Machine learning for meta-analysis

We focus on gathering data from impact evaluations and synthesizing that information through meta-analysis.

The hardest part of doing a meta-analysis is gathering the data. Each paper must be read by two different people, who manually extract information from the papers; if they disagree on any aspect, a third person is called in to arbitrate.

This process doesn’t scale very well and it is a large part of why meta-analyses are often out of date. With many new studies coming out, we would like to be able to streamline this process.

Machine learning allows us to do so. We can extract information about the programs and studies as well as information about the effect sizes they found. For each extracted piece of information, we will also generate a probability that the information is correct.

At the very minimum, this will greatly reduce the amount of time it takes to identify basic characteristics of studies, such as where they were done and which methods they used. It is also the only way to really stay abreast of the latest research for a variety of areas. Given that the methods should be scalable to much of health, education, and economics, we will build this tool in a general way so that its results can inform policy even in developed countries.

You can think of this as a ScienceScape for meta-analysis.

To support this initiative, we have a crowdfunding campaign. Please consider donating.

Researchers wanted

AidGrade is hiring researchers. A master’s degree and experience with impact evaluations is ideal, but others with a quantitative background should apply. The researcher’s main responsibilities will involve data analysis, using the database of impact evaluation results that AidGrade has developed.

Former AidGrade employees have been accepted to PhD programs at Harvard, Yale, Princeton, Stanford, NYU, Columbia and Michigan, and to staff positions at the World Bank.

Work can be done remotely.

Application deadline (revised): May 13, 11:59 pm PDT

To apply, please send cover letter and CV to

New data and papers

We have a lot of new results coming out over the next few weeks – watch this space!

First up: updated data in the meta-analysis app. Data on 10 new interventions were added: contract teachers; financial literacy; HIV education; irrigation; microhealth insurance; micronutrients; SMS reminders; performance pay; rural electrification; and women’s empowerment programs.

Next up: working papers on each of the topics covered to date.

Research assistants wanted

AidGrade is looking for research assistants to help with data work. Applicants should have some quantitative sophistication. Work can be done remotely and is flexible.

Deadline (extended): July 18, 2014.

To apply, please send CV and cover letter to

Get involved!

We are looking for more people to help with our meta-analyses. A master’s degree and experience with impact evaluations and/or meta-analyses is ideal, but we can suggest other options depending on your experience.

In the past, AidGrade employees have been accepted to Economics PhD programs at Yale, Princeton, Stanford and NYU, Political Science PhD programs at Harvard, Princeton, Columbia and Michigan, and staff positions at the World Bank.

We are also accepting intern applications for summer 2014 for both research and outreach roles. This is your chance to learn about impact evaluation and meta-analysis and work with a great team!

One can work remotely. We also have clusters in Washington, D.C., New York, and San Francisco.

Application deadline: April 2, 11:59 pm PDT

To apply, please send cover letter and CV to, highlighting the role(s) for which you wish to be considered.

An intern at AidGrade

Post by Scott Weathers, @scott_weathers.

My first day at AidGrade came with a gift. In our fall team’s first meeting with Eva, we received owl-themed kitchen timers.

But these weren’t stocking stuffers. Eva brought up a set of studies—which everyone was already familiar with—arguing that working for short, fixed blocks of time was the most effective work strategy. Spend less time concentrating and you’re not reaching optimal productivity, spend more and you’re burning yourself out.

This is AidGrade. How can we use our time, money, and energy to the greatest effect? How can development aid be maximized to spur economic growth, cure disease, and save the greatest number of lives?

AidGrade is a risky endeavor. We are less than a dozen employees and interns working to refine the most promising solutions to global disease and poverty.

You’ll see the importance of AidGrade right away. Why isn’t there a database of studies on HIV education, anti-malarial bed nets, micronutrients, or microfinance? If you like to ask big questions, you’ll fit right in.

You’ll also learn quickly. The fine-grained details of Randomized Control Trials (RCTs) may not be too familiar right now, but AidGrade presents an awesome opportunity to dive in.

AidGrade is a development start-up that I believe is working on some of the most important issues the world has today. If you have any aspiration to impact humanity, or at least understand the impact you may have, I encourage you to join.

Scott was a fall term social media intern. Interested in learning more about what opportunities are available? Get in touch: