Elementary schoolgirl reading

Recent results from PISA-D reveal that there is a learning crisis at the top—in many countries even the middle class and “statistical elites” are failing to reach international standards for learning. While much rhetoric around equity in education focuses on bridging gaps within countries, such as between the rich and the poor, the evidence suggests that many education systems are failing to achieve high learning even for the most privileged groups, while less privileged groups do even worse. What role does achieving within-country equality across groups play in reaching universal proficiency goals as highlighted in SDG4?

There’s no question that children in poorer households are disadvantaged in education. Grade attainment is lower for children from poorer households. And household socioeconomic status (SES) is a big factor explaining learning differentials between children from rich and poor households. Furthermore, positive sorting between households and schools can further aggravate inequalities: richer households are able to select better schools (Anand et al., 2018). Work by Das (2018) shows that by Grade 3, gaps in learning emerge between children from households with lower and higher SES, and that these gaps widen dramatically by age 17 due to children from poor households dropping out at higher rates than children from rich households. But a critical question is, how far would closing this wealth gap in learning (achieving learning equality across SES groups) get us towards the international equity goal of universal literacy and numeracy for all?

A new paper addresses this question by decomposing the wealth gap in learning into its key components: differential grade attainment and differential learning achievement per grade. This decomposition is possible because a group of children could have higher literacy or numeracy because more of them have completed a higher grade (the group has a steeper grade attainment profile), more of them are literate in any given grade (the group has a steeper learning profile), or both. Using household survey data from ASER and Uwezo across India, Pakistan, Kenya, Tanzania, and Uganda, the paper explores how close children from poor households would get to universal literacy and numeracy if they had the same grade attainment, learning per grade, or both, as children from rich households. While key international goals focus on equity—defined as universal achievement to a certain standard—many efforts also focus on equality, either of schooling or learning, across groups. Therefore, it is critical to understand the role equality can play, especially across specific groups, in achieving established equity targets.

Equality alone is not sufficient to achieve the equity goal of universal mastery of basics

ASER and Uwezo data show that learning on average is low across the board. Learning is low for both children from poor and rich households relative to any reasonable standard of minimal mastery of basic literacy and numeracy. Achieving an international equity goal, defined as universal literacy and numeracy (which itself is quite a low bar for learning), will require more than just closing the rich-poor learning gap (that is, within-country equality across SES groups)—it will take progress in learning for all.

Figure 1. Even achieving equality with the rich leaves large portions of the poor without basic numeracy skills

Graph showing average growth in learning per year
Source: Akmal & Pritchett. (2019). “Learning Equity Requires More than Equality: Learning Goals and Achievement Gaps between the Rich and the Poor in Five Developing Countries”


A hypothetical scenario where all children from poor households have the grade attainment profiles of children from rich households (while keeping their current learning profiles) would still leave more than 40 percent of the children from poor households innumerate in India, Pakistan, and Uganda (the red bars in figure 1). Equalizing learning profiles between the rich and the poor (while maintaining current grade attainment profiles) would still leave large segments of the poor unable to achieve basic literacy and numeracy (the green bars in figure 1 show this for numeracy). What happens if children from poor households suddenly have the grade attainment and learning achievement profiles of children from rich households, that is, complete within-country equality? Such a scenario would still leave at least a quarter of the children from poor households innumerate in India, Pakistan, Tanzania, and Uganda (the yellow bars in figure 1).

The “hidden exclusion” of lower learning in the same grade is a large part of the equity gap

Breaking this down further, the paper examines the portion of the equity gap that is explained by differential grade attainment and by differential learning within each grade. Excluding Kenya, equalizing grade attainment between children from rich and poor households would only close between 8 percent (India) and 25 percent (Pakistan) of the gap to universal numeracy. Equalizing learning profiles would close between 16 percent (Uganda and Pakistan) and 34 percent (India) of the gap to universal numeracy. Results for literacy show similar projections: excluding Kenya, equalizing grade attainment would close only between 8 percent (Uganda) and 28 percent (Pakistan) of the gap to universal literacy, while equalizing learning profiles would close between 13 percent (Uganda) and 44 percent (India) of the gap to universal literacy.

Children from poor households learning less in each grade is therefore a substantial and often larger part of the equity gap compared to the more widely documented gaps in enrollment and grade attainment. The WDR 2018 calls this “hidden exclusion,” since differential learning within grades typically receives less attention than the easier-to-measure gaps in schooling attainment. Achieving equity, therefore, will require an explicit focus on universal mastery of basic learning outcomes, not only on equality in years of schooling.

Table 1. Learning inequalities at the same grade levels explain a significant proportion of the equity gap

Country Current Learning Levels of Poor (%) Poor-Universal Gap (% points) Percent of Gap to Universal Closed by Having Same Grade Attainment Profile as Rich (%) Percent of Gap to Universal Closed by Having Same Learning Profile as Rich (%) Percent of Gap to Universal Closed by Having Same Grade and Learning Profile as Rich
India 35 65 8 34 42
Kenya 59 41 37 29 54
Pakistan 37 63 25 16 41
Tanzania 52 48 21 33 48
Uganda 38 62 15 16 31

Source: Akmal & Pritchett. (2019). “Learning Equity Requires More than Equality: Learning Goals and Achievement Gaps between the Rich and the Poor in Five Developing Countries”

Learning gaps between children from rich and poor households emerge in early grades

This “hidden exclusion” of learning begins in early grades. In India, the gap in basic numeracy between children from rich and poor households emerges by Grade 3 (also shown by Alcott & Rose, 2017), and persists even after eight years of formal schooling (figure 2). Taking these findings together with those above suggests that both learning improvements targeted at marginalized groups (to address early learning gaps) and broad-based improvements in education systems (that improve learning for all) are needed to achieve international learning targets. One promising approach is addressing problems with overambitious curriculums that assume mastery of grade-level content by teaching at the right level.

Figure 2. The learning gaps between children from rich and poor households emerge early in India

Graph showing learning gaps between children from rich and poor households
Source: Akmal & Pritchett. (2019). “Learning Equity Requires More than Equality: Learning Goals and Achievement Gaps between the Rich and the Poor in Five Developing Countries”

Work using other data sources shows similar results

Wealth is just one dimension of inequality. Other work has examined differences in schooling and learning by gender and found similar results in terms of the contribution of equality to achieving global equity. Kaffenberger and Pritchett (2017) use data on years of schooling and literacy to simulate how women’s literacy would change if (i) women attain the same years of schooling as men (equalizing school attainment), (ii) women achieve the same learning in each year they spend in school (equalizing the learning profile), and (iii) both (equalizing schooling and learning). They find that equalizing both schooling and learning on average would increase women’s literacy by about 9 percentage points, and that this would still leave 30 percent of women illiterate. Achieving universal basic mastery clearly requires more than gender parityit requires substantial improvements for both genders.

Pritchett and Sandefur (2017) examine another “equalization” goal: universal primary school completion. Using DHS data, they examine how literacy levels among young women would be different if all of them had completed primary school. In Nigeria, where 58 percent of the cohort of young women they examine are illiterate, universal primary completion would reduce illiteracy only to 53 percentleaving more than half of the young women functionally illiterate. On average, across nearly 50 developing countries, 40 percent of the young women would have remained illiterate even if all had completed primary school.

Finally, Crouch and Rolleston (2017) look at multiple sources of inequality and find that learning gaps by gender, income, and even urban and rural geography are much smaller than what they call “pure inequality.” Pure inequality is the learning difference between the bottom quartile and top quartile performers, and it is the largest source of learning inequality according to the SACMEQ data (see figure 3). Similar to the findings described above, this indicates that closing the “wealth gap” or the “gender gap” will only go a small way towards closing the “achievement gap.” And, because we know even high performers do poorly by international standards, even closing the achievement gap is but one step on the road towards global equity.

Figure 3. “Pure” inequality between low and high achievers is larger than gaps by gender, wealth, or geography

Graph showing "pure" inequality between high and low achievers and gaps by gender, wealth, or geography
Source: Crouch, L. & Rolleston, C. (2017). “Raising the Floor on Learning Levels: Equitable Improvement Starts with the Tail”


Ensuring equitable quality education (SDG Goal 4)—that is, universal early conceptual mastery of basics—should be a key goal of any education system. Improving equity will require both pro-poor approaches targeting the learning of the marginalized to reduce inequalities, as well as efforts to raise learning levels for all.





Maryam Akmal is a Policy Analyst at the Center for Global Development. She works with Lant Pritchett and Justin Sandefur on effective education systems under the RISE programme. Previously, Maryam worked in Pakistan on education policy. 


Michelle Kaffenberger is a Research Fellow with the RISE Programme where she analyses education systems and learning outcomes. She is also a Senior Research Advisor with the World Bank, focusing on financial inclusion, social protection systems, and international education. She has designed, led, or advised on dozens of quantitative, qualitative, and mixed methods empirical research studies. Michelle has also served as research adviser and consultant to the Bill & Melinda Gates Foundation, the United Nations, the Asian Development Bank, and a variety of private sector and nonprofit organisations. Previously she was research manager and lead analyst at InterMedia, an international development research organisation. She began her career establishing a new base of operations for an international nonprofit in Darjeeling, India.



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