Behaviour and attainment: Understanding the impacts of poverty on children

9 March 2017

By Tom McBride, Director of Evidence, EIF

New analysis published by EIF this week shows that children from less well-off families are more likely to experience emotional and behavioural problems at age 5, and that these problems predict reduced academic attainment later in life. The link between poverty and poor educational outcomes is well established and, on the face of it, presents a significant challenge to the goal of improving intergenerational social mobility. What is crucial, then, is to understand the mechanisms via which the experience of poverty increases the likelihood of achieving poorly at school – not least because these suggest some ways forward for policymakers who are committed to improving social mobility.

Our paper, based on analysis of the 1970 British Cohort Study conducted by Dr Abigail Knight at LSE, indeed finds that social and emotional problems – or poor behaviour – are more prevalent among children from more disadvantaged families. However, crucially, we have also found that this difference can be entirely explained by two factors: mothers’ psychological wellbeing and parents’ education. In other words, this variation in the incidence of behavioural issues seems to be explained by factors which are associated with poverty, rather than poverty itself.

Of course, this does not mean that poverty itself doesn’t matter, but it does suggest that work to address the effects of poverty on children should be informed by an awareness of both the direct impacts of economic hardship on parents and the knock-on impacts within a family. This is at the heart of our ongoing programme of work with the Joseph Rowntree Foundation.

Mothers’ psychological wellbeing is an interesting case: it is only associated with poor behaviour in low or medium-income families, and the effect is strongest for children living in poverty. This suggests that higher family incomes (or related factors) help to protect against the effects of poor maternal wellbeing, and/or that lower incomes exacerbate these effects.

Our work also looked at the relationship between behaviour and attainment. It is perhaps not surprising that children with higher levels of social and emotional problems at age 5 do less well in academic assessments at age 10 or 16. What our analysis highlights, however, is that this relationship remains even when we account for family characteristics such as maternal wellbeing and parental education – these two factors together account for about half of the variation in academic attainment.

In short, poor behaviour at age 5 really is associated with lower test scores later on. Or, to put it another way, this analysis provides specific grounds to believe that addressing children’s social and emotional problems at an early age could have educational benefits down the line.

This analysis provides specific grounds to believe that addressing children’s social and emotional problems at an early age could have educational benefits down the line.

This conclusion is in line with much current thinking: the importance of children’s social and emotional skills for life outcomes is well established. Our analysis adds to the growing body of literature which shows that low levels of parental education are associated with risky behaviour in adults and poor life outcomes for their children. For example, children whose parents have low levels of education on average display slower speech and language development and poorer attainment at school; while in adults, low levels of education have been shown to be a risk factor in prenatal tobacco, drug and alcohol use and teen pregnancy.

But what should policymakers do to redress this imbalance? It is easy to advocate improving educational outcomes and reducing inequalities, but this is clearly a difficult and long-term challenge. Nonetheless, the evidence that poor educational outcomes impact not only on individuals’ life chances but also on those of their children all too clearly increases the imperative to act.

Only time will tell if prime minister Theresa May is able to deliver on her commitment to create a government which works for everyone, but the Opportunity Areas that education secretary Justine Greening has announced could be part of the solution, if they are successful in reducing the socioeconomic gradient in educational outcomes, entry to higher education and household income. However, for this initiative to deliver meaningful change in terms of intergenerational social mobility, it is critical that the Department for Education uses these areas to produce robust evidence on what works to reduce the link between family income and educational outcomes.

In the shorter term, our analysis indicates that programmes which target maternal mental health and early childhood behaviour could support better outcomes. Although there is much to do to improve the evidence base around how we intervene with children and parents who show signals of risk, we are not starting with a blank canvas. For example, EIF’s Foundations for Life work identifies a number of intervention programmes which have been shown to be effective in improving early childhood behaviour, and we are just now concluding a similar assessment project focusing on programmes to bolster children’s social and emotional skills. Information on both sets of programmes will be featured on our Guidebook when we relaunch it in the spring.

  • davidwlkr0

    I know thinktanks, especially those directly or indirectly reliant on incumbent ministers, must mind their ps and qs but really …

    1. ‘Only time will tell if prime minister Theresa May is able to deliver on her commitment to create a government which works for everyone’
    …we’ve just had the Budget; we have OBR and IFS projections to hand. There is not going to be any redistributive moves. Austerity continues so we don’t need ‘time’ to make a judgement about the government’s fiscal commitments and that’s before we even look at social policy in any detail

    2. ‘but the Opportunity Areas … could be part of the solution, if they are successful in reducing the socioeconomic gradient ….household income’. Perhaps I missed the bit that skewed benefits or income in the areas such that the socioeconomic gradient would be touched let alone have its slope lessened. Even if OAs ‘worked’ (and I have missed the evaluation criteria), where is the programme that would scale them up across England/RUK?

    3. …’meaningful change in terms of intergenerational social mobility’ …sounds like it’s slipping into the conventional misunderstanding that sees all mobility as upwards and neglects the obvious point that, without going the full Goldthorpian distance, there must surely be some increase in the odds of falling. When was the last time Justine Greening or any other minister spoke about downwards mobility?