The new National Education Policy (NEP) has correctly laid a lot of emphasis on assessment reforms. The NEP calls for setting up of a National Assessment Centre (PARAKH) to monitor learning outcomes regularly. These are all good, much-needed initiatives.

NEP aims to change the assessment system “from one that is summative and primarily tests rote memorization skills to one that is more regular and tests higher-order skills.” The policy emphasises Board Exam reforms so that the questions in these exams will test “core capacities/competencies”.

‘Assessment Reforms’ are sometimes misinterpreted to mean that there will be more assessments. This raises fears not only of increased pressure on students, but also that the time available for actual teaching will be reduced. Good assessment reforms are not about more assessments but about measuring progress against clear targets – the only way of knowing if and how much improvement is happening.

Further, such assessments are ‘low-stakes’ for the student and teachers. ‘Low-stakes’ means that an individual student’s performance in the test does not count towards her progress or promotion or teacher’s career advancement, but is basically meant to throw light on the overall levels of learning.

The NEP also talks about a ‘light-but-tight’ regulatory framework. This refers to focussing on and strictly regulating a few important matters while providing flexibility on most matters. This is also desirable. If for example, school leaders can be held responsible for the academic and non-academic learning outcomes of their students (which will be an important goal and monitored), they can be given a lot of flexibility in, say, the training of teachers, the organisation of curriculum, school activities, etc. This article proposes such a mechanism that can do this while helping improve learning levels for students country-wide.

The proposed mechanism is this: PARAKH (or the CBSE, till PARAKH is set up), will empanel assessment agencies through a transparent process based on their research, assessment track record and experience of working with schools. These agencies will be contracted to offer quality assessments to schools for classes 3-12. Agencies may offer their services nationwide or in specific states based on their choice and reach but will be empanelled such that the services of at least 3 agencies are available to schools in every state. Agencies offering services nationally will be preferred as they will be able to benchmark learning data more effectively. Agencies will be empanelled for 3 years initially which may subsequently be extended for 5 years at a time if the quality of performance is good. The initial empanelment process may be completed in about a year.

Following the empanelment of the agencies, every non-government-run school in the country will be required to contract with at least one of these empanelled agencies and participate in an independent assessment conducted by that agency once every year. In other words, schools have flexibility about which agency they choose to work with, but not about having to undergo such an assessment once a year. Students of the classes fixed by PARAKH (say 3-12) must participate in the assessment.

Each student will be assessed in Mathematics, Environmental Sciences (only class 3), Science and Social Studies and two languages. The languages will be the medium of instruction and the second language (for example, English and Hindi, or Tamil and English or Gujarati and Hindi) with one being at a higher and the other at a lower level. Agencies may charge for these assessments and schools may either bear the amount themselves or charge parents. An upper limit of what can be charged will be set by PARAKH. This may be about Rs 50 per subject per student initially and is an upper limit (agencies may charge lower fees). Agencies must offer the tests both in paper-pencil and computer formats, though it is expected that the assessments will move to a fully digital format in a few years.

Assessment agencies may offer 1 or 2 fixed time windows during which the assessments will be conducted. Each school must take the assessment in one of those windows every year. The reason for offering 2 time windows is only to accommodate different school calendars. Further, schools are expected to tie with agencies for at least 3 years at a time since measures of improvement will be easier across years if the assessment is conducted by the same agency.

All agencies must share detailed results with schools by a fixed date (say November 30th) every year, and schools must share individual results with parents and summary results with PARAKH by December 31st each year.

Based on their performance relative to all other schools, each school will receive one of 5 ratings (AA+, AA, A, AB and AB-). The ratings will be based largely on the absolute performance of the school but there will be a certain weightage for the improvement compared to the previous year.

For the first 3 years, the ratings will simply serve as a kind of reputation score. Schools will, however, be required to make their rating visible in their prospectus, website and school entrance. In the future, approvals for expansion and fee increases may be linked to the rating a school receives. In cases of consistently poor performance over many years, a school’s accreditation may be negatively affected.

The agencies are expected to provide good services and support to the school (like pre-test video-based sessions and post-test analysis) but they are responsible to PARAKH for the quality and integrity of their assessments. Agencies are expected to use advanced statistical methods to ensure that no unfair means are used in the exams and share these with PARAKH. They must have a forum where schools can ask and receive responses related to the assessments. They must outline clear procedures for handling exceptions like errors found in questions, etc. Each agency must submit an annual report which will include certain sections defined by PARAKH. And finally, each agency must participate in a bi-annual joint conference with the other agencies organised by PARAKH to share learnings and improvement ideas.

Agencies also need to ensure that schools – even though affiliated to different Boards – find value in the process and do it not only out of compulsion, but also see it helping them improve. Agencies thus need to take the schools along and be responsive to their questions, explaining how their assessments measure core learning, which is the same across different Boards. At same time, the tests should be well-designed that they cannot simply be ‘coached’ for. Even if teachers try to ‘teach to the test’, that should lead to improved student learning, not simply inflated scores.

Though customised to the unique realities of India, the concept of non-government schools requiring to assess their learning outcomes regularly through an independent agency empanelled by the government is a practice adopted by many US states as well as many countries in the Middle East. In Dubai, for example, the Knowledge and Human Development Authority (KHDA) similarly mandates annual testing by empanelled agencies and permissions to fix fees are based on their performance in these assessments. Dubai has used this mechanism to ensure that its ranking in international benchmarking assessments like PISA has improved significantly over the years.

In conclusion, the proposed mechanism if implemented, can be a win-win-win for students, schools and the country as whole. Students benefit because the system focuses on useful learning (not just high marks in exams) which aligns with their career and earning goals. Schools are held accountable to constantly improve learning levels, with the best of them getting recognised while others receive specific feedback on how they can improve. The country benefits from an improved educational system at no additional public expenditure, improved assessment capability and alignment of stakeholders towards improving student learning levels and skills. Reliable, anonymised data on student learning – invaluable for policy making and research – will be available. India’s aspirations to do well in PISA and improve our research and innovation capacities will be well-served by building in accountability to student learning through such a mechanism.

Sridhar Rajagopalan

Sridhar Rajagopalan

Co-Founder & Chief Learning Officer at Educational Initiatives
Sridhar Rajagopalan is a co-founder and Chief Learning Officer of Educational Initiatives.
Sridhar Rajagopalan