In February 2014, I posted a blog about the impressive results of Vietnam in the 2012 Program for International Student Assesment (PISA) . Why was I surprised? The reason is simple: Vietnam, despite being one of the poorest countries in that race registered one of the highest scores, breaking the almost omnipresent and negative relationship between socioeconomic status and education quality outcomes.
Fortunately, we also registered surprises in Latin America.
In recent days, the government and the media released the results of the national tests of Peruvian students and presented their evolution. In one year, Peru increased the number of students with satisfactory level in math and reading comprehension in approximately 10%. In addition the percentage of students that were in the initial level was reduced by 12% in math and 3% in reading comprehension. (Read more about the evaluation here)
These results show an acceleration of the previous trend, clearly described by my colleagues Mariana Alfonso and Hugo Ñopo in their last post Peru: The middle-income country in search of better learning outcomes. But above all, it has decreased the relationship between socioeconomic conditions and education outcomes.
Thus, using the Human Development Index (HDI) as an indicator of socioeconomic status, we see that between 2013 and 2014 in the poorest areas of Peru (an HDI of 0.20) the probability of being in the initial level of performance was reduced by almost half, while the probability of being in the satisfactory level was doubled. Traditionally, the HDI has been a measure that predicts the performance of students. In this sense, students of low socioeconomic status consistently tend to learn less.
However, the power of the HDI to predict that a student is at the initial level of learning of the national test, was reduced from 46% to 31%. Likewise, the predictive power of belonging to the satisfactory level went from 55% to 48%. In other words, the HDI matters less and less when it comes to learning outcomes, at least in second grade.
The results show more surprises. For example, there are important differences between the results of student learning reported by the provinces with similar HDI. Thus, the probability of Sánchez Cerro students to obtain a satisfactory level in the national test is four times higher than students form Coronel Portillo, although both provinces registered a similar level of HDI.
Although some conditions are not captured by the HDI, these large differences in results show that there are provinces whose results are more positive regardless of the socioeconomic status of its students. Another surprise is that in the reading comprehension test, Lima is not among the provinces with the top 10 scores.
But is there a free surprise or “free lunch”, like my friends who are economists would say? The answer is no. The progress that Peru has made is due to relatively consistent policies that have been implemented in the education sector for the last years, beginning with the continuous application of tests to measure student performance. These measures have focused on the quality of education and student learning, but also a permanent focus on disadvantaged areas.
The other big challenge Peru faces will be to maintain this trend of improving student learning over time. In other countries of the region, the national tests are also showing important surprises, which I will reveal in my next blog post. Stay tuned!