Assessing education

Thinking of assessment (of any kind) as a primary systemic lever for improvement in education is ineffective

From 2002 to 2008, we ran a large-scale programme across five states in India, in partnership with the respective state governments. It was an “assessment-led reform” programme. The idea was to assess schools on basic conditions such as attendance, and rigorously on the learning levels of the students, and then use this as a lever for improvement.

Schools signed up voluntarily for the programme. The schools that would meet some basic criterion and where student learning was above a certain high level, would be classified as “Learning Guarantee Schools”. These would get publicly felicitated and rewarded, creating incentives for improvement. Over these years, more than 14,000 schools participated in the programme.

For sure, there were some positive outcomes, both intended and unintended. The most visible one was the excitement and engagement that built up in the schools and their villages. This energy was equally shared by people in the education department. In some states, the programme led to improvement of the student assessment methods. In some districts, the programme formed the platform for a more comprehensive effort for improvements in schools, and in a particular state, an institution was set up to keep assessing quality of schools.

But on the basic matter of educational improvement there was no movement, even in the areas where the programme was run very well. Not only in hindsight, but even as the programme progressed, the limitations had become clear and the outcomes had seemed inevitable. So we stopped the programme after 2008.

The experience is no different from other such experiences of ours and others; the issues are quite clear. Let me list some inter-related, general points.

First, external “packaged programmes” like this do not work in education. I would think they don’t work for any social-human process, i.e. can’t address any (small or big) human and ecological challenges. Second, improvement in education needs work on all dimensions of education; this includes organizational culture, curriculum, teacher preparation systems, infrastructure, community ownership etc. Focusing on any one (or few) of these dimensions is inadequate, especially if the idea is that they will cause basic educational improvement.

Third, the fundamental issue is that of human capacity and motivation. This matter trumps all other dimensions; inadequacy in all other dimensions can be compensated by human capacity, but no other dimension can compensate for inadequate human capacity. In schools, human capacity is mostly about the capacity of teachers. And certainly in the case of India, it’s on this fundamental issue that we are most wanting. This matter requires deep and sustained work with existing teachers, and a complete overhaul of our teacher preparation and management system.

Now, specifically on “assessment-led reforms”, which find easy favour and votaries in many places. Two of the largest such programmes have been the US Federal programmes running since 2001: No Child Left Behind and the Race to the Top. It’s not surprising that they haven’t fundamentally improved US education. In reality, their reliance on creating negative incentive structures have caused fundamental problems e.g. malpractice in tests, more segregation in some areas.

By measuring the temperature of individuals in a population, and creating incentives for improvement, population wide improvement in health will not happen. For actual improvement, all the work and investment will have to be on doctors, paramedics, awareness, nutrition, sanitation, hospitals etc. Likewise, thinking of assessment (of any kind) as a primary systemic lever for improvement in education is ineffective; the work lies elsewhere, mostly in building human capacity. In fact, in the case of education, the use of assessment as a lever has definite negative effects.

Assessment in most contexts amounts to testing. Unlike an objective measurement of temperature or blood pressure to assess health, there is a huge “loss of information” in educational testing. A lot of what happens in good teaching and learning cannot be tested effectively. What can be tested is often so narrow (narrowing further with scale) that it has little value. And since testing is poorly understood or because people are bewitched by the illusion of objectivity, systemic use of testing results in distorting teaching and learning. Teachers teach to improve test scores of their students, who in turn learn by rote and study to beat the test; education suffers. In classic economic language, testing distorts incentives. External incentive structures (negative or positive) based on such testing multiply the distortion manifold.

The effective use of assessment in education is for feedback into the learning process by helping teachers and students teach and learn better. Testing is ineffective and deleterious as an administrative tool for direct decision-making or to rank schools and students.

As of now, India has not made the mistake of trying to make assessment a lever for systemic improvement. We must never make this mistake. There are enough cautionary tales from the world of education, big and small, near and far. We must continue to keep (and improve) assessment as a part of the classroom-level educational process.

Source | Mint = The Wall Street Journal | 11 June 2015