CUET needs tweaks to ensure quality testing
While a welcome step towards a high-quality entrance exam, cracks in its implementation exist. These must be sealed to ensure that college aspirants are not pushed towards rote-learning over critical thinking
The University Grants Commission (UGC) has announced that starting from the 2022-23 academic year, all admissions to undergraduate courses in central universities will be based on a Common University Entrance Test (CUET) in July. Colleges may use the Class 12 Board exam marks as an eligibility criterion – requiring a minimum percentage, for example – but not in the actual selection process.
This announcement is based on the National Education Policy (NEP)’s recommendation to create a “high-quality common aptitude test, as well as specialised common subject exams”. It is commendable that UGC moved quickly to implement this welcome recommendation. However, some deviations may notably dilute NEP’s intent and impact. For example, NEP recommends that the choice to use the test should stay with colleges and that the test should be conducted twice a year. UGC has made the test mandatory, while not making any mention about it being available twice a year.
The most important deviation, however, is that while NEP emphasises quality, conceptual understanding and testing of application skills, UGC has stressed that the test “will be strictly as per the NCERT [National Council of Educational Research and Training] syllabus” and its difficulty level will be “moderated so that students feel comfortable by just studying” class 12 content. Narrow adherence to the textbook can make a test rote-based, as we have seen with Board exams. Higher order thinking skills are best tested in a context unfamiliar to the student and not taken straight from the textbook.
Another trend, witnessed in Board exams too, is that of giving a lot of choice in the questions. CUET allows students to attempt only 80% of the questions in any section. While intended to make the paper easier and discourage coaching, choice reduces the psychometric validity of a test and its ability to discriminate effectively between students as two students doing the “same” test may attempt only 60% questions in common. A good test should have questions at varying levels of difficulty including some that only the best-performing students can answer – with all questions being compulsory. Otherwise, there will simply be a clustering of students at the cut-offs, making the process less fair and reducing the quality of student selection. Eventually, this may encourage top students to try and join more discerning education systems.
In India, the use of multiple-choice questions (MCQ) is favoured as they are seen as objective and easy to score. However, with critical thinking becoming important and specialisation increasing, there is a need to measure deeper skills. Obviously, an only-MCQ test cannot test students’ ability to write or present a convincing answer. The “extended essay” is one of the most important parts of many international assessments these days. There are challenges to scoring subjective questions, no doubt, but those can be addressed, as many large international examination boards demonstrate every year.
CUET is also signalling that Board exams are less important. This seems to be a devaluation not only of school examinations but also of schooling itself. The reason quoted for this step is that Board exams conducted by different states are not comparable. But there is a simple, immediate solution to that, and also a deeper, longer-term one.
The simple solution is to have every Board provide percentile scores to students. Even with papers of different difficulty, students scoring, say, 92.4 percentile in Assam and Rajasthan would be comparable. The deeper, longer-term solution is standardisation of Board exams across the country. This is a qualitative and quantitative process that will ensure that all Boards test and grade students uniformly. Most advanced countries with multiple Boards attempt this; we should as well.
A criticism of CUET has been that it will lead to increased coaching and a number of coaching courses have already sprung up. In our experience, demand for coaching depends less on the toughness of an exam and more on the stakes involved and the competition. There is a lot of coaching for the Central Teacher Eligibility Test (CTET), which is not tough but seen as a step to a “permanent government job”. Admissions to central universities neither have a very high stake nor witness intense competition – the UGC chairman said the selection rate would be one out of five or six. Hence, the difficulty level of CUET need not be diluted with an intent to reduce coaching.
Finally, between now and June – in less than 90 days – high-quality papers have to be set in 60 different subjects. Papers in 26 domain subjects have to be translated into 12 languages each – a total of almost 350 papers. Exam creators will be under tremendous pressure to deliver both quality and quantity for crucial examinations in this short span. This, compared to international Boards which take up to two years to prepare highquality papers. Sometimes, the way to get quality is simply to give time.
CUET is potentially a welcome step in the right direction. If it can become a high-quality test that measures conceptual understanding in a psychometrically valid way, I believe it will be accepted voluntarily by a majority of colleges and play a key role in transforming education in India.
Sridhar Rajagopalan is co-founder and chief learning officer of Educational Initiatives, a leading Assessments and EdTech player in India. He was a member of the committee that drafted guidelines for PARAKH, the National Assessment Council recommended by the NEP 2020 The views expressed are personal