Dr. Keith Wiley is a Senior Research Associate at HAC with extensive experience analyzing housing, development, and lending data in rural communities.
Why is it important to collect data to support things that I already know about my community?
It’s true that people often understand their community’s needs better than outsiders. But the power of data lies in its ability to reinforce and validate those insights. Anecdotal claims, even when accurate, are easy to dismiss. Data lends credibility and weight to your argument.
For example, I might write in a grant proposal that my service area has both an aging population and an aging housing stock—factors that clearly justify funding for a home repair project to support aging in place efforts. However, without numbers, this claim lacks depth. If instead I write that there has been a 25% increase (over the last decade) in the share of service area residents aged 70 or older (500 people), in my community where over 70% of occupied dwellings were built before 1980 (compared to 50% nationally), I’ve added measurable support to the argument that housing rehabilitation is needed.
This same principle applies to research. It may be generally well known that a problem exists, but adding data helps reinforce that understanding, provides information about the scope of the issue, and puts it in context.
How can I know if the data I use is trustworthy?
There’s a lot of data online, and much of it appears conflicting or inconsistent. I completely agree that this makes the work of data users more difficult, but I think if one follows the general rule of relying only on well-respected sources, this problem can be alleviated.
Federal, state, and local governments provide the most trustworthy data. When their data sources align, I always prioritize them over private data. For example, the U.S. Census Bureau is a highly trustworthy resource that applies rigorous standards for its data collection and estimation. Its products—like the decennial census population counts and the American Community Survey (ACS)—are extremely reliable. The same can be said for other federal agencies such as the Department of Housing and Urban Development (HUD), the Consumer Financial Protection Bureau (CFPB), and the Federal Reserve Board. State data centers, agencies, and local governments also provide trustworthy information, such as property tax assessment data.
I am not saying this information is always 100% accurate, but official sources do make every effort to ensure accuracy.
Private data sources can also be acceptable—often they are based on information from government agencies. But, as with all data, it is essential to cite the source and provide any important context about its origin and how you are using it. Transparency is key: if there is an issue, it can be discovered and corrected.
Can secondary data—such as ACS housing estimates—help me with my own data collection?
Yes. While external data is often used to describe community needs or inform planning, it can also guide what you collect for your own projects.
For example, in a housing rehabilitation program, you might collect data on structure age, housing type, household type, and ownership status—using ACS categories as a model. This not only aligns with your project’s purpose but also allows you to compare your data to ACS estimates to measure progress.
There’s no need to reinvent the wheel when deciding what information to collect. It can always be modified if necessary, but the key point is to use reliable data sources to help guide those efforts.
What is one thing that you think people misunderstand about data and its importance?
People often assume data provides one clear answer—just one number that tells you everything. But that’s rarely the case. What makes data valuable is its nuance. Each data point is a piece of a larger puzzle, and the more pieces you have, the better your understanding.
Take a county with limited home lending activity. Many might assume the issue is simply a lack of down payment funds. But an analysis of Home Mortgage Disclosure Act (HMDA) data might show that poor credit history is the most common reason for mortgage denials. That doesn’t mean down payment assistance isn’t needed—it just means other issues, like credit access, also need attention.
In most cases—especially in the social sciences—nuance matters. This isn’t a lab experiment with beakers and controlled conditions.
I find data work to be tedious, not very interesting, and something I put off doing until the last minute. Is there another way to think about it that might motivate me?
I get that working with data can be a challenge—and one that many people prefer to avoid.
I recommend thinking of it more like detective work, with each piece of data representing a clue that will ultimately improve our understanding. It’s more like Indiana Jones searching for the Holy Grail than it is a mundane work task.
It really can be fun—if you shift how you view the work. It’s all about mindset.
