On average, children in poorer districts of the country tend to learn less in schools, shows an analysis of two large data sets on learning outcomes
The analysis shows that large parts of Uttar Pradesh, Bihar, and Madhya Pradesh have the poorest learning outcomes in the country
New Delhi: There is a stark divide across rural India in learning outcomes, with better performing districts clustered in a few parts of the country, a Mint-HT analysis of two large data sets on learning outcomes show.
The first data set is the unit-level data from the annual state of education report (ASER) 2016 survey, sourced from the non-governmental organization Pratham which conducts the ASER surveys. The second data set is the set of district-level report cards published by the National Council of Educational Research and Training (NCERT) based on the results of the National Achievement Survey (NAS) conducted in 2017.
The analysis shows that large parts of Uttar Pradesh, Bihar, and Madhya Pradesh have the poorest learning outcomes in the country, according to both the surveys analysed. These parts of the country are also among the poorest, suggesting that India’s school system is failing children precisely in those parts of the country where it is needed the most.
On basic numeracy, most of the better-performing districts are clustered in peninsular India, in the extreme north, and in a slim strip along the east coast extending all the way to the North-East, ASER data shows. Barring the extreme north, the other two belts also feature in the top performing NAS districts (charts 1a and 1b).
The analysis is based on the district-wise aggregate performance of rural students of the third standard in math questions asked by NAS. For ASER, the analysis considers aggregate district-wise performance of primary students in basic math (subtraction). To ensure sizeable samples at the district level, the performance of students in the classes III-V bracket were aggregated.
District-level comparisons of ASER and NAS numbers don’t always converge. For instance, consider the extreme West region: NAS data suggests the region—especially districts in Rajasthan—is among the top performer in basic math. But the ASER numbers suggest that these districts are considerably worse than the national average.
To be sure, the two surveys have different methodologies. ASER is a rapid assessment survey conducted across rural households covering both private and government school students while NAS employs a more detailed test and is conducted in classrooms of government schools across the country. The analysis also considers a larger cohort from the ASER survey than from NAS, as described above. All of these differences could be driving at least part of the variation between the trends from the two surveys.
Overall, 34% of the districts in the bottom quintile (bottom 20%) ranked, according to the ASER survey results, feature in the bottom quintile of NAS. And more than half of the districts in ASER’s bottom quintile feature in the bottom two NAS quintiles (bottom 40%), suggesting significant convergence in trends when it comes to performance in basic math at the primary level of schooling.
The trends in reading ability and language scores are broadly similar to the trends in math performance. South India is an outperformer when it comes to language scores (NAS) and reading ability (ASER). But barring pockets in the North-East and West Bengal, the east fares considerably worse than the western half of the country when it comes to language proficiency and reading ability (charts 2a and 2b).
Several districts of Punjab, Haryana, and Jammu and Kashmir fare well in terms of reading ability (ASER) although they don’t seem to be among the top districts in language scores, according to NAS. Rajasthan, again is an outperformer according to NAS but not according to ASER.
While there are differences in the two surveys, both surveys indicate significant geographic inequality in learning outcomes, with prosperous parts of the country, on average, performing far better than the poorer parts. A Mint analysis of the data from the National Family Health Survey (NFHS) conducted in 2015-16 had shown how richer districts of the country are clustered in the Western half and the poorer districts in the eastern half of the country. With some notable exceptions (such as in the North-East), the trends in learning outcomes are roughly similar to the trends in affluence, this analysis shows. One reason for this is that the affluent and the privileged tend to learn better, as the first part of this series showed, and they tend to be clustered in the relatively more affluent districts. The other reason for this could be the low level of investments in school infrastructure and weak monitoring in the poorer districts. This could be the other big driver of poorer learning outcomes in the districts that need better schooling the most.
This is the concluding part of a two-part data journalism series on learning outcomes in India. The first part examined how household inequality shapes inequities in learning outcomes.