Critical Reflections on the Learning Metrics Task Force

written by: David Archer published 3 February 2014 updated 3 February 2014

Over the past two years I have been an active member of the Learning Metrics Task Force (LMTF), an international effort convened by the Unesco Institute of Statistics and the Center for Universal Education at the Brookings Institution. The overarching objective was“to create a shift in the global conversation on education from a focus on access to access plus learning”. It aimed to make learning “ a central component of the post-2015 global development agenda and to make recommendations for common goals to improve learning opportunities and outcomes”. As I re-read the various reports from LMTF (see for example the summary report “Towards Universal Learning: recommendations from the LMTF“ - www.brookings.edu/learningmetrics),  I found myself critically reflecting on what we collectively produced and feel compelled to share some thoughts.

First of all it is important to highlight some of the positive contributions of the LMTF.  As a whole I think the task force helped to challenge the complacency that was setting in within some international policy circles that global education goals had been largely met and that therefore a post-2015 education goal was not needed. By highlighting the shockingly poor learning outcomes achieved in many countries the LMTF has helped to refresh the case for a focus on education.

Another major contribution has been to move the debate on from a focus on narrow literacy and numeracy outcomes, by highlighting seven domains of learning: physical well-being, social and emotional, culture and the arts, literacy and communication, learning approaches and cognition, numeracy and maths, science and technology. There have been some influential voices arguing for a post-2015 goal framed around early grade reading or writing and the Global Partnership for Education has set a strategic objective on learning which is similarly narrowly framed. By asserting a broad, holistic framework LMTF is much more consistent with the human rights commitment that the aims of education should be about the development of the full human personality. In the context of tendencies to focus only on primary schooling it is also positive that LMTF’s framework starts with early childhood education and continues through to at least lower secondary / post-primary education. There are some concerns that can be raised about the logic of the seven domains: they do not organically link with cognitive development theory or learning psychology and some key areas of learning such as history are not very evident – but these are not proposed as a curriculum and their inclusiveness does represent an advance against those advocating a narrow framework.

The LMTF should also be welcomed for its focus on equity, recognising that the inequalities within countries are often masked by national level data. There is a recognition that countries need to map access and learning against diverse characteristics of children in order to ensure equitable learning opportunities – and this is often overlooked. Similarly the attention placed on supporting country systems is important as it suggests that there is not a one-size fits all solution and that diverse contexts will require diverse investments to improve assessment and learning.

However, there are some concerns that have become clearer to me with the passing of time. The LMTF calls for a global paradigm shift towards access plus learning. On reflection this is a problematic framing of the task. Whilst the MDGs focused only on two access-related education goals the education community has always been concerned with learning. The quality of education and learning outcomes are explicitly part of the Education For All framework agreed initially in 1990 in Jomtien and reinforced in Dakar in 2000. It is problematic to suggest that somehow the education community was previously obsessed with access – indeed there were widespread laments at how the MDGs reduced the EFA agenda and many strong critiques of the World Bank and others for promoting a reductive agenda. So, yes we need to address learning but this is not a sudden new discovery for most educators and it is not really a paradigm shift.

The LMTF report claims that“the education community has reached a consensus on the skills and competencies that are important …. and a small set of indicators that are feasible and desirable to track at the global level.”  I fear this is an over-statement. It may be that 1,700 people in 118 countries made some contribution at some moment to the process - but this was a very ad hoc process. There were no Ministers of Education, there was no mechanism to ensure people were representative of the wider education community and the vast majority who did participate were given no opportunity to endorse or vote on the conclusions reached. It may be right to say there is consensus on the importance of learning – but that was largely pre-existing. But it is wrong to claim consensus on a small set of indicators and the desirability of global tracking. Indeed this is widely contested. Indeed, there are many people across the education community who see this approach reinforcing a narrow focus on testing, which they would vigorously oppose.

