The last two decades have seen a tremendous expansion of the government school system (and probably an even more rapid expansion of private schools) in India. Unfortunately, learning levels have not shown any such improvement, and are, more worryingly, showing a decline[1].  Researchers and policy makers are struggling to find ways to improve learning levels in a systemic way. It is in this context, the phenomenon of private tuitions has started getting some attention[2].

An important question that emerges here is – who provides these private tuitions? In many instances, it is the school teacher who also provides private coaching or tuitions after school hours. This creates a potential distortion in teacher incentives. It is quite possible for these teachers to finish only part of the syllabus in the classroom and generate demand for their own tuitions outside the classroom. As a result, the students who don’t demand these tuitions (because, say, they can’t afford it) are clearly worse-off.

In an interesting new paper, Seema Jayachandran tests precisely this hypothesis using survey and test scores data from secondary school students in Nepal, to empirically assess the effects of tutoring on student achievement.

Education in Nepal is divided into primary school (grades 1 to 5), lower secondary school (grades 6 to 8) and secondary school (grades 9 to 10). At the end of grade 10, students appear for national school leaving certificate (SLC) exams, which must be cleared to continue further education.  Given the importance of this exam for furthering education, tutoring is common at secondary level, especially for Math, Science and English, the subjects where the pass rates are quite low. Tutoring is provided by school teachers post school-hours, as well as by others. But the share of these other third-party providers is quite low.

The author tests a number of hypotheses in her paper[3]. The first is that the teachers who offer tuitions, cover less material during the school day in order to generate demand for their tutoring.  She finds that offering tuitions make government school teachers 7.1 percentage points less likely to teach for the whole period compared to their counterparts in private schools.  Second, what consequences does this have for student test performance? Not surprisingly, the results reported in the paper suggest that offering tuition reduces the passing rate by 10 percentage points in government schools as compared to private schools.  Third, the author shows that household wealth is the most important predictor of higher take-up of tutoring. Thus, the existence of tutoring by government school teachers is likely to be most harmful for the poorest students, which is what the evidence seems to suggest.

The implications of this finding are quite clear: discouraging teachers from tutoring can actually be beneficial and can improve student’s academic performance. This is what the Right to Education Act (RTE) has done in India. The Act clearly states that, “no teacher shall engage himself or herself in private tuition or private teaching activity”. But do we know how well is this implemented on the ground?

[1] See ASER Reports for various years, available on http://www.asercentre.org/#qd70g

[2] There is some literature on prevalence on private tuitions but not much on impact of tuitions on learning level. See ‘The Shadow Education System: Private Tutoring and Its Implications for Planners’, Bray Mark (2007), UNESCO: International Institute for Educational Planning.  

[3] I have not gone in details of empirical techniques employed in the paper. Those interested in that aspect are requested to read the paper.