Over the last decade, Turkey has achieved relatively high economic growth at just over 5% per year. As was discussed in a recent blog post, this translated into broadly inclusive growth- not only did income grow for all segments of the population, but there were also improvements in a number of non-income indicators of well-being, including in health and education. Last week the OECD launched the latest results from its Program for International Student Assessment (PISA), a triennial international assessment that evaluates education systems worldwide by testing the skills and knowledge of 15-year-old students. This new data allows us to revisit the question of how Turkey has fared in terms of the performance of its education system in general and with regards to inclusiveness in particular. So what does the data tell us?
First-overall performance. As Figure 1 shows, in all three test subjects Turkey has narrowed the gap with the OECD in the last 9 years, with reading demonstrating the most striking rate of improvement. One can reasonably argue that such improvements aren’t much of a surprise given Turkey’s initially low results in PISA. It’s just convergence! True, but a comparison of the correlation between initial average PISA scores in 2003 and improvements in PISA scores between 2003 and 2012 reveals that Turkey did much better than one would expect based on its initial results. In fact, Turkey has the highest annual change in average PISA scores over this period of any participating country, and it did better than countries that had poorer results than Turkey in 2003, such as Brazil, Mexico or Indonesia.
So how inclusive was this improvement? Again Turkey performed impressively. If any student, irrespective of their background has an equal chance of performing well in school then this is a strong indicator of a more equitable education system. The importance of the socio-economic background of one’s parents in Turkey in determining performance in mathematics fell significantly between 2003 and 2012. In fact, Turkey is the most successful country, except for Liechtenstein, in reducing the effect of family background in education success. Moreover, the improvement in scores was faster for the poorer students than for the better-off students. The average PISA scores of students in the poorest quintile increased by 56 points (or 14.5 percent) between 2003 and 2012 as compared with an increase of 25 points in the richest quintile (5 percent). As a result, the “achievement gap” between the richest and the poorest socioeconomic groups has shrunk from 123 points to 92 points.
How did Turkey manage to achieve these striking improvements in education? Many factors probably combined to produce this effect and further work would be needed to disentangle the causes, but here are some likely candidates. Firstly, increasing financing probably played a part. Between 2004 and 2012, public per capita expenditure on education increased by 7.4 percent annually. In 2012, Turkey spent almost 80 percent more on education per capita in real terms compared to 2004. As a result, the number of teachers increased fast. And teacher salaries doubled in real terms over the last decade, which will have made the teaching profession more attractive and probably encouraged better applicants to apply for teaching positions. Together with the fact that from 2002 teachers were recruited based on the results of a central examination, we would expect this to have resulted in an increase in teacher quality. Financing also supported an increase in resources devoted to physical and IT infrastructure. Although more money is strongly correlated with improved education outcomes, we know that this is not the only factor. More resources can be good, but how you use them is as important, if not more so. What else did Turkey do right? Starting in 2006/07 several curricula changes were introduced that seem like a likely candidate. The new curricula not only changed the content of school education but encouraged innovative teaching methods with an emphasis on student-centered-learning. This provided a more active role for students instead of memorizing information, which was the main approach prior to the changes.
This is all very positive. But despite significant improvement in the last 9 years, it’s important to recognize that Turkey’s PISA scores are still relatively low and Turkey’s relative performance with respect to its income level is not striking. Moreover, inequality in education outcomes is still a challenge. Turkey needs to do more- we’ll use the next blog to explore what PISA tells us about priority areas for policy reform going forward.