Blog March 2021

Poverty-related attainment gap: Time to think differently

Former Professor of Education at Strathclyde University, Sue Ellis, responds to our recently-commissioned evidence review of the poverty-related attainment gap.

"To address the poverty-related attainment gap, we need to understand learning, teaching and attainment as deeply social and woven into the ecologies of institutions and communities."

The Poverty-related Attainment Gap: A review of the evidence report (The Robertson Trust/Poverty Alliance) will surely be compulsory reading for every Scottish educator and policy maker. It summarises the evidence-base for interventions at pre-school, school, post compulsory education and into-work. It acknowledges that delivering equity is complex, that education is not solely responsible for the link between poverty and attainment, but also insists that education has value and the capacity to impact on life-outcomes.

The authors, Laura Robertson and Fiona McHardy, shine a welcome light on the complexities facing those supporting the drive for equity. They highlight some of the evidence that policy and teaching communities need to think about and point to the value of:

  • Parental engagement programmes
  • High-quality early years provision
  • High-quality teaching and whole-school approaches
  • Personalised and persistent careers advice, work and mentoring arrangements
  • The importance of wider community engagement

An extremely useful summary table details the types of intervention study and the evidence gaps, particularly the evidence gaps in relation to implementation in Scotland. The affordances and constraints in Scotland's education system are very different from those of the USA or England, and Scottish practitioners and local authorities need evidence about how and why interventions have an impact in the unique 'context of implementation' that is Scotland. The focus the report puts on the need for this is important for the pragmatic, practical knowledge such evidence studies can generate.

So, what this evidence review doesn’t provide is interesting: it doesn’t say whether the attainment-gap in Scotland has narrowed; it doesn’t say how interventions worked in Scotland, for whom, in which circumstances and why. Scotland offers a unique framework to identify and address the multi-faceted ways whereby poverty impacts attainment, but operationalising this requires sociological, ethnographic and quantitative evidence as well as evidence about how to effectively scope and design achievable, multi-layered, holistic, approaches.

This is no fault of the authors; they can’t review evidence where none has been offered. 

This is the situation: Scotland’s policy framework emphasises prevention and integrated public service provision (see the Christie Commission 2011). To address the poverty-related attainment gap we need to understand learning, teaching and attainment as deeply social and woven into the ecologies of institutions and communities. Yet international research continues to accord highest value to experimental evidence from randomised controlled trials, which deliberately strip-out contextual factors and focus on the impact of stand-alone interventions. Anyone wishing to weaken the link between attainment outcomes and poverty must gamble on programmes with only limited input-output, de-contextualised, information about their impact. What they really need is nuanced evidence addressing questions about the context and detail of implementation, and what kind of mix works in different circumstances.

The Attainment Scotland Fund has spent £750 million to address a complicated situation. Sensible spending decisions require an evidence-framework that more effectively echoes the public policy model but, so far, no such framework has emerged.

This lack of useful evidence reflects how difficult it is to swim against international rip-tides. It also reflects the challenges and political realities of Scotland’s policy model: the need to recognise and track the influence of different players; the need for analysis versus the pressure for quick results; the human desire to only report success stories; and the time, knowledge and rigor required to understand, document and learn from past efforts. 

Einstein said, “If I had an hour to solve a problem I'd spend 55 minutes thinking about the problem and 5 minutes thinking about solutions.”  An effective evidence-base to raise attainment and weaken the relationship with poverty would trial tools to scope the affordances and constraints in the learning landscape. It would support analyses of the existing institutional assumptions and routines that amplify disadvantage.  It would explore the evidence on prevention, on how to use methodologies like Theory of Change to ground change-narratives and enable institutional learning, and it would detail the conditions under which service-users can become key levers for change.

Instead, the international focus on ‘proven’ programmes has encouraged multiple interventions and additional staffing to be layered over existing institutional practices. it is both an exhausting and ineffective route to meaningful change. Larry Cuban sums it up with the following metaphor:

“The hurricane whips up twenty-foot high waves agitating the surface of the ocean yet fathoms below the surface, fish and plant life go undisturbed by the uproar above.”


This report provides a good basis for the new kind of journey Scotland now needs. It presents a wide view of attainment – from early years to work – and details a range of inherently social interventions. But current attainment initiatives remain trapped by the narrow range of evidence on offer. To do things differently, we need to think differently. And to do that, we need to re-focus and re-boot with a different evidence-base.



Sue Ellis retired as Professor of Education at Strathclyde University in 2020.  Her most recent publication is: Ellis, S., & Rowe, A. (2020). Literacy, social justice and inclusion: a large‐scale design experiment to narrow the attainment gap linked to poverty. Support for Learning, 35 (4), 418-43.