It looks as though males are doing better at English than females this year, but as Melanie Butler explains, it’s not a clear result
The EF index, the annual ranking of countries by English language levels, has come up with a remarkable finding: for the first time, this year, men across the globe beat women on their language test.
The EF ranking rarely features in the EL Gazette because the actual evidence provided by an online multiple-choice test of reading and listening, taken by a random group of people who self-select to do so really doesn’t tell us that much – not even how good the test takers are at actually speaking English, because speaking isn’t tested.
Year on year very little changes: the Dutch come in at number one, followed by almost all the other Germanic-speaking countries plus Singapore and South Africa. The Slavs beat the Latin speakers, with the exception of the Portuguese and Romanians. The Philippines and the Malays are up with the Europeans, but most non-Europeans with non-Indo- European languages struggle. Apart from the Francophone Lebanese, Arabs fare particularly badly, which should be no surprise, as the CIA ranks Arabic as the hardest language for English speakers to learn. Besides, the one language skill which the Arabs are unusually strong in – speaking – isn’t tested by EF.
Countries go up a bit one year and down a bit the next, depending on the language level of the people from that country who took the test that year. Sample size is also key. If a relatively small number of people from one country take the test, results can go up and down like a yo-yo.
Additionally, if there’s a sudden jump in the sample size it may lead to a drop, as less- educated people go online and take the test.
Language distance remains the biggest predictor for how well someone from a particular country does on any language test.
This year, though, EF found one stunning change. For the first time since its test was launched in 2014, male candidates outscored their female peers across the world. This phenomenon was noted in Europe last year, but it has now spread globally, with Africa the only region where women were still in the lead.
As any linguistic researcher will tell you, girls routinely outperform boys at language learning, not only in every national and international exam, but in almost every piece of language learning research ever published. Males beating females on a language test is like England winning the World Cup at football – it almost never happens.
However, a few recent studies have also noticed the boys- do-best phenomenon. The studies were all from northern Europe and involved boys who enjoyed computer games. One from Finland noted that boys who are computer gamers were beginning to outperform girls, while a Norwegian paper found that boy gamers from disadvantaged backgrounds could read in English better than in Norwegian. Perhaps long hours of practice playing games in English has given boys the edge on the EF test.
Or maybe they just watch more subtitled television in English. Several studies have shown that children from countries which subtitle English TV shows score better at English than those that dub them, even when both countries speak the same language. This may explain why the subtitling Austrians came second on this year’s EF global ranking, with the dubbing Germans coming in 11th place and why Portugal, in at number 7, beat Brazil, which ranks at 60.
However, not dubbing films and TV cannot explain how Portugal, the highest riser this year, has improved its score by 116 points since last year.
Since countries cannot significantly improve the average English language level of their entire population in 12 months, it probably just shows that a higher proportion of wealthy, well-educated Portuguese – the demographic most likely to speak English well – took the test.
And how do we explain why boys are still trailing girls in all the other exams, including IELTS and TOEFL, which publish their annual results by gender, nationality and first language?
One factor in the design of the EF test may skew the results in favour of males: it relies solely on the results of multiple choice questions and MCQs, as they are known, have been shown in many studies to favour boys.
A recent study from Silvia Griselda, at Melbourne University, examined the results of 500,000 of the OECD’s PISA tests in maths and found that the higher percentage of MCQs in the test, the lower the scores of girls compared to boys.
So, are boys really beating girls at English? We won’t know until 2026, when the results of the first PISA language test comes out. The international exam run by the OECD will test the English skills of listening, reading and speaking in a representative sample of 600,000 15-year-olds in 79 countries. Until then we’re really just guessing.