Advanced Search Techniques

A Review

These techniques are especially helpful when doing Literature Reviews.

Ensure that you use the best keywords in your search by checking the database's:

  • Subject Headings

To narrow your results, i.e., to get fewer results, but ones which are more focused and relevant to your specific topic, try:

  • Boolean AND, NOT
  • Field Searching
  • Limiters
  • Phrase Searching
  • Proximity Searching

If you are not getting enough results, you may want to broaden your search by trying:

  • Boolean OR
  • Truncation / Wildcards

Boolean searching, i.e. using Boolean Operators in between your search words, is the single most effective 'advanced' search technique you can use to improve your search results, making them more relevant and precise.

Boolean operators that help you narrow your results:

  • Boolean AND
    • Instructs the search engine to only bring back results that contain BOTH search words
    • Databases usually recognize AND as the primary operator, and will connect concepts with AND together first
  • Boolean NOT
    • Instructs the search engine to NOT bring back any results that contain a specific search word
    • Take care when using this one - results containing the search terms you are interested in may be left out of your results if you are using NOT - simply because they also happened to contain the unwanted search term.

Boolean operators that help you broaden your results:

  • Boolean OR
    • Instructs the search engine to bring back results that contain either one search word OR the other - it will bring back results containing either word - so will usually end up bringing back more results.
    • This can be a good choice when your search topic word has several synonyms and you aren't sure which one will be used in the documents you seek.
    • Take care when using OR - sometime it is necessary to use parentheses in conjunction with OR in order to get the search to run correctly. (See the 'Parentheses/Nesting' tab above for more details.)

How to Perform Boolean Searches

    • Use a database's advanced search interface:

      • Many databases offer several search boxes (and the ability to add more if needed) to accommodate all your search terms.
      • Here, to add a Boolean operator to your search, you simply choose the one you want from the pull-down list (such as in the image above) that is available in between each pair of search boxes.

OR

  • Type your Boolean operator, in all caps, in the search box  between your search terms
  • Why type the Boolean operators in capital letters?  Some database search engines can't 'see' those operators unless they are capitalized.
  • Some databases automatically 'assume' a Boolean AND to be in between two or more search words.  But not all do, so it's safest to type the Boolean connectors in capital letters.

  It often helps, and sometimes is quite necessary, to combine the use of Boolean connectors with parentheses in order to carry out a truly effective search query.  See the 'Parentheses/Nesting' tab above for more details.

Field Searching is a very important technique that you can use to focus your search results. Most databases will give you the option to limit your search to your keywords having to have been found in one of a database document's fields - in other words, the metadata  (information about the document) that the database indexes / labels each document with, in order to be able to quickly retrieve the document when asked to do so. These choices of fields to search can most often be found as a pull-down list right next to the search box.  Choose which field to search in before running your search:

The exact fields available will vary by database.  However, most will allow you to limit your search to results where your keywords were in a document's abstract, its title, or allow you to specify that you want results where your search term was an author's name or perhaps in the title of a publication.

Most databases also offer the option to specify which field(s) to search in by adding Field Codes to your search query (i.e., by typing them in by hand).  These specific codes, and exactly how you have to type them into your search for it to work properly, will of course vary by database. This way of doing field searching was the norm in the early days of databases, and some searchers still prefer it.

Since many databases by default search the entire text of the documents they contain, focusing your search to the words having to have been found in the title, keywords (supplied by either the database or the author)  or abstract is one of the best and easiest ways to make sure that your  search results are articles that are truly focused on your topic - as opposed to getting articles that contain your search words in random places in the text or perhaps somewhere in the list of references, as can often happen.

Article and Book databases allow their users to limit their searches in various ways, both before and after the search query has been run.  These choices, and how you pick, them will vary by database, but will tend to be found close to the search box.  In many case, post-searching limiters will be found to one side (most often the left) of the list of search results.

The most common limiters that are available include:

Publication Date
Topic
Journal Title
language
Author
Peer Reviewed
Full Text
Source Type
Document Type

Many, though not all, databases support using parentheses in search queries.  This technique is also known as nesting.

