AN ‘anti-maths mindset’ is holding the economy back, declared Prime Minister Rishi Sunak in April as he announced the formation of a new maths advisory group. I don’t know if that’s the main problem, but I do know that a good grasp of basic maths is important for everyday life, and I hope the advisory group will look at the crucial influence of the family on exam results. I especially hope they will read the analysis of ‘Marriage and GCSE results’ I produced with Professor Steve McKay at the University of Lincoln last month for Marriage Foundation.
To me it is self-evident that family plays a big role in how children do at school. One of my daughters managed to improve from a lowly D in her first year of Maths A-level to finish with an A*. She so loved her school that she wrote to the then PM David Cameron to say how wonderful the state system was. I reminded her that part of the reason she did so well in the second year was that her parents told her she needed to knuckle down. Her school taught her well. But we challenged and encouraged her at home. She’s a chartered accountant now. Family matters.
Yet I am not holding my breath. A 264-page paper entitled the ‘Influences on students’ GCSE attainment and progress at age 16’ written for the Department for Education in 2014 talked a lot about family but only in the context of income, education and ethnicity. What it did not mention was anything to do with family structure (whether two parents or one, married or otherwise), parent-parent relationships, or parent-child relationships. This seems an extraordinary omission.
By an act of serendipity, we happen to have a huge up-to-date national dataset which allows us to look at the role of family. The Millennium Cohort Study has followed thousands of children born in the years 2000-2002 to today. I am deeply familiar with the Millennium Cohort because I’m using it for my PhD research at Bristol into family structure and stability.
The most recently released data was from when these children were aged 17 and had just done their GCSEs. In our education study, we looked at how many of our sample of 2,400 teenage boys and 2,600 teenage girls did not achieve GCSE passes in both Maths and English at the new grade 4 or old grade C. We took into account a range of factors which represented family background and home life. This included the mother’s ethnicity, religion and education, factors from when the child was young, such as the mother’s early employment status, family income, type of housing and how happy she was with her relationship with the father, other family characteristics such as how long they’d been together as a couple and whether there were older siblings when the child was born, and finally whether the parents ever married, whether they stayed together and how close the child was to mother and father at age 14 just before exams.
Using the latest techniques we were able to identify which of these factors made a difference and which did not. The results were fascinating.
For boys, only two factors made a unique contribution to getting these two GCSEs: the mother’s level of education and the child’s closeness to mother at age 14. For girls, just three factors mattered: not living in council housing as a baby, having parents who married at some stage whether before or after their birth, and having parents who still lived together.
So in both cases, socio-economics definitely mattered. For boys, it was the mother’s education. For girls, it was the kind of housing she lived in when young. Perhaps surprisingly, whether the mother worked and how much the family earned when the child was young had no unique impact on later exam results for either boys or girls.
But family environment also mattered. For boys, it was about a good relationship with mother as a teenager. For girls, it was about family structure, whether her parents ever married and whether they stayed together.
So let’s get this straight. This is a serious analysis of up-to-date UK data using state-of-the-art techniques. It was as interesting finding out what didn’t affect exam results as what did. Things that didn’t matter included the mother’s ethnicity, religion, employment and income, her age and relationship happiness when the child was born. Tell me you’re not surprised by this. We certainly were.
We were equally surprised that the key influences boiled down to so few ingredients which were mostly to do with the child’s relationship with the opposite-sex parent. Boys who did well tended to get on well with their mothers. Girls who did well had a father around in the house.
So what’s the message to the PM’s advisory group? I am in full agreement that more children need to do better at school. In our study we found that 27 per cent of boys and 22 per cent of girls did not have basic passes in both Maths and English at GCSE level. Add in those who started as lone parents – which we didn’t in our main analysis – and the proportions were 30 per cent of boys and 24 per cent of girls. That’s pretty shocking.
I suspect that different or better teaching at school will be a difficult policy goal, although they’re bound to recommend something along those lines. Having more mothers with degrees and fewer families dependent on state housing would also help. These too are difficult and expensive policy goals.
But what happens at home often costs the state little to nothing and could be where the biggest gains can be made. Getting married, getting on with each other as parents and getting on with your children are major pointers to how well children do at school. Our evidence is robust on this. As ever I won’t expect much. We’ve had decades of policy that either ignores family altogether or is outright antagonistic to stable family life. But wouldn’t it be wonderful if one day the government recognised the importance of marriage and family relationships and actually did something about it?