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.

Peace-Work is a volunteer cooperative of statisticians, data scientists, and other researchers applying analytics to issues in issue-driven advocacy and social justice. Although Peace-Work has been around for just a few years, formally organized in 2014, its founding members have a long history in individual statistical advocacy. An all-volunteer organization, Peace-Work projects are often in academic and policy research, with volunteers as likely to be found working with government economic data to write a position paper as they are working hand-in-hand with a social justice organization.

A distinctive feature of Peace-Work’s statistical research and advocacy is crowd-sourcing time. This is like crowd-sourcing dollars from supporters, except Peace-Work crowd-sources volunteer hours from the statistics and data science community. With no paid staff or other financial resources, larger projects are broken into small pieces to accommodate volunteers’ busy schedules. This business model is reflected in the play on words in the name: Peace work is what we do, and piece work is how we operate.

Peace-Work connects volunteers with organizations and data sets, often from governments and other official sources, to address issues they care about deeply. Projects can be as varied as the interests of the volunteers. With a practice focusing on advocacy, projects have included drivers of education success and failure, root cause analysis of homelessness, descriptive statistics of privilege and the impact of racial bias, environmental research, promoting and participating in hackathons, and policy analysis. The focus on crowd-sourcing hours means projects can be in any area of Data for Good. Some volunteers contact us to find an organization in their particular area of interest, and organizations reach out to find statistical support to further their good work.

Community involvement is strongly encouraged. Volunteers regularly participate in hackathons and ASA events such as DataFest, visit colleges to talk to student groups, and present at conferences. They rely on the software tools of their choosing, and open source tools are common. People are encouraged to share their applications, methods, and source code with the wider D4G community.

This also gives Peace-Work a character of intersectionality. For example, food deserts intersect with disproportional poverty among persons of color, which in turn relates to homelessness, which is connected to structural issues in education, which leads to the school-to-prison pipeline, and on and on.

Peace-work volunteers often collaborate with subject matter experts in different areas to combine their expertise on complex, multifaceted problems. Once, when presenting on human trafficking at a large international conference, a Peace-Work researcher saw several people come into the room who all knew each other but didn’t know the others planned to attend. Their area of interest was money laundering. They came to the talk because they discovered a strong connection to human trafficking—something even the presenter didn’t know previously.

While the problems and issues D4G researchers address are often studied in some degree of isolation, Peace-Work’s community-based model shows how a broad-based community approach helps develop holistic analytic solutions for data-driven policy to address intersectional issues.

Since the outbreak of COVID-19, much of the research at Peace-Work has focused on aspects of pandemics. Projects have included disproportionate impact of the pandemic on marginalized communities such as BIPOC [Black, indigenous, and people of color], the homeless, and prisoners. We have considered the impact of differences in practices reflected in the widely varying outcomes in different states. At the Conference on Statistical Practice next February, the annual ethics panel will focus on the ethical issues raised in both data and analysis. Statistical issues raised by these questions include biases in detection and reporting, best metrics for tracking the pandemic, geospatial analysis and visualizations, and pooled testing.

Ongoing areas of interest for Peace-Work are often the most severe challenges afflicting society—homelessness, addiction, the rise of internet hate speech, school shootings, and even genocide. One major initiative is human trafficking research, with studies on risk factors, geographic distribution, and data-driven policy advocacy.

Affiliated with the Global Association of Human Trafficking Scholars and partnering with local organizations fighting human trafficking organizations, Peace-Work researchers combined demographics from federal agencies with victim counts from the national hotline operated by Polaris. This was used to perform a meta-analysis, combining Polaris data from each state to identify factors predictive of a high level of human trafficking while accounting for variations between the states. The fixed effects found by the state-level meta-analysis were then applied to data on large cities to create a decision tree model to identify potential unidentified centers of human trafficking activity.

Peace-Work recently developed a model to identify legislative and government practices most effective at identifying perpetrators and victims. This was done by comparing actual per capita confirmed victim rates to the model. The practices of states consistently outperforming the model are compared with those most likely to be missing in poorer performing states. Identifying these best practices allows Peace-Work volunteers to partner with state-level advocacy groups to comment on proposed legislation so it is more effective.

Peace-Work’s unusual business model—crowd-sourcing hours instead of dollars to support a 100% volunteer organization—offers some surprising advantages. Remaining an independent research group—not tied to any university, organization, or government agency and broke by design—certainly has its challenges. However, it also means Peace-Work can’t be de-funded and shut down—certainly a plus in today’s highly polarized environment where even hard science can become a political football. We will never stop a project due to losing a grant because we don’t have any. And we aren’t afraid of becoming homeless because we already are. The organization perseveres not despite its unusual operational strategy, but because of it and the volunteers who make this important work a reality.