"Moneyball" Giving, Data Philanthropy Review and "Seeing like a State" #Data Digest

keishataylor
See all of keishataylor's blog posts

This week  the revolutionary data driven philanthropy of the Laura and John Arnold Foundation is featured and a call for beneficiary inclusion in Markets for Good goes out. Data philanthropy efforts are reviewed and advice for writing about data is offered. The rational for "seeing data as a state" rather than a donor in Africa to help answer policymakers questions is also explained.

Data and Philanthropy
The New Science of Giving
This Wall Street Journal post discusses how John Arnold a former natural-gas trader at Enron and hedge fund founder is giving away his $4 billion fortune through a new “Moneyball” approach to giving.  Through the Laura and John Arnold Foundation he set up with his wife, he plans to use the money to ambitiously solve some of America’s biggest problems like obesity and crime through data analysis and science, with an unsentimental focus on results. 
He is using his background as a trader to forward the foundation's approach and spends a lot of time doing research and evaluating data in the pursuit of high risk social change projects.

Why Markets for Good may go wrong
In this Alliance Magazine article David Bonbright, chief executive of Keystone Accountability explains that while the Markets for Good initiative intended to develop a philanthropic ecosystem and social change information marketplace that gave the most marginalised a voice, in reality this has not been the case. According to him, Markets For Good has a flawed understanding of beneficiaries as primarily consumers of information. He also says that it focuses on service availability and eligibility requirements rather than those it serves. He advocates a Feedback Principle of Public Reporting where organizations publish beneficiary feedback so that the information flow is not top down, but instead people centered.

Data Philanthropy: Where are we now
Andreas Pawelke and Anoush Rima Tatevossian reflect on the progress of data philanthropy (private sector companies sharing data for public benefit) since it first emerged at the World Economic Forum in 2011. They found that the concept has developed and there is a lot more discussion and debate about it and more evidence to support it.  Private companies are also testing the concept by actually sharing data. For example, France Telecom-Orange has made anonymized records of five million mobile phone users in Cote d'Ivoire available to the research community as part of the Data for Development Challenge. However the search is still on in communities around the world for legal frameworks, ethical guidelines and technology solutions that preserve data privacy.

Aid Data
Seeing Like a State in Africa: Data Needed
Justin Sandefur, a member of a “Data for African Development” working group that is collaborating with the African Population Health Research Centre in Nairobi argues for "seeing like a state" - i.e. collecting data to answer policymakers questions and not like a donor when collecting data in Africa today. In this Center for Global Development post, he says that African governments are wrongly being paid to collect statistics to populate donor database and considers how this affects the formulation of more effective education policies in Kenya. He says that “seeing like a state” is more similar to “seeing like a citizen”, and that citizens are more likely to ask for information like comparative school performance, for instance, since it is of direct importance to them, on open data platforms.

Big Data
Pundits: Stop Sounding Ignorant About Data
Challenging some of the predictable backlash against big data hype Andrew McAfee describes some of the errors people make when writing about data and gives a short list of things to consider when writing about it in his Harvard Business Review blog post. He says it is important to recognise that absolute certainty is not the goal (because It's impossible) and people are not inherently better at making decisions, predictions, judgments, and diagnoses. He also notes that quantification is useful in every field of inquiry and big data's advocates don't think everything can (and should) be determined by computers.