Skip to main content
October 13, 2023 | News

How Can Improved Data and Evidence Improve Lives in Connecticut?

Tobin Partners with State of Connecticut to Support its Nationally Recognized Data Strategy

Data Strategy Partnership

Nat McLaughlin, Tobin Senior Program Manager, with Scott Gaul and Rachel Leventhal-Weiner

The US Treasury recently recognized Connecticut for its efforts to study the impact of COVID relief funds in the state. This is just one recent component of a nation-leading research and data strategy that will deliver near-term benefits to families, communities, and the state. More recently, the Governor’s Office and Tobin Center announced a joint commitment to launch evidence-based policy initiatives, and identify ways to leverage universities’ substantial research and analytic capacities.

To explore how local policymakers are incorporating rigorous research into their decision making—and how the Tobin Center is supporting these efforts—we recently sat down for a conversation with Scott Gaul, Connecticut’s Chief Data Officer, and Rachel Leventhal-Weiner, the Director of Evaluation and Impact at the Office of Policy and Management (OPM).

Both Gaul and Leventhal-Weiner work closely with state agencies, state contractors, applied researchers, and evaluators to develop program evaluation plans, coordinate access to state data resources, and synthesize evidence for reporting and communications. “We’re making sure that data are doing work to improve the way the state operates,” said Gaul, “and ultimately benefit the millions of people who rely on Connecticut’s public services.”

The Tobin Center and State of Connecticut partnership grew out of the Connecticut Governor’s Fellowship Program, a public-private partnership that brings together diverse and talented early-career professionals with senior state leaders to solve Connecticut’s most pressing policy problems. The cornerstone of our collaboration involves Tobin Center experts helping to identify priority projects that could benefit the state through enhanced data and evaluation capabilities. The aim is to enhance the state’s ability to conduct advanced analytics on programs that are meaningful to the state, and allows the state to use evidence to develop and implement more effective policies.

Our commitment to collaborate with the state and support its efforts to build an outcomes-oriented data infrastructure runs deep: it represents a central strategy of Yale to support Connecticut data and research needs and to activate higher education in support of state research and policy goals.

- Dave Wilkinson, Tobin Center Executive Director

During the conversation, Gaul pointed to the expanded set of programs and resources from the 2021 American Rescue Plan Act, which provides many great opportunities to develop and build evidence around the state’s investments:

Our goal is to make sure that we’re investing these resources in the right places, and that we're using evidence as much as possible to allocate them. If we aren't sure about the evidence base for something, we need to make sure we're taking the opportunity over the next few years to really learn about what’s happening with these investments: Whether they benefited residents of Connecticut, how much, and what sort of lessons we can apply to our future work.

Connecticut’s Nationally Recognized Data Strategy

Connecticut’s statewide data strategy includes three major components. Through Gaul and Lenenthal-Weiner’s work, they aim to increase data access for researchers and the public so that agencies and services they provide are transparent and accountable to citizens. Along the way—through resources, training, and partnerships like the one with Tobin—they’re also working to ensure that state agencies have the capacity to collect, manage, and use data. Finally, and most importantly, Gaul noted, “our goal is to inform both policy and practice in a way that actually improves lives.”

As Director of Evaluation and Impact, Leventhal-Weiner’s role is to help agencies ask questions about their work, generate actionable information, and help them to understand what's going on with their programs and services.

The overarching goal is to make data more accessible and useful. That process looks different depending on the agency. Some agencies have a really robust data collection system and robust data use within their agency, but many do not. We’re trying to meet each agency where they are and see what’s most useful. We’re asking questions like: What does it look like when we have greater access to services? What happens when we're scaling services in places that used not to have them? What is the best way to structure service provision in the first place?

