Session Schedule

New Approaches to Evaluating Workforce Training Programs Without Running Experiments
Dr. Jamie Fogel, Opportunity Insights

Many nonprofits and government agencies want to know whether their programs are actually working, but running a formal experiment to measure that is often too costly or impractical. This session presents new, early-stage methods from Opportunity Insights that link administrative data, rather than running experiments, to estimate the causal effect of workforce development programs on participant outcomes. While most programs cannot access federal data, Dr. Fogel will also discuss what they are learning about how these methods and insights might eventually be made usable for nonprofits or state agencies who have access to outcome data. 

AI for Data Professionals: Practical Lessons in Governance and Tool Use

Institutions that serve the public have high standards that we need our technology to meet. Our data practices must protect privacy while still improving decision-making. During this session, Bhukima Shah will share an approach to achieve potential benefits from AI without feeding actual sensitive records by providing metadata instead. This lets AI tools help organize, document, and assess data without ever seeing the sensitive content underneath. Discover how to transform unstructured state agency reports into objective accountability scorecards by leveraging AI to map narrative updates directly against statutory mandates. Using a real-world test case evaluating Connecticut's Office of the Cannabis Ombudsman, attendees will explore how to scale this civic tech framework across other government agencies to drive unprecedented public transparency. Participants will leave with strategies and practices they can use as they seek to integrate ethical AI use into their data toolbox.  

Data Tools: Create Impactful Data Visualizations

How can you turn your raw, messy data into eye-catching visualizations? What makes a good visualization? How do you build a map with zero technical skills? Which tools are right for your project?

In this session, join CTData staff who will answer all your data visualization questions. You'll learn how to work with raw data, design clear and effective visualizations, and tell your organization's story using simple, approachable tools like Excel, Datawrapper, AI tools such as Claude, and CTData’s own geographic resources, and more.

Brand new to creating data visualizations? Already familiar with the basics? No matter your skill level, you’ll gain practical data skills and learn visualization best practices. Come ready to share your ideas and ask questions!

Data Linkage in Practice: Connecting Systems to Understand Outcomes

Organizations serving the same individuals often collect overlapping information but see only their own piece of the picture, limiting understanding of how programs and institutions affect people over time across an entire geography. This session presents two examples of data linkage in practice. Julie Plano (Yale University) will discuss the Emerging Infections Program's approach to connecting surveillance systems, electronic health records, immunization registries, environmental monitoring, and national datasets such as the Census and BRFSS to build a fuller picture of health outcomes. Next, Cynthia Willner will share results from a CTData Collaborative project linking administrative records from Hartford nonprofits and state agencies to understand educational and employment outcomes for Hartford youth, identify gaps in youths’ access to supportive services, and understand the outcomes of young people served by Hartford nonprofits, guided by the learning priorities of organizations serving Hartford's youth. Attendees will leave with a deeper understanding of the possibilities of data linking to inform research and evaluation questions. 

Facilitation Skills for Creating a Strong Data Culture

Data professionals often face blank stares when explaining findings, or watch staff enter data incorrectly even after dropdowns and other safeguards were built to prevent it. Ensuring your work is understood takes more than good analysis: It takes facilitation skills, a tool that's often overlooked but essential for any data professional. In this session,Sarah Eisele-Dryrli and Jazmin Fermin introduce key facilitation techniques to encourage more effective conversations with staff about data. Attendees will have the opportunity to practice these skills live and will leave equipped with new methods to enhance staff confidence and engagement with data, leading to higher-quality data for better decision-making. 

Census 2030: The Time to Prepare is Now

While 2030 seems far away, one of the foundational aspects of the decennial census is in the planning phase right now (Local Update of Census Addresses, or LUCA) and will begin in March, 2027. With multiple, untested changes to the census form, along with an unprecedented level of mistrust in providing information to government agencies, this may be the most challenging census of our lifetimes. However, this constitutionally-required activity serves as the denominator for countless datasets that we all use, in addition to being used in the formulas for up to $12 billion in funding to the state of Connecticut. Panelists will discuss how they are preparing for Census 2030 and what participants can do in their organizations and communties starting today.   

