It’s better to be a poor pupil in a rich country than the reverse
THERE ARE some things money can’t buy. Education, however, does not appear to be among them—at least as measured by performance on international exams. On average, pupils in wealthy countries obtain vastly higher test scores than those in developing ones. In turn, strong students tend to become productive workers, making the mostly rich economies they join richer still.
The exact mechanism by which knowledge is bought remains unclear. Do students in the rich world fare better because their governments provide superior schools? Or is the reason that they tend to have richer parents, and enjoy more educational resources at home? A new working paper by Dev Patel of Harvard University and Justin Sandefur of the Centre for Global Development, a think-tank, offers strong evidence that the wealth of a country affects exam results just as much as the wealth of a pupil’s household does.
Evaluating test scores around the world is harder than it sounds. Although pupils in the rich world mostly take one of a few big international exams, many developing countries rely on regional tests. This precludes apples-to-apples comparisons.
The authors surmounted this obstacle by fielding an exam in 2016 for 2,314 children in Bihar, in northern India. It included both questions from the leading tests and ones taken from smaller exams. Using answers from the same pupils on the same day to questions from different tests, they built a statistical model they called a “Rosetta Stone”. It can translate scores from a range of exams—such as one used solely in west Africa—into an equivalent mark on other common international tests.
Messrs Patel and Sandefur then used these equations to estimate how pupils in 80 different countries would fare on the benchmark Trends in International Mathematics and Science Study (TIMSS). Their data show that the wealth of a student’s country and family have similar impacts on test scores—meaning that big gaps in GDP per person matter more than small ones in household income do. For example, pupils from families that are very poor by rich-world standards—those earning $5,000 a year (measured in 2005 dollars)—were expected to score around 500 out of 1,000 on the TIMSS in America, and 560 in Japan. In contrast, those whose parents make $10,000 a year in an upper-middle-income country, such as Costa Rica, still manage only the equivalent of a 475.
The study also found that the influence of parental earnings is not constant. Instead, an extra $1,000 in family income “buys” a larger increase in test scores in highly unequal countries than it does in ones that split their economic pie more evenly. One possible reason is that elites tend to educate their children privately in places where wealth is concentrated, such as Brazil. In contrast, in countries with relatively flat income distributions, like Croatia or Armenia, pupils from different social classes are more likely to attend the same schools. This could reduce the impact of family wealth on test scores. ■
Sources: “A Rosetta Stone for Human Capital”, by D. Patel and J. Sandefur; World Bank; The Economist
This article appeared in the Graphic detail section of the print edition under the headline "A rising tide"
- Bitcoin Outlook: Bullish Scenario May Play Out if Key Technical Support Holds
- AUD/USD Rate Faces Australia CPI Report & Fed Rate Decision
- Trading Earnings Season: 3 Steps for Using Earnings Reports
- Forex Vs Stocks: Top Differences & How to Trade Them
- How to Learn Technical Analysis with DailyFX