The Intersect: A CHFA Housing Blog

Housing and Data: An Interview with Fionnuala Darby-Hudgens

By: Marcus Smith 

“Housing and…” is CHFA’s interview series that explores how housing connects with and influences various sectors that shape our communities. From health to education to economic opportunity and beyond, each installment features insights from thought leaders in their respective fields, helping to highlight the critical role housing plays in shaping broader societal outcomes. In this installment, we talk with Fionnuala “Finn” Darby-Hudgens, Director of Data Strategic Planning at CTData Collaborative. 

 

Marcus Smith: I’d love to hear from you what your journey was like getting to where you are now. How did you find yourself interested in data and in housing at the same time?

Finn Darby-Hudgens: I went to college thinking I was going to teach second grade. I was like, “Yes, I’m going to come in Michelle Pfeiffer-style and solve public education.” By junior year, I realized there was nothing I could do about education unless I worked in housing.

I had the opportunity to do research my senior year at Trinity. In 2010 I read and quantified all 169 municipal zoning ordinances for exclusionary variables. At the time, I had a thesis – “your two-acre minimum lot sizes are the problem” – and thought, “Here you go, we fixed it.” But it wasn’t accessible; it was really technical.

At the Fair Housing Center we used data all the time. One number that still cracks me up: my colleague Dave Lavery and I were asked, “How much do you think rental assistance is going to cost during a global pandemic?” We calculated it and came up with $440 million – and somehow that’s the number folks stuck with. That’s what they allocated. It was probably never enough, but it was close. It helped to meet the immediate need.

 

MS: Housing and homelessness are two parts of the industry where data gets messy. There’s lag: we make future-facing projections with data that can be a year old – or older. What’s your advice for making data-informed decisions in the face of uncertainty and lag?

FDH: An early social science teacher told me: when your data is a mess, start writing simple sentences about it. I still use that strategy. Let’s say you’re a homeless service provider and your data is messy. Start with, “We served X people in January.” Then add complexity: “We served X people in January, and Y were men,” or “Z were over 65.” Scaffold your learning so it makes sense before moving on to more sophisticated analysis. From there, the next step would be to pair messy data with expertise—yours or someone else’s. Staying with the example, if homelessness data is telling you something that doesn’t feel accurate, investigate why.

For lagging data – rather than trying to identify discrete phenomena, focus on trends and use longer-term analysis. And be transparent about limitations. For example, HMDA data is notoriously laggy; it might not tell you about this minute’s market, but it can tell you about end users over time. Say that clearly: “A limitation of these data is that they’re five years old.”

And don’t guess. It’s more powerful to say, “We don’t have that information,” or “That’s unclear,” than to make an incorrect assumption.

 

MS: I'm reminded of a recent conversation we had where we were contemplating where to find housing quality data. And we ultimately agreed that we may not have tight housing-quality data here in, say, Hartford, but we can point to studies elsewhere – say, Philadelphia – and say something like, "we know it’s not apples-to-apples, but here’s what others found.”

FDH: We’ve done a good job of that in some cases, and had challenges in others. In your example, we must be clear about what’s apples-to-apples (they’re both Eastern seaboard cities, they have industrial histories) and what makes Connecticut unique and harder to compare (we have no regional or county governance).

 

MS: I’ve heard you mention CTData is working on a dozen or so housing-related initiatives, and a priority is helping organizations build internal capacity: “So you’re collecting data, now what?” What data gaps do you see?

FDH: No gaps – everyone has too much data. My first response is: slow down and analyze what you’re collecting. Ask yourself, “How do you use it?” I suspect we collect a lot we don’t use. That’s a burden to staff, and a burden on the people you collect from – data collection is extractive. Make sure you’re using everything you collect. If not, it’s okay – that just means you probably don’t need to keep collecting it.

Nonprofits – and funders, too – are good at legacy work: “We do this because we’ve always done this.” We’re also good at building programs, and we don’t often look at administrative data behind them: staff hours, setup time, ROI, outcomes. I saw a “post-session report” [following the last legislative session] that said this nonprofit followed 64 bills. That’s… a lot. Was it helpful? What did you do to follow them – testimony, workshops, meetings, convenings? What happened as a result? We’re not amazing at looking at our own administrative data.

 

MS: How can data be used to close gaps between opposing viewpoints? Maybe not change people’s minds head-on, per se, but rather build public will.

FDH: In prepping for this conversation, I reviewed testimony from H.B. 5002. So many people wrote things like, “We need 93,000 affordable units,” “There are 149,000 extremely low-income renters,” “We need 51,000 affordable and available rental homes.” Those numbers don’t mean much to most people. But if you say, “One in four families can’t afford their current housing,” that lands. We have to reduce jargon – terms like “AMI” and “subsidy” – and be clear.

One piece of testimony that stood out was from the CEO of TVCCA in Norwich. It started: “On one night, 79 people were homeless in Norwich.” That’s a number I can understand—and a number that feels actionable. Clear, accessible, high-quality data storytelling matters. If you tell people we need 93,000 units of affordable housing, you’ll immediately get, “That’s impossible.” So what else can we say? What can our data stories say to move the needle?

 

MS: Who’s doing this well – using data to inform program development, storytelling, or strategy?

FDH: Something we practice at CTData – and this should be the first step in every really good data effort – is that it starts with a question. Not “it would be nice to know,” but something we truly need to answer with data.

I’ve been deeply involved in eviction data collection in Connecticut – both at CTData and earlier at the Fair Housing Center. The question has been: “Who is most impacted by eviction in Connecticut?” When we analyzed filings and mapped them, we could answer clearly. That analysis was used in the legislature to help pass Right to Counsel laws.

 

MS: If you had a magic wand to improve the way Connecticut collects or uses housing data, what one thing would you fix or create?

FDH: My “magic wand” would be to support municipal data infrastructure. Imagine integrating rental unit conditions, certificates of occupancy, landlord registrations – statewide. We could tell a lot about the affordable housing landscape. But many municipalities are stretched thinner than small nonprofits. If we invested in municipal infrastructure – especially administrative services like COs or  permits – we could analyze things like zoning appeals. Right now, the data often lives in meeting minutes or, if you're lucky, in PDFs on town websites. You can’t analyze data across 169 places when it’s not digitized. You could enlist recent college grads to do it. Like a modern New Deal – job training plus digitizing records, helping municipalities. That would be powerful.

 

Sign up to receive Intersect blog posts right in your Inbox!

****

Marcus Smith is the Director of Research, Marketing, and Outreach at the Connecticut Housing Finance Authority (CHFA), where he helps drive statewide impact through data-informed strategies and compelling storytelling. With over two decades in mission-driven roles, Marcus has dedicated his career to expanding access to affordable, stable housing. His prior experience includes leadership in healthy housing initiatives at Connecticut Children's Medical Center and community development with Sheldon Oak Central. He holds a BA from the University of New Hampshire and an MBA from the University of Connecticut.