“Teaching at the right level” (TaRL) is a pedagogical approach developed by the Indian NGO Pratham that assesses children’s reading and mathematics skills using a simple tool and then regroups students according to learning level rather than age or grade. Trained instructors then target teaching to the learning level of each group using tailored learning activities and materials. A series of randomized evaluations by J-PAL affiliated researchers over the past fifteen years have shown that this method of reorienting teaching to the level of the student, rather than the rigid expectations of a curriculum, consistently improves learning outcomes. This iterative process of innovation and evaluation has helped Pratham to refine and adapt the TaRL approach over time, which has now reached over 50 million children in India and Africa.
The Problem: Despite record high enrollment rates, 60 percent of primary school children in developing countries still fail to achieve minimum proficiency in reading and arithmetic.
Over the past decade, many developing countries have expanded access to primary school, but improvements in school access and enrollment have not translated into actual improvements in learning outcomes for all students. This is especially true in rural India, where enrollment rates in primary schools in 2016 reached nearly 97 percent, but over half of children in grade five could not read a grade two level text and over a quarter could not solve simple division problems.1
Several challenges could account for low learning levels. Achieving annual enrollment targets is easier than improving attendance throughout the year or strengthening pedagogy to improve student learning outcomes. Many newly enrolled children are first-generation students who differ in age, school preparedness, and support at home. Half of all classrooms do not meet pupil-teacher ratio norms and there is a wide heterogeneity of learning levels in classes.
The typical age-grade organization of schools, combined with unrealistically ambitious curriculum standards, seem to be major constraints in helping children learn effectively. Teachers may not have the pedagogical tools to identify learning levels and then calibrate their teaching accordingly, and students who are far behind the scheduled curriculum may lose interest and fall behind even further.
The Research: Interventions that restructure classes by learning level, rather than grade or age, can produce large gains in learning outcomes.
Since 2001, several J-PAL affiliated researchers—Abhijit Banerjee (MIT), James Berry (University of Delaware), Shawn Cole (Harvard Business School), Esther Duflo (MIT), Leigh Linden (University of Texas Austin)—and colleagues—Rukmini Banerji, Rachel Glennerster, Harini Kannan, Stuti Khemani, Shobhini Mukherji, Marc Shotland, and Michael Walton—have partnered with Pratham to evaluate a range of programs aimed at improving student learning through pedagogical interventions.
This partnership demonstrates how findings from an evaluation can provide the implementer with important inputs into subsequent program modifications, which in turn are rigorously evaluated. Pratham has continued to test, learn, and modify its programs in partnership with J-PAL affiliated researchers for over a decade. These include:
- Balsakhi (“child’s friend”) program, which recruited young women to lead remedial education classes for two hours during the school day. The program targeted children in grades two to four who were identified as falling behind their peers, and involved pulling them out of the classroom for a portion of the day for TaRL activities.
- Reading camps, which trained community volunteers to teach basic reading skills outside of schools for two to three months over the summer.
- Read India, which tested the relative impact of four different interventions in rural, government-run primary schools. These included (1) training government teachers to provide remedial classes over the summer; (2) providing schools with teaching and learning materials to help students master basic literacy and numeracy; (3) providing teaching and learning materials, as well as training teachers on how to use them; and (4) training community volunteers to support teachers either during the school day or after hours.
- Teacher-led Learning Enhancement Program, which trained government teachers in the state of Haryana to assess students’ Hindi literacy levels at the beginning of the year and then restructured the class according to those levels, rather than grades, for one hour per day within school hours. Teachers were monitored and received mentoring to support their work to teach students at their level. This was the first evaluation to demonstrate how government teachers can successfully implement the program during the school day.
- 40-day Learning Camps, which involved Pratham instructors leading short-burst learning camps during school time in the state of Uttar Pradesh. Pratham instructors assessed and regrouped students and used level-appropriate teaching and learning materials to focus on basic math and reading. The camps effectively doubled the number of children who could read paragraphs or stories by the end of the program.
Results from these evaluations show that programs that reorient classroom instruction to teach at the level of the student are consistently effective, raising test scores by between 0.07 and 0.70 standard deviations. Combined with evidence from Kenya, which showed the efficacy of organizing classrooms by students’ initial learning levels, these findings led to the emergence of “teaching at the right level” as a general, scalable lesson within J-PAL’s portfolio of education research.