Nesting is when you surround parts of your search query with parentheses. This is done to tell the database's search engine which part of your query to group together first, before before paying attention to the rest of the terms in your search query.

If you are not using parentheses, search engines will by default group together the search words in your query that have AND in between them first.

You can use parentheses to tell a search engine to do things such as:

  • Look for multiple words that have to be close to each other in the document. (This is not as precise as a phrase search, where the two search terms must be right next to each other in that exact order, nor as precise as a proximity search, where the two words must be no more than a certain number of words away from each other), but tells the search engine that the words need to be fairly close to each other, instead of one word being in one part of the document, and the other being way off in some other part.
    • e.g.: (human AND rights) AND NGOs
  • Treat two groups of multiple words as two synonyms, or otherwise fairly closely related.  You are in effect saying: “I would like results about at least one of these search terms.” You'll take either, or both. This is done by using Boolean OR in between two parentheses that contain multiple words.
    • e.g.: (therapy OR treatments) AND schizophrenia
    • Warning: Running the search in the above example without using the parentheses [i.e. if it looked like: therapy OR treatment AND schizophrenia ] would result in a completely different set of results - and a very imprecise one. Why? The search engine would interpret it as: 'Bring back results about either a) therapy...or about b) treatments and schizophrenia.' One set of results would just be anything that mentioned therapy (of any kind!) and the other would be results about treatments related to schizophrenia. So sure, you'd get some results that happened to be what you were looking for, but you'd also get a ton of results that had nothing to do with what you really wanted...

This is the most precise search technique of all.  When you enclose multiple words with double quote marks (") on either side, you are telling the search engine that you want only results with those words right next to each other, in that exact order.

This is a good technique to use when you are searching for a 'term of art' (i.e., a term, in this case, one containing multiple words, that has a specialized meaning in a particular field or profession).  In other words, everyone when communicating about that topic uses that exact phrase. 

Another example of when using a phrase search would be useful is when you are seeking something like a book or movie title, or someone's name.

e.g.:

  • "Return on investment"
  • "Peer tutoring"
  • "Myocardial infarction"
  • "Shakespeare in Love"
  • "Of Mice and Men"
  • "Twin Peaks"
  • "President Lincoln"
  • "Steve Jobs"

Proximity searching allows you to search for two words that occur in the document within a certain number of words from each other. To do this, you usually type a proximity operator (will vary by database) as well as a number, between two search terms.

e.g. (from Ebsco databases):

(tax OR tariff) N5 reform
oil W3 (disaster OR clean-up OR contamination)
(baseball OR football OR basketball) N5 (teams OR players)

The databases we have that allow proximity searching include:

Subject headings are an example of a Controlled Vocabulary. They are, in effect, the virtual "file folder labels" that a database uses to organize its files. So, if you are having trouble finding results in your topic area, it may help to see how the database you are using 'files' its information. In other words, you may have a search term or even a few about your topic, but the database may use a completely different term to file the information you are seeking.  If that is the case, it would be more efficient to use the database's subject term while searching there.

Most databases have a link to their Controlled Vocabulary (also called: Subject Headings, Thesaurus, Subject Terms, Headings, Index Terms, MeSH), near or at the top of the search page. There, you can type in your search word, then see if the database has a preferred term for your topic that you can use when searching.

List of subject headings from an Ebsco database:

Truncation and Wildcards are great ways to expand your search, as they broadens your search to include various word endings and spellings.

It is especially useful when you are researching a topic where there are alternate spellings (such as variations between American and British spellings), or where your topic makes use of words where there can be both singular and plural endings, or word variations where there can be multiple forms.

To use truncation, enter the root of a word and put the truncation symbol (which will vary by database, though many tend to use an asterisk (*)) at the end. The search engine will then search for any variant form that can be made from that root word.  Take care exactly what root you use as your search term, in case there might be wildly unrelated words that can be made from it which have nothing to do with your topic.

e.g.: child* will bring back results including:  child, childs, children, childrens, childhood

Wildcards are when you replace a letter in your search term with the database's wildcard symbol. This is useful for asking the search engine to find spelling variants.

e.g.: colo?r will bring back results containing either the word color or colour as results.