Children of another tongue

The OECD’s new PISA exam scores were announced this week. The PISA test is a useful index to compare the educational attainment of children from different countries. This year, two things marked the announcement of the 2012 exam scores in over 60 countries for me. 

One was the discussion of the meaning of PISA comparisons between countries. China’s results are a part of the furor. The claim is that the Chinese conducted tests only on students around Shanghai, where the school system is better than in other parts of the country. That would mean that the data is skewed. Regardless of whether or not this is true, there is no question that sampling is important. Note that PISA tests are conducted in official languages. Some things will inevitably be lost in translation.

Despite all that, the PISA scores do have meaning. If China’s data is inflated, others are likely to be as well. You could compare your unusually good school results with that of the Chinese to get a better idea. 

At the end of the day, the whole idea of testing the analytical abilities of children around the world is superb. That is how we can place Turkey’s education system relative to those in the rest of the world. It turns out that our system is still in shambles. Nothing new there. Turkey is improving, but it still ranks 44th among 65 countries. 

Let me touch on to the second issue that marked this year’s PISA score announcement. In Turkey, the percentage of students declaring that their first language is different from the test language rose to 6.3 percent. PISA tests are conducted in the official language of every country, so children were able to note whether this was their first language or not. This led me to thinking whether we have a sampling error in the case of Turkey, too. Let me elaborate. 

The first language response is interesting for two reasons. One is that an increasing number of kids in Turkey are born into native languages other than Turkish. We appear to be in a Kurdish awakening process. In the 2003 PISA test, Turkey’s share of young test takers of another native tongue was around 1.8 percent. This increased to 2.4 percent in 2006. Then, with the first Kurdish opening process, this rose to 3.9 percent in 2009. The 2012 test finally saw the number reach 6.3 percent. Few have focused on this rising trend of children having another native tongue. Ali Sökmen, from the Economic Policy Research Foundation of Turkey (TEPAV), noted the trend at a meeting on native language education this week. 

The rising number must be a direct result of the political peace process. It is, however, rather slow.
Could this be a bounded awakening process? TEPAV surveys show that the percentage of Turkish citizens declaring their first language to be Kurdish is around 15 percent, while 81 percent declare Turkish to be their first language. That is why 6.3 percent still looks too small to me, and why the awakening seems bounded.

So, why it is still so low? It could be a sampling error of the Chinese sort, as Turkey’s Kurds may be underrepresented in the PISA sample. We would need to check the sample for an accurate answer to that. Another possibility is that Turkey’s Kurds simply cannot yet speak Kurdish and do not consider is to be their first language. According to TEPAV surveys, only 21 percent of Turkish citizens who declared Kurdish as their first language are literate in it. That is around 3 percent of the total population. Forty-six percent said they had no Kurdish literacy at all. Thirty-four percent said they could only barely speak Kurdish. 

I do not think that there is a sampling error in the case of Turkey, just a bounded political awakening process. It is bounded because Turkey’s Kurds simply do not speak Kurdish.