Last year, we introduced Birds of a Feathers lunch tables and in our post-conference survey, you let us know you’d love to see it offered again! We also heard that you wanted a say in the tables offered, so this year, we’re inviting you to suggest topics! 

Birds of a Feather tables are opportunities for you to have lunch with others who share the same curiosity, challenge, or passion – maybe you’re the only data person in your org, you want to connect with others around a shared identity, or you love tending to your garden. It doesn’t have to be about data, it’s a chance to connect with others on whatever speaks to you.  

Learn more and submit your ideas here! 

Early Childhood Data in Connecticut: What We're Learning and Where to Look

Connecticut has several ongoing efforts to collect and use data on early childhood, and this session introduces three of them. Attendees will learn about three initiatives that increase access to information and data on early childhood. First, Suzannah Holsenbeck, Kimberly Montero, and Rep. Robin Comey will share an initiative led by Read to Grow to compile and utilize early childhood and early literacy data, including maps and tools that will be useful to a range of stakeholders. Cynthia Willner will highlight how the Connecticut RAPID Survey raises the voices of parents of young children across the state on policy issues including childcare affordability, financial security, housing, and healthcare. Finally, attendees will learn from Allison Logan about CT 359's State of the Babies report, which compiles public data on the well-being and needs of Connecticut’s youngest children and their families and provides recommended policy solutions. State Rep. Robin Comey will serve as a discussant, identifying key themes across the presentations and discussing implications for early childhood policy and practice. Attendees will leave with key findings, data resources, and opportunities to inform their own planning and advocacy for young children and families. 

Filling the Gaps Left by Federal Data Instability on Food Insecurity

When the federal government stopped collecting the USDA Household Food Security Survey in 2025, organizations like Connecticut FoodShare lost a key source of information for targeting scarce resources. This session tells two connected stories of how Connecticut organizations responded to that gap. Melvette Hill, Yukiyo Iida, and Megan Baker with The Commission on Women, Children, Seniors, Equity and Opportunity will share how it formed a food data workgroup and adapted its approach to culturally inclusive data practices in response to the loss of federal measures and pre-existing gaps in data methodology. Tyler Kleykamp of Connecticut Foodshare will describe how it converted its own administrative data into a usable dataset to guide service locations, grants, and partnerships. Attendees will leave with a practical framework for spotting data gaps early, building coalitions around data preservation, and sustaining community-centered data efforts without large infrastructure.

Unique Challenges of Community-Based Organizations in AI Policy and Implementation

AI is touted as a solution to many data woes including governance, management, quality, and security. Whether AI is truly strengthening or disrupting the data ecosystem may not yet be clear, but its impact is undeniable. However, solutions are often designed and optimized for profit. Community-based organizations that want to take advantage of this technology make different considerations and face different constraints when developing their AI policies. Those challenges may include reputational risk, privacy and security of data from marginalization and vulnerable communities, technology expertise, budgeting for technology, monitoring and enforcement capacity, etc. Mica Smith (COMPASS), CTData staff, and attendees will pose their own challenges and lessons with each other. This session will give space to discuss those challenges and brainstorm solutions within the nonprofit and public-sector community. 

Strengthening Prevention Data: The SEOW Portal and Mixed-Methods Evaluation

Strong prevention work depends on data that reflects not just numbers, but the priorities and lived experience of the communities it serves. This session shares two examples of prevention data projects grounded in community decision-making.  First, Jennifer Sussman, Jason Cheung, and Jill Walsh discuss the CT SEOW Prevention Data Portal, developed in 2018 and re-designed in 2026, through a partnership between the Statewide Epidemiological Outcomes Workgroup, UConn Health, CTData, and the Department of Mental Health and Addiction Services. This session explores the Portal as both a resource and a process, covering needs assessment, data prioritization, governance, the re-design process, and a brief demonstration of its features. Next, Kaitlyn Maurais will present the Middletown Prevention Council's mixed-methods evaluation approach, grounded in the Strategic Prevention Framework. By integrating quantitative data, including youth surveys, with qualitative stakeholder feedback, the coalition informs data-driven prevention planning, fosters cross-sector collaboration, and strengthens community ownership. Attendees will leave with practical strategies for engaging stakeholders and translating evaluation findings into sustainable, community-driven action.

