New study on PAL: How taking tech to classrooms is helping kids in India, what are the challenges

What is personalised adaptive learning?

It’s learning personalised to suit the needs of each student. It involves the use of a software on a tablet or computer to first determine the learning level of a student, and then tailor learning material and questions accordingly. It is designed to be dynamic – each student may get different material and questions based on what their initial level is.

The software might take the form of a story or game, where students need to accumulate points or deal with ‘monsters’, and climb up levels.

EdTech companies across the world that provide versions of this software, including AI-based ones. For instance, Squirrel AI, a Chinese EdTech company, focuses on adaptive learning.

Why is it used?

The use of the system is grounded in the understanding that each class has students at different levels of learning. For instance, in India, the Annual Status of Education Report (ASER) 2024 report showed that only 44.8% of Class 5 students in government schools could read a Class 2 text.

Kremer said PAL “meets the student where they are”.

A paper published earlier this year in the journal ‘Computers and Education: Artificial Intelligence’, by researchers at Nanyang Technological University, Singapore, likened PAL to a personal tutor. It pointed to how adaptive learning is a modern, digitised version of an instructional technique referred to by American educational psychologist Benjamin Bloom, called ‘mastery learning’. Bloom pointed out in a 1968 paper that individual students may need different types and qualities of instruction to achieve mastery, and that “if every student had a very good tutor, most of them would be able to learn a particular subject to a high degree.”

Where is PAL in use, and what does research say about its use in India?

Use in India is scattered. Supported by different agencies, and promoted by the companies that develop the software, PAL has been in use in some government schools in Rajasthan, Uttar Pradesh, and Telangana, besides Andhra Pradesh, where the state rolled it out for Math, English and Telugu in 2019.

Research on the use of PAL in India points to an improvement in learning outcomes among students, but that does not necessarily reflect in school exam performance.

A paper this year by researchers at the University of California, San Diego, and the Stockholm School of Economics evaluated 40 government schools and 6,500 students in Rajasthan to determine the impact of PAL. After 18 months, it was found to have improved learning outcomes in Math and Hindi. In school exams, however, the study found no evidence of improvement. It pointed out that the PAL instruction was targeted at the student’s actual learning level, which was often many years below the grade. “Even meaningful increases in learning from this low base are unlikely to be captured by grade-level school exams,” it noted.

Another evaluation published in 2018–by the two researchers from the Rajasthan study, and a third researcher at New York University – looked at the use of the same PAL tool that was used in Rajasthan, for 619 students in Delhi, who used it in after-school sessions rather than the Rajasthan system of using it in school. The researchers conducted tests to evaluate performance and found that students who used the tool scored higher in Math and Hindi than those who didn’t. They found that the relative impact was better for weaker students. In regular school exams, they found improvements in Hindi, but not in Math.

Last year, a panel constituted by the Education Ministry and headed by former ISRO chairman K. Radhakrishnan to suggest reforms in the conduct of national entrance exams, recommended the adaptive model for these exams. It suggested a move to ‘computer adaptive testing’ in the long run, which is that exam takers will be tested on questions given to them based on their individual abilities, instead of a standard question paper for all.

What does research say about its use and efficacy abroad?

It has been in use in several parts of the world for years, in school and higher education. Research points to differences in impact depending on factors like the time for which the tool is used. Researchers have, over the years, suggested a closer examination of the efficacy of the tool in different settings.

A few researchers have also pointed to differences in gains among students – students at lower learning levels/girls/students in lower classes may have gained more in terms of learning outcomes. Kremer said that the study in Andhra Pradesh also found that girls and students in lower classes, and those at lower learning levels gained more from PAL. About the gender difference, he said girls had used the tools more than boys.

What are the challenges?

An official in Andhra Pradesh said that procuring and replacing hardware – tablets in this case — can be a challenge. Only 63.5% of schools in the country have internet facilities, and around 65% have computers, according to the latest UDISE+ data.

The official also pointed out that it is important to get teachers on board and ensure training.

Research on an adaptive learning system in China in 2018 referred to the need to see it as a blended tool – both the teacher and technology are used for effective classroom learning. Kremer, similarly, said that the tech and data from it can be used to send messages to the teacher on where the student needs help.