On reflection, perhaps the biggest problem with LMTF was that whilst the overarching objective and aims were about learning, its focus in practice was purely on measuring and assessing learning, not actually improving learning. This of course might be obvious from the use of the word “metrics” in the title but the task force did frame itself to be more broadly about “learning” and there is no parallel or linked effort on the same scale which does actually focus on improving learning. Crucially no systematic effort has been made by LMTF to establish the connections between measuring learning and improving learning. As the saying goes, you don’t make a pig heavier by weighing it more often.  Improved statistics do not inherently contribute to improved learning. How you can maximise the links between assessment and improved learning is a rich area to explore - but this was not even touched upon..

One issue on which serious continuing discussion is needed concerns the LMTF’s work on learning indicators for global tracking. Seven areas of measurement for global tracking are proposed, each of which is a composite of different indicators, many of which are not clearly established (e.g. breadth of learning / citizen of the world / readiness to learn). If elaborated this would end up involving dozens of actual indicators that it is suggested should be globally tracked. This is in significant tension with the focus on supporting country systems that operate in diverse contexts and risks creating an overwhelming pressure for standardisation.

Of course some groups benefit from global standardisation of indicators and testing, perhaps most obviously those large scale private providers who can develop common tests and offers common services on a high volume basis, removing smaller competitors and facilitating higher profits. But many others are likely to be losers, perhaps most obviously children themselves who face assessments that are culturally inappropriate or teachers who find their performance judged without contextual factors being taken into account. Those teaching or learning in difficult circumstances, in minority languages with complex scripts, in poor areas, with large class sizes of inadequate facilities will tend to come predictably bottom of the resulting league tables.

Whilst most people involved in LMTF would not want to support a culture of standardised testing, this is perhaps one of the most likely unintended consequences. There were some enlightened discussions within the task force about the dangers of testing, but these are not articulated strongly in the final reports. Indeed, assessment is presented as a “public good”. The intention behind this statement is positively intended – that tools, documents and data should be made freely available and any cost barriers to assessment should be eliminated. But the interpretation of this risks something very different – a celebration of assessment and testing as inherently worthwhile (whether or not it results in improvements to learning).

I do not need to cover here all the dangers associated with a testing-led education system as others have articulated this powerfully enough over the years. But it is worth reminding ourselves that obsessive “teaching to the test” can actively undermine learning. It can destroy the joy of learning and mean learners might be driven to pass a test but will not develop transferable or practical skills.  The more focus given to testing within a system the more likely it is that things that cannot be easily tested will be overlooked. In its worst forms testing can create horrendous stress for children and for teachers, creating a climate that is far from conducive. I am convinced that this is not the intention of those involved in the LMTF but because this is not explicitly challenged in the LMTF reports those who read them could easily get the wrong impression and may forget that testing is not teaching.

One way in which LMTF could avoid being associated with high-stakes classroom testing is to highlight the value of household based surveys for collecting data. There are many examples of surveys which can generate some powerful information about the overall education levels of the population – helpfully capturing information about children out of school as well as those who are in school. There are challenges in how these are best designed and how the data is best used. Household level surveys can provide useful contextual information about socio-economic status – though they cannot easily provide contextual information about the school experience or learning environment of children. As such these surveys can raise issues around levels of learning but cannot be used for diagnostic purposes to work out what needs to be done to improve learning. The important thing is not to let this data be mis-used.

To be diagnostic about how to improve learning we need a massive complementary investment in formative assessment by teachers themselves. LMTF makes no serious mention of this and yet this is surely the most crucial step in linking assessment to improved learning. Teachers need to be trained to identify the progress that individual learners are making and to adjust their teaching methods accordingly. In some contexts this is a routine part of teacher development but in many parts of the world there are major threats to the teaching profession. In Africa there is a rapid spread of non-professional teachers who are not trained even in the basics let alone how to do formative assessment. Indeed, for far too long global policy dialogue and financing of education has overlooked the critical role of teacher training and professional development. This has meant that old models of pre-service residential training colleges have been left to decline rather than attention being paid to how to reinvent teacher training with a greater focus on classroom practice, in-service training and support for mentoring programmes.