Housing Data in Action: Tools for Advocacy and Preservation

Connecticut has a wealth of housing data and tools that can support planning, advocacy, and preservation work. This session covers three of them: Alysha Gardner (Parntership for Strong Communities) will share a hands-on training on using American Community Survey data at the state legislative district level for advocacy, including targeted emails, factsheets, testimony, the Partnership for Strong Communities' Housing Snapshot Maps; Kelly McElwain and Cate Asp (Public and Affordable Housing Research Corporation) discuss the National Housing Preservation Database, which helps track affordable housing at risk of losing its affordability; and a CTData tool for identifying property owners and relevant contacts. Presenters will cover how to match geographic data with the right decision-makers, from federal to local levels, and how to choose the best data to use for your advocacy and decision-making. Attendees will leave with concrete tools and data sources for their own housing work.

Census 2030: Preparing Amidst Uncertainty

While 2030 may seem far away, Connecticut is already preparing for what could be an even more challenging decennial census than the one in 2020. Rule changes, reduced funding, and declining public trust all threaten the accurate count of population and housing units that the state needs to guide policy, planning, and programs. Preparation is already underway on three fronts: engaging community leaders to improve housing counts ahead of the 2027 Local Update of Census Addresses (LUCA), starting discussions with organizations that work with groups most likely to be undercounted — including immigrant families, children under five, and renters — and working toward launching a Complete Count Committee. In this session, Michelle Riordan-Nold and Sarah Eisele-Dyrli will walk through these efforts, and attendees will leave with concrete action items they can take to support an accurate count of Connecticut. 

From Five CT’s to Six: A New Framework for Understanding Connecticut’s Towns

The Five Connecticuts framework captures the wide differences among Connecticut towns by grouping them into five categories (Urban Core, Urban Periphery, Wealthy, Suburban, Rural) using indicators such as a town’s income, education, population density, and more. Building on this work, CTData developed our own methodology that groups towns into six categories based on socioeconomic indicators, setting aside population density and urbanicity. Jason DeBari-Cheung and Cynthia Willner (CTData) will discuss how our new groupings give researchers and data users a simple way to look at towns that have similar characteristics, rather than relying on county, planning region, or state level data that can mask how much towns differ within those boundaries. Attendees will learn about the methodology behind the new categories and how to apply the six Connecticuts to their own needs assessments, planning, and research. 

Who Benefits? Teacher Pension Financing and School Funding Equity

The State of Connecticut pays the full employer contribution for teacher pensions, even though costs are set at the local district level and vary widely across towns, accounting for roughly a quarter of all education spending each year. In two earlier reports, Equable Institute used Governmental Accounting Standards Board data and state enrollment figures to convert these costs into per-pupil spending, producing a Per Pupil Pension Subsidy metric that shows inequities in state-level resource allocation across districts. This session introduces a new report, releasing this fall, that examines district-level pension costs on a per-teacher basis and compares them to teacher attrition rates. Anthony Randazzo (Equable Institute) will describe how Connecticut's pension financing structure reinforces inequities across district lines, particularly amid ongoing teacher shortages. The session will close with an overview of how keeping these costs outside the state's school funding formula affects overall school funding. 

Engaging with the Equitable AI Adoption Framework
Sarah Di Troia, Managing Director, OutcomesAI & Chief Innovation Officer at Project Evident

As AI becomes more common, and as the stakes get higher, data professionals are seeking to ensure we maintain our commitment to ethics and accuracy as we consider adoption. Developed in partnership with 34 nonprofit practitioners, school district leaders, and subject matter experts, the Equitable AI Adoption Framework provides actionable guidance across the full breadth of AI—from data readiness and stakeholder engagement to safe and fair practice and technology implementation. The framework addresses top-of-mind questions about AI implementation directly, including where to start, how organizations can use AI responsibly and safely, and how to best align AI with organizations’ mission and outcomes, grounding AI adoption in their values, program logic, and evidence-based practice. Di Troia will discuss the framework and how you can start applying this framework in your organization.