PISA Scores for 10 new Education Systems Announced

This post is also available in: Spanish

The Australian Council for Educational Research (ACER) just released the document “PISA 2009 Plus Results: Performance of 15-year-olds in reading, mathematics and science for 10 additional participants,” which presents results from the Program for International Student Assessment (PISA) for countries that participated after the initial round of examinations.

The new participants include two Latin American education systems: Costa Rica and Venezuela’s Miranda Province. It is relevant to note that Miranda-Venezuela did not meet the PISA standard on schools response rate, and that Himachal Pradesh-India and Tamil Nadu-India did not meet PISA standards for student sampling, so caution should be exercised when interpreting results from these education systems.

Mean Scores on PISA Reading Test, 2009

Key: Dark blue: LAC; Green: Second Round Countries; Red: Costa Rica and Venezuela-Miranda

Sources: OECD (2010) and ACER (2011). Note: Not all differences between countries are statistically significant.

In August, PREAL summarized the first round of PISA results in “Measuring Up? How Latin America and the Caribbean Performed on the 2009 PISA.” We were curious whether the performance of Costa Rica and Miranda-Venezuela would follow the trend of the nine Latin American and Caribbean (LAC) countries that participated previously. Here are some key observations from the new results:

• Low average scores among the 10 new participants bumped the relative results of a few LAC countries out of the lowest third of all participating countries.

• Chile remains the highest-performing LAC country in reading and science, and continues to be tied with Uruguay and Mexico in math. But Costa Rica’s mean scores in reading are so close to Chile’s that the differences are statistically insignificant. Costa Rica also came close to matching Chile’s performance levels in science but scored behind in math.

• At just 3 percent, Trinidad & Tobago and Uruguay continue to have the highest share of top performers in the region in all subjects, while Panama and Peru maintain the highest share of low-performing students at 65 percent. In other words, the new scores do not change the fact that an alarming proportion of students in the region fail to demonstrate minimum levels of performance.
• Costa Rica and Venezuela-Miranda follow the regional trend for gender gaps: girls outperform boys in reading and boys outperform girls in math. Additionally, Costa Rican boys outperformed girls in science (all differences are statistically significant).

• In reading, both Costa Rica and Venezuela-Miranda performed above what their levels of education investment would predict, and Costa Rica performed better than what its GDP per capita would predict (for Miranda, please use caution given its low school response rate and the fact that the results are not representative of the country as a whole).

Key: Dark blue: LAC; Red: Costa Rica and Venezuela-Miranda

Sources: OECD (2010) and ACER (2011). Note: Lowest levels include Leve 1 and below.

Outside of Latin America and the Caribbean, two states in India—Tamil Nadu and Himachal Pradesh—also participated in PISA for the first time. While these two states are not representative of India overall and did not meet PISA standards for student sampling, initial indications seem to suggest that their scores were low, with few high performers. This is contrary to what Pritchett and Viarengo predicted in 2009, when they suggested that India would have low overall scores with significant pockets of high performers. In fact, the share of underperformers for these two states ranged from 82-90% in all subjects.