Subject Headings vs. Plain-Language Keywords: Building a Mixed Query for Mature and Emerging Topics

When you are searching for literature under a deadline, the choice between controlled vocabulary (subject headings) and plain-language keywords can make the difference between a focused set of relevant papers and a flood of noise. This article compares the two approaches on the same task—finding studies for a course paper or early manuscript—and shows where each fails. You will learn how to combine them into a mixed query, with a decision table and a workflow that fits a student timeline. The guidance draws on established search methodology from the Cochrane Handbook and reporting standards from PRISMA 2020, as well as NIH guidance on rigorous research practice.

Searching is not a single step; it is a planning task. Cochrane recommends planning searches from eligibility criteria, using multiple sources, combining free-text terms with subject headings, documenting the process, deduplicating records, screening in stages, and rerunning searches before publication. For a student paper, the same principles apply at a smaller scale. A well-built query saves time during screening because it retrieves fewer irrelevant records. A poorly built query forces you to wade through hundreds of off-topic results, eating into the hours you need for reading and writing.

NIH guidance treats careful assessment of prior knowledge, strengths, weaknesses, knowledge gaps, transparent methods, validation, and reproducibility as core parts of rigorous research practice. Your search strategy is part of that transparent method. Documenting which terms you used and why helps you—and your supervisor—understand how you arrived at your evidence base.

Controlled vocabulary refers to standardized subject headings assigned by indexers. These headings describe the content of an article using a consistent term, regardless of the author's wording. For mature topics with a stable body of literature, subject headings can be powerful: they capture synonyms and related concepts under one label, reducing the need to list every possible variant.

Cochrane explicitly recommends combining free-text terms with subject headings. This implies that subject headings alone are not sufficient; they are one component of a broader strategy. The strength of controlled vocabulary lies in its consistency for well-indexed concepts. However, it depends on the indexing process: newer articles may not yet have headings assigned, and emerging topics may lack appropriate headings altogether.

Plain-language keywords are the words you would naturally use to describe your topic. They are flexible and can be adapted quickly. For emerging topics, where standardized headings may not exist or may lag behind, keywords are often the only way to retrieve relevant papers. Cochrane's recommendation to use free-text terms alongside subject headings acknowledges this flexibility.

But keywords have a downside: they retrieve any record containing the word, regardless of context. Homonyms and polysemy can introduce noise. You may need to use truncation, phrase searching, and field tags to control precision. For a student under deadline, the trade-off is between recall (finding everything potentially relevant) and precision (avoiding irrelevant results).

Imagine you are writing a course paper on a mature topic—say, the effect of a well-studied intervention—and you have two weeks. You could search using a subject heading for the intervention, or you could search using keywords like the intervention name and its synonyms. On the same task, the subject heading search might retrieve a focused set of articles that are centrally about the intervention. The keyword search might retrieve more articles, including those that mention the intervention only in passing, but it might also miss articles that use a different term for the same concept.

Now imagine an emerging topic—for example, a new technology applied to an old problem. Subject headings may not yet exist for the technology. A keyword search is likely your primary tool. But you may still find some relevant subject headings for the underlying problem or the old problem domain. Combining a heading for the problem with keywords for the new technology can improve relevance.

The key insight is that neither approach is universally superior. The choice depends on the maturity of the topic and the state of indexing. Cochrane's guidance to combine both reflects this reality.

Controlled vocabulary fails when:

• The topic is too new for headings to be assigned.
• The database does not index the concept with a suitable heading.
• The heading is too broad or too narrow for your question.
• You rely solely on headings and miss articles that use different terminology.

Plain-language keywords fail when:

• The same word has multiple meanings, leading to irrelevant results.
• Authors use different terms for the same concept, and you fail to include all variants.
• The keyword is too common, retrieving thousands of records.
• You rely solely on keywords and miss articles indexed under a relevant heading but using different wording.

Cochrane's recommendation to combine both is a direct response to these failure modes. By using both, you compensate for the weaknesses of each.

A mixed query combines subject headings and keywords using Boolean operators. The exact syntax depends on the database, but the logic is consistent. Here is a workflow you can follow under a deadline.

Cochrane recommends planning searches from eligibility criteria. Write down the key concepts your paper must address. For a student paper, this might be the population, intervention, comparison, and outcome (PICO). These concepts become the building blocks of your query.

For each concept, look up the appropriate subject heading in your database's thesaurus. For mature topics, you may find a precise heading. For emerging topics, you may need to use a broader heading or skip this step for that concept.

For each concept, brainstorm all the ways authors might describe it. Include spelling variants, acronyms, and related terms. Use truncation to capture word stems. This step is especially important for emerging topics where terminology is not yet standardized.

Combine synonyms within a concept using OR. Combine different concepts using AND. For a mixed query, include both the subject heading and the keywords for each concept, connected with OR. For example: (Subject Heading OR keyword1 OR keyword2) AND (Subject Heading2 OR keyword3 OR keyword4).

Run the query and examine the first few pages of results. If you see too many irrelevant records, add more specific terms or use field tags. If you see too few, remove the least essential concept or add more synonyms. Cochrane recommends documenting the process, so keep a record of your final query and any changes.

Cochrane recommends using multiple sources. For a student paper, this might mean searching two or three databases plus a search engine for grey literature. Each source may have different indexing, so your mixed query may need adaptation.

Cochrane recommends deduplicating records and screening in stages. Start with title and abstract screening, then full-text screening. This staged approach saves time and keeps you organized.

WisPaper is an AI research assistant for finding, screening, organizing, and working with academic papers. You can use it at several points in the workflow above.

When you are ready to run your mixed query, you can use WisPaper search to find academic papers through its scholarly search interface. If you prefer a public entry point, the WisPaper Scholar Search page explains how to find academic papers. For Chinese readers, the WisPaper Scholar Search Chinese page provides a Chinese-language entry point.

As you collect papers, you can save or upload them to My Library to organize them. This is useful for keeping track of which papers came from which query and for deduplication. If you need to ask questions based on the papers you have collected, Library QA lets you ask questions based on papers uploaded to My Library or added from search results. This can help you quickly check whether a paper addresses your eligibility criteria.

For longer research tasks, such as a literature review, Scholar Agent fits WisPaper's Scholar Agent area and its literature review, inspiration, or reproduction use cases. If you need to manage your overall project, Research Projects is the area where you can create a new project.

Suppose you have three weeks to write a course paper on a mature topic, and your supervisor has asked for a reproducible search strategy. You have limited access to databases through your university library. You also have a rubric that emphasizes thoroughness and transparency.

Week 1: Define eligibility criteria and identify subject headings and keywords for each concept. Build a mixed query. Run it in two databases. Document the query and the date of search. Save results to My Library.

Week 2: Screen titles and abstracts. Use Library QA to ask questions about papers you are unsure about. Deduplicate records. Rerun the search to capture any new articles.

Week 3: Write the paper, including a methods section that describes your search strategy. Cite the papers you retained. If your supervisor asks for a flow diagram, you can adapt the PRISMA 2020 flow-diagram templates, which are designed for transparent reporting of search, selection, and synthesis methods.

At the end of this process, you have a documented, reproducible search that meets the rubric's emphasis on transparency. You also have a clear answer to the question: did you use subject headings, keywords, or both? The answer is both, and you can explain why.

Before you submit, review your search strategy against your eligibility criteria. Did you include both subject headings and keywords for each concept? Did you document your sources and search dates? Did you screen in stages and deduplicate? If you can answer yes to these questions, your search is likely rigorous enough for a course paper. If not, spend an extra hour refining your query and documenting your process. The time you invest in a mixed query now will save you time during screening and writing.