This post is also available in: Spanish
Each time PISA results are released, those of us who analyze them focus on the average performance of countries and, given our line of work, the most disadvantaged groups (women, rural students, and the poor). Yet the finding that never ceases to surprise my Argentine audiences is this: the most privileged students perform much worse than the public would expect.
There seems to be an impression among businessmen, diplomats, and Latin American politicians that, by sending their children to private, bilingual schools in which students take a battery of international exams, they can shield them from the education crisis their countries are in. However, the data indicate not only that there are few (in fact, increasingly fewer) students who excel in Latin America, but that even students from higher socio-economic backgrounds perform poorly.
Very few students excel in reading, math, or science
First and foremost, high school students who excel are few and far between. As Soledad Bos, Emiliana Vegas, and I demonstrated in a new brief from the Inter-American Development Bank (IDB), the percentage of students performing at the highest levels on the PISA exam—i.e., levels 5 and 6 of a total of 6 levels—is small. For example, in math, it is 0.3% in Argentina, 0.8% in Brazil, 1.6% in Chile, 0.3% in Colombia, 0.6% in Costa Rica, Mexico, and Peru, and 1.3% in Uruguay. By comparison, the percentage of students at these levels is 55% in Shanghai-China, 40% in Singapore, 16% in Canada, 15% in New Zealand, and 13% in the average Organization for Economic Co-operation and Development (OECD) country.
The percentage of top-performing students in math and reading has declined
The percentage of students in the region scoring at the higher levels has remained stagnant in the last decade, and when it has changed, it has actually shrunk. In math, the percentage of students at the higher levels dropped from any benchmark since 2003. In reading, this percentage dropped in Argentina (since 2000), Brazil and Chile (since 2009), and Uruguay (from any year since 2003). In science, it has not changed in any country for any reference year.
Almost no Latin American student excels in more than one subject
Even those Latin American students who excel do so in only one subject. Those privileged parents in the region who think their children are receiving a holistic education will be surprised to know that the percentage of students performing at the highest levels in all three PISA subjects is 0.2% in Chile and Uruguay, 0.1% in Brazil and Costa Rica, and 0% in Argentina, Colombia, Mexico, and Peru.
Large Latin American cities are no exception
These national percentages can lead some to think that Latin American cities, home to some of the best universities in the region, are the exception. Unfortunately, this is not the case. In math, the percentage of students at the highest levels is 0.3% in Bogota, 1% in the city of Buenos Aires, 0.9% in Mexico City, 0.3% in Rio de Janeiro, 1.1% in São Paulo.
The wealthiest students in the region perform at the same level as the poorest in OECD countries
This is perhaps the news that most contradicts the expectations of privileged Latin American parents who believe their children receive a first class education. PISA has its own socio-economic and cultural level index that includes parents’ occupation and education, as well as the number and types of resources in a student’s family (including resources that indicate family wealth, cultural possessions, and educational resources). With the exception of Uruguay, the richest 25% of students across Latin America perform worse than the poorest 25% of students in the average OECD country in math. In fact, in Colombia, the richest 25% perform almost a grade level below the poorest 25% of the OECD countries in math.
Schools with the richest students in the region fare just as badly
Although wealthy students do poorly, one could expect that the schools that serve these students, on average, perform like the best schools in developed countries. However, this is not the case. Again, with the exception of Uruguay, 25% of the “wealthiest” schools (defined as schools in which the average student has a socio-economic and cultural level above that of the average student in the country) perform worse than 25% of the “poorest” OECD schools in math.
Even the Latin American “super-nerds” are not all that “nerdy”
The denial of privileged Latin American parents can lead them to believe that their children are the exception. But this is highly unlikely. Even if we take the students from the region who scored in the 95th percentile (meaning they performed better than 95% of the students in their country), with the exception of Chile, they performed like students in the 75th percentile (meaning a student who surpassed 75% of his/her peers) in the average OECD country. Even the Chilean “super nerds” only beat those in the 75th percentile of the OECD by a few points.
These data are good and bad news for the region. The bad news is that the low level of top-performing students on PISA is troubling because it is these students who can exercise critical thinking skills and base decisions on evidence—two qualities we would like to see in the next generation of entrepreneurs and policymakers. The good news is that perhaps upon realizing that the low quality of education in Latin America does not exclude them, leaders in the region will see it is in their best interest to insist on improving education in their countries.