Start with the concepts, not the whole question
A research question gives you a direction, but its exact wording may not match the language used in articles, reports, or catalogs. Suppose your question is: “How do urban trees affect summer heat in neighborhoods?” A source might discuss “street trees,” “tree canopy,” “air temperature,” or “heat exposure” without using the phrase “summer heat in neighborhoods.” A keyword list helps you find those sources without guessing a new search from scratch each time.
Break the question into two or three essential concepts. Here, they might be urban trees, heat, and neighborhoods. Put each concept in its own column or group. Leave out words such as “how” and “affect” at first: they explain the relationship you want to investigate, but they may unnecessarily restrict a search. You can add more specific relationship terms later if the results are too broad.
Add synonyms and alternative wording
For each concept, write down terms that someone else might use for roughly the same idea. For urban trees, possibilities include city trees, street trees, and urban tree canopy. For heat, try high temperatures, urban heat, and heat exposure. For neighborhoods, consider residential areas, communities, or city districts. These are starting points, not interchangeable labels in every context: “tree canopy,” for example, describes overhead cover rather than individual trees.
Pay attention to spelling variants, abbreviations, and differences between everyday and technical language. A researcher may use “vegetation” where a resident would say “trees,” but vegetation also includes plants that are not trees. Keep that broader term on the list, marked as broader, rather than silently treating it as an exact synonym. This small distinction makes it easier to understand why a search returns unexpected results.
Map broader, narrower, and related terms
Synonyms are only one part of a useful list. Broader terms help when a search finds too little: urban greenery may lead to sources that include trees alongside parks or other plants. Narrower terms help when results are overwhelming: street trees or tree canopy cover may bring the search closer to a particular feature. Related terms describe a connected idea rather than the same thing. Shade, surface temperature, and heat islands may be relevant to the example question, but each points to a different aspect of it.
A simple label beside each term—alternative, broader, narrower, or related—keeps the list organized. It also prevents the question from drifting. If you search for shade, ask whether the sources still help explain urban trees and heat, or whether they address shade from buildings instead. A related term is useful when it opens a promising path, not merely because it produces more results.
Turn the list into searches
Search one concept at a time to see what language appears in useful results. Then combine concepts. Many search tools allow OR between alternatives and AND between concepts: (“street trees” OR “tree canopy”) AND (“urban heat” OR “air temperature”). Quotation marks may keep a phrase together, though search tools handle quotation marks and grouping differently. Check the help information for the database or search engine you are using rather than assuming the same syntax works everywhere.
Begin with a small number of strong terms instead of pasting in the entire list. If the search is too narrow, remove a concept or try a broader term. If it is too broad, choose a narrower term or add a concept such as neighborhoods. As you scan promising sources, note the words used in titles, abstracts, headings, and subject labels. Those terms can reveal vocabulary you did not think of, especially when a field uses specialized language.
Keep a record and revise with purpose
Your keyword list should change as you learn. Record the searches you tried, where you tried them, and a short note about the results: “mostly about parks,” “useful neighborhood studies,” or “too focused on indoor heat.” This is more helpful than keeping a long, unexplained list of words. It lets you return to a productive search and see why an unproductive one missed the mark.
Before adding a new term, decide what problem it should solve. Are you looking for different wording, a broader view, or a more precise topic? In the urban-trees example, adding “surface temperature” may help explore a particular way heat is described, but it should not replace searches about people’s heat exposure if that remains central to your question. A good keyword list is a working map of the topic: flexible enough to uncover unfamiliar sources, and clear enough to keep the research focused.