Sadly, perhaps unintentionally, LMTF has added its loud voice to the resounding silence on teachers. There is no analysis of the threats to learning outcomes posed by non-professional teachers or the under-investment in renewing teacher training systems. There is no analysis of the crucial role that teachers play in assessing and improving learning. If the serious intent of the LMTF is to put learning on the global agenda it has missed the crucial ingredient. We should be talking about teaching and learning together – as the latest EFA Global Monitoring report helpfully does.

Unfortunately the silence on teachers is perhaps not entirely unintentional. There are many people who really do see teachers as the problem, who persistently highlight ghost teachers and absenteeism of teachers or who blame teachers for poor learning outcomes. There is rarely any contextual analysis to these attacks on teachers and rarely is there any solution posed. Of course ghost teachers should not be on the payroll and teachers should turn up. Stronger accountability systems are often needed and many of us in civil society are working hard to do this, linking local, district and national efforts to improve the accountability of the public education system. But rather than focus on them as the problem we need to focus on professional teachers as the key part of the solution. We need to listen to them and give them a voice in policy dialogue at all levels because they are the ones on the front-line in the classroom and they are the ones who can make actual improvements in learning happen.

One of the reasons why LMTF ignores teachers is that the framing around “learning outcomes” is sometimes presented as being contrasted with a focus on “inputs”. It is suggested that people who talk about “inputs” are dinosaurs – that this is the old way of doing things that got us into this mess, producing poor learning outcomes - and that it is time to move on. I find this rather bizarre as it suggests a lack of interest in how we might actually improve learning outcomes – what are the levers that might bring about positive changes to outcomes? It seems to me to be self-evident that we should balance talk about outcomes with continued attention to inputs and processes. We need to look at how we can develop well trained teachers with the ability to assess the progress of individual children and we need to keep a close eye on class sizes, on adequate infrastructure, on the relevance of the curriculum, on ensuring there are decent textbooks that arrive in classrooms and get used, on promoting participatory teaching-learning processes rather than rote learning, on deepening relationships and accountabilities between schools and parents and communities. When assessments suggest that some schools are performing well and others less so we need to get under the skin and look at the inputs and processes which may have led to these outcomes.  By definition outcomes arise from these inputs and processes and you cannot make adjustments directly to the outcomes in themselves. LMTF should have done more to highlight such connections rather than feed into the cult of self-contained learning outcomes.

The re-framing of the education agenda around learning outcomes is also deeply problematic because it undermines the framework of education rights. Diverse international conventions and treaties have built a complex understanding of the right to education, perhaps best captured by Katarina Tomasevski’s (late UN Rapporteur on the Right to Education) in her articulation of 4 As – that education should be available, accessible, acceptable and adaptable (see www.right-to-education.org). These rights are indivisible and inter-dependent and provide a powerful framework for promoting quality education. A focus on learning outcomes is not only problematic because it is truncated but also because there is no right as such to a fixed set of learning outcomes. A child with severe disabilities may not achieve the same learning outcomes as other children but they do have precisely the same right to a quality education.

In conclusion the LMTF has been an influential actor over the past two years and there are some positive contributions that have been made but, perhaps unintentionally, there are also some serious limitations and problems with the outcomes to date.  As we move towards setting a post-2015 development goal on education we need to have a much broader and more systematic process to build consensus on the future priorities – and it is extremely important that we do not end up with a narrowly framed goal on learning outcomes, based on the agendas of some dominant Northern voices. This would be to repeat the mistake of the past when a narrowly focused MDG on access to education undermined the more inclusive and collectively agreed EFA vision. We need a much more holistic goal which is consistent with human rights frameworks, promoting quality public education for all – and we need indicators that will balance quality inputs, quality processes and a broad range of quality outcomes.

[email protected]

The opinions expressed in this blog are those of the author and do not necessarily reflect any official policies or positions of Education International.