"Using Evidence For Better Policy"

A new Karthik Muralidharan piece in Wednesday's Wall Street Journal on improving schooling outcomes in the Indian context. As always, I am super impressed by the author's attention to the need for a broad-based, tested outcomes-based approach to improving one of the worst schooling systems in the world. Muralidharan focuses on what works and what does not work. A quick recap:

"What Does Not Work" Simply adding more teachers to classrooms, building more classrooms, mid-day meals, better teacher training, better teacher salaries, and more student materials does not necessarily correlate to improved schooling outcomes. The author does note that there are inherent benefits to providing these inputs for students and families, which cannot be denied from a humanist viewpoint, but the actual, tested correlation with improved schooling is weak.

"What Works" Early education interventions ("supplemental remedial instruction"), focused on relevant, learner-centered materials, is critical to improving outcomes for these children. The key here is the materials themselves-they must focus on the ACTUAL level of the children, not the level that the bureaucrats and textbook publishers in far away capital cities dream up as appropriate. As I have mentioned many times on this website, the idea of relevance and connection to learners' lives is absolutely critical.

Muralidharan points to the importance of school governance, mainly in improving teacher absence problems, is a critical step in improving outcomes. While strengthening these management systems is obviously an important aspect of teacher management, we must also remember that teaching is an inherently autonomous profession, perhaps more so than any other; thus, we must focus on individual motivation inside the classroom (and related factors such as class size, appropriate materials, and creative capacity) in addition to the structure systems.

Looking at the area of teacher salaries, though the reseach shows that base salaries do not necessarily improve outcomes, performance-based incentives do. The question that I continue to have is the government's capacity in developing nations to handle performance-pay programs, in lieu of the fact that even some of the best managed educational systems in the developing world regularly struggle to pay even regular salaries on time. The nature of the current initiatives, in the Indian context, point to this, as they are all managed and run by NGO's.

The author stresses moving from the "business as usual" approach in educational development to an outputs/trial based approach. While this is certainly the path foward, we must walk with caution, and remember that we are dealing with the very human issues of motivation, satisfaction, change, and performance, in improving educational delivery in the developing world. Additional attention has been focusing on these specific areas, but much more remains to be done.