DC act of kindness

Snow was falling, and sidewalks were already slick with a half-inch of wet snow. A woman walked with her cane up to the crosswalk, clearly working hard not to fall, and assessed her options for getting across once the light turned. A young woman came up behind her, and the light turned. The young woman, seeing the other struggling, reached out her hand, and said a word. A connection was made. The one helped the other across the puddle against the curb, across the street, and over the opposite puddle. All the way, two mouths moved and two heads nodded, affirming human connection and human dignity with friendly words. Then they were on the far sidewalk. The initial reason for their connection gone, I thought I saw the young woman hesitate for a split second, as if making a decision. Just a split second. Then, her choice made, the two figures continued slowly together, the connection still visible even through the snow, until they were out of sight. 

When I moved to DC, I was worried that everyone would be focused on themselves and on the “who you know” game. I didn’t expect there was much room for simple kindnesses to strangers.

Perhaps this was unfair. Last night waiting for the bus at Cleveland Park, watching this scene unfold, my belief in the good inherent in human nature was reinforced. Even in this town.



Benefits of being top of the class?

Is that arrogant, smiling guy or girl giving your graduation speech actually going to be more successful? An interesting study from the UK suggests that, maybe they will, but perhaps not for the reasons you think… or maybe we all just knew it all along.

Essentially, they found that being a big fish in a small pond is better than being a equally big fish in a pond full of huge fish. The study finds that (take a breath…) kids who were in the top of their class in primary school had better test scores in secondary school, than kids who had the same scores in primary school but were in better schools, putting them in the middle of their class.

The money quote: “Non-cognitive skills such as confidence, perseverance and resilience have big effects on achievement.”

But. I would also be very interested to see how they controlled for the “inherent motivation” factor – the question of whether that confidence is inherent and not a result of their environment, and those kids were more inclined to achieve better test scores in both primary and secondary school anyway. I guess I should read the whole paper…

I am a Development Handyman. Here is my Toolbox.

Monitoring tools, systems strengthening tools, tools to build capacity of community-based groups, tools for gender assessment. A complete set of screwdrivers and wrenches.

We in development sometimes throw around this word as if we were a carpenter, to describe the things we use to help us do our job better. They might be guidelines, training curricula, or lists of indicators with a description of the best way to measure them. And though the title of this post is sarcastic, I don’t mean to disparage the incredible work, rigorous evidence, and highly-informed expertise that have gone into creating – and successfully using – many of them.

But what does the word “tool” imply? That they are standardized and will work the same anywhere? That we have them in our belt and, when the situation arises, we can pull out the right one for the job? That by banging a few nails and tightening a few screws, we can predict what will (or won’t) happen as a result?

Working in international development from the faraway land of Washington, DC, I am always interested by the little tricks of language that are used here – intentionally or not – that give us the impression (illusion?) of our own mastery and control over what happens in our programs. Nobody has devious intentions (or at least, very few in the NGO world), and everybody will affirm the importance of adapting to local context, and be genuine about it. Everybody knows that folks in DC can only have a certain amount of control over what happens on the ground in developing countries. Still, we use these words that let us believe (and perhaps more to the point, let donors believe) that we do have control. That the problems are linear and un-complex. That technical approaches are enough, and political considerations are irrelevant. That we have the right tools for the job.


Information can be powerful

This is becoming a theme of my ramblings recently, so much so that I’m seeing evidence everywhere. Reading today that the FDA is proposing to ban all artificial trans fats, what stood out is not the strength of the proposal (though does seem to be a pretty unequivocal ban). It’s that a lot of potential resistance to a ban – big companies that used trans fats – has already been neutralized back in 2006, when the FDA required that artificial trans fats be listed in food labels. Imagine if McDonald’s and Kraft still used huge quantities of trans fats. They would be fighting tooth and nail against a ban! But make them tell their customers how much trans fats they’re using? They find ways to eliminate or slash the content. And now, banning it is no big deal for them.

So, can we apply this to international development? This example shows how facilitating access to information (amount of trans fat on labels or menus), combined with mechanisms for accountability (choice of where to spend their dollars), can lead people to change their behavior (consumers buy less food with trans fat, and producers reduce trans fat content). The contexts of course are very different, but the theory at least is the same – access to useful information, that people can interpret and act on, can empower them to create positive change.

A few thoughts on adapting this theory to development, will keep updating:

  • Strengthen mechanisms for acting on information. In the trans fats case, it already existed – the competitive market. In our field, these mechanisms are weak or nonexistent in many cases.
  • Listen to people. Outsiders won’t necessarily know what information will be useful, or how certain types of information will be useful.
  • Mechanisms should be adaptive to the extent possible. So that people can keep making them more useful!


Is Boston’s new mayor anti-data?

No, but he knows there’s more to policy than what the stats say.

The money quote from the New York Times article: “Mr. Walsh also said he would judge his success in reducing crime not by statistics but by ‘how the community feels.’ ”

I get a kick out of this, especially because right across the river are the data wunderkinds of Harvard, MIT, J-PAL, etc. The consensus at these institutions, I suspect, would be that the word “feel” has no place in policy analysis. Or perhaps, they will start coming up with complicated, ingenious, and expensive ways to measure feelings.

I say good for Marty Walsh. And I hope that he doesn’t just look at opinion polls asking “do you feel safer?” I hope that he “uses some qualitative evidence”, which is policy language that translates in normal English to, “listens to people about how safe they feel and why.” As a good mayor should.


