Computer-aided learning programmes raised Maths test scores 2 times and 2.5 times for Hindi in just 4.5 months
India may finally have a solution to its serious learning deficit problem. Despite primary school enrolments jumping to around 95% today, as one ASER study after another shows, students simply aren’t learning. At an aggregate level, the average Grade 6 student is around 2.5 grade levels below 6th grade standards in maths, and this rises to 4.5 grade levels by Grade 9. While private schools fare better than government ones, once the higher education levels of parents are factored in, the difference narrows. Different states have used different approaches to fix the problem—hiring of younger para-teachers improves outcomes somewhat and dramatically lowers costs. A control-group study by Karthik Muralidharan of UC San Diego, in 2012, found providing individual-level performance bonuses to teachers led to test-score gains of 0.54 standard deviations (SD) and 0.35 SD in math and language for students exposed to the programme for five years. A new study by Muralidharan, Abhijeet Singh (University College, London) and Alejandro Ganimian (J-PAL) examines the impact of a computer-aided learning programme called Mindspark. Developed by Educational Initiatives, it has a bank of 45,000 test questions and, because it has been used by over 400,000 students, it is able to generate questions based on the ability levels of students and then take them up the curve. Unlike a classroom where teaching levels are determined by the poorest student, Mindspark enables dynamic ‘teaching at the right level’ for each student. And the teaching can be scaled up at the pace the student is learning. Using a low-income neighbourhood around Delhi as the catchment area, the study found Mindspark raised maths test scores 2 times and 2.5 times for Hindi in just 4.5 months. In the context of Muralidharan’s 2012 study, the new one says, “we estimate that regularly attending Mindspark could yield similar gains in one-tenth the time”.
You may also like to watch this video
The economics is even more appealing. While the government spending per student was around `1,500 per month (the students studied 240 minutes per week) in the traditional method, using computer-aided learning cost just `1,000 per month and cut student time to 180 minutes. But since it is not going to be possible to eliminate the teachers, Mindspark—or some alternative—will have to be an add-on. For India’s 30 crore students in Grade 1-12, that means a cost of `28,000 crore per year—compared to the current budget of `46,000 crore, that’s a lot of money but in terms of the outcomes and overall education-spend (including private tuitions) of 4% of GDP, it’s a tiny fraction.