Searching Library Databases Effectively

Why database searching differs from web searching

A general web search engine guesses what you want from a few words. Library databases index particular collections, such as journal articles, conference papers, books or reports, and many reward precise, structured queries. They also hold material ordinary search often cannot reach, such as subscription articles your institution pays for.

Different databases suit different subjects, and what your library offers depends on your institution. A library website’s subject guide is usually the best place to see which databases are recommended for your field. Interfaces vary, so the details below may not match every system exactly. Check the help page of the database you are using.

Step 1: Turn your question into concepts

Start with your research question and break it into its main ideas. For example, take the question: “How does sleep affect exam performance in university students?” The main concepts are:

1. sleep
2. exam performance
3. university students

Write these down. A strong search usually combines two or three concepts. Using the whole question as a sentence rarely works well in a database.

Step 2: List keywords and alternatives

For each concept, brainstorm other words that authors might have used. Different writers use different terms for the same idea, so a search with only one word will miss relevant work.

Consider spelling variants (such as “behavior” and “behavior”) and singular and plural forms. Reading a few good articles on your topic also helps, since they show you the vocabulary the field uses. Some databases offer subject headings or thesaurus terms, which are standard labels assigned by indexers. Searching with those can be more consistent than guessing words.

Step 3: Combine terms with Boolean operators

Boolean operators tell the database how your terms relate. The three main ones are AND, OR and NOT.

• AND narrows a search. Every term must appear. sleep AND undergraduates returns results that mention both.
• OR widens a search. Any of the terms may appear. undergraduates OR "college students" finds either. Use OR to join the alternatives within one concept.
• NOT excludes a term. sleep NOT infants removes results containing “infants”. Use it carefully, because it can accidentally remove relevant material.

A typical structure is to join the synonyms for each concept with OR, wrap each group in parentheses, and join the groups with AND:

(sleep OR "sleep deprivation" OR "sleep quality") AND ("exam performance" OR "academic achievement") AND (undergraduates OR "college students")

Parentheses matter because they control the order in which the database interprets the query. Without them, results can be unpredictable.

Step 4: Use phrases and truncation

Phrase searching. Quotation marks usually tell a database to look for words together in that order. “Academic achievement” will find that phrase rather than pages with “academic” in one place and “achievement” somewhere else. Most databases support this, but check yours.

Truncation and wildcards. Many databases let you use a symbol, often an asterisk, to find word endings. For example, student* can retrieve student, students and studentship. Symbols differ between systems, and over-truncating (such as stud*) brings in unrelated words, so use it with care.

Step 5: Narrow with filters

After you run a search, databases usually offer filters, sometimes called limiters. Common ones include:

• Date range. Useful when recent work matters. Choose a range that suits your topic, because in some subjects older studies are still important.
• Peer-reviewed or scholarly. Restricts results to journals using peer review, though the rigor of review varies.
• Source or document type. Journal article, book, review, conference paper, thesis.
• Full text available. Helpful if you need immediate access, but be aware it can hide relevant items that your library can obtain another way, such as through interlibrary loan.

Add filters one at a time so you can see their effect.

Step 6: Judge and refine

Skim the first page of results. Are they about your topic? If not, change something.

Too many results? Add another concept with AND, use more specific terms, use phrase searching, search in title or abstract fields instead of full text, or apply a date or document-type filter.

Too few results? Remove a concept, add synonyms with OR, loosen phrases, use truncation, or try a different database.

Search is rarely one-and-done. Keep a note of the queries you try and which databases you used; a simple document is enough.

Step 7: Follow the trail

Once you find a useful article, treat it as a map to others.

• Check its reference list for earlier relevant work.
• Use “cited by” features, where available, to find newer work that refers to it.
• Look for review articles, which summarize a field and point to key studies. These are secondary sources, as explained in the comparison of primary sources and secondary ones.

Step 8: Evaluate and record

Finding a result is not the same as judging it. Before you rely on any source, apply questions of authority, purpose, evidence and date. For academic articles, the abstract tells you what the paper claims, but you should read enough of the body to check how those claims were supported, and reading papers as a non-specialist can help.

Save details as you go: author, title, journal, year, volume, pages and a stable link if available. Sensible file naming helps when you store PDFs.

Getting help

Librarians are trained in searching and often happy to review a search strategy. If a database blocks you from a full text, ask your library before paying for an article, because there are often free routes, such as open access versions.