Slow ideas – convincing people to change

In development we try for a lot of behavior change. It’s easy to train people. We can sit a bunch of mothers down and explain why all the latest research shows that giving their baby water instead of breastmilk, what they’ve always done, is unhealthy. But it’s a rare that explanation wins out against imbedded practice.

A recent New Yorker article called “Slow Ideas” by Atul Gawande asks why some innovations spread fast, and others slow? He pulls examples from surgery. Anesthesia caught on like wildfire once it was discovered in the mid-1800s, making once-agonizing surgery virtually painless. Antiseptics, on the other hand, took decades to be widespread despite being discovered around the same time and tackling the very real problem of infection. Why?

His answer twofold: “First, one combatted a visible and immediate problem (pain); the other combatted an invisible problem (germs)…  Second, while both made life better for patients, only [anesthesia] made life better for doctors” who no longer had to operate on people while they thrashed around. So, to catch on quickly, the change had to solve a problem that was inconvenient for the person who could make the change.

Another key, for Gawande, is the source of the information. Sales reps use repeated contact to establish trust with potential clients; the “rule of seven touches.” This human contact may be even more important in many developing countries, where trust is often more rooted in social networks.

There are two elements that Gawande hints at in his article, which I think merit a little more discussion.

First: the visibility of cause-and-effect. With anesthesia, it’s easy to see that ether works. A “trial run” would show dramatic results: with it, patients lie calmly; without, they trash in agony. But with antisepsis, the treatment is applied during the procedure but the effects don’t manifest themselves until later. And, just one tiny mistake in the complicated process of sterilizing everything can lead to an infection just the same. A “trial run” of antisepsis might reduce the likelihood that a patient would get an infection, or might reduce the seriousness of that infection; far less dramatic.

So, being able to see clearly that the change works to fix the problem is an important factor in adoption.

Second: the beliefs that underlie people’s behavior. Gawande briefly suggests that skepticism about germ theory in the 1800s may have slowed the spread of antisepsis. But the influence of deeply-held beliefs – or even general suspicion of new ideas – has the potential to be a powerful, and hidden, barrier to behavior change. A good example here is female genital mutilation, a practice that in many places is integral to people’s idea of womanhood. You can go around talking up the proven health benefits of ending the practice itself, but people will see that as an attack on no less than their cultural identity.

So, the social and cultural factors that underlie current behavior are also an important factor for development actors working to change behavior.

With all this in mind, below are five questions that should be considered in any behavior change efforts:

  • How visible / inconvenient is the problem for people who could make the change?
  • How easy is this change to do properly, compared to the inconvenience of the problem?
  • How easy to see is the cause-and-effect of the change fixing the problem?
  • How deeply held are current beliefs about the nature of the problem?
  • How credible / relatable is the source / medium of information for the recipient?

I’m sure that I’m not the first person to think of these things, and that others have developed these concepts much more in depth. I will keep looking into this & very happy to hear any suggestions!

Data revolution… for whom?

A ton of data isn’t much good if you can’t use it for anything. Or if you can’t even get to it.

So with the data and accountability revolution growing in international development, we need to be thinking about who needs what data to hold whom accountable?

Large scale data for tracking comparative data across countries is certainly important. But what about the people of poor countries who could use data to hold government officials – national and/or local – accountable?

Both are important, but for people anywhere to be truly empowered they need more than the international community looking out for them. They need the information to look out for themselves – and not just to have information, but to have mechanisms for consistent access to useful information.

This statement contains two parts. I’ll start with the end: “useful information.”

Big STATA datasets or meta-evaluations are probably not much use in rural areas. Nor are national-level indicators. Local accountability means that people need to know how to track how their kids are doing in school, not the national average test scores disaggregated by gender.

The second part of that sentence is: “mechanisms for consistent access.”

Creating real mechanisms for people to hold local officials and national governments accountable is something that’s received scant attention in most high-level development circles, at least until recently. Ironically, the “data revolution” and pivot towards “results” in development may actually be working against this – students’ test scores are a lot easier to measure than the effectiveness of community-based school management mechanisms, and the money follows the measures these days. Of course, this is damn near impossible in some countries – Zimbabwe comes to mind, but also tricky in a tightly-controlled place like Ethiopia. But that doesn’t mean that there aren’t places where the development community can pursue local accountability and learn something about it.

And there is interest, both from above and below. Examples include:

  • The Million Voices project put out its report recently. Two key phrases for me (from the summary findings – i’m not that much of a nerd, and i recently got netflix) were: 1) “people consistently ranked honest and responsive government” as one of their highest priorities; and 2) People “see the data revolution as the foundation for an accountability revolution.” But not everyone is a data jock. Not all data is useful to people with limited education or resources; and even when you do have those things, all data must be interpreted. For this, people need consistent access to useful information that they can interpret themselves; eg, their kid’s report card.
  • Lant Prichett, a brainy professor from Harvard, has a new book about education – haven’t read it, but I did see him speak about it. His basic thesis is: development orgs & national policymakers need to stop focusing on inputs (teacher training, books, classrooms, ICTs, graduations) and start focusing on what kids are learning. Key to this for Pritchett is more adaptive systems, which means local control of schools, which means consistent access to useful information that they can interpret themselves… 
  • Last, just a quick mention of development blog gold, in the From Poverty to Power blog, on social accountability NGO Twaweza. Will write more reaction to this when I make the time.

At risk of being repetitive: consistent access to useful information that they can interpret themselves