While the implementation of this model can take many forms (e.g., with government teachers or lightly trained volunteers/para-teachers, during the school day or outside of school hours), there are three key components that are consistent across the various programs studied: (1) students’ learning levels are assessed at the beginning of the school year or program, (2) students are grouped based on their learning levels, and (3) students are taught in these groups using level-appropriate materials rather than following a curriculum they are not yet prepared to learn.
From Research to Action: Teaching at the right level emerged as a policy lesson through an iterative process of research and implementation, and Pratham has continued to develop and tweak its programming based on evidence.
Starting in 2012, Pratham began to focus on two specific programs grounded in the evidence and designed for implementation at scale.
(1) Learning camp model: The “Learning Camp” model is implemented directly by Pratham staff with assistance from locally recruited and trained volunteers. These camps provide intensive bursts of instruction in math and Hindi for eight to ten days at a time for up to two months. Students in grades three to five are grouped according to learning level (i.e. whether they can read letters, words, sentences, or paragraphs) and are taught using level-appropriate materials tailored for each group. The environment of the camp is different from a standard classroom as it is interactive; children engage in math and language games in groups designed to strengthen basic skills.
(2) Government partnership model: Drawing on the Learning Enhancement Program in Haryana, Pratham is now working with state governments in India to integrate teaching at the right level into the government school system. In the government partnership model, government teachers set aside one to two hours per day to reorganize their classrooms by ability rather than age or grade. Groups of students are then taught using level-appropriate materials. Pratham trains school supervisors (cluster-level education staff) to train and provide ongoing mentoring to teachers, including in-classroom demonstrations of the approach.
Following the success of Pratham’s programs in India, the Government of Ghana piloted its own version of TaRL to see how the approach might work in the Ghanaian context. Teacher Community Assistant Initiative (TCAI) trained teachers and teacher community assistants (TCAs) provide instruction at the learning level of their pupils with a variety of different implementation schemes among a nationally representative sample of schools. Results from a randomized evaluation conducted by Innovations for Poverty Action (IPA) suggest that having TCAs provide remedial instruction targeted to the lowest performing pupils, both during school and after school, can improve basic literacy skills after just a few months.
Drawing on the success of the TaRL approach in India and Ghana, several other countries in sub-Saharan Africa are formulating or rolling out TaRL programs with technical support from Pratham and J-PAL. The Zambian government has launched the program in 470 schools, and plans to scale to 1,800 schools by 2020. With support from Pratham and J-PAL Africa, Governments in Côte d’Ivoire and Nigeria have launched TaRL pilots to adapt the approach to their local contexts. Organizations such as Evidence Action, IPA, the Japanese International Cooperation Agency (JICA), UNICEF, USAID, and Young 1ove are working to support governments to scale TaRL in other countries across the African continent.
Banerjee, Abhijit V., Shawn Cole, Esther Duflo, and Leigh Linden. 2007. "Remedying education: Evidence from two randomized experiments in India." The Quarterly Journal of Economics 122 (3): 1235–1264. https://doi.org/10.1162/qjec.122.3.1235.
Banerjee, Abhijit V., Rukmini Banerji, Esther Duflo, Rachel Glennerster, and Stuti Khemani. 2010. “Pitfalls of participatory programs: Evidence from a randomized evaluation in education in India.” American Economic Journal: Economic Policy 2 (1): 1–30. http://www.aeaweb.org/articles.php?doi=10.1257/pol.2.1.1.
Banerjee, Abhijit, Rukmini Banerji, James Berry, Esther Duflo, Harini Kannan, Shobhini Mukherji, Marc Shotland, and Michael Walton. 2016. “Mainstreaming an effective intervention: Evidence from randomized evaluations of ‘Teaching at the Right Level’ in India.” NBER Working Paper No. 22746. https://www.nber.org/papers/w22746.
Banerjee, Abhijit, Rukmini Banerji, James Berry, Esther Duflo, Harini Kannan, Shobhini Mukerji, Marc Shotland, and Michael Walton. 2017. "From proof of concept to scalable policies: challenges and solutions, with an application." Journal of Economic Perspectives 31 (4): 73–102. https://doi.org/10.1257/jep.31.4.73.