There has been a substantial increase in access to education in India, but such access has not been simultaneously met by an increase in quality. This has two consequences.
First, the learning levels of children in government schools are abysmally low so the expanded access is unlikely to have a major impact on future earnings or provide access to high-paying occupations. For instance, in 2018, 55 percent of fifth-grade children in public schools in India could not read a second-grade textbook (ASER, 2018). Second, many families with the capacity to pay for a private school are switching their children to the private sector. However, evidence on whether private schools in India lead to test score gains is mixed.
Thus, from a policy perspective, it is important to understand if raising the quality of government schools can lead to improved schooling outcomes. My research exploits a natural experiment in education policy in India to examine the effects of creating high-quality government schools.
India’s effort to improve school quality
With the intention of improving primary and secondary education, India designated 3,479 out of 5,564 blocks as educationally backward blocks (EBBs). Addressing the state of EBBs and public education in India, then-Prime Minister Manmohan Singh, in his Independence Day Speech in 2007, called for states “to give priority to education, as education alone is the foundation on which a progressive, prosperous society can be built.” To accomplish this, it was proposed that the government would establish 3,500 “model” schools, one for each EBB. Although funding for the Model School Scheme was split between the states and the federal government, state governments were responsible for the scheme’s implementation. However, as of 2016, the Government of India has stopped funding the Model Schools Scheme and the program is transferred to the state governments.
Model school scheme
The Model School Scheme, launched in 2008, established government schools that have a superior infrastructure, high accountability, English as the default medium of instruction, and contract teachers. The objective was to start one exceptionally good government school in each of the EBB that could serve as an archetype for traditional government schools to emulate. A block is considered educationally backward if its female literacy rate was below the national average and its gender gap in literacy was above the national average in 2001. I look at Karnataka, a southern state in India, where model schools — called ’Adarasha Vidayalas‘ — start at grade 6 and end at grade 10. Karnataka has a total of 74 EBBs and the first cohort of model schools was admitted in 2009.
Admission into a model school in Karnataka is determined through an entrance exam where students are tested on languages, math, science, social science, general knowledge, and cognitive ability. The entrance exam is conducted at the block level; hence, students residing in a particular block compete for the model school in that block.
Each model school can admit up to 80 students in total. Using the admission lists prepared by the examination authority, the principal of each model school will admit students in descending order, based on their entrance exam score and caste category. The nature of the selection process creates a cutoff for each caste category within each model school. As a result of this admission process, nearly identical students are either admitted to or rejected from a model school. For example, if a school’s cutoff score under the SC category is 70 points, an SC category student who scored 70 can attend the model school but an SC category student who scored 69 cannot.
With a data set of over 63,000 students from 74 model schools across three cohorts, I am able to investigate three dimensions of schooling outcomes: (i) academic achievement as measured by test scores and final grades; (ii) educational attainment indicators using years of schooling; and (iii) career choice using choice of major in pre-university college.
For academic achievement, attending a model school increases math test scores by 6.8 points (0.38 standard deviations), science test scores by 4.1 points (0.26 sd), and social science test scores by 4.7 points (0.26 sd).
Attending a model school also increases the probability of obtaining an A or A+ grade in tenth-grade by 20 percentage points. For educational attainment indicators, attending a model school increases the probability of passing tenth-grade by 5.3 percentage points and increases the probability of joining pre-university college by 11.9 percentage points.
I also explore heterogeneity in the scheme’s effects by gender since geographic blocks were classified as educationally backward based on the gender gaps in education. I find that attending a model school increases girls’ and boys’ test scores in math and social science, and the likelihood of scoring an A/A+ by about the same amount.
Interestingly, attending a model school has a positive effect on females when it comes to the probability of joining pre-university and choosing science as a major compared to an almost zero effect on males. In general, the results suggest that model schools work for girls as well as boys.
Potential change mechanisms
First, traditional government school teachers are civil-workers who are hired on a permanent basis and the model school teachers are recruited on a contract basis. From a pure effort-based perspective, the temporary-contract structure leads to model school teachers exerting high effort levels either to ensure the renewal of their contract or in order to become a permanent government school teacher (Muralidharan and Sundararaman, 2013; Duflo et al., 2015).
Second, the primary objective for launching the Model School Scheme was to create schools that could serve as an archetype for traditional government schools to emulate. Therefore, the Department of Education governed the model schools very closely by increasing the number of inspections, increasing the number of meetings with school principals, and holding the schools accountable for properly performing their daily functions. Additionally, the targets set for model schools to achieve were much higher than those given to traditional government schools.
Third, attending a model school can influence student psychology in a positive way through the medium of instruction and infrastructure. There is well-documented evidence suggesting high returns to learning in English. Traditional government school students, who are mostly low-SES or low-income students, maybe demotivated by the prior belief that they cannot compete with their counterparts at private schools either for higher education or for high-level jobs. If this is the case, then learning in English in a model school may boost the esteem of government school students. Similarly, the improved school infrastructure may also make students believe that the education they are receiving is comparable to that of their private school counterparts.
Relevance to policymakers
First, the Model School Scheme is yet to be fully implemented by all state governments. For instance, 12 out of 21 states with Educationally Backward Blocks (EBB) did not have functional model schools as of 2016. For example, Odisha, an eastern state with 173 blocks out of 315 classified as EBB, only implemented the Model School Scheme in 2017.
Second, Karnataka is planning to create new traditional government schools and/or consolidating the existing schools based on the model school framework. In one policy, Karnataka is planning on introducing an English medium track starting from grade 1 in 1,000 traditional government schools in the 2019-20 academic year. The government plans to gradually add an English medium track to all government schools in future years. This proposed move has divided the pro-regional language activists and the leaders of SC and ST groups, majority of whose children rely on traditional government schools.
Finally, turning to the costs of running the model schools, back of the envelope calculations imply that per-pupil expenditure in model schools is between Rs 9,315 – Rs 11,632 and per-pupil expenditure in traditional government schools is between Rs 11,848 – Rs 16,914. Therefore, the costs of operating model schools are comparable to that of traditional government schools.
Raising the quality of government schools can have significant positive effects on several dimensions of student outcomes.
With 75 per cent (about 1 million) of schools being government schools and 65 per cent (approximately 120 million) of the children who are in school attending a government school, quality of government school in India is a first-order policy issue. Improving the quality of government schools is at the core of the current education reforms that are being introduced by various state governments in India. Uncovering the effects of improved government schools prior to their state-wide implementation can be vital to their success.
Personal note: A lot of credit has to been given to the Government of Karnataka, The Department of Public Instruction, all the government officials in the education department, and BEO, CRC, BRCs working in these taluks to ensure proper implementation of the scheme. All I have done is analyze the effects of a well-implemented program and think critically about what policymakers and economists can learn from it.
This is an edited excerpt from the author’s job market paper “Public Schools Can Improve Student Outcomes: Evidence from a Natural Experiment in India”