| Filename | Latest commit message | Latest commit date |
|---|---|---|
| README.md | ||
Building a Research Glossary to Keep Terms Consistent Across a Project
Every field has terms that mean slightly different things to different people. In health research, "adherence", "compliance" and "persistence" overlap but are not identical. In education, "engagement" can mean attendance, attention, emotional investment or all three. In machine learning, "validation" and "test" sets are swapped between papers more often than anyone would like. When a project draws on many sources and involves several people, these differences cause real problems: inconsistent coding, misinterpreted findings, confusing writing and analyses that compare things that are not comparable. A project glossary, built deliberately and kept linked to sources, prevents many of these problems. This article explains how to build one and how a source-linked assistant such as Yila can help trace how each term is used in the literature.
Why a glossary matters
Terminological inconsistency is rarely obvious. Two team members may use the same word for weeks before discovering they meant different things. A systematic review may pool studies that measured "quality of life" with instruments capturing quite different constructs. A paper may use a term in its introduction in one sense and in its results in another. Each of these undermines the reliability of the work.
A glossary makes definitions explicit and shared. It forces decisions about which meaning the project will use, documents alternatives and provides a reference for everyone involved, including future readers.
Identify the terms that need defining
Not every term needs a glossary entry. Focus on terms that are central to the research question and analysis, that are used inconsistently in the literature, that are technical and may be unfamiliar to some team members or readers, or that the project uses in a specific or restricted sense.
Start with the key concepts in your research question and the main outcomes and exposures. Add terms as they arise during reading, coding or discussion. A useful signal is any moment when someone asks "what exactly do we mean by this?"
Write useful entries
Each glossary entry should include:
- The term and any common synonyms or variants.
- The project definition: the meaning the project will use, written precisely.
- Alternative definitions: other meanings found in the literature, with sources.
- Operationalisation: how the term is measured or identified in the project, where relevant.
- Notes: decisions, boundary cases and examples.
- Sources: key references for each definition, with locations.
The alternative definitions and sources are what make the glossary a research tool rather than a simple list. They show that the project's choice was informed and allow readers to understand how your usage relates to others'.
Trace how terms are used in the literature
Finding how different sources define and use a term can be time-consuming. Definitions may appear in introductions, methods, footnotes or supplementary materials, and many papers use terms without defining them at all.
A source-linked assistant shortens this work. Asking Yila how a set of papers defines a term returns answers linked to the passages where definitions or usages appear. You can quickly see the range of meanings, identify papers that use the term without definition and copy precise locations into the glossary entry. The decision about which definition the project adopts remains a matter of judgement and discussion; the assistant makes the evidence for that decision easy to assemble.
Agree definitions as a team
Definitions should be agreed, not imposed. Circulate draft entries, discuss disagreements and record the outcome. Where the team cannot agree, note the competing views and the reason for the final choice. This process often surfaces deeper differences in how team members understand the research question, which are better resolved early.
Use the glossary actively
A glossary that sits unread in a project folder has little value. Make it part of daily work. Link to it from protocols, coding manuals and data dictionaries. Refer to it during team meetings when terms are debated. Ask new team members to read it as part of onboarding. Check drafts against it before submission.
Keep it current
Projects evolve. New terms emerge, definitions are refined and boundary cases are discovered. Assign someone to maintain the glossary, review it at regular intervals and record changes with dates and reasons. If a definition changes after data collection or coding has begun, assess the consequences and document how they were handled.
Format and tools
The glossary does not need special software. A shared document or spreadsheet with one row or section per term works well for most projects, provided it has a clear owner and a change log. Larger projects may prefer a wiki, which allows linking between entries and from other project documents. Whatever format you choose, make the glossary easy to search, easy to cite internally and accessible to everyone who needs it, including collaborators outside your institution.
Use it in writing
When writing papers, define key terms on first use, consistent with the glossary. For terms used differently in the literature, briefly acknowledge the variation and state your usage. Consider including a glossary or definitions table as supplementary material, particularly for interdisciplinary audiences. Consistency across papers from the same project helps readers follow the work and compare results.
Glossaries in interdisciplinary teams
Interdisciplinary projects face the hardest terminology problems, because the same word can carry established but different meanings in each discipline. "Model" means something different to a statistician, an engineer and a psychologist; "validity" has distinct technical senses in measurement, law and logic. In these teams, the glossary should record each discipline's usage explicitly, not just the project's chosen definition. Early workshops where members explain their own discipline's terms to each other are a good way to populate it. The resulting glossary becomes a bridge between disciplines and a lasting resource for anyone joining the project later.
Linking the glossary to data
Where the project produces datasets, connect glossary entries to variables in the data dictionary. Each variable that measures a defined concept should reference the glossary entry, and each entry should list the variables that operationalise it. This connection helps analysts choose the right variables, helps reviewers check that analyses match definitions and makes shared or archived data much easier for others to reuse correctly. When the glossary changes, the linked variables show immediately what may need to be reviewed.
Common pitfalls
- Defining terms only informally in conversation.
- Listing definitions without sources or alternatives.
- Failing to record how terms are operationalised.
- Letting the glossary go out of date.
- Using terms inconsistently between protocol, analysis and publication.
Conclusion
A project glossary is a small investment that pays off throughout a project. By identifying terms that need defining, writing entries that include alternatives and sources, agreeing definitions as a team, using the glossary actively and keeping it current, you reduce misunderstanding, improve consistency and make your work easier for others to interpret. Tools that trace how terms are used across sources make building such a glossary faster, and keep every definition tied to the literature behind it.