Once you have selected a variety of keywords to use, it is time to build your search query. Remember that not all databases use the same search terms, but most use some form of Boolean searching, and many also use truncation and phrase searching. Using all three techniques in one search query can produce very precise results. Keywords can also be combined with defined subject headings to increase the precision of your search. Many databases have help pages so you can learn what their specific features are. If you would like assistance with developing your search query, please request a consultation with a reference librarian.
- Boolean Searching: Use the connector terms AND, OR, & NOT to structure your search query. AND requires that both keywords appear in the same document; OR requires that one or the other keyword appears; NOT requires that the keword is absent. You can construct a particularly powerful search query by using parentheses ( ) to nest concepts together.
- Truncation Searching: Many databases will let you search for multiple keywords with the same root by using a truncation symbol such as * or ? at the end of the root (ex: using handl* will find handle, handled, handler, handles, and handling). There are two caveats here:
1) Make sure you know what the truncation symbol is for the database you are searching by checking the help pages
2) Choose a root that will return a sufficiently small number of relevant words. For example, the root "rat*" will find not only rat and rats, but also ratio, rate, and rationale.
- Phrase Searching: If you are looking for two or more words that are almost always next to each other, you can force most databases to search for them as a phrase by surrounding them with double quotation marks.
Example: "blood sampling"
- Keyword vs. Subject Heading Searching: Our focus so far has been on searching by keywords, or natural language. Subject headings, a system of controlled vocabulary used to classify information into specific categories, provide a very powerful way to locate focused, on-topic results. The articles in PubMed, for example, are classified using Medical Subject Headings, or MeSH. PubMed provides a searchable database of MeSH terms; these suggested headings are particularly useful when conducting research in animal studies.