Ready or Not, Start a Data Glossary

Counting students at an institution sounds simple—until you actually try to do it. That’s what I thought when I started my Institutional Research (IR) career. I quickly learned just how many ways a “student” could be defined and counted. For instance, do dual enrollment students count as “students”? What about students taking only non-credit courses? How should we count students in exchange programs who begin a medical degree at one institution but finish elsewhere? And how do we account for them without making it appear they “dropped out” rather than completed their degrees?

As IR work expands, it can seem unnecessary to sit down and define terms as basic as “student” or “full-time employee.” It’s easy to work in silos and miss the fact that other campus departments may use the same data differently, producing conflicting conclusions. Taking the time to clarify those definitions now makes the work easier later by reducing confusion and inconsistency.

Yet, starting a data glossary can feel overwhelming. I tend to over-research projects, trying to gather every detail before I begin. Although that instinct comes from a good place, it often leaves me stuck on the details and waiting for perfection before moving forward. If that sounds familiar, these tips may help.

Find Case Studies

You probably already have examples of how inconsistent definitions have caused problems in your work. Maybe your office and the Registrar’s office produced conflicting reports, or you uncovered an error after taking over as the IPEDS keyholder. Perhaps leadership questioned why your data did not match their expectations. Gather those examples and show how a data glossary could help avoid similar issues in the future. These case studies can be a useful way to get others on board.

Don’t Reinvent the Wheel

Writing a definition can feel daunting. Where do you begin, and how do you know it is accurate? You do not have to start from scratch! As I build my institution’s data glossary, I am using the IPEDS Data Glossary as a starting point. Because our data needs to align with these definitions, beginning there helps support federal compliance.

Other colleges with strong data governance programs also offer public data glossaries that can provide ideas—or even a foundation—for your own terms. I especially like the Institutional Data Dictionary at the University of Arizona and George Washington University’s Business Glossary. These examples show how institutions can make definitions easy to find and use as a single source of truth for stakeholders.

Pace Yourself

With the IPEDS Glossary as a starting point, I spend 15 minutes each day reviewing key definitions and noting comments or suggestions. I also compare those definitions with how we describe our data in dashboards and reports. For example, if a state report uses a slightly different definition of student FTE than IPEDS, I add a separate glossary entry with a modifier such as “state report” so both versions are clearly documented.

This low-tech approach is a good place to start. Over time, you may want to explore online tools that support the process. Many include approval workflows and can connect to your information systems, linking each term to the data and code behind it.

Don’t Do It Alone

You may be an IR shop of one, or perhaps data governance is only part of your job. Even in a big department, you may feel bad asking for help because everyone is just as busy as you. However, don’t do it alone. Consider asking one or two co-workers or colleagues from other departments to review the definitions with you. For example, one of my co-workers knows human resources (HR) data better than I do and is working with our HR department to align our definitions. That collaboration is helping lay the foundation for our HR data glossary.

This teamwork can also help resolve any questions you may have in adapting IPEDS definitions to your institution. For example, the IPEDS Glossary’s first line for “Full-time staff (employees)” is “As defined by the institution.” This is a great opportunity to work with your HR department to fill in those details and get that definition down in a centralized location.

Ready, Go, Set!

I recently heard a keynote speaker stress the idea of “Ready, Go, Set!” (a twist on “Ready, Set, Go!”). Rather than waiting until everything is perfectly “set” before you begin, you start moving (“go”), learn from reality, and then adjust (“set”) as you proceed. In other words, instead of waiting for the perfect plan or a list of all definitions, you and your team should move forward with the first set of terms, gather real-world feedback, and then refine your approach as you go.

This approach does not suggest that we act recklessly or skip preparation. It recognizes that important insights often emerge only after we take those first few steps. Progress yields information that planning alone cannot provide. The way becomes clearer once you’re in motion.

After I finish reviewing the top 50 IPEDS definitions my IR office uses, we plan to bring those to our newly formed data governance council for approval. This allows for more feedback and increases collaboration across the university. If your institution does not have a similar council, a small working group can review the definitions internally. If this is a success, you can recommend more formal systems later on as you develop a data strategy for your institution.

Conclusion

Clear data definitions might not be glamorous, but they are essential for data integrity, interdepartmental consistency, and informed decision-making. Just documenting a few of your institution’s most utilized terms can create trust and start a practice of filling out the glossary over time, as a team.

Mary Kate Blake is Associate Chief Data Officer in University Data & Analytics at Montana State University in Bozeman, where she spends a lot of time thinking about data so others don’t have to. She is passionate about building clear data definitions and improving how student metrics are understood and used across organizations.