Bad (and unequal) education encore

The Education Reform Initiative (ERI), an autonomous think-and-do-tank within Sabancı University in its director Batuhan Aydagül’s own words, showed inequality in Turkish students’ academic performances in its analysis of the 2011 International Mathematics and Science Study (TIMMS) exam, which was released in September.

While the best Turkish students scored higher than the best students in Singapore, the country with the highest marks, the worst scored lower than the worst students in Morocco and Ghana, which were at the bottom of the rankings. Işıl Oral, one of the authors of the study, noted at the time that the next step was to analyze the determinants of this inequality. A new ERI report co-written by her does just that.

In “Equality and Academic Success in Turkish Education,” which was released on May 22, ERI researchers Oral and Eileen Joyce McGivney conduct statistical analyses to determine the relationship between academic success and socioeconomic status. Their results are generally not surprising, but still very striking:

There are significant regional differences in access to education. While girls’ secondary school net schooling rate is above 80 percent in certain parts of western Turkey, it is below 50 percent in all of eastern Turkey. In these regions, boys’ schooling rates are 7-8 percentage points higher than girls’.

These regional results are partly due to a language and ethnicity effect: Students from Kurdish-speaking households are two years behind their peers from Turkish-speaking households in mathematical skills. But there are other factors at play as well.

Istanbul think tank Betam’s earlier research showed that fathers’ education has a positive impact on high school children’s test scores. The ERI study finds that heads of households who are college grads spend three times as much on their kids’ education as high school grads, who in turn spend twice as much as middle school grads. This cannot be explained by income differences alone.

You could argue that more privileged kids have access to better schools. That is true, partly because schools in rich neighborhoods receive more informal funding from parents, as Aydagül explained to me back in September. However, two thirds of TIMMS 8th grade math score variance is due to intra-school student differences.

Socioeconomic-induced inequality is actually aggravated by the system. While primary school students from high socioeconomic classes continue on to high-quality science high schools, their underprivileged peers attend poor vocational schools. As a result, although socioeconomic status is the main factor affecting a student’s performance in primary school, the problem has morphed into one of system-wide inequality.

Oral and McGivney note in both the introduction and conclusion of the report, probably to make sure that Prime Minister Erdoğan and his education minister get it, that the best educational systems are those that combine quality and equality. By this measure, the Turkish system is one of the worst.