children are not left behind in their schooling.
The sad and uncomfortable truth is that for South Africa - with low and unequal achievement scores - the longer that social distancing is in place, the bigger the learning losses for learners, especially the most disadvantaged, thereby deepening inequalities.
Authors: Dr Vijay Reddy, a distinguished research specialist, Prof Crain Soudien, CEO, and Dr Lolita Winnaar, a senior research specialist at the HSRC