How to use evidence to improve student learning

See if you can spot the pattern:

• “Although the quantity of schooling has expanded rapidly, quality is often abysmal.” (Kremer et al.)
• “Between 1999 and 2009, an extra 52 million children enrolled in primary school…Yet the quality of education in many schools is unacceptably poor.” (Krishnaratne et al.)
• “Progress over the last decade in regards to school access and enrollment has been promising.” But “current learning levels for primary as well as secondary school students are extremely low in much of Sub-Saharan Africa” (Conn)
•  “The most consistent focus of investment has been on increasing primary and secondary school enrollment rates… More recently, however, attention has begun to swing toward the quality of schools and the achievement of students—and here the evidence on outcomes is decidedly more mixed.” (Glewwe et al.)
• “Over the past decade, low- and middle-income countries have made considerable progress in increasing the number of children and youth who enroll in school and stay long enough to learn basic skills… Learning in many low- and middle-income countries remains appallingly low.” (Murnane & Ganimian)

Again and again, we hear the refrain: access is improving, but learning lags. Thankfully, an increasing number of studies reveal interventions that work – and those that don’t – to improve learning around the world.

At the same time, an increasing number of reviews seek to synthesize what works most consistently. In a paper just published in a peer-reviewed research journal, the World Bank Research Observer, we studied six of these reviews to see where there was the most consensus on what works. We found two areas. Of course, these aren’t the only types of interventions that work to improve learning. However, they are areas with evidence from multiple studies and multiple countries, so these are good places to start.

What works?

Pedagogical interventions that match teaching to students’ learning. Matching teaching levels to the level of learning of the student doesn’t sound like rocket science but, in overcrowded classrooms and overambitious curricula, it can be difficult. Interventions that help to achieve this have consistently made a difference. What does this look like?

• In Kenya, a program that separated students into classes by initial ability to allow teachers to focus their pedagogy led to gains for both low- and higher-achieving students.
• In Liberia, a program that trained teachers to carry out diagnostics of their students in order to teach to their learning gaps (+ provided materials) led to gains in a range of reading outcomes, although note that another program giving diagnostic feedback on student performance – this one in India – increased teacher effort but didn’t translate into learning gains.
• In India, a program that separated students into classes by ability for just one hour of the day led to significant gains in learning.
• Sometimes, programs use technology to teach students at their level. Again, in India, there’s a  program using math software that let students learn at their own pace, which improved student outcomes with high cost-effectiveness.

Individualized, repeated teacher training associated with a specific method or task. Not all teacher training is created equal. However, massive resources are spent on in-service teacher training (i.e., training teachers who are already on the job) and a collection of evaluations have used teacher training to deliver significant learning gains. Teacher training programs that enable teachers to effectively use new pedagogies and those which provide follow-up training (i.e., not just a one-off) have repeatedly been effective.

• Training teachers to use flashcards, storybooks, and libraries led to student learning gains in India, while providing these resources without this training was less effective.
• Training teachers in a scaffolding approach to literacy instruction led to significant gains in Uganda and smaller gains in Kenya.
• Training teachers in structured and systematic lessons and activities for literacy led to gains in reading in early grades in Mali.
• Training teachers in another early grade literacy program led to large gains in Uganda.

Not all teacher training delivers results ( see an example with actual negative impacts here, from Costa Rica), but training teachers in specific pedagogical methods (which often come with supplementary materials) and following up with them afterwards can deliver gains.

Remember…

There is a consistent stream of new evidence. A summary of evidence that came out last month cites many studies from the last few years, and we will continue to add to what works in many contexts. What will work in a specific country (or region of a country) will always be affected by local factors. This evidence points us in the right direction, but it is crucial to continue to experiment locally to find the best solutions for learners worldwide.

Bonus materials

Our new paper – in the World Bank Research Observer – is also available in an earlier form as a Policy Research Working Paper. At that time, we wrote a blog post exploring why reviews of what works to improve learning often come to different conclusions. We also had a separate post – back then – on what works to improve learning. This post is updated with a little bit of additional evidence.

 
The order of authors on this blog post was randomized .

Find out more about the World Bank Group’s work on education on  Twitter  and  Flipboard.