We Cannot Ignore the Reality of Global COVID Learning Loss (News and Research 318)

‘We cannot ignore the reality of global Covid learning loss’| During the pandemic, school closures ranged from almost no closure at all for a handful of countries, such as Sweden, to a total between 10 and 30 weeks for most countries reporting solid school data.

World Bank and Unesco analysis shows many countries with learning loss proportional to the duration of school closures – and around the globe, schools for 168 million young people closed for almost a full year.

Indeed, for Brazil, Norway and Poland, the learning losses are almost equivalent to a full school year.

But the topline figures do not tell the whole story: the full picture is one of significant variation in impact – mathematical attainment is particularly hard-hit, and there has been a tendency for young people from lower socioeconomic backgrounds to be most affected.

Even in single nations, the effects are highly distributed and with uneven impact.

All this does not mean we had “a normal interruption of normal school activities”. We did not stop to now resume as if nothing more than a “short vacation from schooling” had happened.

During remote learning, a few students progressed normally, other students halted their progress and most students progressed but with many voids, gaps and flaws in their knowledge and skill, and in their learning identity and “school connectedness”.

Even more serious is the fact that these voids are not uniform: where some students may have progressed, others did not. Unfortunately, education agents – such as boards of education and ministries – may not have a full understanding of these deficiencies.

Instead of facing this reality, some tend to close their eyes and deny the facts. The rhetoric of “back to normal” is strong.

The denials of learning loss came early. Prominent voices doubted the existence of learning loss. The children are resilient, they will recover, they won’t forget what they have learned, and they will catch up.

But, if absence from schooling has no negative consequences, then why go to school at all?

What can we do now? Three ways to take action

First, we must understand the situation in each school, each class and each student. We need assessment tools at all levels. This has already happened in countries in which exams and other forms of assessment happen regularly. But countries and systems that have shunned serious evaluations have no other way to face the reality and act. Only through assessment can the distribution and scale of learning loss be understood and action taken.

Second, this needs increased attention to the curriculum learning goals. There cannot be serious assessment if goals are not clear. Assessment is always done against criteria and standards; sometimes implicit, but better when they are explicit, and public. Wherever curricular goals are loosely defined, recovery is more challenging.

Third, we need to understand that the effects of Covid are highly individual; they are widely and unpredictably distributed – extreme in some, mild in others. The best policy action is targeted, focussed on those adversely affected, and matching actions to specific needs. Tutoring, special classes and increased schooling all need to be considered, even in the context of global financial pressures. While we need to focus on how each young person has been affected, some approaches to differentiation can increase inequity if they lead to lowered or heavily personalised learning standards. Ambitious and well-defined curricular goals are for all, and assessment using these can correct misalignments and inequalities, resulting not only in support to individual young people but to collective improvement.

What part of all this provides the biggest challenge for public policy? The most affected are likely to be less visible to services, and without voice. Our schools must attend to these highly vulnerable young people.

Projections of adult skills and the effect of COVID-19 | This paper projects Skills in Literacy Adjusted Mean Years of Schooling (SLAMYS) for the working age population in 45 countries and quinquennial time periods until 2050 according to various population scenarios. Moreover, the effect of school closures due to the COVID-19 pandemic on these projections is integrated. Adult skills are projected using the cohort components method. They can help in assessing the potential consequences of the recent trends for the adult population, particularly the workforce, whose skills are essential for the jobs contributing to economic growth and development outlooks. Our projections are novel as they take into account both the amount of schooling and quality of education and also consider the changes in adult skills through lifetime. Projections show that the adult skills gap between countries in the Global North and countries in the Global South will likely continue to exist by 2050, even under very optimistic assumptions–but may widen or narrow depending on the demographic development trajectories specific to each country. Moreover, the loss of learning due to school closures during the COVID-19 pandemic further exacerbates inequalities between countries. Particularly, in countries where schools have been closed for a prolonged period of time and the infrastructure for effective online schooling is lacking, the skills of cohorts who were in school during the pandemic have been severely affected. The fact that the duration of school closures has been longer in many low- and middle-income countries is a serious concern for achieving global human capital equality. The impact of the COVID-19 pandemic is projected to erase decades-long gains in adult skills for affected cohorts unless policies to mitigate learning loss are implemented immediately. [Tes Magazine, December 7, 2022, by Tim Oates, Nuno Crato, Harry Patrinos]

Heterogeneity in School Value-Added and the Private Premium | Using rich panel data from Pakistan, test scores based measures of quality (School Value-Addeds or SVAs) are computed for more than 800 schools across 112 villages and verify that they are valid and unbiased. With the SVA measures, we then document three striking features of the schooling environment. First, there is substantial within-village variation in quality. The annualized difference in learning between the best and worst performing school in the same village is 0.4 sd; compounded over 5 years of primary schooling, this difference is similar in size to the test score gap between low- and high-income countries. Second, students learn more in private schools (0.15 sd per year on average), but substantial within-sector variation in quality means that the effects of reallocating students from public to private schools can range from -0.35sd to +0.65sd. Thus, there is a range of possible causal estimates of the private premium, a feature of the environment we illustrate using three different identification approaches. Finally, parents appear to recognize and reward SVA in the private sector, but the link between parental demand and SVA is weaker in the public sector. These results have implications for both the measurement of the private premium and how we design and evaluate policies that reallocate children across schools, such as school closures and vouchers.

Women in the pipeline: A dynamic decomposition of firm pay gaps | This paper proposes a new decomposition method to understand how gender pay gaps arise within firms. The method accounts for pipeline effects, non-stationary environments, and dynamic interactions between pay gap components. The decomposition is applied to a new data set covering all employees at the World Bank Group between 1987 and 2015 and shows that historical differences in the positions for which men and women were hired account for 77% of today’s average salary difference, dwarfing the roles of entry salaries, salary growth, or retention differences. Forward simulations show that 20% of the total gap can be assigned to pipeline effects that would resolve mechanically with time.