Skip to Main Content

Literature Searching: Search techniques

Subject heading and Keyword searching

This section introduces some key concepts for searching databases.

The first is subject heading v keyword searching. Both have strengths and weaknesses, so we would suggest combining both together for the most thorough search. 

Next is Boolean Operators - AND, OR and NOT. These are used to combine concepts within your search to find the most relevant items.

How to use phrase searching, truncation and database limits/filters.
 

Subject Headings

Subject headings are standardised terms used to classify articles. They are assigned by human indexers. These indexers tag the article with subject headings that relate to the content.


Advantages

  • Because it looks at concept rather than exact words it tends to offer more focused search
  • Will locate relevant articles even when keywords are not in the article title or abstract
  • Retrieves articles on your subject no matter what spelling or synonym is used

Disadvantages

  • Different databases use different subject headings (Medline uses MeSH, EMBASE uses EMTREE, etc)
  • Not all databases provide subject headings or a thesaurus (e.g. Web of Science and Scopus do not)
  • There is often a lag time to index articles. So most recent material may not be found.
  • Some concepts may not be contained in the index or are dealt with poorly. New concepts take time to be added.              
  • Human error can lead to mis-categorisation


 

Phrase Searching

Putting a phrase in “quotes” searches for those exact words in that specific order.

Most databases automatically place the AND operator between your search terms. So Perioperative Care will retrieve results where both terms appear on their own as well as where they appear together. This will give lots of irrelevant articles.

By placing the phrase in quotes “Perioperative Care" will look for these words in this exact order. This gives a much more focused search.

Truncation

Truncation is a technique that broadens your search to include various word endings and spellings. Truncation is very useful when you know one of your search terms has several endings, but all of the variations represent basically the same idea. Using truncation will help you complete your search faster because you will not have to manually type in and search every variation of the word.

Most databases use the * (asterix)  to search for all forms of a word. Results will include any ending of that root word. 

eg. therap*  will find therapy, therapies, therapeutic, therapeutics

surg* will find surgery, surgeries, surgeon, or surgical

Use with some care - If the word stem is too short, there may be too many possible variations and you might pick up unrelated terms.

Boolean Operators - AND, OR, NOT

You can use Boolean operators to link different concepts in order to find exactly what you're looking for. They can be used to either narrow or broaden your set of results.

Boolean Operators, AND OR ,NOT

  • AND gives only results giving both/all terms. Use for linking different ideas/concepts in search. Narrows your search. In many, but not all, databases, the AND is implied. Though all your search terms are included in the results, they may not be connected together in the way you want. eg, cancer tumour neoplasm  is translated to cancer AND tumour AND neoplasm. This will only give results where ALL three terms are included rather than results where ANY are included. 
  • OR gives results including either terms. Use for alternative ways of representing the same concept eg. (cancer OR tumour OR neoplasm). Broadens your search.
  • NOT excludes terms from search. If not used carefully it can exclude relevant items. Rarely used. 

When using more than one Boolean Operator in a search you should use parentheses when linking them. This will avoid getting unexpected results. For example use (cancer OR tumour OR neoplasm) AND (surgery OR radiation OR chemotherapy).

 

Subject heading v keyword searching

Subject Heading

Keyword

Because it looks at concept rather than exact words it tends to offer more focused search

Can pick up a wider range of articles than subject searching

Will locate relevant articles even when keywords are not in the article title or abstract

Useful especially when not be an exact subject match

Retrieves articles on your subject no matter what spelling or synonym is used

Limit fields to Title and Abstract for most focused search

Different databases use different subject headings (Medline uses MeSH, EMBASE uses EMTREE, etc)

Searches regardless of its contextual meaning. So can pick up the term in irrelevant contexts Eg. stress as synonym for anxiety but will also find articles on stress fracture.

Not all databases provide subject headings or a thesaurus (e.g. Web of Science and Scopus do not)

Using all fields includes publication name etc which can pull up strange results

There is often a lag time to index articles. So most recent material may not be found.

Looks for exact terms so may miss synonyms or alternative spellings

Some concepts may not be contained in the index or are dealt with poorly. New concepts take time to be added.             

more flexible to search by - can combine together using Boolean or proximity operators

Human error can lead to mis-categorisation

Can miss synonyms and alternative spellings

Search techniques

Phrase searching

Putting a phrase in “quotes” searches for those exact words in that specific order. Most databases automatically place the AND operator between your search terms. So Perioperative Care will retrieve results where both terms appear on their own. Lots of irrelevant articles. “Perioperative Care" will look for these words in this exact order.

Truncation

Truncation using the * (asterix) allows you to search for all forms of a word. You could list them all (therapy OR therapies OR therapeutics OR therapeutic) but an easier way is by the use of truncation therap*. Use with some care - If the word stem is too short, there may be too many possible variations and you might pick up unrelated terms.

Limiting by Field
Field searching allows you to limit your search to a specific part of a bibliographic record such as the author, article title, etc. This can be particularly helpful when you are trying to find an article that you read previously and can only remember bits of the reference. Limit to Title/Abstract for focused search.

Limits/Filtering Options
You can limit searches by language, publication date, article type, gender, age groups. A useful way of focusing the question. They do not necessarily ensure that your search results are of greater relevance. Think which limits refine your research question, and which are simply arbitrary ways of reducing the number of results.

Keyword searching

Keyword search looks for your exact term or phrase in all fields (by default) or specified record fields (usually title, abstract). Limit to title and abstract fields for most focused search.

  • Can pick up a wider range of articles than subject searching 
  • Useful especially when not be an exact subject match
  • Searches regardless of its contextual meaning. So can pick up the term in irrelevant contexts Eg. stress as synonym for anxiety but will also find articles on stress fracture.
  • Using all fields includes publication name etc which can pull up strange results
  • Looks for exact terms so may miss synonyms or alternative spellings
  • Flexible to search by - can combine together using Boolean or proximity operators for more complex concepts

Proximity Searching

Proximity Operators

Many databases allow you to specify that the words you are searching are within a certain proximity of each other. Proximity operators are more specific than Boolean operators and make your search more precise.

Pubmed

For example, to search PubMed for citations where the terms "hip" and "pain" appear with no more than two words between them in the Title/Abstract search field, try the search:

"hip pain"[Title/Abstract:~2]

Search results may include hip pain, hip-related pain, hip joint pain, hip/groin pain, hip biomechanics and pain, pain after total hip arthroplasty, pain in right hip, and more.

EBSCO (Cinahl)

Near Operator (N) finds results in which search terms are within a specified number of words of one another in any order.
Example: tax N5 reform finds results that contain tax reform as well as reform of income tax.
Within Operator (W) finds results in which search terms are within a specified number of words of one another and in the order in which you entered them.
eg. hiking W5 trails finds results that contain hiking trails and hiking of easy trails, but does not find results containing easy trails for hiking

OVID (Medline, Embase, Psycinfo)

The defined adjacency operator (ADJn) retrieves records that contain search terms within a specified number (n-1) of words from each other in any order (stop-words included). To use the adjacency operator, separate your search terms with ADJ and a number from 1 to 99 as explained below:

ADJ1     Next to each other, in any order
ADJ2     Next to each other, in any order, up to 1 word in between