How researchers get heard
Abstract lines

Esther Duflo’s TED Talk

Sometime soon — today, tomorrow, this weekend — you should watch Esther Duflo’s TED talk. It’s a model for how to frame your disruptive research and ideas.

This is Esther Duflo who just won the Nobel Prize in economics for her research to find the most effective interventions against poverty and associated, preventable diseases. Duflo helped pioneer the use of randomized control trials (RCTs) in these studies, elevating an entire field’s approach to these issues. She’s the youngest person to win the Nobel in economics, and just the second woman.

Her TED talk is a masterpiece: someone who a) isn’t a great public speaker in English and b) has difficult material to explain using c) a twist on a classic TED formula to d) give you a new frame on a common problem you won’t forget.

First: she doesn’t get bogged down explaining her own methodology. Instead, she focuses on the mental shift it represents for the audience — a new, more effective way of looking at the world and solving problems. She makes the status quo way the enemy, so that we’re all eager to abandon it.

The coverage of Duflo’s Nobel win (with her husband, fellow MIT economist Abhijit Banerjee, and Harvard’s Michael Kremer) has concentrated on how the trio revolutionized poverty economics with RCTs. Fascinating to all of us — but I doubt many other people care about methodological shifts.

One of Duflo’s major tricks in the TED talk: she mentions RCTs twice, in consecutive sentences, and then never again. If we could study poverty as medical researchers study health, she says, then we could make similar amazing strides. That’s all she needs to say.

Instead, she starts her talk with an intriguing hypothetical: If you were a politician and had a few million dollars, how would you best spend it toward fighting poverty?

Most people begin arguing about the virtues and demerits of foreign aid. But that’s a distraction, Duflo says, because there are too many variables and there’s no counterfactual. We don’t have another Africa that didn’t get aid to compare this one to.

In essence, the question is too big — and we’re like medieval doctors about it: attributing success to leeches but having no evidence.

Instead, she goes on, what if we asked smaller questions — but still important ones — that we could get evidence-based answers for? Answers that would impact tens of millions?

Questions such as: what will help get the 25 million children around the world who aren’t getting immunized their shots? Which price should anti-malarial bed nets be to get people to use them and buy others? What are the best incentives to get parents to send their kids to school?

Notice the move Duflo has made. The problems she wants to investigate are the enemy. But the enemy she’s targeting in the talk is more intimate: a way of thinking that, because it is susceptible to ideology and not grounded in science, amounts to guesswork and reactivity that actually obstructs solving big problems. “More ideological than practical,” she keeps saying in various ways throughout her talk.

Second: Duflo’s presentation of her findings is masterly: she picks the ones that are astonishingly counterintuitive, and frames them all as options, clearly pointing to the best one.

This is such an important lesson for researchers. Her charts are not about what she foundthey’re about what she found and how to act on them. About the clarity and memorability of options a policymaker or funder needs to choose the best one. That, of course, is the point of her research design — but it’s also beautiful research-to-policy communications.

How do you get parents to take their children for immunizations? Duflo’s chart shows: you give them a kilo of free lentils — just enough of a nudge to get them to go and keep coming back.

How do you get them to use bed nets? Give them away, the evidence demonstrates, against anecdotes that people will use them for fishing.

What’s the best way to get kids to school? Tell them the benefits of education, and deworm them, says the evidence — for every $100 you spend, you get between 30 and 40 extra years of education.

“This is not your intuition, this is not what people would have gone for, and yet these are the programs that work,” Duflo says. Again: she doesn’t talk about her methodology. The results she obtains speak for it. In making the case for the clear policy choices her work has uncovered, she’s also made the case for the approach that generated them.

Third: After these astonishing results, Duflo twists the usual TED talk ending (i.e., here’s the evidence-based takeaway that will make your life so much better):

“There is no miracle cure, but modern medicine is saving millions of lives every year, and we can do the same thing…I can’t tell you whether the aid we have spent in the past has made a difference, but can we come back here in 30 years and say, ‘What we have done, it really prompted a change for the better.’ I believe it can and I hope we will.”

I love her dizzying swaps of modesty and ambition — I can’t answer the ridiculously unanswerable big question that keeps tripping us up (modesty in the face of conventional ambition), but I can follow an approach that, in a couple of decades, will give us a bevy of answers that make the world much better…which is what we all want (my ambition that, through a collection of modest, science-informed intervention, we can collectively achieve our ambitions to greatly improve the lives of the poor).

It’s easy to rip on TED talks — even more so now since the MIT Media Lab scandal and Anand Giridharadas’s book, “Winners Take All”, which fingered them as a vector of neoliberalism.

But more than 1 million people have watched this 2010 TED talk, and walked away with a new way of thinking and a new appreciation of the power of research to ask and answer critical questions. That’s more than research communications — that’s turning an idea from research into a paradigm.