This column is written for anyone engaged in or interested in statistical consulting. It includes articles ranging from what starting a consulting business would entail to what could be taught in a consulting course. If you have ideas for articles, contact the ASA’s Section on Statistical Consulting’s publications officer, Mary Kwasny.

With a PhD in statistical astrophysics, David Corliss works in analytics architecture at Ford Motor Company while continuing astrophysics research on the side. He serves on the steering committee for the Conference on Statistical Practice and president-elect of the Detroit Chapter. He is the founder of Peace-Work, a volunteer cooperative of statisticians and data scientists providing analytic support for charitable groups and applying statistical methods to issue-driven advocacy in poverty, education, and social justice.

As small businesses providing a service, consultants often are deeply embedded in the communities they serve—whether defined by a local geographic area, industry, or other characteristic. One way in which a growing number of statistical consultants serve their communities today is through pro bono work for good causes and organizations.

In many professions, pro bono work is an ordinary—even expected—part of the job. While physicians and attorneys are well known for their professional commitment to volunteer work, many other professions have strong connections to serving their communities in this way. In recent years, volunteer work in statistics and data science have become much more common. This “Data for Good” movement has been embraced by individuals, companies, and professional associations.

There are many organizations working in Data for Good, including the ASA volunteer outreach group Statistics without Borders. Working with a volunteer organization offers many advantages: connecting people to volunteer projects and groups needing assistance and providing mentoring, software support, project planning, and opportunities to work on larger projects. However, working as a consultant offers substantial advantages.

Independent consultants have complete creative control of methodology and analytic practices, greater flexibility in which analytic tools are used, and the ability to work with a circle of friends in familiar surroundings. Individual volunteer projects can be less formal, without a definite timeline, and often can be put on the shelf for a while and restarted when more time is available.

After many years of volunteering as an individual consultant, I founded Peace-Work to crowd-source volunteer hours within the statistics community to support projects in issue-driven advocacy. Many of Peace-Work’s analyses focus on the ugly side of Data for Good—current projects include using mixed models to analyze human trafficking by state and mining hate speech in social media to predict violence. However, much of the work is a direct descendant of a consultant project on a more cheerful subject: helping a local Habitat for Humanity chapter find more volunteers and donors.

Individual consultants make up the largest part of the Data for Good movement. Often, a personal connection to a group needing statistical help is the most important factor in matching people and projects. The Habitat for Humanity project began in 2005 while I was consulting at Ford Motor Company, which has a long history of community involvement. One of the marketing managers and I volunteered for the local Habitat for Humanity chapter. After volunteering on a number of construction projects and getting to know the chapter’s board of directors, a conversation began with my Ford colleague about how statistical analysis could help the organization.

Usually with statistical volunteering projects, an organization is doing really good things and has data, but not the analytic resources required to leverage it. The organization knows its mission, people, and data, but is unfamiliar with statistical methodology. They usually lack statistical software, so the consultant needs to bring this to the table. Also, because these charitable organizations need to focus their few hard dollars on other needs, paying to acquire additional data sources usually is not an option. This requires the consultant to blend the organization’s data with publicly available sources.

The process with Habitat for Humanity was conducted in much the same way as for ordinary for-pay consulting projects. Discussion with the Habitat for Humanity team identified the points they were concerned about the most. Like many organizations, Western Wayne County Habitat for Humanity needed better ways to find more volunteers and donors. As a result, this case study may be especially helpful to other organizations where statisticians can provide analytics support.

After showing the Habitat for Humanity team how to anonymize a contact list using mocked-up data, the team stripped PII (personally identifiable information) from their volunteer and donor lists and provided a copy. As the data never left the organization, this step may not have been strictly necessary, but it is strongly recommended as a best practice. Nothing can go wrong with the data you never receive. Summary counts of Habitat for Humanity construction volunteers and donors at the municipality level were combined with Census Bureau data.

Geographic analysis quickly revealed most of the support for the chapter came from a small portion of their geographic area, with many cities largely unreached. An affluence scale was developed to combine demographics data into a single score that best correlated with the per capita number of donors and volunteers. A cluster analysis on this score binned the communities into three groups based on level of charitable need in the community and their ability to meet it. Looking at the entire area, instead of one small part where most of the board members lived, produced a list of communities to target where the organization had little presence.

In addition to finding places to recruit more donors and volunteers, the analysis revealed the houses were very poor, with weak infrastructure and especially poor schools, where the local Habitat for Humanity group was building. This produced a recommendation to build houses in a different, neighboring community, where the families would receive more support and the children would attend better schools—doing more good for the families in new homes, even if the land prices there were somewhat higher.

This project tells the story of how most statistical volunteering by consultants and other unaffiliated individuals happens: A person is already helping an organization in a nonstatistical way—working at the public library; walking dogs at the local animal shelter; or volunteering at a school, house of worship, or community center. Individual statistical volunteering happens when people with analytic skills know how those skills can be used to help the organizations and causes they already support.

Participation in the growing Data for Good movement provides opportunities to use our statistical skills to support the organizations, projects, and causes we care about the most. As the movement continues to grow, we can envision a day when pro bono activity becomes normative—an ordinary part of a statistical career. Participation from everyone is encouraged, from students to the most experienced and entry-level analytic workers to top executives. Sharing your skills for the greater good can be part of everyone’s career journey. Where is your place?