Ideas to Action:

Independent research for global prosperity

Working Papers

Primary Schooling, Student Learning, and School Quality in Rural Bangladesh - Working Paper 349

Mohammad Niaz Asadullah and Nazmul Chaudhury
December 17, 2013
Download PDF

From the Foreword by Lant Pritchett

The Center for Global Development is committed to the expansion of education as a key component of development. But while the focus on expanding education has been on the expansion of schooling the Center has been an early promoter of the pivot from an exclusive emphasis on schooling to greater emphasis on learning (see Filmer, Pritchett, and Hasan 2006). As part of that effort, CGD published The Rebirth of Education: Schooling ain’t Learning in September 2013. One of the key concepts in this book is a “learning profile” which is the association between years of schooling and mastery of skills. Unfortunately there are, as yet, very few empirical estimates of learning profiles as most assessments are done in school and for limited ranges of grades (or ages). This means that while we can examine how much 15 year olds can do from the PISA or what sixth graders can do from SACMEQ or what eighth graders can do from TIMSS these are static pictures and do not show the trajectory of learning over time as students progress through grades.

This present paper, by Mohammad Niaz Asadullah and Nazmul Chaudhury therefore makes an important contribution to the literature in a key area of CGD concern. Using a representative sample of 2400 households producing data on 3323 children aged 10 to 17 they assess ability to answer simple arithmetic question (either oral or written). The striking finding of their paper is just how little having completed an additional year of school increases the likelihood the child has mastered basic skills. Of children who had completed primary school (but no more) the percent who could get 3 or more of 4 questions correct on the oral assessment was only 56 percent for males and 42 percent for females. Even more striking is that regressions show the likelihood of getting 3 or more correct is only 9 percentage points higher for children with grade 5 complete than for children with no schooling at all. Taken at face value this suggests that five full years of primary school taught less than 1 in 10 children how to do simple math. The accumulation of results like this is central to keeping the pressure on for a new set of development goals in education that focus on learning outcomes and particularly on early mastery of key literacy and numeracy skills.

Related Topics:

Rights & Permissions

You may use and disseminate CGD’s publications under these conditions.