Both Gaul and Leventhal-Weiner outlined a number of examples of how state agencies generate evidence that will help the state make meaningful programmatic and budgetary decisions. For example, the State Department of Education’s (SDE) Learner Engagement and Attendance Program (LEAP) is a home-visiting program that targets districts with high levels of chronic absenteeism, and works to bring kids back to school. Through an evaluation of both the model and training, researchers talked to providers, families, and children, and found that attendance rates increased from 4 to 16 percentage points following LEAP home visits.

Based on the success of the program, it's become a national model, and now they're extending it using American Rescue Plan funds into the next year. “We’re excited to see how data and evidence are being used to improve attendance and bring students back to class. Through the SDE’s evaluation, they were able to better address the extreme learning loss in places that weren't equipped to support students through the pandemic,” Leventhal-Weiner said.

Making Data Accessible and Available

With how much newly available information is becoming available throughout the state government ecosystem, Gaul and Leventhal-Weiner emphasized how critical it is to ensure that employees and legislators actually know how to think about and use this data. The same is true for advocates or just regular citizens and residents, said Leventhal-Weiner:

There are some really basic questions that could be answered by state administrative data that aren't answered because maybe the data aren’t in a good format, or it's never really been queried, or the data quality is not great—or it is great, but agencies are nervous to share it openly because they're worried about criticism or scrutiny. So there are a host of barriers to overcome. Through the work my team is doing, and the partnership with Tobin, we hope that more data will be used by a broad range of audiences.

The Tobin-Connecticut Partnership

From the state’s perspective, one of their recent priorities is to build better pathways for research partnerships. While Gaul and Leventhal-Weiner’s offices are making strides to build internal capacity, they noted that they don’t necessarily have all the domain expertise to be able to do rigorous impact evaluation. “Building up a way in which we can have more effective partnerships with researchers is important. For example we have issues delivering data for research purposes in a timely fashion within the legal constraints for how that data can be used. That is one of the main issues on the state side,” Gaul said, “We also often need the perspective that an outside entity brings when evaluating a program. The partnership with the Tobin Center will help us figure out some really important questions from an independent perspective.”

While the partnership is still in the early stages, the Tobin Center and the State are moving forward to test, learn, and strengthen the state’s data infrastructure, with a particular focus on childcare, medicaid, and climate infrastructure.

In one such project, announced through the Governor’s Office, the State of Connecticut, New Haven Public Schools, and the Tobin Center will produce a first-of-kind study on the the impact of childcare on young children and on parental employment outcomes. This collaboration will produce a first-of-kind intergenerational study that will provide causal evidence on how access to free public pre-kindergarten (pre-K) affects near-term and long-term outcomes for kids and their parents.

The research will link the results of 20 years of school choice assignment lotteries data with state data on educational and labor market outcomes for parents and kids. This study will provide comprehensive evidence on the costs and benefits of public pre-K programming—something that is both a national policy and research priority and of immediate importance to a state focused on improving its childcare system. “They're in the middle of evaluating child care programs through the Blue Ribbon Commission right now. So it's a good moment to focus on childcare services,” Gaul said.

A Model for Successful Collaboration Between State Governments and Higher Education

In terms of what a successful long-term partnership would look like, both Gaul and Leventhal-Weiner voiced hope that the work that's underway in the evaluation and impact space takes a greater hold in the conversation, and becomes a part of the way the state makes its budget and policy.

Right now we're working towards that and no one's really fighting the idea that it isn't great to think about what works and invest in something that's going to have good outcomes, but it's not the way that all state business has been conducted before,” Gaul said, “To me, success in a few years really means having the idea of impact or the need for evaluation or even just the questions around data are just a part of our regular conversation in budget and policy discussions.

This partnership fits into the Tobin Center and the State’s collective vision of identifying ways that agencies can leverage universities’ substantial research and analytic capacities. With state and local governments increasing their demand for data and evidence, the Tobin Center is excited to collaborate on projects that help “level-up” the state’s data and analytics capacity. Through this work, Connecticut can create a model that other states will value, will find helpfully instructional, and can follow.