Context

Low learning levels are a major challenge in rural India, where schooling does not often translate into learning. Although enrolment rates for children in the 6-to-14-year-old age group were 97 per cent in 2018, just around half of the children enrolled in grade 5 could read a grade 2 text, and only 28 per cent were able to solve simple division problems. Compounding the problem is a lack of tools to help teachers identify student-learning levels and customise their teaching, together with high pupil-to-teacher ratios.

Pratham, a large education NGO in India, developed the Teaching at the Right Level (TARL) methodology to improve learning outcomes. The model involves grouping children (typically in grades 3 to 5) based on their learning levels. Teachers use appropriate activities and materials to help children acquire foundational skills in reading and arithmetic. Instead of relying on an end-of-year exam, teachers assess students’ progress more regularly using easily administered assessment tools.

For more than a decade, Pratham and the Abdul Latif Jameel Poverty Action Lab (J-PAL), a leading research organisation specialising in randomised impact evaluations, have conducted evaluations of this methodology, thereby establishing it as an evidence-informed approach for improving children’s learning outcomes.

Although TaRL had been found to be effective when implemented by volunteers and government schoolteachers outside school hours, the partners wanted to test its effectiveness when delivered through the government system. The 3ie-supported evaluation had four treatment arms:

1. the teacher-led Learning Enhancement Programme (LEP) for Hindi that used Pratham’s TaRL model and the routine official teacher monitoring system;
2. the government’s Continuous and Comprehensive Evaluation (CCE) system, which involves teachers evaluating student performance regularly through various techniques, instead of one final exam;
3. a combination of LEP and CCE; and
4. control schools, which received no intervention.

The evaluation was conducted in 400 primary schools in two districts of Haryana, India.
 

Evidence

The impact evaluation showed CCE did not have any impact on learning outcomes when compared to the status quo. However, LEP had a large, positive and statistically significant effect on students’ basic Hindi reading and writing abilities. The effect was larger for girls than boys. Combining LEP and CCE had no significant effect on test scores relative to LEP alone. Also, LEP had no impact on mathematics scores.

Process monitoring showed that CCE did not lead to any change in teaching practices in terms of implementing any CCE-recommended techniques. Conversely, LEP had high levels of compliance and was well implemented.

Evidence impacts

Suggested citation

International Initiative for Impact Evaluation (3ie), 2020. Using evidence to improve children’s foundational skills in India [online summary], Evidence Impact Summaries. New Delhi: 3ie.

Related

From proof of concept to scalable policies: challenges and solutions, with an application, NBER, Dec 2016
This working paper describes how the 3ie-supported Haryana evaluation evidence contributed to take-up of the TaRL model by strengthening the theory with practical insights.

Case study: Teaching at the Right Level to improve learning
Abdul Latif Jameel Poverty Action Lab’s case study of how research has informed decision-making in the context of foundational learning interventions.