Many organizations are committed to understanding their impact: collecting data on their services and tracking their participants’ outcomes. But it is not easy to collect high-quality data, analyze it efficiently and accurately, and then use that information to refine and improve one’s services. MDRC has extensive experience helping organizations use their data to understand how their programs and services work—and then translate what they learn into meaningful action. This spotlight illustrates an example of how MDRC partnered with Actionable Intelligence for Social Policy, Chapin Hall, and the Coleridge Initiative to work with state and local Temporary Assistance for Needy Families (TANF) agencies to make better use of their existing data. 

The Opportunity

TANF programs, like other human service agencies, collect data on participating families to meet federal reporting requirements. But the full potential of these data—their potential to help organizations improve program services and participant outcomes—is often untapped. By analyzing the data, TANF (and other human service agencies) can gain new insights into participants’ experiences and needs. These insights can inform decisions about daily program operations, policies, and budgets that might lead to better employment outcomes and self-sufficiency for participants.

MDRC’s Solution: Learning by Doing

MDRC’s Center for Data Insights and its partner agencies launched the TANF Data Collaborative (TDC) which offered eight TANF state agencies a variety of forms of training and technical assistance using a “learning-by-doing” approach. This approach allowed the agencies to build their data analytics knowledge and skills by completing a project using existing TANF data.

To build its data analytics capabilities, each agency was encouraged to create a cross-disciplinary team of staff members from different units, drawn from all levels of the organizational hierarchy. The existence of such a team was intended to foster communication and collaboration across units through regular meetings, the creation of a shared vocabulary, and opportunities to draw on each staff member’s expertise. Participating pilot teams identified research questions that were feasible to undertake and could be completed successfully during the 30-month pilot period. By bringing data and program staff members together, teams were able to produce more accurate analyses and interpretations that could be translated into action.

The TDC team offered an Applied Data Analytics course to lay the foundation for each pilot team to design and conduct their own analytics project and engage in hands-on practice with TANF and employment data. The TDC team engaged all types of learners, balancing consistency across all participating agencies with customization that differentiated learning experiences for staff members who had a wide range of skills and competencies. Pilot teams worked with a dedicated coach, who guided teams through the five phases of their analytics project: laying groundwork, accessing data, preparing data, analyzing data, and communicating findings. Additionally, pilot teams participated in a wide array of activities to promote learning, including 14 webinars that combined instructional content with small-group discussion and problem-solving exercises, as well as formal and informal peer-to peer interactions such as cross-pilot conferences and “hangouts”—opportunities for pilot teams to gather online, informally, to share their experiences. The TDC team also organized group discussions by staff role, added new learning modules, and invited other experts to offer guidance and instruction.

The TDC team designated milestones and associated deliverables for the teams, so they could keep momentum over the course of the 30-month engagement and have moments for reflection and documentation. This approach to overall project planning also enabled the TDC team to align the learning and peer events, resources, and coaching topics it offered with each milestone, so the support agencies received was always relevant and timely.

It was important throughout the partnership for the TDC team to remain flexible and respond to agencies’ changing needs. That flexibility was needed especially during the launch of TDC, which coincided with the beginning of the COVID-19 pandemic. The TDC team was well prepared to adjust in response to new information, shifting leadership priorities, and staff turnover.

Implications for the Field

This pilot initiative offers a real-world model for federal agencies, policymakers, foundations, and other funders interested in investing in efficient, long-lasting approaches to improve data analytics capabilities among state agencies. MDRC’s learning-by-doing approach is broadly relevant to a wide range of public-sector agencies seeking to use data collected in the normal course of administering programs to improve their practices, their services, and the impact they make in participants’ lives. MDRC knows from this work that training and technical assistance—carefully designed and tailored, and supplied through supportive collaboration—can help partners use their data to answer the questions most important to them. MDRC’s focus on sustainability from the beginning of the engagement, communication and collaboration from its technical assistance teams, and a consistent approach that allows for flexibility and different learning styles sets up partners for long-term success. In the TANF Data Collaborative, after the pilot period ended, some of the agencies continued to perform analyses independently, applying what they learned to other data sources, such as data arising from the Workforce Investment Opportunity Act and statewide longitudinal data systems. Other organizations working with MDRC can expect to benefit from the learning-by-doing approach similarly.