Joshua Keating suggests that it is unfair to compare the educational performance of geographically compact city-states to other countries, and that China is “cheating” by releasing PISA data for Shanghai but not for the country as a whole:
The three “countries” at the top of the PISA rankings are in fact cities—Shanghai, Singapore, and Hong Kong—as is No. 6, Macau. These are all big cities with great schools by any standards, but comparing them against large, geographically dispersed countries is a little misleading.
Shanghai’s No. 1 spot on the rankings is particularly problematic. Singapore is an independent country, obviously, and Hong Kong and Macau are autonomous regions, but why just Shanghai and not the rest of China?
As Tom Loveless for the Brookings Institution wrote earlier this year, “China has an unusual arrangement with the Organization for Economic Co-operation and Development (OECD), the organization responsible for PISA. Other provinces took the 2009 PISA test, but the Chinese government only allowed the release of Shanghai’s scores.”
1. China has a GDP per capita (PPP) of $9,300, roughly the same as Jamaica. It is a middle-income country that is somewhat more affluent than Egypt and Ukraine and somewhat less affluent than Brazil and South Africa. Shanghai has a GDP per capita (PPP) of $20,313, in the same ballpark as Poland, Hungary, and Chile. Shanghai’s PISA results are not labeled as China’s results. Rather, they are labeled as Shanghai’s results, and when we compare Shanghai (with a population of 14.35 million) to Poland (38.54M), Hungary (9.944M), and Chile (17.46M), we see that Shanghai blows them out of the water, just as it blows countries and jurisdictions that are far more affluent out of the water (e.g., US GDP per capita is $50,700, or over twice that of Shanghai). So yes, let’s compare apples to apples … and see that Shanghai is still very impressive. And when we compare Shanghai to U.S. jurisdictions, like the cities of New York city and Los Angeles, or even affluent suburban jurisdictions like Fairfield County, Connecticut, my guess is that Shanghai would still fare well. Note that I don’t think we ought to mimic Shanghai’s schools, and I’m well aware of the fact that methods that work well in a middle-income society undergoing a rapid economic transformation are not necessarily well-suited to a mature market democracy with a diverse population and a large share of children raised in post-marital families.
2. If we dismiss Singapore and Hong Kong on the grounds that they are cities, we might also consider dismissing other highly urbanized countries. Singapore and Hong Kong are both 100 percent urbanized, as you might expect. But should we also discount Belgium (97.5), Israel (91.9), Japan (91.3), Australia (89.2), Denmark (86.9), France (85.8), Sweden (85.2), the Netherlands (83.2), South Korea (83.2), the U.S. (82.4), or Canada (80.7)? Does anyone believe that excluding the relatively small rural minorities in these countries would make much of a difference in comparisons with Singapore and Hong Kong? This might seem silly, as there is a big difference between having your urban population concentrated in one city (as in Singapore and Hong Kong) as opposed to many different metropolitan areas (as in the U.S.). But Canada represents a middle ground, as its urban population is concentrated in a relatively small handful of metropolitan areas, many of which have seen a consolidation of local governments in recent decades (e.g., in Toronto and Montreal). The truth is that Singapore and Hong Kong aren’t as strange as we make them out to be. Yes, they are entrepôt cities situated along key waterways, and entrepôt cities have economic advantages over other regions. Yet many modern cities can be characterized as entrepôts surrounded by vast rural expanses that are of limited economic significance, and most rich market societies are essentially collections of cities. There is no reason in principle why the US and Europe can’t have a dozen or so cities among the 200 or so mid- to large-sized cities in both regions that perform as well Singapore and Hong Kong. And though the data is not readily available, it is not at all clear that this is the case.
Any way you slice it — compare jurisdictions by income level, compare cities to other cities, factor in household income, limit the comparison to students raised by college-educated parents, or native-born students or students from a country’s largest ethnocultural group — you will find that the U.S. is almost never among the top performers. This suggests that we ought to do more to improve performance among the weakest U.S. performers, clearly. But it also suggests that we ought to do more to improve performance among the strongest U.S. performers, a cause that has been neglected by K-12 education reformers, as Rick Hess has argued.