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.

2018 was a banner year for Data for Good, full of important projects and new opportunities. The D4G community grew larger, stronger, and more connected. It also started developing growing pains.

As the concept of Data for Good has gone mainstream, the number of projects has grown dramatically—but the number of people looking to become connected with a project has grown even faster! Also, many projects today are larger, with hundreds or even thousands of participants in some of the larger civic hackathons.

The good news is there are now legions of people wanting to use their analytic skills on projects that make a real difference in their community and our world. The challenge today is capturing this lightning in a bottle and directing it and our expertise to make the most impact.

One of our greatest challenges is creating the projects, themselves, and helping people plug into them. This need highlights a growing problem of leadership in Data for Good—not in direction, but in the skills needed to design, set up, and lead the projects. Developing well-trained analytic project managers is just as critical to making an impact in Data for Good as learning the statistical skills needed to do the analysis.

Hackathons are growing in popularity, and for good reason: They usually have the following characteristics of projects that pull in volunteers:

• Time-bound: people need to know when it will start and end
• Expectations: people need to know specifically what they will be asked to do
• Skill-led: people need to know their particular skill set is needed and will be useful
• Actionable: people need to know they can make a direct impact without a lot of extra training in advance

Hackathons can also be a great place to learn the leadership skills needed for D4G projects. One way to start is by networking with the people running the event. After participating in a few, ask if you can participate on the team for the next one. There are a number of good resources for running hackathons available online, but it’s best to combine them with hands-on experience.

As people gradually take on larger roles and projects, many leaders may struggle to overcome the jump from a small team personally led by the organization’s head to a larger, more corporate-sized organization. Direct, personal leadership only works up to about a dozen people, which limits the tasks that can be addressed. Leaders in these organizations can move forward by collaborating on a large project involving dozens or even hundreds of people and by taking ownership of just one piece of a large project—one data source, one type of modeling method, one part of a multi-faceted problem (e.g., a survey in support of a larger effort). In the process, developing leaders learn about corporate-sized projects and develop the skills needed to collaborate with other teams on larger initiatives.

By this point, I expect a number of readers are thinking, “But we’ve got a great project! Where are all these people who want to get involved?” This actually points to another area of project leadership: promoting your work. While having great ideas is a critical first step, developing an understanding of communication, advertising, networking, and recruiting will go a long way toward finding the people needed to make a project successful. Website design and development, working with the media, and recruiting sponsors are important skills leaders need.

Our statistics defines a place in the “For Good” space, but organizational, project management, and leadership skills are also needed to enable more projects and help the growing number of volunteers get involved in well-organized projects that make a difference. Please send me your thoughts about this and other concerns you think the Data for Good community should address.

I hope to be able to talk with many of you February 14–16 in New Orleans at the Conference on Statistical Practice, which will again feature a Data for Good networking dinner out and a panel discussion about ethics, including the role of Data for Good.