The lack of effective education is a significant problem faced by developing countries. While many of these countries boast high levels of enrollment, students often advance through school without learning basic concepts in reading and arithmetic, creating a learning gap that is detrimental to the productivity of the working population. A 2016 paper entitled, “Mainstreaming an Effective Intervention: Evidence from Randomized Evaluations of “Teaching at the Right Level” in India” (Banerjee et. al) aims to locate the causes of the learning gap faced by students in India (specifically four states: Bihar, Uttarakhand, Haryana, and Uttar Pradesh) and determine a model for a program that can be scaled and implemented to the entire Indian education system.
Causes of the Learning Gap
Government-run primary schools in India feature automatic promotion through grade 8. Each grade level teaches a common syllabus to every student, regardless of their ability. Therefore, even students who lack a fundamental understanding of concepts taught early in school advance to the next grade with no checks in place. For example, an Annual Status of Education Report (ASER) for Indian schools in 2014 shows that “39 percent of fifth graders could not read at a second-grade level” (Banerjee et. al 4). Banerjee’s paper concludes that automatic promotion, along with a lack of targeted teaching based on current learning levels of students, are some of the reasons for the learning gap in India. Solutions to these problems are readily available. However, the authors mention that while “simple changes in pedagogy can lead to significant improvements in learning levels” (4), the government is mostly unwilling to experiment with changes in teaching style and remedial education for struggling students. Thus, most of the burden falls on NGOs to reform education.
With evidence to support the positive effects of pedagogical changes on student success, the authors of this paper set out to produce a replicable model that would reform the government school system and improve the quality of existing government schoolteachers. The experiment conducted in this paper is based on an instructional model devised by Pratham, an NGO that aims to improve learning on a large scale while remaining cost-effective. Pratham implemented a system in Uttar Pradesh where students would be organized into groups based on their current ability rather than grade level. These groups would be taught by “carefully monitored but lightly trained community volunteers” (5), and would focus on basic language and arithmetic skills. A study from 2010 conducted by Banerjee and his colleagues (Banerji, R., Duflo, E., Glennerster, R., and S. Khemani) showed that Pratham’s program had significantly improved the reading levels of children who had attended.
The Pratham program had shown signs of progress, but its success on a national scale was uncertain. To determine whether this type of intervention was worth pursuing at the national level, the authors of the paper set up experiments in four Indian states: Bihar, Uttarakhand, Haryana, and Uttar Pradesh. In Bihar, a “summer camp” was conducted where government teachers were recruited for remedial classes during the students’ summer vacation. After the summer camp, test scores improved by .07-.09 standard deviations. This program carried on to Haryana and Uttar Pradesh, where test scores also improved (in Uttar Pradesh, on-site monitoring was a likely factor in the successful implementation of the summer camp). In Uttarakhand, the experiment involved three different scenarios. In the first scenario, schools were given learning materials (not specified in the paper) without any additional assistance. In the second scenario, teachers were trained to use the Pratham strategy and the school received the same materials. In the third scenario, materials were provided, teachers were trained, and additional in-school volunteers were hired to use the Pratham method in conjunction with the teachers. Schools were randomly selected to receive one of these three treatments, and results were compared after two years.
The results from the experiments after the two year observation period were interesting, if not unexpected. Exact replication of the Pratham method (conducted in Bihar, then Uttarakhand as scenario two) proved successful (see Appendix for results of the experiment), but the other scenarios failed. Scenario one failed simply because providing additional teaching materials is not sufficient to improve test scores. Scenario three also failed, but for an unexpected reason: instead of the volunteers functioning as valuable assets to aid teachers with implementing the Pratham method, teachers were instead using them only as assistants and reverting back to their old teaching style.
Conclusions and Importance to Future Efforts
The successes and failures of the experiments conducted in Bihar, Haryana, Uttar Pradesh, and Uttarakhand led the authors of the paper to note two “key ingredients” that play a major part in the Pratham model and its ability to scale and improve government-run schools. The first ingredient is the grouping of children by their current learning level as opposed to grade level, and the second ingredient is focusing on concepts that are suitable for all children at a given learning level. However, these ingredients only seem to be followed when additional support from Pratham staff is available and rigorous on-site monitoring is frequent.
Studies like this provide much-needed evidence to present to the governments of developing countries, in the hopes that they will be able to reform their schooling systems and immensely improve the quality of their education. This paper in particular shows that simple changes in pedagogy and the method in which students are grouped can lead to a hugely improved education system (Uttar Pradesh’s summer camp presented a staggering .61-.70 standard deviation increase in test scores after only two years). Education is one of the most important problems faced by developing countries, but the methods described in this paper detail a way forward that is simple enough for widespread implementation, and effective enough to make the effort worth it.
Banerjee, A., Banerji, R., Berry, J., Duflo, E., Kannan, H., Mukherji, S., . . . Walton, M. (2016,
October). Mainstreaming an Effective Intervention: Evidence from Randomized
Evaluations of. Retrieved May 08, 2017, from http://www.nber.org/papers/w22746
Banerjee, A., Banerji, R., Duflo, E., Glennerster, R., & Khemani, S. (2008, September). Pitfalls of
Participatory Programs: Evidence From a Randomized Evaluation in Education in India.
Retrieved May 08, 2017, from http://www.nber.org/papers/w14311