By Paula Korsnakova, Senior Research and Liaison Advisor, IEA
Reflecting on the results of providing instruction in a language other than the one spoken at home
Did you know that apparently 66% of children in the world are raised to speak more than one language? Countries where more than one language is commonly spoken have demands for both linguistic and cultural diversity in their curricula.
Reading comprehension is perhaps the most critical foundation for improved attainment in most school subjects, including mathematics and science, supporting an improved and enhanced overall school experience.
A recent policy brief by Sarah Howie and Megan Chamberlain investigated the effect of instruction in a second language on reading performance in nine countries using the results of the 2011 round of IEA’s Progress in International Reading Literacy Study (PIRLS). This taken place in regular five-year cycles since 2001 at grade 4 level. The authors used information reported by students on whether they spoke the language of instruction (which is also the language of assessment) frequently at home (always or almost always) or not (sometimes or never). Such information is available for all countries that took part in PIRLS.
Using the same PIRLS data, the World Inequality Database on Education has shown the differences between students based on whether they were instructed and assessed in a language other than the one they speak at home.
The 2016 Global Education Monitoring Report showed that, on one estimate, 40% of the global population is not educated in the language they speak at home. Of the nine countries examined by Howie and Chamberlain, three had an even higher percentage of students not learning in their home language: Botswana (90%), Singapore (68%), and South Africa (57%). In all three, those learning in their second or third language performed significantly below the native speakers.
Singapore uses English as the medium of instruction for all school subjects except civics and moral education. While there are four official languages (Malay, Chinese, Tamil, and English), of which Malay is the official national language, English is the language of administration and commonly spoken by Singaporeans. Thus, students are constantly exposed to English in their daily environment. Nevertheless, those students who spoke English at home did significantly better than their peers.
Botswana and South Africa use an additive bilingualism model, where reading capacity is supposed to start developing in the mother tongue and the additional language is introduced in parallel or at a later stage of education. In fact, South Africa has 11 official languages of instruction. Botswana assessed students in English (at grade 6). South Africa administered the test in English and Afrikaans. Those students who were assessed in their home language did better than their peers even after accounting for socioeconomic status, which is proxied by the number of books at home.
This finding provides an example of the need to apply indicator 4.5.2 of the SDG 4 monitoring framework, the percentage of students in primary education whose first or home language is the language of instruction, when we interpret the results of reading assessments, such as PIRLS.
Teaching in the home language needs to be properly resourced to benefit students
It is not enough to simply teach in the home language. The policy needs to be fully funded and implemented. Another study by Surette Van Staden used prePIRLS (PIRLS Literacy) 2011 data for South Africa. PrePIRLS was a slightly easier version of the PIRLS assessment, designed to test basic reading skills, for example whether students could recognize words and phrases, read sentences and simple paragraphs, retrieve explicitly stated information, and make straightforward inferences. The study assessed students in all 11 official South African languages at Grade 4.
Being assessed in an African language also spoken at home, predicted significantly lower results than being assessed in English. This finding underlines the complexity and difficulty in developing quality education within disadvantaged communities, where those who are learning in a second language that is better established may be achieving higher scores than those who learn in a language of instruction that is inadequately supported by teaching resources of a comparable quality. This lack hinders student learning just as being taught in a language not spoken at home.
I invite you, the curious reader, to access the PIRLS data to test your own hypotheses. We are always interested in the resulting findings and interpretations!
The PIRLS 2016 results will be released on 5 December 2017 in partnership with UNESCO in Paris. The database will be available in February 2018.