This column is written for those interested in learning about the world of Data for Good, where statistical analysis is dedicated to good causes that benefit our lives, our communities, and our world. If you would like to know more or have ideas for articles, contact David Corliss.

With a PhD in statistical astrophysics, David Corliss leads a data science team at Fiat Chrysler. He serves on the steering committee for the Conference on Statistical Practice and is the founder of Peace-Work, a volunteer cooperative of statisticians and data scientists providing analytic support for charitable groups and applying statistical methods in issue-driven advocacy.

Environmental dangers are often described in terms of tipping points—a maximum greenhouse gas level, the total human population our planet can support, or the increase in temperature triggering a large rise in sea levels. As much as we hear about tipping points for environmental disasters, the concept itself is benign: a mathematical threshold in a small value that, when crossed, leads to a large change in something else.

Tipping points can also be for good things, like the number of trees saved and planted for a sustainable oxygen supply or the number of statisticians participating in a hackathon to form a nexus of cross-pollinating creative ideas for a better world.

Examples of statistics at the tipping point are all around us, and all of us can participate in making a difference for good. Here are some examples.

Hackathon Focuses on Climate Change
The C40 Climate Leadership Group is a network of major cities around the world working together to fight climate change. C40 brings together mayors, city departments, business executives, and thought leaders around a mission to “collaborate effectively; share knowledge; and drive meaningful, measurable, and sustainable action on climate change.” Ninety-four cities with a combined population of more than 700 million people participate in the network. A particular focus of C40 is practical, substantive, and effective implementation of the 2016 Paris Agreement on climate change.

Last year, C40 partnered with Qlik and the city of Boston, a C40 member, to sponsor a hackathon. Boston supplied data energy use in buildings across the city, which accounts for almost three-fourths of greenhouse gas emissions in Boston. Teams participating in the hackathon developed innovative solutions to help understand, measure, and limit greenhouse gas sources. Many of these creative data science applications can be applied to other cities.

The C40 hackathon is a great example of an event using statistical science to fight climate change, showing many best practices needed to host a high-quality event that makes the most impact: a core group or local volunteers collaborated with a larger organization championing the cause they want to promote. A small core group of organizers developed careful plans and carried the project through. A trusted organization provided data that can be easily transferred to the laptops of hackathon participants. A centrally located site with parking available nearby facilitated participation, and social media was used to promote the event. Altogether, the C40 hackathon was both a great meeting of minds inspiring creative solutions and a roadmap for future events.

Protecting Endangered Species with Machine Learning
Another area in which Data for Good is making a difference in our natural world is animal conservation. Monitoring the movement of animals provides critical data for conservations of threatened and endangered species. However, the devices used for tracking can be expensive, difficult to implement, and short lived. They can also present a risk of interfering with animal behavior, possibly altering the very qualities researchers want to measure.

The conservation group WildTrack has developed a system to identify individuals from several species of animals from footprints. Photographs of the footprints are digitized, and a machine learning algorithm identifies the animal that made it with 90 percent accuracy.

These algorithms are being developed by more than a dozen university students around the world. To date, there are algorithms for 13 species.

JMP, from the SAS Institute, supports the project and provides its data visualization software. The result of this collaboration between an NGO, universities, and industry is an inexpensive, noninvasive, and effective method of counting and monitoring threatened and endangered animals.

Identifying Climate Change Vulnerabilities in the Third World
Whether due to deforestation for farmland, pollution, overpopulation, or other mechanisms, harmful practices in poor countries are often contributors to climate change. Helping to meet this need, the Global Facility for Disaster Reduction and Recovery (GFDRR) collaborated with Data for Good leader DataKind.

The GFDRR is a World Bank agency providing funding for projects addressing the risk of disasters around the world. A UK-based DataKind team of volunteer data scientists developed a machine learning algorithm to analyze satellite images, identifying buildings and their type and use.

This information can be used following a natural disaster to identify population centers, critical infrastructure, and surviving facilities to help aid agencies respond more quickly and effectively. This analysis makes the countries using them more resilient to natural disasters.

More projects are in the works, connecting volunteer statisticians and data scientists to address UN sustainable development goals.

Data for Good opportunities to help preserve and protect the world around us are as varied as the concerns that need to be addressed. When we participate in Data for Good, each and every one of us has the power to become a tipping point—a point of leverage to bring about beneficial change and create data-driven sustainable practices for better communities and a healthier world.

Get Involved

In Data for Good opportunities this month, check out the student awards.

Sponsored by the Statistical Computing and Statistical Graphics sections, two student awards will be presented at JSM in Philadelphia.

The John M. Chambers Statistical Software Award honors software written by, or in collaboration with, an undergraduate or graduate student. The Student Paper Competition recognizes a paper in the area of statistical computing and statistical graphics.

Neither are specific to Data for Good, but it’s a great opportunity to show off your work while making a positive difference with statistics. I hope the D4G community will be well represented.

The submission deadline for both awards is December 15.

Also of interest to Data for Good advocates and activists is the new ASA Leader HUB. It’s a central location for resources leaders in ASA chapters, sections, and interest groups can use. Of particular interest is the ASA Chapter Stimulus program, supporting activities that develop chapter interest and membership—and Data for Good projects are a great way to do it.