Girls in year 5 performed “significantly” worse than boys in mathematics in England, amid a mixed global picture for closing gender gaps in attainment, according to an analysis of international data.
The Trends in International Mathematics and Science Study (TIMSS), released by analysts in Massachusetts today, assessed 57 countries on their performance between 2012 and 2015. It is the sixth such analysis over the last two decades.
Whereas “nothing much changed” in England’s relatively solid performance for maths and science among nine- to 10- and 13 -to 14-year-old pupils since the last study in 2011, there was an unexpected drop among younger girls, said experts.
Despite data in England repeatedly showing boys trail girls in reading, maths and writing on entry to primary school – with lower literacy rates persisting into secondary school – the TIMSS study showed girls in England performed worse than their male counterparts in maths at year five.
“In 2015, year 5 boys performed significantly higher than girls in maths, indicating the reopening of a gender gap that had almost disappeared by 2007.”
While boys obtained on average 549 in maths in year 5, girls were statistically significantly behind by six points on 543.
If something new has happened, that gender gap with follow through
Prior to that, there had been no gender differences in mathematics for either of the age groups tested by TIMSS, in either the 2011 or 2007 cycles.
But the government document said that, according to separate national analysis, when the same age group moved into year 6 “there were no differences by gender in boys’ and girls’ maths attainment”.
Based on “provisional 2016 data”, 70 per cent of both boys and girls achieved the expected national standard at the end of key stage 2 – raising questions about differences in the way TIMSS and national tests evaluate pupils.
Ben Durbin, head of international education at the National Foundation for Educational Research (NFER), said the result was a “leftfield” finding of the TIMSS data for England.
“If you look at the trends, there has actually always been a marginally better performance of boys than girls in maths, and that gap appears to have opened further in 2015 to significant difference.”
There was no significant gender gap between boys and girls in mathematics in year 9, according to the TIMSS data.
As such, younger girls coming through to secondary level would need to be carefully watched in four years, said Durbin.
“The cohort who are currently in year 5, it could be that if something new has happened, that gender gap with follow through. But we need to wait four years to look at that.”
England’s results were in the global context of gender gaps narrowing, with 20-year trends of longstanding higher performance of boys than girls in maths and science closing.
Yet boys still took the lead in maths with higher scores in 18 countries at an average difference to girls of nine points – showing the three point difference in England’s year 5 pupils is at least smaller than average.
Twenty-three countries had no difference in the average mathematics achievement between boys and girls – and in the eight countries where girls overtook boys, they were on average ahead by a much larger gap of 18 points.
There was also a considerable narrowing of the gender gap in physics, which was tested among those pupils taking advanced physics in the TIMSS Advanced study. Yet more boys than girls continue to enrol in advanced maths and physics globally.
The gender gap in national test results
National results for boys in England have previously showed that while 77 per cent of girls under five-years-old reach the expected standard in maths, social skills, communication and language, that figure is much lower for boys at 62 per cent.
Boys are also being outperformed by girls in literacy at primary and secondary level because they are more likely to skip sections of reading, another study at the University of Dundee showed.
At GCSE, however, maths remained the only subject in which boys outperformed girls this year, by 0.7 percentage points at A*to A and 0.5 percentage points at A* to C.