Tiger Photo courtesy of Wong Mei Teng

We’ve received comments from readers before about the PISA: how it isn’t an accurate test for judging differences between populations, how it’s a surprisingly good test for measuring change (or lack thereof) over time, how it shouldn’t be used to support cases for academic reform or teaching efficacy. Others have pointed out that low PISA scores don’t correlate with slow economies (although this claim is problematic too; see note at bottom). Now new research from Ben-Gurion University in Israel has thrown another wrench into the mix: they suggest that PISA scores aren’t about school at all, but about home culture.

The study compared test scores of Chinese students in Shanghai, often (and controversially seen as the best educated students in the world, with their peers who had immigrated to Australia and New Zealand. They also compared those two groups with non-Chinese Australians and New Zealanders as a group. (Australia and New Zealand were chosen because their students were the easiest to differentiate by ethnicity.) The researchers found a small difference of 15 points (out of 1000) between students abroad and students in China, and a considerably larger difference between immigrants and non-immigrant students in the English-speaking countries. Because the statistical methods of PISA scores are believed to be unreliable, the difference in immigrant and native Chinese students’ scores is probably not significant; it is less clear whether the larger difference found in Australia and New Zealand is.

The authors of the report claim that their study shows that schooling is less important in PISA scores than ‘culture’, construed broadly: Chinese families are more likely to emphasize the importance of schoolwork and spend time studying. But this is a problematic idea, too. First of all, there is the “tiger mothers” stereotype, which is doubly harmful: stigmatizing high academic achievement and drawing attention to cultural difference. Numerous studies have shown that Chinese parents do not perceive their parenting as tyrannical; in fact, harsh parenting may negatively impact academic achievement and self-esteem, and there is no evidence that a change in parenting style affects academic achievement. Finally, the firestorm of criticism that has been attached to Amy Chua, the original ‘tiger mother’, might make schools think twice before pointing to her children’s success as an example.

Moreover, immigrant students — or children of immigrant families — may be more likely to perform well in school on the whole, regardless of where they come from. While some studies have found that language difference is a barrier for some students, others have seen no difference at all. It is possible that immigrants in themselves have more intrinsic motivation to succeed, or that more advantaged — intellectually, socioeconomically, or in some other way — families are able to immigrate.

A question that the study was unable to answer, but which could potentially help tease out these issues, is that of study abroad. Study-abroad students are often more motivated and more advanced than their peers. If a similar difference were seen in visiting and newcomer students, ‘culture’ may not be at work at all; instead, we should see immigrants as a specific subgroup with a high likelihood of academic achievement. In other words, they would succeed in any academic environment.

The study does highlight one wrong response to PISA: changing instruction to be more like China. Without further evidence of the efficacy of such changes, they may simply be a waste of students’ time and government money. Since simple strategies like reading aloud can help boost student achievement, a cheaper and simpler first step may be to help parents recreate ‘immigrant culture’ at home.

Note: The ACSD argues that the US’ midrange PISA scores don’t correlate with its economic health or strength — that is, that the US economy has continued a strong growth pattern since the 1970s, despite its “stagnant” (their words, not mine) test performance. This is true in fact, but somewhat misleading. First, it is sad that despite the increasing amount of time and attention being devoted to education reform, test scores measuring basic reading and mathematical literacy have not improved. And second, GDP per capita is not the best measure of the economic health of the average American — especially in 2014, with similarly “stagnant” high unemployment. How much of that economic health is the 1% skewing our data?

About the Author: Jaclyn Neel is a visiting Assistant Professor in Ancient History at York University in Toronto, Ontario.